Você está na página 1de 383

Anxiety in Health Behaviors

and Physical Illness

Series Editor: Martin M. Antony, Professor, Department of Psychology,
Ryerson University, Toronto, Ontario, Canada


Conceptualization and Treatment
Edited by Susan M. Orsillo and Lizabeth Roemer


Edited by Jonathan S. Abramowitz and Arthur C. Houts


Characteristics, Assessment, and Psychological Treatment
Christopher A. Kearney


A Practitioner’s Guide
Patricia Furer, John R. Walker, and Murray B. Stein

Cognitive-Behavioral Therapy for Hairpulling and Related Problems
Martin E. Franklin and David F. Tolin


Edited by Michael J. Zvolensky and Jasper A. J. Smits

A Continuation Order Plan is available for this series. A continuation order will bring delivery of each
new volume immediately upon publication. Volumes are billed only upon actual shipment. For further
information please contact the publisher.
Anxiety in Health
Behaviors and Physical

Michael J. Zvolensky
University of Vermont
Burlington, Vermont, USA

Jasper A. J. Smits
Southern Methodist University
Dallas, Texas, USA
Michael J. Zvolensky, Ph.D. Jasper A. J. Smits, Ph.D.
Department of Psychology Department of Psychology
University of Vermont Southern Methodist University
John Dewey Hall 6424 Hilltop Lane
2 Colchester Avenue Dallas, TX 75205
Burlington, VT 05405-0134 USA
USA jsmits@smu.edu

ISBN 978-0-387-74752-1 e-ISBN 978-0-387-74753-8

Library of Congress Control Number: 2007935078

© 2008 Springer Science+Business Media, LLC

All rights reserved. This work may not be translated or copied in whole or in part without
the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring
Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or
scholarly analysis. Use in connection with any form of information storage and retrieval,
electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks, and similar terms,
even if they are not identified as such, is not to be taken as an expression of opinion as to
whether or not they are subject to proprietary rights.

Printed on acid-free paper.

9 8 7 6 5 4 3 2 1


Research has been accumulating on the prevalence and nature of the co-
occurrence between various forms of anxiety disorders and problematic health
behaviors as well as physical illness. This research has significant implications
for both those interested and affected by anxiety as well as physical health
factors. Yet, it is striking that there has been little systematic integration of this
health-oriented research in contemporary science and practice on anxiety and
its disorders. This relative neglect is unfortunate given that the co-occurrence of
anxiety and health problems is a major public health priority when measured
both in human and financial terms.
The overarching aim of this book is to provide a single resource that offers
current theoretical perspectives and cutting eZdge reviews of scientific research
on health behaviors and physical illness in relation to anxiety and its disorders.
A critical analysis of this emerging literature is needed to help move this field
forward, making this proposed volume timely. The specific objectives of this
edited book are to (1) provide a review of the literature on the link between
anxiety and certain health behaviors and processes as well as physical illness;
(2) present contemporary theories of their co-occurrence and interplay (e.g.,
onset, maintenance, and relapse); and (3) provide an analysis of recent research
in regard to therapeutic models for targeting these problems.
The book is organized into two general sections. In the first part of the book,
prototypical health behaviors – smoking, alcohol, illicit substance use, exercise,
and sleep – are discussed in relation to anxiety and its disorders. In the second
part of the book, the association between anxiety psychopathology and physi-
cal health conditions – chronic pain, cardiovascular disease, asthma, HIV/
AIDS – and their treatment are covered. In this same section, the potential
role of puberty and the menstrual cycle in the onset and maintenance of anxiety
psychopathology are discussed.
Inspection of the excellent and comprehensive works has yielded a number of
broad-based conclusions relevant to informing research and practice for anxi-
ety disorders. First, there is consistent empirical evidence that medical problems
and poor health behaviors are overrepresented among persons with anxiety
disorders, and vice versa. Thus, there is a pressing need to marshal information
on anxiety-health processes to better serve this population. Second, as each

vi Preface

contribution makes clear, there is uniform evidence that both health behaviors
and physical illness can, and do, affect the nature of anxiety psychopathology.
Yet, the exact nature of these associations depends on the specific disorder and
health factor in question. And finally, a variety of the chapters make clear that
persons suffering from anxiety psychopathology and poor health behaviors or
medical illnesses may need specialized interventions to prompt clinical change.
That is, traditional interventions may not be ideally suited or maximize clinical
benefit for this population.
For us, the present book offers the opportunity to appreciate the importance
and complexity involved with the study of anxiety disorders. For many years,
health behaviors and medical illnesses have been a neglected facet of anxiety
disorder research and practice. The contributions in this book help drive home
the message that such neglect is unwarranted, and that by working to better
understand the enigmas between health status and functioning and anxiety
psychopathology, significant clinically-relevant strides can likely be achieved.
We hope the present book helps move such work forward and bring a better
quality of life and reduced morbidity to persons with anxiety disorders in the
We owe gratitude to many people who have helped us complete this project.
First among these are the experts who authored the chapters. We would like to
thank them for their hard work and dedication. We also appreciate the com-
ments and suggestions of Dr. Martin Antony, the editor of the Series in Anxiety
and Related Disorders, and the assistance of Anna Tobias of Spinger with the
publishing of this book. Lastly, we continue to be appreciative of our respective
family members, Heidi and Jack Zvolensky and Jill and Stella Smits, for their
love and support.

April 2007 Michael J. Zvolensky, Ph.D.

University of Vermont
Jasper A. J. Smits, Ph.D.
Southern Methodist University

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Part I: Health Behaviors and Anxiety Disorders

Tobacco Use and Panic Psychopathology: Current Status and Future

Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Michael J. Zvolensky, Theresa Leyro, Amit Bernstein,
Matthew T. Feldner, Andrew R. Yartz, Kimberly Babson, and
Marcel O. Bonn-Miller

Alcohol Use and Anxiety Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

Brigitte C. Sabourin and Sherry H. Stewart

Illicit Drug Use Across the Anxiety Disorders: Prevalence, Underlying

Mechanisms, and Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Matthew T. Tull, David E. Baruch, Michelle S. Duplinsky,
and C. W. Lejuez

The Promise of Exercise Interventions for the Anxiety Disorders . . . . . . . . 81

Jasper A. J. Smits, Angela C. Berry, Mark B. Powers, Tracy L. Greer,
and Michael W. Otto

Anxiety and Insomnia: Theoretical Relationship and Future

Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Thomas W. Uhde and Bernadette M. Cortese

Part II: Physical Conditions and Anxiety Disorders

Anxiety Disorders and Physical Illness Comorbidity: An Overview . . . . . . 131

Tanya Sala, Brian J. Cox, and Jitender Sareen

viii Contents

The Relation Between Puberty and Adolescent Anxiety: Theory

and Evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Ellen W. Leen-Feldner, Laura E. Reardon, Chris Hayward, and
Rose C. Smith

Anxiety, Anxiety Disorders, and the Menstrual Cycle . . . . . . . . . . . . . . . . 181

Sandra T. Sigmon and Janell G. Schartel

Pain and Anxiety Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207

Gordon J.G. Asmundson, Murray P. Abrams, and Kelsey C. Collimore

Asthma in Anxiety and Its Disorders: Overview and Synthesis . . . . . . . . . . 237

Lisa S. Elwood and Bunmi O. Olatunji

Cardiovascular Disease and Anxiety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279

Kamila S. White

HIV and Anxiety. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317

Conall O’Cleirigh, Trevor A. Hart, and Carolyn A. James

Physical Illness and Treatment of Anxiety Disorders: A Review . . . . . . . . . 341

Norman B. Schmidt, Meghan E. Keough, Lora Rose Hunter, and Ann
P. Funk

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367

Murray P. Abrams, B.A., Anxiety Bernadette M. Cortese, Ph.D.,

and Illness Behaviours Laboratory, Department of Psychiatry, Penn State
University of Regina College of Medicine and Hershey
Medical Center
Gordon J.G. Asmundson, Ph.D.
Anxiety and Illness Behaviours Brian J. Cox Ph.D., Department of
Laboratory, University of Regina Community Health Sciences and
Department of Psychiatry,
Kimberly Babson, B.A., Department University of Manitoba
of Psychology, University of
Arkansas Michelle S. Duplinsky, Center for
Addictions, Personality, and Emotion
David E. Baruch, Center for Research and the University of
Addictions, Personality, and Emotion Maryland
Research and the University of
Lisa S. Elwood, M.A., Department of
Psychology, University of Arkansas
Amit Bernstein, Ph.D., Veterans
Affairs Palo Alto Health Care System
and University of Vermont Matthew T. Feldner, Ph.D.,
Department of Psychology,
Angela C. Berry, M.A., Department University of Arkansas
of Psychology, Southern Methodist
University Ann P. Funk, M.A., Department of
Psychology, Florida
Marcel Bonn-Miller, B.A., State University
Department of Psychology,
University of Vermont Tracy L. Greer, Ph.D., Department of
Psychiatry, University of Texas
Kelsey C. Collimore, B.S., Anxiety Southwestern Medical Center at
and Illness Behaviours Laboratory, Dallas
University of Regina

x Contributors

Trevor A. Hart, Ph.D., Mark B. Powers, Ph.D., Department

Department of Psychology, of Psychology, University of
York University Amsterdam

Chris Hayward, M.D., M.P.H., Laura E. Reardon, M.A.,

Department of Psychiatry and Department of Psychology,
Behavioral Sciences, Stanford University of Arkansas
Brigitte C. Sabourin, B.A.,
Lora Rose Hunter, B.A., Department Department of Psychology,
of Psychology, Florida State Dalhousie University

Carolyn A. James, M.A., Department Tanya Sala, M.D., FRCPC,

of Psychology, York University, Department of Psychiatry, University
Toronto of Manitoba

Meghan E. Keough, M.S., Jitender Sareen B.Sc., M.D., FRCPC,

Department of Psychology, Florida Department of Community Health
State University Sciences and Department of
Psychiatry, University of Manitoba
Ellen W. Leen-Feldner, Ph.D.,
Department of Psychology, Janell G. Schartel, M.A., Department
University of Arkansas of Psychology, University of Maine

Carl W. Lejuez, Ph.D., Center for Norman B. Schmidt, Ph.D.,

Addictions, Personality, and Emotion Department of Psychology, Florida
Research and the University of State University
Sandra T. Sigmon, Ph.D.
Teresa Leyro, B.A., Department of Department of Psychology,
Psychology, University of Vermont University of Maine
Conall O’Cleirigh, Ph.D.,
Massachusetts General Hospital/ Rose C. Smith, B.A.,
Harvard Medical School and Fenway Department of Psychology,
Community Health University of Arkansas

Bunmi O. Olatunji, Ph.D., Jasper A. J. Smits, Ph.D.,

Department of Psychology, Department of Psychology, Southern
Vanderbilt University Methodist University

Michael W. Otto, Ph.D., Center for Sherry H. Stewart, Ph.D.,

Anxiety and Related Disorders, Departments of Psychiatry and
Boston University Psychology, Dalhousie University
Contributors xi

Matthew T. Tull, Ph.D., Center Kamila S. White, Ph.D., University

for Addictions, Personality, of Missouri-Saint Louis
and Emotion Research and the
University of Maryland
Andrew R. Yartz, Ph.D., Department
of Psychology, University of Vermont
Thomas W. Uhde, M.D.
Department of Psychiatry, Michael J. Zvolensky, Ph.D.,
Penn State College of Medicine Department of Psychology,
and Hershey Medical Center University of Vermont
Part I
Health Behaviors and Anxiety Disorders
Tobacco Use and Panic Psychopathology: Current
Status and Future Directions

Michael J. Zvolensky, Teresa Leyro, Amit Bernstein, Matthew T. Feldner,

Andrew R. Yartz, Kimberly Babson, and Marcel O. Bonn-Miller

Recently, there has been increased effort to better understand linkages between
tobacco use and the anxiety disorders (Feldner, Babson, & Zvolensky, 2007;
Morissette, Tull, Gulliver, Kamholz, & Zimering, 2007; Morrell & Cohen,
2006; Zvolensky, Bernstein, Marshall, & Feldner, 2006; Zvolensky, Feldner,
Leen-Feldner, & McLeish, 2005; Zvolensky, Schmidt, & Stewart, 2003). These
efforts are theoretically and clinically important because substance use pro-
blems frequently co-occur with anxiety psychopathology, and anxiety-related
factors often play a role in tobacco use and dependence (Morissette et al.,
2007; Morrell & Cohen, 2006; Zvolensky, Bernstein et al., 2006). However, our
understanding of the explanatory nature of these comorbid relations is only
beginning to emerge. The purpose of the present chapter is to provide a current
review of extant empirical work pertaining to the inter-relations between
tobacco use and panic psychopathology. We expressly and specifically focus
on panic psychopathology, rather than anxiety disorders more broadly, as
most of the work on tobacco and anxiety relations to date has focused on
This chapter is organized into four sections. First, we briefly describe panic
psychopathology. Second, we review risk factor terminology developed by
Kraemer, Kazdin, and Offord (1997), in order to establish a nomenclature for
conceptualizing interrelations between tobacco and panic psychopathology.
Third, we describe tobacco use and common patterns of use. Fourth, we discuss
the nature of comorbidity between tobacco use and panic psychopathology and
review and evaluate the related empirical evidence. We focus both on the role of
tobacco use in the onset and maintenance of panic psychopathology, and the
role of panic factors and processes in the onset and maintenance of smoking.
Within each of these sections, we identify gaps in the existing literature and
highlight formative questions for future research.

Michael J. Zvolensky
University of Vermont, John Dewey Hall, Burlington, VT 05405-0134, Tel: 802-656-8994,
Fax: 802-656-8783

M. J. Zvolensky, J. A. Smits (eds.), Anxiety in Health Behaviors and Physical Illness. 3

Ó Springer 2008
4 M. J. Zvolensky et al.

Panic Psychopathology

The term ‘‘panic psychopathology’’ is used in this chapter to denote panic

attacks, panic disorder, and agoraphobia (with or without panic disorder).
Panic attacks are a subjective sense of extreme fear or impending doom accom-
panied by an autonomic nervous system surge and a strong flight-or-fight
action tendency (Barlow, Brown, & Craske, 1994). Recent estimates of unex-
pected (‘‘out of the blue’’) panic attacks in representative samples suggest that
approximately 20% of individuals experience such attacks at one point in their
lives and 11.2% in the past 12-months (Kessler, Chiu, Jin, Ruscio, Shear, &
Walters, 2006). Thus, panic attacks are a relatively common psychological
experience and many people experience panic attacks without necessarily devel-
oping panic disorder (i.e., nonclinical panic attacks; Norton, Cox, & Malan,
1992). Typically, individuals who experience nonclinical panic attacks do not
experience these attacks as ‘‘spontaneous’’ or ‘‘uncued’’ (as is generally the case
in panic disorder), but rather in certain contexts such as stressful or threatening
social situations (Norton, 1989). Panic attacks also occur among those
with other types of psychopathology (i.e., beyond panic disorder; Bryant &
Panasetis, 2005). Panic attack onset can occur across the lifespan, but early
onset tends to first occur between the ages of 12–13 years (Hayward et al., 1992;
Warren & Zgourides, 1988; please see ‘‘Developmental Course’’ section below
for further details).
Panic disorder involves recurrent unexpected panic attacks and anxious
apprehension about the possibility of experiencing future panic episodes
(American Psychiatric Association, 2000). Lifetime estimates of panic disorder
without agoraphobia are 3.7% and 1.1% for panic disorder with agoraphobia
(Kessler et al., 2006). Twelve-month estimates for panic disorder (with or
without agoraphobia) are approximately 2.8% (Kessler et al., 2006). Thus,
panic disorder is a relatively common psychiatric disorder both in terms of
lifetime and 12-month prevalence rates. This clinical condition is generally
regarded as a disorder of adulthood with a median age of onset of 24 (Burke,
Burke, Regier, & Rae, 1990), although emerging research has noted that
another possible ‘‘peak onset period’’ may be between ages 45–54 years
(Burke et al., 1990). Panic disorder with and without agoraphobia is associated
with a chronic, fluctuating course and high rates of both psychiatric comorbid-
ity and substance use disorders (Zvolensky, Bernstein et al., 2006).
Although not all persons with panic disorder will meet diagnostic criteria for
agoraphobia, individuals with panic disorder often show signs of avoiding
potentially threatening situations in which a panic attack might occur (Feldner,
Zvolensky, & Leen-Feldner, 2004). Agoraphobia reflects a pattern of behavior
characterized by consistent avoidance of threatening situations where a panic
attack or high anxiety is perceived to be likely (e.g., limited options to escape),
or experiencing marked emotional distress when in such situations. Avoidance
behavior can be multifaceted, with a wide range of stimuli that are perceived as
Panic and Tobacco 5

threat-relevant (e.g., anything from being in crowds to certain substances like

caffeine; Feldner et al., 2004). Although agoraphobia does not necessarily
require the presence of panic attacks or panic disorder (Fava, Grandi, &
Canestrari, 1988), researchers frequently conceptualize agoraphobia as a com-
plication of (severe) panic disorder (Barlow, 2002). However, it is noteworthy
that this approach has been increasingly called into question in recent years, as
research indicates that there may be differing forms of agoraphobic avoidance
(Hayward & Wilson, in press). Regardless, agoraphobia with or without panic
disorder often is related to higher rates of clinically significant life impairment
and severity of illness (Kessler et al., 2006). The onset of agoraphobia with or
without panic disorder is not as firmly established as that of panic attacks and
panic disorder; though research suggests it likely occurs later in life than the
onset of panic attacks and panic disorder (Lindesay, 1991).

Vulnerability Nomenclature

Explicit delineation of terminology facilitates efforts to understand the nature

of associations between tobacco use and panic psychopathology. Led by the
work of Kraemer and colleagues, groundbreaking conceptual strides have
created a clearer understanding of various risk processes (Kazdin, Kraemer,
Kessler, Kupfer, & Offord, 1997; Kraemer et al., 1997; Kraemer, Lowe, and
Kupfer, 2005; Kraemer, Stice, Kazdin, Offord, & Kupfer, 2001). Specifically, as
is reviewed in this section, Kraemer and colleagues have worked to standardize
operational definitions for risk processes to increase the clarity and consistency
with which these factors are communicated across studies.
Risk factors. A risk factor is a variable that is related to, and temporally
precedes, an unwanted outcome (Kraemer et al., 1997). Causal risk factors
reflect variables that, when modified in some way (e.g., through an interven-
tion), produce change (increase or decrease) in the dependent variable of
interest (Kraemer et al., 1997). Controlled research designs are necessary to
document causal effects because they can rule out competing alternative
explanations (e.g., ‘‘confounding variables’’). Proxy risk factors are variables
that are related to an outcome of interest, but this association is due to the proxy
risk factor’s relation with another causal risk factor (Kraemer et al., 2001).
Thus, change in a proxy risk factor would not yield corresponding systematic
change in an outcome variable; accordingly, a proxy risk factor may ‘‘mark’’
risk, but not explain or account for such risk.
Due to the importance of the ability to change a risk factor, both risk and
proxy risk factors often are further categorized on the basis of whether or not
they are malleable (i.e., can be changed or altered). When a risk factor cannot be
changed, it can be classified as a fixed marker, whereas when it can be changed,
it can be classified as a variable risk factor (Kraemer et al., 2005).
6 M. J. Zvolensky et al.

A risk factor also can be contrasted with a maintenance factor. A mainte-

nance factor is a variable that predicts the persistence of an existing condition
over time among individuals already demonstrating the outcome (Stice, 2002).
In theory, the same categorization scheme could be applied to maintenance
factors in terms of whether or not they are causal or proxy maintenance factors
(Kazdin et al., 1997). Moreover, a risk factor also may subsequently function as
a maintenance factor.
Qualifying conditions for risk factor effects. Clarifying relations between
vulnerability processes and outcomes represents only the ‘‘first step’’ in a larger
process of risk factor research. That is, this step represents a focus on ‘‘main
effects,’’ but does not explicate how, when, or among whom a specified risk
process unfolds (see Zvolensky, Schmidt, Bernstein, & Keough, 2006). We do
not delve fully into these explanatory processes within the present chapter, as
extant work has largely focused on questions of main effects at this early
developmental stage of this area of study.

Tobacco Use: Definition, Nature, and Prevalence

Cigarette smoking. Cigarette smoking is widely recognized as the most popular

form of tobacco use and as a major public health problem (Windle & Windle,
1999). Indeed, cigarette smoking remains a leading preventable cause of death
and disability in the United States (Centers for Disease Control and Prevention
[CDC], 1994). Smoking is considered a key factor in various types of medical
illness, including heart disease, a variety of pulmonary diseases (e.g., chronic
obstructive pulmonary disease), and many types of cancer (CDC, 1994, 2002).
For instance, smoking is responsible for almost 31% of all cancer-related deaths
(American Cancer Society [ACS], 2006a). Although smoking increases one’s
risk for developing many of these lethal medical diseases, quitting smoking
decreases the risk of developing such problems and may increase the survival
time among persons who have already developed such medical problems
(Samet, 1992).
Despite a reduction in smoking prevalence over the past 25 years, approxi-
mately 45–48 million (approximately 22% to 25%) adults in the U.S. currently
smoke (CDC, 1996). Though nearly 70% of these smokers are motivated to quit
(CDC, 2002), approximately 90–95% of smokers who do try to quit smoking on
their own (Cohen et al., 1989), and 60–80% who attend treatment programs,
relapse to smoking (CDC, 2002). Additional work suggests that 64% of youth
(adolescents) report having tried cigarettes and 14% have smoked frequently in
the past month (i.e., 20 out of the past 30 days; CDC, 2002). Thus, there is
evidence that smoking is not only a major source of death and disability, but
that once started it is often difficult to stop.
Smokeless tobacco use. Although cigarettes are the most common type of
tobacco use, there also is a significant population of smokeless tobacco
Panic and Tobacco 7

users in the U.S. and other regions of the world (e.g., India; ACS, 1999).
Epidemiological data in the U.S., for example, suggest that approximately
3% of individuals have used some form of smokeless tobacco – either snuff
(finely ground, shredded tobacco) or chewing tobacco – in the past month
(ACS, 1999). These rates of use are elevated among young Caucasian males
compared to females, and among those in southeastern and north central states
compared to other regions, as well as rural compared to urban settings
(Hatsukami & Severson, 1999). For example, past work suggests that the
highest rates of current use (16.6%) are among Caucasian males 18–25 years
of age (CDC, 1994).
In general, rates of smokeless tobacco use are noteworthy because smokeless
tobacco contains carcinogens (e.g., tobacco-specific nitrosamines [TSNAs];
National Cancer Institute and National Institute of Health (NCI/NIH), 2006)
known to be causal agents in lung, oral, esophageal, liver, and pancreatic
cancer (About: Smoking Cessation, 2006b). In addition, smokeless tobacco
use is associated with greater nicotine absorption and for longer periods of
time (e.g., stays in the bloodstream for greater durations of time) than cigarette
smoking (ACS, 2006; NCI/NIH 2006b). Similar to cigarettes, smokeless
tobacco use can lead to increased rates of physical disease (e.g., oral cancer,
gum disease) and nicotine addiction (ACS, 1999). Whereas the risks associated
with cigarette use have become well-publicized in the U.S., public awareness
about the dangers of smokeless tobacco remains limited. Indeed, many perceive
smokeless tobacco as a ‘‘risk-free alternative’’ to cigarette smoking (CDC,
1994). Although as many as 50% of smokeless tobacco users report wanting
to quit (ACS, 2006a), as with cigarettes, rates of relapse remain high (Hatsu-
kami, Jensen, Allen, Grillo, & Bliss, 1996). In a recent review of smokeless
tobacco treatment programs, Hatsukami and Severson (1999) estimated a 3–6
month abstinent rate of 12%–30% with intensive behavioral treatments fairing
better than other intervention options. Thus, similar to cigarette use, smokeless
tobacco use is an addictive behavior that is difficult to quit and more intensive
care appears to yield relatively better outcomes.

Prevalence of Comorbid Tobacco Use and Panic Psychopathology

There have been both representative surveys and community-based studies

focused on addressing the extent of the co-occurrence between tobacco use
and panic psychopathology. To date, this work has largely centered on cigarette
smoking rather than smokeless tobacco. Additionally, from a historical per-
spective, the vast majority of early work in this domain did not focus on panic
or other specific anxiety conditions, but rather, addressed anxiety disorders as a
single ‘‘class of problems’’ (e.g., Breslau, 1995; Brown, Lewinsohn, Seeley, &
Wagner, 1996; Costello, Erkanli, Federman, & Angold, 1999; Degenhardt,
Hall, & Lynskey, 2001; Hughes, Hatsukami, Mitchell, & Dalgren, 1986;
8 M. J. Zvolensky et al.

Kandel, Huang, & Davies, 2001; Kandel et al., 1997; Merikangas et al., 1998;
Tilley, 1987). Although such work has importantly directed scientific attention
to tobacco-anxiety relations at the broadest level, it is limited in demarcating
specific rates of co-occurrence for (particular) disorders of interest, such as
panic. Thus, for the sake of explanatory specificity, we focus our summary on
investigations that expressly distinguished panic psychopathology from other
anxiety disorders.
Comorbidity prevalence. The majority of studies have focused on document-
ing rates of smoking among persons with panic psychopathology. The criteria
for smoking behavior has varied across investigations. Moreover, treatment-
based recruitment strategies have been most commonly employed, perhaps
making these data somewhat less generalizable to the overall smoking popula-
tion. Nonetheless, among treatment-seeking adults, several studies have
reported that current daily smoking among patients with panic psychopathology
(either panic disorder or agoraphobia or both) ranged from 19% Panic
(Baker-Morissette, Gulliver, Wiegel, & Barlow, 2004) to 57% (Himle, Thyer, &
Fischer, 1988), with the vast majority of investigations falling between 30% to
50% (Amering et al., 1999; Lopes et al., 2002; McCabe et al., 2004; Pohl,
Yeragani, Balon, Lycaki, & McBride, 1992). These rates of daily smoking are
typically higher than comparison groups involving persons without psychiatric
problems and typically higher, or as high as, rates among persons with other
anxiety or mood disorders (McCabe et al., 2004). Thus, these data collectively
suggest that smoking is a relatively common unhealthy behavior among treat-
ment-seeking individuals with panic psychopathology.
Studies focused on non-treatment seekers are currently limited. Of the
available work, one study focused on youth (Hayward, Killen, & Taylor,
1989; 95 9th graders in public schools) and the other on college students
(Valentiner, Mounts, & Deacon, 2004, n = 337). In both investigations, indi-
viduals with panic attacks, but not necessarily panic disorder or agoraphobia,
had higher rates of cigarette use on a ‘‘regular basis’’ (Hayward et al., 1989;
Valentiner et al., 2004). For example, Hayward et al. (1989) reported that of
those with a lifetime history of panic attacks, 77% had engaged in ‘‘experi-
mental’’ or ‘‘regular’’ cigarette use compared with 48% of adolescents without a
lifetime history of panic attacks. These results, albeit highly limited in overall
scope, generally parallel those of the treatment-oriented investigations noted
earlier in terms of documenting elevated use prevalence among individuals with
panic problems.
Another set of investigations has utilized representative sampling methods to
explore the nature of tobacco use among those with panic psychopathology
(Covey, Hughes, Glassman, Blazer, & George, 1994; Farrell et al., 2001; Lasser
et al., 2000). In perhaps the most comprehensive and well-known of these
investigations, Lasser et al. (2000) examined smoking status according to
psychiatric diagnoses using data from the National Comorbidity Survey (NCS),
a nationally representative study that used structured clinical interviews to docu-
ment mental illness (Kessler et al., 1994). Participants were 4,411 individuals aged
Panic and Tobacco 9

15 to 54 years. Among individuals diagnosed with panic attacks, panic disorder,

and agoraphobia in their lifetime, 38%, 35%, and 38% were current smokers,
respectively. These rates were significantly greater than rates of current smoking
among individuals without mental illness. By comparison, 36% of individuals
with a lifetime history of major depression and 49% of individuals with a lifetime
diagnosis of drug abuse or dependence were current smokers. Rates of lifetime
smoking among persons with a lifetime history of panic psychopathology
(i.e., panic attacks, panic disorder, or agoraphobia) ranged from 58% to 61%.
When diagnostic status in the past month was used as the grouping variable,
current rates of smoking were 46% among persons with panic attacks, 42%
among persons with panic disorder, and 48.1% among persons with agorapho-
bia. It is noteworthy that as number of mental diagnoses increased (ranging from
0 to 4 or more), the percentage of heavy (i.e., peak consumption exceeding
24 cigarettes a day) compared to relatively lighter (i.e., peak consumption less
than 24 cigarettes per day) smokers increased. Overall, these data, coupled with
the treatment-seeking data noted earlier, suggest that smoking occurs at relatively
higher rates among those with panic psychopathology compared to those with no
mental illness.
Whereas the studies just reviewed focused on smoking among those with
panic psychopathology, other investigations have sought to evaluate rates of
panic psychopathology among smokers (Black, Zimmerman, & Coryell, 1999;
Breslau, Kilbey, & Andreski, 1991; Goodwin, Zvolensky, & Keyes, 2007;
Nelson & Wittchen, 1998). All of these investigations except that by Black
and colleagues (1999), which involved community-based recruitment, involved
some sort of representative sampling strategy. In contrast to the studies
reviewed earlier, these investigations attempt to understand panic within the
context of tobacco dependence and severity. Here, across studies, results
indicate that among those persons meeting criteria for more addictive use of
cigarettes (e.g., nicotine dependence), there is a greater prevalence of panic
psychopathology (Breslau et al., 1991). For example, Breslau and colleagues
(1991) found that 6.6% of persons meeting criteria for moderate dependence,
4.8% of those with mild dependence, and 2.4% of those with no dependence
had a lifetime diagnosis of panic disorder. Nelson and Wittchen (1998) similarly
found that among participants endorsing a lifetime history of smoking (yes/no),
7.6% met lifetime diagnostic criteria for panic attacks, 2% for panic disorder,
and 4.4% for agoraphobia. These rates of panic psychopathology were
significantly greater than those reported among nonsmokers; 2.4% had a
panic attack history, 0.7% had panic disorder, and 1.6% had agoraphobia.
Smokers with a lifetime nicotine dependence diagnosis compared to smokers
without such a diagnosis evidenced greater rates of panic attacks (11.3% versus
4.0%), panic disorder (2.2% versus 1.8%), and agoraphobia (6.4% versus
2.5%). It should be noted that a similar, albeit not uniform, pattern of findings
was apparent for individuals with other psychiatric disorders (e.g., alcohol
dependence, drug dependence).
10 M. J. Zvolensky et al.

In the only study to focus on cigarette and smokeless tobacco use, Goodwin
and colleagues (2007) found that among a representative sample from the U.S.,
rates of past-year panic attacks (with or without agoraphobia) were greatest
among smokers with nicotine dependence (6.7%), followed by cigarette use
with no dependence (2.2%), both of which were greater than among those with
no past year cigarette or smokeless tobacco use (1.5%). Being dependent on
smokeless tobacco (1.9%) was largely comparable to no past year tobacco use,
both of which were greater than smokeless tobacco use without nicotine
dependence (0.6%).
Overall, the extant literature suggests that heavier rates of cigarette use
(greater degrees of dependence) are associated with a greater rate of comorbid-
ity with panic psychopathology. Although limited, this pattern of findings does
not yet seem to be apparent for smokeless tobacco use, suggesting that factors
related to the ‘‘mode of administration’’ may be an important domain to further
understand in tobacco-panic linkages.
Future directions. Though there are many avenues for future inquiry into the
nature of tobacco-panic comorbidity, here we highlight a few domains of
primary importance based upon the gaps in the existing literature. Before
specific recommendations are made, it is striking to point out that, to the best
of our knowledge, none of the past work focused on comorbidity issues has
been a priori oriented on tobacco-panic relations. Thus, it is, perhaps, not
surprising that some of the assessment approaches used in past work may not
be fully comprehensive or geared towards maximizing information about the
nature of the co-occurrence of these specific behavioral problems. As such, a
first-step in improving research in this domain would be to design evaluations
specifically focused on better understanding tobacco-panic comorbidity.
Beyond this general issue, there are at least three specific points within this
domain that would be particularly useful to address.
First, only one study has provided data on smokeless tobacco and panic
comorbidity. Thus, to foster further empirical knowledge in this domain, it is
necessary to complete investigations wherein multiple forms of tobacco use are
assessed to provide information on both cigarette use and smokeless tobacco.
Aside from providing much needed data on smokeless tobacco and
psychopathology, this type of work would help to define the parameters of
tobacco-panic relations more generally. In this same context, it would be
advisable to clarify the extent to which the observed co-occurrence rates
between tobacco use and panic psychopathology are similar to, or different
from, other health behaviors (e.g., alcohol use, physical exercise). In general,
research suggests smoking is strongly positively related to alcohol and other
substance use and negatively related to exercise (Zvolensky & Bernstein, 2005).
This work is necessary because it would further explicate the degree to which
tobacco use is or is not unique to the co-occurrence of panic psychopathology.
As the present book illustrates, research in the health-anxiety linkage is only
now emerging.
Panic and Tobacco 11

Second, there are a number of issues central to the generalizability of the

reviewed investigations. The large majority of these studies have nottilized
representative samples and therefore selection biases may be operative. It also
is noteworthy that only one study (Hayward et al., 1989) focused exclusively on
youth. Therefore, it is not possible to generalize the present tobacco-panic
relations to other segments of the lifespan (e.g., adolescents) or the various
stages of tobacco use (e.g., initiation, maintenance) that would presumably be
apparent across different age ranges. Additionally, there are very limited data
on tobacco-panic psychopathology linkages from a cross-national perspective.
As factors that govern tobacco use may vary across communities and cultures,
and in conjunction with the world-wide public health impact of both smoking
and panic problems, it is important to extend work in this area to more diverse
global populations.
Finally, existing work has largely utilized limited assessments of smoking
and panic psychopathology. Due to the focus on the ‘‘presence or absence of
daily smoking,’’ little is known about the nature or topography of smoking
behavior in terms of its association with panic psychopathology (e.g., age
of onset, age of daily use, amount used when smoking the heaviest). This
work would be improved by broadening smoking assessments to include a
more detailed account of smoking history. It also may be productive to
incorporate a multidimensional approach that takes into consideration
theoretically-relevant motivational processes underlying cigarette use
(Piper et al., 2004). This type of approach could be particularly valuable when
examining linkages between smoking and panic psychopathology, whereby
motivation to smoke to avoid negative affect might be a formative psychologi-
cal process (Zvolensky, Schmidt, Antony et al., 2005).

Nature of the Associations Between Tobacco Use

and Panic Psychopathology

Developmental course. As a basis for understanding the nature of the relations

between tobacco use and panic psychopathology, it is important to first clarify
their developmental course. Representative data on the age of onset of panic
attacks and smoking provide a means to evaluate temporal sequence.
Research suggests that the onset of daily smoking typically occurs between
the ages of 15 and 20 and rarely after age 25 (Breslau, Johnson, & Hiripi, 2001).
For example, the CDC reports that in the United States, approximately 3,900
adolescents between the ages of 12 and 17 years initiate cigarette smoking each
day (CDC, 2004), and an additional 1,500 become daily cigarette smokers each
day (Substance Abuse and Mental Health Services Administration, 2005).
Early studies of smokeless tobacco use indicate a mean age of onset between
10 and 12 years (i.e., Gottlieb, Pope, Rickert, & Hardin, 1993).
12 M. J. Zvolensky et al.

Studies examining the typical age of onset for panic attacks also suggest that
such problems often first occur in adolescence. For example, Goodwin and
Gotlib (2004) reported that the mean age of panic attack onset was 13.4 years
(n = 1,285; age range 9–17). Other studies based on community or school
samples have found similar results, with the modal age of onset of panic attacks
being 12 years old (Hayward et al., 1992; Warren & Zgourides, 1988), and
clinical samples report a slightly younger age of panic attack onset (Alessi &
Magen, 1988; Black & Robbins, 1990). These data suggest that, across studies,
panic attack onset tends to first occur between the ages of 12–13 years. One
important interpretative caveat to these investigations is that they focus
exclusively on youth and expressly do not sample from a larger age range.
Thus, it is possible that the ‘‘average’’ age of onset of panic attacks may be
different if the sampling strategy incorporated adults.
Based on available indirect data from smoking and panic attack age of onset
studies, it appears that in many instances the typical age of onset of panic
attacks precedes the typical age of onset of smoking. However, retrospective
reports of smokers with ‘‘active panic problems’’ are not entirely consistent with
this perspective. For example, Amering and colleagues (1999) examined 102
consecutive panic disorder patients with or without agoraphobia attending an
academic treatment clinic in Austria. Participants were diagnosed using the
SCID-III-R (First, Spitzer, Gibbon, & Williams, 1995) and interviewed about
their smoking status. Individuals presenting with ‘‘severe somatic illness’’
and comorbid depression and other psychiatric illnesses were excluded
from the study. Amering and colleagues (1999) reported that the onset of
smoking preceded the onset of panic disorder (cf. panic attacks) by 12.3 years
(SD = 9.4) in a community sample of individuals with the condition (n = 102).
Bernstein, Zvolensky, Schmidt, and Sachs-Ericsson (2007) directly evaluated
onset patterns among 4,409 adults (Mage = 33.1, SD = 10.7, females = 2,221)
from the NCS (Kessler et al., 1997). Results indicated that among cases with a
lifetime history of comorbid daily smoking and panic attacks (n = 167), the
onset of daily smoking (M = 16.0 years, SD = 3.0) preceded the onset of panic
attacks (M = 27.8 years, SD = 7.6) in the majority, but not among all, of the
comorbid cases (63.7%, n = 106). A relatively large minority of comorbid cases
(33%; n = 55) reported that panic attacks (M = 11.4 years, SD = 5.2) preceded
the onset of daily smoking (M = 18.2 years, SD = 4.7). The concurrent (same
year) onset of these two problems appeared rarely (3.3%, n = 6). Also, as the
pattern of ages of onset above illustrate, daily smoking demonstrated a rela-
tively consistent mean age of onset (mid to late adolescence) across comorbid
sub-samples and the uni-morbid sub-sample of smokers (age 18.5 years). In
contrast, the mean ages of onset of panic attacks differed markedly between the
comorbid sub-samples and the uni-morbid sub-sample of nonsmokers with
panic attacks (age 20.3 years). Overall, while data focused expressly on devel-
opmental course and smoking-panic psychopathology is limited, extant studies
suggest that the majority of cases may involve smoking preceding panic attacks.
Panic and Tobacco 13

Future directions. Again, it is important to highlight that work in this

domain, although important for illuminating basic facets of tobacco-panic
relations, is currently limited. First, to the best of our knowledge, no studies
have examined the developmental course for smokeless tobacco. Future pro-
spective work is therefore needed to expand this area of knowledge. Second,
future study is needed to more carefully examine the putative smoking-to-panic
and panic-to-smoking developmental courses. Third, smoking often occurs in a
context of other substance use patterns. Thus, it may be useful to understand
the developmental relations of tobacco and panic within the larger develop-
mental context of panic-substance use comorbidity. In this sense, understand-
ing the relative degree of ‘‘specificity’’ of initial findings vis a vis other substance
use patterns and problems as well as health behaviors (e.g., physical exercise)
may be a fruitful next research step.
Current knowledge regarding tobacco use and panic psychopathology.
Cross-sectional studies that have utilized interview and self-report methods
have uniformly indicated that smoking, compared to non-smoking, is asso-
ciated with more panic-relevant symptoms and impairment among nonclinical
(Zvolensky, Forsyth, Fuse, Feldner, & Leen-Feldner, 2002) and clinical
(McCabe et al., 2004; Zvolensky, Eifert, Feldner, & Leen-Feldner, 2003)
samples. For example, Zvolensky, Schmidt and McCreary (2003) found that
treatment-seeking smokers with panic disorder compared to nonsmokers with
panic disorder reported more severe and intense anxiety symptoms, greater
interview-based overall severity ratings of panic symptoms, and more social
impairment. In these investigations, effects did not vary by gender, age, or other
forms of substance use. Moreover, there is emerging evidence that these types of
effects are relatively specific to panic disorder and psychopathology that
frequently co-occurs with panic (e.g., posttraumatic stress disorder; Feldner
et al., 2007). For example, Morissette and colleagues (2006) found that smokers
with anxiety disorders, as compared to their non-smoking counterparts,
reported higher levels of anxiety sensitivity (i.e., fear of anxiety and
bodily-related sensations; McNally, 2002), anxiety symptoms, and agoraphobic
avoidance. However, this association was specific to panic disorder and not
evident for any of the other studied anxiety disorders, which did not include
posttraumatic stress disorder (Morissette et al., 2006).
Laboratory studies, although less common, have yielded similar findings.
Zvolensky and colleagues (2004), for example, employed a voluntary hyperven-
tilation paradigm to examine associations between smoking and panic-relevant
fearful responding to bodily sensations. Results indicated smokers with panic
disorder reported greater levels of anxiety than smokers without panic disorder
at baseline, and also showed greater increases in anxiety during the
post-challenge assessment and recovery periods. Although smokers with, versus
without, panic disorder did not differ on baseline or post-challenge anxiety,
smokers with, compared to without, panic disorder, demonstrated slower
affective recovery from the challenge. These results indicate that smoking
compared to not smoking is related to greater affective distress in response to
14 M. J. Zvolensky et al.

panic-relevant cues even among those with panic disorder. Less attention has
been focused on determining the relation between smoking rate and level of
affective distress or impairment. Yet, a number of studies, some involving
prospective measurement (discussed in greater detail below), have found that
smoking rate is related to greater degrees of panic-specific emotional symptoms
(e.g., panic-relevant avoidance; Breslau & Klein, 1999; Goodwin, Lewinsohn, &
Seeley, 2005; Johnson et al., 2000; McLeish, Zvolensky, & Bucossi, 2007;
Zvolensky, Kotov, Antipova, & Schmidt, 2003). Thus, there is empirical
evidence that both smoking status and rate are related to increased risk for
panic-relevant emotional vulnerability.
Cross-sectional tests also have helped to clarify factors that may affect the
smoking-panic relation. In one study of epidemiologically-defined (i.e., repre-
sentative) adult residents of Moscow (n = 95 daily smokers from a larger
sample of about 400 persons; Zvolensky, Kotov et al., 2003), anxiety sensitivity
moderated the effects of cigarettes smoked per day (m = 15) on level of
agoraphobic avoidance. This significant interaction accounted for approxi-
mately 10% of unique variance after controlling for their respective main effects
and the theoretically-relevant factors of problematic alcohol use and negative
affectivity. No interaction, however, was found for panic attacks, potentially
due to the fact that assessment of this factor was restricted to the past (most
recent) week to enhance the validity of panic reports (but probably truncating
variability). Similar moderating effects have been evident for perceived health
among young adult daily smokers (McLeish, Zvolensky, Bonn-Miller, &
Bernstein, 2006), and for neuroticism among a representative sample of adult
smokers (Zvolensky, Sachs-Ericsson, Feldner, Schmidt, & Bowman, 2006).
Overall, these findings suggest smokers are not a homogeneous group in regard
to their risk for panic problems and that individual differences in anxiety
sensitivity (or other cognitive-affective factors like perceived health or neuroti-
cism) may be key factors in accounting for such differences.
Moderating effects for anxiety sensitivity also have been evident in
between-group tests involving smokers and nonsmokers. For example, the
combination of high levels of anxiety sensitivity and a positive current smoking
status predicted panic symptoms and somatic complaints, but not depressive
symptoms in a biological challenge test (Leen-Feldner et al., 2007). Again, such
findings suggest that anxiety sensitivity (and possibly other factors) may mod-
erate the relation between smoking and prototypical panic psychopathology
variables (panic attacks and somatic complaints) even after controlling for
gender and negative affectivity. Moreover, these associations are specific to
panic-relevant processes. In a re-analysis of the Russian epidemiological study
reported earlier, Zvolensky and colleagues extended this smoking and anxiety
sensitivity effect (Zvolensky, Kotov, Bonn-Miller, Schmidt, & Antipova, in
press). Here, anxiety sensitivity, again, moderated the association of smoking
status with indices of anxiety symptoms; effects were evident after controlling
for the variance accounted for by alcohol use problems, environmental stress
(past month), and gender.
Panic and Tobacco 15

Although cross-sectional data are informative in the study of tobacco-panic

psychopathology relations, their utility also is limited. Prospective studies offer
unique insight into the nature of the observed relations over time, and by
extension, the order or temporal sequence of the associations. Researchers
have evaluated the association between smoking and risk of panic
psychopathology in a number of studies. Breslau and Klein (1999) tested the
association between daily smoking and risk for panic attacks and panic
disorder. Participants were drawn from two separate epidemiologically-defined
data sets. Across both data sets, results indicated that there was a significant
lifetime and prospective association between daily smoking and onset of panic
attacks and panic disorder; daily smokers were almost 4 times more likely to
experience panic attacks and 13 times more likely to develop panic disorder
after controlling for major depression and gender. Additionally, among
individuals who continued to smoke, compared to those who had quit, there
was a significantly increased risk for experiencing a panic attack and panic
disorder. Johnson and colleagues (2000) also found that anxiety disorders
during adolescence were not significantly related to smoking in young
adulthood. However, smoking in adolescence increased the risk for developing
agoraphobia and panic disorder during early adulthood. These effects were
observed above and beyond the variance accounted for by temperament, family
history of psychopathology, drug/alcohol use and other theoretically-relevant
factors. Specifically, adolescents who were heavy smokers were 15.6 times
more likely to develop panic disorder in early adulthood than non-smokers.
Interestingly, adolescents who smoked fewer than 20 cigarettes per day were not
at elevated risk for the development of (later) anxiety disorders,
potentially suggesting, once again, that heavier smoking levels impart greater
panic-related risk.
In another prospective study recently completed in Germany, 2,500 partici-
pants (ages 14–24 years at baseline) were evaluated over 4 years (Isensee,
Wittchen, Stein, Höfler, & Lieb, 2003). Compared with all other levels of
smoking, dependent regular smokers at baseline were significantly more likely
to develop panic attacks and panic disorder, and a similar pattern was observed
for agoraphobia. Similarly, Breslau, Novak, and Kessler (2004) evaluated daily
smoking and subsequent onset of psychiatric disorders. Results indicated
that the onset of panic disorder (odds ratio = 2.6) and agoraphobia
(odds ratio = 4.4) were associated with pre-existing daily smoking after
controlling for age, gender, ethnicity, and educational level. Additionally,
after controlling for pre-existing psychiatric disorders and sociodemographic
characteristics, current nicotine dependent smokers were significantly more
likely to have panic disorder compared to current non-dependent smokers
and former smokers. Importantly, the likelihood of panic disorder and
agoraphobia was significantly reduced as time since quitting increased; these
effects were specific to these conditions and not other psychiatric disorders
(e.g., major depressive disorder), suggesting quitting smoking likely decreases
the risk of developing panic problems, an issue that is discussed in greater detail
16 M. J. Zvolensky et al.

later in the chapter. More recently, Goodwin and colleagues (2005) replicated
the results of Breslau and Klein (1999), Johnson et al. (2000), and Isensee et al.
(2003) by finding that daily smoking during adolescence was associated with an
increased risk for panic attacks and panic disorder in young adulthood.
Moreover, the observed effects were no longer evident after controlling for
parental smoking and anxiety disorder status, suggesting that these family
history characteristics may be formative in the linkages between smoking and
panic psychopathology.
Prospective tests examining moderating factors in the tobacco use-panic
relation are very limited. In the only study to date on this topic, McLeish
and colleagues (2007) evaluated the moderating role of anxiety sensitivity in
the relation between smoking rate and panic vulnerability variables among a
community-based sample of 125 daily smokers (60 females; Mage =
26.02 years). Findings indicate that the interaction between anxiety sensitiv-
ity and smoking rate significantly predicted concurrent agoraphobic avoid-
ance (3.2% of unique variance) and change in levels of anticipatory anxiety
about bodily sensations during the 3-month follow-up period (4.7% unique
variance). Smokers high in anxiety sensitivity who also smoked at greater
rates reported the highest levels of avoidance and greatest increase in antici-
patory anxiety. These data, in accord with cross-sectional findings (Leen-
Feldner et al., 2007; Zvolensky, Kotov et al., 2003), once again suggest that
anxiety sensitivity is an important individual difference factor that, when
coupled with higher rates of smoking, is associated with greater levels of
avoidance and anticipatory anxiety among daily smokers, both of which
contribute to the development of panic psychopathology.
Overall, research co-addressing smoking and panic psychopathology
suggests that smoking can be considered a variable risk factor for panic
problems. Indeed, existing work provides evidence regarding relations with
panic problems based on cross-sectional and prospective studies, but it is
noteworthy that this work is rarely multi-method in its approach. To have
more confidence in smoking-panic psychopathology relations, the incorpora-
tion of multi-method assessment protocols would be an important next
research step. Additionally, evidence from cross-sectional, and to a lesser
extent, prospective studies indicates that fears of internal sensations (anxiety
sensitivity) and perhaps other ‘‘affect-amplifiers’’ (e.g., perceived health,
neuroticism) may moderate smoking-panic processes.
Future directions. There is a rapidly developing empirical literature on
tobacco-use and panic psychopathology relations. Such scientific interest in
this work underscores its public health relevance and potential clinical implica-
tions (see Zvolensky, Bernstein, Yartz, McLeish, & Feldner, in press, for an
expanded discussion of treatment implications of tobacco-panic relations). At
the same time, this literature remains relatively under-developed and there are a
number of key areas in need of future study.
First, as in the area of comorbidity prevalence studies reviewed earlier, there
is a dearth of data on smokeless tobacco-panic relations. Virtually no scientific
Panic and Tobacco 17

data exists on this important topic, making it a fertile area for future explora-
tion. Second, available data suggest daily smoking tends to precede the onset of
panic attacks in the majority of cases, although direct evaluations with panic
disorder and agoraphobia have not been completed. Given that smoking can be
changed via intervention (Abrams et al., 2003), there is evidence of its potential
malleability, and hence, possible application to prevention programs for panic
psychopathology. Overall, then, evidence that changing cigarette smoking rate
or smoking cessation will alter the future risk of panic psychopathology from a
preventative standpoint is lacking. Thus, it is currently not clear if smoking
represents a variable marker or a variable causal risk factor for panic psycho-
pathology. To clarify this issue, it is important for future research to examine
changes in smoking prospectively following experimental manipulation
(e.g., smoking cessation intervention; Zvolensky, Schmidt, Bernstein, &
Keough, 2006). Third, research has yet to examine the possibility that shared
or common risk factors may further explain the development of comorbid
tobacco use and panic. It is theoretically possible that certain biological,
psychological, and social factors may partially underlie the etiology and
maintenance of these behavior problems. And finally, while there is a growing
literature on moderating factors, there has been little scientific attention to
mediators of smoking-panic associations and therefore almost no empirical
knowledge exists pertaining to the putative causal mechanisms of interest.
Intensifying the focus on mediators of smoking-panic relations is a clinically-
relevant and timely task. Specifically, clarification of key mechanisms through
which smoking achieves its panicogenic effects will stimulate the development
of targeted interventions focused on therapeutic processes, and help to establish
such processes (e.g., emotional reactivity) as important in the etiology and/or
maintenance of panic-related problems. Only Breslau and Klein (1999)
conducted exploratory analyses of possible mediators by evaluating the
role of lung disease. Although medical illness is one useful process to better
understand, other factors such as perceived health, affect tolerance, various
trajectories of emotional distress (e.g., delayed recovery), withdrawal
symptoms, and avoidance-oriented smoking patterns are all examples of
theoretically-relevant factors deserving of future study (see Zvolensky &
Bernstein, 2005, for an expanded discussion).
Current knowledge regarding the relation between panic psychopathology.
pre-morbid panic risk variables, and tobacco use. Although much of the most
highly publicized work on smoking and panic psychopathology pertains to the
potential role of smoking in the onset or maintenance of panic problems, panic
vulnerability characteristics, broadly encompassing both pre-morbid variables
and full-blown panic problems, may conversely impact smoking behavior
(Zvolensky & Bernstein, 2005). Work in this domain has focused on empirical
evidence related to smoking cessation outcome, expression of withdrawal
symptoms, and motivational and cognitive processes related to smoking
behavior (e.g., outcome expectancies). A major strength of work in this domain
is that study of smoking has involved measurement of various facets of
18 M. J. Zvolensky et al.

smoking-related problems and processes (i.e., motivational processes) as

opposed to focusing more narrowly on rates of tobacco use. This multidimen-
sional conceptualization and measurement of smoking behavior is central to
theoretical and clinical advances relevant to tobacco-panic relations. Moreover,
it provides Moreover, it provides a means by which to further examine the role
of panic-specific, and other anxiety factors (e.g., traumatic event exposure,
negative affectivity), in smoking initiation (Bernstein, Zvolensky et al., 2007;
Feldner et al., 2007; Isensee et al., 2003).
Research has shown that affective vulnerability factors like panic psycho-
pathology may be related to problems in quitting smoking. Lasser et al. (2000),
for example, reported quit rates (i.e., proportion of lifetime smokers who were
not current smokers) in relation to psychiatric diagnosis among a representative
sample from the U.S. Using 1-month diagnostic status as a criterion point, the
quit rate was 29% for persons with panic attacks, 32% for those with panic
disorder, and 23% for persons with agoraphobia. Individuals with both panic
attacks and agoraphobia in the past month were significantly less successful in
quitting smoking compared to individuals with no mental illness (42%). Covey
and colleagues (1994) reported conceptually similar findings that panic psycho-
pathology may be related to poorer success in quitting, although it is not clear
whether this effect is more robust than the types of association(s) between
psychiatric disorders more generally and quit success. Other cross-sectional
field and laboratory work using a community sample has found that daily
smokers with a history of panic attacks, but no axis I histories, reported
significantly shorter average quit attempt histories, measured in days, com-
pared to smokers without panic (Zvolensky, Lejuez, Kahler, & Brown, 2004).
Similar results have been observed in laboratory investigations (Zvolensky,
Feldner, Eifert, & Brown, 2001). Although limited by cross-sectional design,
and by extension, possible reporting errors (e.g., recall biases), these data
provide evidence of a relation between panic psychopathology variables and
problems in quitting.
A related line of work has focused on anxiety sensitivity and success in
quitting smoking. Anxiety sensitivity tends to be elevated among individuals
who fear anxiety and arousal-related sensations such as panic disorder
(Bernstein & Zvolensky, 2007). In the earliest study in this domain, Brown,
Kahler, Zvolensky, Lejuez, and Ramsey (2001) examined a subset of data from
a randomized controlled clinical trial comparing standard smoking cessation
treatment versus standard smoking cessation plus cognitive-behavioral treat-
ment for depression in smokers with past major depressive disorder. In this
investigation, the association between anxiety sensitivity and relapse during the
early stages of a quit attempt (e.g., first week), when individuals are most apt to
experience symptoms of anxiety (Hughes, Higgins, & Hatsukami, 1990), was
examined. Anxiety sensitivity was significantly associated with increased odds
of lapsing during the first week after quit day (odds ratio = 2.0). Subsequent
work has conceptually replicated and extended the results of Brown and
colleagues (2001). For example, Zvolensky, Bonn-Miller, Bernstein, and
Panic and Tobacco 19

Marshall (2006) found anxiety sensitivity was significantly associated with

increased risk of early smoking relapse among a community sample of daily
smokers; these effects were evident above and beyond smoking rate and nega-
tive affectivity. Such work has recently been extended to low-level smokers from
Mexico, adding cross-national empirical support (Zvolensky, Bernstein, et al.,
in press). Collectively, there is a growing amount of empirical evidence suggest-
ing that panic psychopathology or pre-morbid panic-relevant variables such as
elevated anxiety sensitivity is related to early relapse problems, and possibly,
lower rates of overall success in quitting. Here again, controlled, prospective
studies are an important next research step, as they would remove concerns that
observed effects to date are attributable to reporting biases.
A closely related line of inquiry has suggested that anxiety sensitivity is
related to motivation to quit, barriers to quitting, and reasons for quitting.
For example, Zvolensky, Baker and colleagues (2004) found anxiety sensitivity
was related to higher levels of current motivation to quit smoking among adult
daily smokers (Mage = 20.4], Mcigarettes per day = 10.2); effects were not attri-
butable to other theoretically-relevant factors (e.g., gender, smoking rate;
Zvolensky, Baker et al., 2004). These findings may at first seem counterintuitive
in that it seems logical that individuals with high levels of anxiety sensitivity
would be less likely to express interest or motivation in quitting due to the
feared negative consequences related to quitting (e.g., withdrawal symptoms,
emotional dyscontrol). Yet, related work suggests that smokers who worry
about the negative health-related effects of smoking may engage in more quit-
ting behavior (Dijkstra & Brosschot, 2003). From this perspective, high anxiety
sensitivity smokers may be more apt to perceive a personal vulnerability to the
negative effects of smoking (e.g., health risks), and as such, express greater
motivation to quit (Zvolensky & Bernstein, 2005) despite their greater difficulty
in successfully doing so (Brown et al., 2001). In line with this reasoning,
Zvolensky, Vujanovic and colleagues (2007) more recently examined the rela-
tions between anxiety sensitivity and (1) motivation to quit smoking, (2) barriers
to smoking cessation, and (3) reasons for quitting smoking among 329
(160 females; Mage = 26.08 years, SD = 10.92) adult daily smokers. After
covarying for theoretically-relevant variables (negative affectivity, gender, axis
I psychopathology, non-clinical panic attack history, number of cigarettes
smoked per day, and current levels of alcohol consumption), anxiety sensitivity
was significantly incrementally related to level of motivation to quit smoking, as
well as perceived barriers to quitting smoking. Additionally, after accounting
for the variance explained by other theoretically relevant variables, anxiety
sensitivity was significantly associated with self control reasons for quitting
smoking (intrinsic factors) as well as immediate reinforcement and social influ-
ence reasons for quitting (extrinsic factors). These results provide empirical
evidence that anxiety sensitivity is uniquely related to level of motivation to quit
smoking, perceived barriers to quitting, and certain intrinsic and extrinsic
reasons for quitting.
20 M. J. Zvolensky et al.

Panic psychopathology or pre-morbid risk factors also appear to be related

to severity of acute nicotine withdrawal. In an early study in this domain,
Breslau, Kilbey, and Andreski (1992) found tobacco withdrawal symptoms in
a sample of young adults were significantly elevated among smokers with ‘‘any
anxiety disorder’’ compared to individuals without a history of such disorders;
however, specific anxiety diagnoses were not provided, rendering unclear the
specificity of such results to panic psychopathology per se. Zvolensky and
colleagues (2004) found that daily smokers with a history of panic attacks
reported significantly more intense anxiety-related withdrawal symptoms
(anxiety, restlessness, difficulty concentrating, and irritability) compared to
smokers without such a history; no differences were evident for the other
tobacco withdrawal symptoms (e.g., increased appetite). In another study,
Zvolensky, Baker et al. (2004) tested whether anxiety sensitivity predicted the
intensity of withdrawal symptoms during the first week of daily smokers’ most
recent quit attempt. Results indicated that anxiety sensitivity predicted the
intensity of nicotine withdrawal symptoms during the first week of smokers’
most recent quit attempt, and this effect was above and beyond variance
accounted for by negative affectivity, panic attack history, gender, cigarettes
per day, and age of smoking onset, accounting for 16% of unique variance in
withdrawal symptoms. This work is promising in suggesting panic-specific
factors are related to an enhanced reactivity to nicotine withdrawal symptoms.
Experimental work, which is now underway in our laboratory, is necessary to
provide an additional degree of confidence in such conclusions.
Another facet of evidence in support of a panic-tobacco relation is apparent
from motivational and outcome expectancy research. In regard to smoking-
related motivational processes, there is a large empirical literature documenting
that smokers often attribute their smoking, at least in part, to its mood-regulating
functions and believe that smoking will reduce negative affect states (Parrott,
1999). Due to their affective vulnerability, smokers with panic-relevant vulner-
abilities (i.e., high anxiety sensitivity) may be particularly motivated to smoke to
escape from emotional distress elicited by acute nicotine withdrawal or
non-withdrawal states (e.g., anticipatory anxiety; Zvolensky & Bernstein,
2005). A number of cross-sectional studies support this theory. Specifically,
studies have indicated that anxiety sensitivity is associated with coping-oriented
smoking motives among young adults with no history of psychopathology
(Novak, Burgess, Clark, Zvolensky, & Brown, 2003; Stewart, Karp, Pihl, &
Peterson, 1997; Zvolensky, Bonn-Miller et al., 2006), adolescents (Comeau,
Stewart, & Loba, 2001), and individuals with a past history of major depression
(Brown, Kahler et al., 2001). Zvolensky, Feldner, Leen-Feldner et al. (2004)
report conceptually similar findings for relations between anxiety sensitivity
and negative-reinforcement outcome expectancies for smoking. The Comeau
et al. (2001) investigation, in particular, is noteworthy in that anxiety sensitivity
moderated the relation between trait anxiety (frequency of anxiety symptoms)
and use of cigarettes to cope with affective distress, reporting a stronger relation-
ship between anxiety and use of cigarettes to cope with negative emotions
Panic and Tobacco 21

among high anxiety sensitive compared with low anxiety sensitive youth. Using a
sample of panic disorder patients, Zvolensky and colleagues (2005) also found
that smokers with panic disorder reported higher levels of smoking to reduce
negative affect than their counterparts without such a history. These
cross-sectional studies are not capable of elucidating the direction of the effects.
Theoretically, coping-oriented smoking motives may have bi-directional effects,
influencing, and being influenced by, affective vulnerability. An initial investiga-
tion exploring this possibility was consistent with such an account (Gregor,
Zvolensky, Bernstein, Marshall, & Yartz, 2007), reporting that coping-oriented
motives were incrementally related to a variety of negative affective and cognitive
Overall, there are a variety of separate, but related, lines of inquiry indicating
panic psychopathology and a select number of pre-morbid risk factors are
meaningfully related to smoking behavior. These lines of work differ in their
focus, but broadly indicate that panic factors (full blown disorders and certain
pre-morbid risk factors) are related to abstinence duration during smoking
cessation outcome expectancies related to smoking, perceived barriers and
reasons for quitting, nicotine withdrawal symptom severity, and motivational
bases for smoking. Thus, it is most appropriate to conceptualize many of the
studied variables (e.g., anxiety sensitivity) as variable risk factors. Although
definitive prospective work has not been conducted to firmly establish temporal
precedence in many, if not most, of the investigations, theoretically, panic
variables would precede the smoking factors. It also is theoretically possible
that panic and panic-relevant risk factors may not necessarily developmentally
precede smoking, but nevertheless meaningfully influence the course and nature
of smoking behavior over time via many of the processes described throughout
this chapter. Further prospective work will delineate the possibility that these
panic-smoking relations may be transactional over development.
Future directions. As in the earlier sections of this chapter, a first observation
and recommendation for future research is to better understand the relations
between panic psychopathology and smokeless tobacco use. There is no empiri-
cal work completed in this domain to the best of our knowledge, leaving this
facet of the tobacco-panic linkage undocumented. Second, essentially all of the
existing work on panic psychopathology (and related factors) and smoking
behavior is focused on main effects. This approach seems appropriate given
the currently limited knowledge in the area, but represents only a ‘‘first step’’ in a
larger scientific effort. Future work is needed to increase understanding about
linkages among these factors beyond main effects by including moderational
and mediational tests of theoretically relevant variables. Similarly, the possibi-
lity that shared or common risk factors may underlie panic problems,
panic-relevant risk factors, and these smoking-related problems and processes
has received little theoretical or empirical evaluation. Third, the generalizability
of panic psychopathology and smoking research is limited in that it has focused
largely on adults from the U.S. Furthermore, there is very little information on
the nature of these relations among youth. Given the early age of onset of these
22 M. J. Zvolensky et al.

behaviors and their health relevance, as well as the international scope of this
public health problem, research development in these domains is needed.
Finally, little research has directly targeted panic vulnerability factors in smok-
ing cessation interventions. Similarly, there is little work addressing smoking in
the context of panic-related treatments. Given the consistent empirical evidence
of bi-directional associations between these often comorbid behavioral pro-
blems, it is important to develop specialized treatments, as generic interventions
may not target the affective vulnerability processes functioning to maintain
smoking in this population. For example, it may be useful to integrate inter-
oceptive exposure, cognitive restructuring, and psychoeducation exercises
developed for panic prevention and treatment programs with standard smoking
cessation strategies and nicotine replacement therapy. These therapeutic tactics
may be most effective when they target theoretically-relevant panic risk factors
like anxiety sensitivity in order to facilitate cessation. As a second illustration, it
may be useful to target smoking cessation as part of evidence-based panic-
problem treatment strategies. While there have been some inroads made in this
domain, with successful case reports and pilot studies now being reported
(Zvolensky, Bernstein et al., in press; Zvolensky, Lejuez, Kahler, & Brown,
2003; Zvolensky, Schmidt et al., 2006), much work is yet to be addressed in this


The present chapter provides an updated review of extant empirical work

pertaining to the inter-relations between tobacco use and panic psychopathol-
ogy. Although a relatively nascent area of research, a recent spate of empirical
evidence indicates that theoretically and clinically important associations exist
between tobacco use and panic psychopathology. Although promising, there
are a multitude of critical gaps in the research literature that need to be
addressed in future studies. We hope that such ongoing research may help
translate knowledge about basic processes to the development and dissemina-
tion of powerful clinical intervention strategies for tobacco users with
panic-related vulnerabilities or psychopathology.


About: Smoking Cessation (2006). Retrieved October 1, 2006 from http://quitsmoking.about.

Abrams, D. B., Niaura, R., Brown, R. A., Emmons, K. M., Goldstein, M. G., & Monti, P. M.
(2003). The tobacco dependence treatment handbook: A guide to best practices. New York:
Guilford Press.
Alessi, N. E., & Magen, J. (1988). Panic disorder in psychiatrically hospitalized children.
American Journal of Psychiatry, 145, 1450–1452.
Panic and Tobacco 23

American Cancer Society. (2006a). Cancer Facts & Figures – 2006. American Cancer Society.
American Cancer Society (2006b, February 13). Prevention and Early Detection:
Quitting Spit (Smokeless) Tobacco. Retrieved September 18th from http://www.cancer.
American Cancer Society (1999). Cancer facts and figures –1999. American Cancer Society.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental
disorders (4th ed. text revision). Washington, DC: Author.
Amering, M., Banker, B., Berger, P., Griengl, H., Windhaber, J., & Katschnig, H. (1999).
Panic disorder and cigarette smoking behavior. Comprehensive Psychiatry, 40, 35–38.
Baker-Morissette, S. L., Gulliver, S. B., Wiegel, M., & Barlow, D. H. (2004). Prevalence of
smoking in anxiety disorders uncomplicated b comorbid alcohol of substance abuse.
Journal of Psychopathology and Behavioral Assessment, 26, 107–112.
Barlow, D. H. (2002). Anxiety and its disorders: The nature and treatment of anxiety and panic
(2nd ed.). New York: Guilford Press.
Barlow, D. H., Brown, T. A., & Craske, M. G. (1994). Definitions of panic attacks and panic
disorder in the DSM-IV: Implications for research. Journal of Abnormal Psychology, 103,
Bernstein, A., & Zvolensky, M. J. (2007). Anxiety sensitivity: Selective review of promising
research and future directions. Expert Review in Neurotherapeutics, 7, 97–101.
Bernstein, A., Zvolensky, M. J., Schmidt, N. B., & Sachs-Ericsson, N. (2007). Developmental
course(s) of lifetime cigarette use and panic attack comorbidity: An equifinal phenom-
enon. Behavior Modification, 31, 117–135.
Black, B., & Robbins, D. R. (1990). Panic disorder in children and adolescents. Journal of the
American Academy of Child and Adolescent Psychiatry, 29, 36–44.
Black, B., Zimmerman, M., & Coryell, W. H. (1999). Cigarette smoking and psychiatric
disorder in a community sample. Annals of Clinical Psychiatry, 11, 129–136.
Breslau, N. (1995). Psychiatric comorbidity of smoking and nicotine dependence. Behavior
Genetics, 25, 95–101.
Breslau, N., Johnson, E. O., & Hiripi, E. (2001). Nicotine dependence in the United States.
Archives of General Psychiatry, 58, 810–816.
Breslau, N., Kilbey, M., & Andreski, P. (1991). Nicotine dependence, major depression, and
anxiety in young adults. Archives of General Psychiatry, 48, 1069–1074.
Breslau, N., Kilbey, M. M., & Andreski, P. (1992). Nicotine withdrawal symptoms and
psychiatric disorders: findings from an epidemiologic study of young adults. American
Journal of Psychiatry, 149, 464–469.
Breslau, N., & Klein, D. F. (1999). Smoking and panic attacks: An epidemiologic investiga-
tion. Archives of General Psychiatry, 56,1141–1147.
Breslau, N., Novak, S. P., & Kessler, R. C. (2004). Daily smoking and the subsequent onset of
psychiatric disorders. Psychological Medicine, 34, 323–333.
Brown, R. A., Kahler, C. W., Zvolensky, M. J., Lejuez, C. W., & Ramsey, S. E. (2001).
Anxiety sensitivity: Relationship to negative affect smoking and smoking cessation in
smokers with past major depressive disorder. Addictive Behaviors, 26, 887–899.
Brown, R. A., Lewinsohn, P. M., Seeley, J. R., & Wagner, E. F. (1996). Cigarette smoking,
major depression, and other psychiatric disorders among adolescents. Journal of the
American Academy of Child and Adolescent Psychiatry, 35, 1602–1610.
Bryant, R. A., & Panasetis, P. (2005). The role of panic in acute dissociative reactions
following trauma. British Journal of Clinical Psychology, 44, 489–494.
Burke, K. C., Burke, J. D., Regier, D. A., & Rae, D. S. (1990). Age at onset of selected
mental disorders in five community populations. Archives of General Psychiatry, 47,
Centers for Disease Control and Prevention (CDC). (1994). Preventing Tobacco Use Among
Young People: A Report of the Surgeon General. Atlanta, GA: Department of Health and
Human Services.
24 M. J. Zvolensky et al.

Centers for Disease Control and Prevention (CDC). (1996). Cigarette smoking among
adults–United States, 1994. Morbidity and Mortality Weekly Report, 45, 588–590.
Centers for Disease Control and Prevention (CDC). (2002). Cigarette smoking among
adults–United States, 2000. Morbidity and Mortality Weekly Report, 51, 642–645.
Centers for Disease Control and Prevention (CDC). (2004). State-specific prevalence of
current cigarette smoking among adults–United States, 2002. Morbidity and Mortality
Weekly Report, 52, 1277–1280.
Cohen, S., Lichtenstein, E., Prochaska, J. O., Rossi, J. S., Gritz, E. R., Carr, C. R. et al. (1989).
Debunking myths about self-quitting: Evidence from 10 prospective studies of persons
who attempt to quit smoking by themselves. American Psychologist, 44, 1355–1365.
Comeau, N., Stewart, S. H., & Loba, P. (2001). The relations of trait anxiety, anxiety
sensitivity, and sensation seeking to adolescents’ motivations for alcohol, cigarette, and
marijuana use. Addictive Behaviors, 26, 803–825.
Costello, E. J., Erkanli, A., Federman, E., & Angold, A. (1999). Development of psychiatric
comorbidity with substance abuse in adolescents: Effects of timing and sex. Journal of
Clinical Child Psychology, 28, 298–311.
Covey, L. S., Hughes, D. C., Glassman, A. H., Blazer, D. G., & George, L. K. (1994).
Ever-smoking, quitting, and psychiatric disorders: Evidence from the Durham, North
Carolina, epidemiologic catchment area. Tobacco Control, 3, 222–227.
Degenhardt, L., Hall, W., & Lynskey, M. (2001). Alcohol, cannabis, and tobacco use among
Australians: A comparison of their associations with other drug use and disorders,
affective and anxiety disorders, and psychosis. Addiction, 96, 1603–1614.
Dijkstra, A., & Brosschot, J. (2003). Worry about health in smoking behaviour change.
Behaviour Research and Therapy, 41, 1081–1092.
Farrell, M., Howes, S., Bebbington, P., Brugha, T., Jenkins, R., Lewis, G., et al. (2001).
Nicotine, alcohol and drug dependence and psychiatric comorbidity. British Journal of
Psychiatry, 179, 432–437.
Fava, G. A., Grandi, S., & Canestrari, R. (1988). Prodromal symptoms in panic disorder with
agoraphobia. American Journal of Psychiatry, 145, 1564–1567.
Feldner, M. T., Babson, K. A., & Zvolensky, M. J. (2007). Smoking, traumatic event
exposure, and post-traumatic stress: A critical review of the empirical literature. Clinical
Psychology Review, 27, 14–45.
Feldner, M. T., Zvolensky, M. J., & Leen-Feldner, E. (2004). A critical review of the empirical
literature on coping and panic disorder. Clinical Psychology Review, 24, 123–148.
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1995). Structured Clinical
Interview for DSM-III-R Axis I Disorders. New York: New York State Psychiatric
Goodwin, R. D., & Gotlib, I. H. (2004). Panic attacks and psychopathology among youth.
Acta Pyschiatrica Scandinavica, 109, 216–221.
Goodwin, R. D., Lewinsohn, P. M., & Seeley, J. R. (2005). Cigarette smoking and panic
attacks among young adults in the community: The role of parental smoking and anxiety
disorders. Biological Psychiatry, 58, 686–693.
Goodwin, R. D., Zvolensky, M. J., & Keyes, K. (2007). Nicotine dependence and mental
disorders among adults in the United States: Evaluating the role of mode of administration.
Manuscript submitted for publication.
Gottlieb, A., Pope, S. K., Rickert, V. I., & Hardin, B. H. (1993). Patterns of smokeless tobacco
use by young adolescents. Pediatrics, 91, 75–78.
Gregor, K., Zvolensky, M. J., Bernstein, A., Marshall, E. C., & Yartz, A. (2007). Smoking
motives in the prediction of affective vulnerability among young adult daily smokers.
Behaviour Research and Therapy, 45, 471–482.
Hatsukami, D., Jensen, J., Allen, S., Grillo, M., & Bliss, R. (1996). Effects of behavioral and
pharmacological treatment on smokeless tobacco users. Journal of Consulting and Clinical
Psychology, 64, 153–161.
Panic and Tobacco 25

Hatsukami, D. A., & Severson, (1999). Oral spit tobacco: addiction, prevention, and treat-
ment. Nicotine & Tobacco Research, 1, 21–44.
Hayward, C., Killen, J. D., Hammer, L. D., Litt, I. F., Wilson, D. M., Simmonds, B., et al.
(1992). Pubertal stage and panic attack history in sixth- and seventh-grade girls. American
Journal of Psychiatry, 149, 1239–1243.
Hayward, C., Killen, J. D., & Taylor, C. B. (1989). Panic attacks in young adolescents.
American Journal of Psychiatry, 146, 1061–1062.
Hayward, C., & Wilson, K. (in press). Anxiety sensitivity: The missing piece to the agorapho-
bia-without-panic puzzle. Behavior Modification.
Himle, J., Thyer, B. A., & Fischer, D. J. (1988). Prevalence of smoking among anxious
outpatients. Phobia Practice and Research Journal, 1, 25–31.
Hughes, J. R., Hatsukami, D. K., Mitchell, J. E., & Dalgren, L. A. (1986). Prevalence of
smoking among psychiatric outpatients. American Journal of Psychiatry, 143, 993–997.
Hughes, J. R., Higgins, S. T., & Hatsukami, D. (1990). Effects of abstinence from tobacco:
A critical review. In L. T. Kozlowski, H. M. Annis, H. D. Cappell, F. B. Kalant, E. M., &
E. R. Sellers, Vingilis (Eds.), Research advances in alcohol and drug problems (Vol. 10).
Plenum publishing.
Isensee, B., Wittchen, H-U., Stein, M. B., Höfler, M., & Lieb, R. (2003). Smoking increases
the risk of panic: Findings from a prospective community study. Archives of General
Psychiatry, 60, 692–700.
Johnson, J. G., Cohen, P., Pine, D. S., Klein, D. F., Kasen, S., & Brook, J. S. (2000).
Association between cigarette smoking and anxiety disorders during adolescence and
early adulthood. JAMA: Journal of the American Medical Association, 284, 2348–2351.
Kandel, D. B., Huang, F. Y., & Davies, M. (2001). Comorbidity between patterns of
substance use dependence and psychiatric syndromes. Drug and Alcohol Dependence, 64,
Kandel, D. B., Johnson, J. G., Bird, H. R., Canino, G., Goodman, S. H., Lahey, B. B., et al.
(1997). Psychiatric disorders associated with substance use among children and adoles-
cents: Findings form the Methods for the Epidemiology of Child and Adolescent Mental
Disorders (MECA) Study. Journal of Abnormal Child Psychology, 25, 121–132.
Kazdin, A. E., Kraemer, H. C., Kessler, R. C., Kupfer, D. J., & Offord, D. R. (1997).
Contributions of risk-factor research to developmental psychopathology. Clinical
Psychology Review, 17, 375–406.
Kessler, R. C., Anthony, J. C., Blazer, D. G., Bromet, E., Eaton, W. W., Kendler, K., et al.
(1997). The US national cormorbidity survey: Overview and future directions. Epidemio-
logiae Psichiatria Sociale, 6, 4–16.
Kessler, R. C., Chiu, W. T., Jin, R., Ruscio, A. M., Shear, K., & Walters, E. E. (2006). The
epidemiology of panic attacks, panic disorder, and agoraphobia in the national comor-
bidity survey replication. Archives of General Psychiatry, 63, 415–424.
Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S., et al.
(1994). Lifetime and 12-month prevalence rates of DSM-III-R psychiatric disorders in the
United States: Results from the National Comorbidity Survey. Archives of General
Psychiatry, 51, 8–19.
Kraemer, H. C., Kazdin, A. E., & Offord, D. R. (1997). Coming to terms with terms of risk.
Archives of General Psychiatry, 54, 337–343.
Kraemer, H. C., Lowe, K. K., & Kupfer, D. J. (2005). To your health: How to understand what
research tells us about risk. New York: Oxford University Press
Kraemer, H. C., Stice, E., Kazdin, A., Offord, D., & Kupfer, D. (2001). How do risk factors
work together? Mediators, moderators, and independent, overlapping, and proxy risk
factors. American Journal of Psychiatry, 158, 848–856.
Lasser, K., Boyd, J. W., Woolhandler, S., Himmelstein, D. U., McCormick, D., & Bor, D. H.
(2000). Smoking and mental illness: A population-based prevalence study. Journal of the
American Medical Association, 284, 2606–2610.
26 M. J. Zvolensky et al.

Leen-Feldner, E. W., Zvolensky, M. J., van Lent, J., Vujanovic, A. A., Bleau, T., Bernstein, A.,
et al. (2007). Anxiety sensitivity moderates tobacco smoking in regard to panic attack
symptoms and bodily complaints: A concurrent test among adolescents. Journal of
Psychopathology and Behavioral Assessment, 29, 69–79.
Lindesay, J. (1991). Phobic disorders in the elderly. British Journal of Psychiatry, 159,
Lopes, F. L., Nascimento, I., Zin, W. A., Valenca, A. M., Mezzasalma, M. A., Figueira, I.,
et al. (2002). Smoking and psychiatric disorders: A comorbidity survey. Brazilian Journal
of Medical and Biological Research, 35, 961–967.
McCabe, R. E., Chudzik, S. M., Antony, M. M., Young, L, Swinson, R. P., & Zvolensky, M. J.
(2004). Smoking behaviors across anxiety disorders. Journal of Anxiety Disorders, 18, 7–18.
McLeish, A. C., Zvolensky, M. J., Bonn-Miller, M. O., & Bernstein, A. (2006). Perceived
health moderates the association between smoking rate and panic vulnerability among
daily smokers. Depression and Anxiety, 23, 257–265.
McLeish, A. C., Zvolensky, M. J., & Bucossi, M. M. (2007). Interaction between smoking rate
and anxiety sensitivity: Relation to anticipatory anxiety and panic-relevant avoidance
among daily smokers. Journal of Anxiety Disorders, 21, 849–859.
McNally, R. J. (2002). Anxiety sensitivity and panic disorder. Biological Psychiatry, 52,
Merikangas, K. R., Mehta, R. L., Molnar, B. E., Walters, E. E., Swendsen, J. D., Aguilar-
Gaziola, S., et al. (1998). Comorbidity of substance use disorders with mood and anxiety
disorders: Results of the International Consortium in Psychiatric Epidemiology. Addictive
Behaviors, 23, 893–907.
Morissette, S. B., Brown, T. A., Kamholz, B., & Gulliver, S. (2006). Differences between
smokers and nonsmokers with anxiety disorders. Journal of Anxiety Disorders, 20,
Morissette, S. B., Tull, M. T., Gulliver, S. B., Kamholz, B. W., & Zimering, R. T. (2007).
Anxiety, anxiety disorders, tobacco use, and nicotine: A critical review of interrelation-
ships. Psychological Bulletin, 133, 245–272.
Morrell, H. E. R., & Cohen, L. M. (2006). Cigarette smoking, anxiety, and depression. Journal
of Psychopathology and Behavioral Assessment, 28, 281–295.
National Cancer Institute and National Institute of Health (NCI/NIH). (2006). Smokeless
Tobacco and Cancer: Questions and Answers. Retrieved September 18th, 2006, from
Nelson, C. B., & Wittchen, H-U. (1998). Smoking and nicotine dependence. European
Addictive Research, 4, 42–49.
Norton, G. R. (1989). Panic attack questionnaire. In M. Hersen and A. Bellack (Eds.),
Dictionary of behavioral assessment techniques (pp. 332–334). New York: Pergamon Press.
Norton, G. R., Cox, B. J., & Malan, J. (1992). Nonclinical panickers: A critical review.
Clinical Psychology Review, 12, 121–139.
Novak, A., Burgess, E. S., Clark, M., Zvolensky, M. J., & Brown, R. A. (2003). Anxiety
sensitivity, self-reported motives for alcohol and nicotine use and level of consumption.
Journal of Anxiety Disorders, 17, 165–180.
Parrott, A. C. (1999). Does cigarette smoking cause stress? American Psychologist, 18,
Piper, M. E., Piasecki, T. M., Federman, E. B., Bolt, D. M., Smith, S. S., Fiore, M. C., et al.
(2004). A multiple motives approach to tobacco dependence: The Wisconsin Inventory of
Smoking Dependence Motives (WISDM-68). Journal of Consulting and Clinical Psychol-
ogy, 72, 139–154.
Pohl, R., Yeragani, V. K., Balon, R., Lycaki, H., & McBride, R. (1992). Smoking in patients
with panic disorder. Psychiatry Research, 43, 253–262.
Samet, J. M. (1992). The health benefits of smoking cessation. The Medical Clinics of North
America, 76, 399–414.
Panic and Tobacco 27

Stewart, S. H., Karp, J., Pihl, R. O., & Peterson, R. A. (1997). Anxiety sensitivity and
self-reported reasons for drug use. Journal of Substance Abuse, 9, 223–240.
Stice, E. (2002). Risk and maintenance factors for eating pathology: A meta-analytic review.
Psychological Bulletin, 128, 825–848.
Substance Abuse and Mental Health Services Administration. (2005). Science-based preven-
tion programs and principles. Rockville, MD: U.S. Department of Health and Human
Services, Center for Substance Abuse Prevention.
Tilley, S. (1987). Alcohol, other drugs and tobacco use and anxiolytic effectiveness: a com-
parison of anxious patients and psychiatric nurses. British Journal of Psychiatry 151,
Valentiner, D. P., Mounts, N. S., & Deacon, B. J. (2004). Panic attacks, depression and
anxiety symptoms, and substance use behaviors during late adolescence. Journal of
Anxiety Disorders, 18, 573–586.
Warren, R., & Zgourides, G. (1988). Panic attacks in high school students: Implications for
prevention and intervention. Phobia Practice and Research Journal, 1, 97–113.
Windle, M., & Windle, R. (1999). Adolescent tobacco, alcohol and drug use: Current findings.
Adolescent Medicine, 10, 153–163
Zvolensky, M. J., Baker, K. M., Leen-Feldner, E. W., Bonn-Miller, M. O., Feldner, M. T., &
Brown, R. A. (2004). Anxiety sensitivity: Association with intensity of retrospectively-
rated smoking-related withdrawal symptoms and motivation to quit. Cognitive Behaviour
Therapy, 33, 114–125.
Zvolensky, M. J., & Bernstein, A. (2005). Cigarette smoking and panic psychopathology.
Current Directions in Psychological Science, 14, 301–305.
Zvolensky, M. J., Bernstein, A., Marshall, E. C., & Feldner, M. T. (2006). Panic attacks, panic
disorder, and agoraphobia: Associations with substance use, abuse, and dependence.
Current Psychiatry Reports, 8(4), 279–285.
Zvolensky, M. J., Bernstein, A., Sachs-Ericsson, N., Schmidt, N. B., Buckner, J. D., &
Bonn-Miller, M. O. (2006). Lifetime associations between cannabis, use, abuse, and
dependence and panic attacks in a representative sample. Behaviour Research and Therapy,
44, 907–924.
Zvolensky, M. J., Bernstein, A., Yartz, A. R., McLeish, A., & Feldner, M. T. (in press).
Cognitive-behavioral treatment of comorbid panic psychopathology and tobacco use and
dependence. In S. H. Stewart & P. Conrad (Eds.), Comorbidity of anxiety and substance use
disorders. New York: Springer.
Zvolensky, M. J., Bonn-Miller, M. O., Bernstein, A., & Marshall, E. C. (2006).
Anxiety sensitivity and abstinence duration to smoking. Journal of Mental Health,
15, 659–670.
Zvolensky, M. J., Eifert, G. H., Feldner, M. T., & Leen-Feldner, E. W. (2003). Heart-focused
anxiety and chest pain in post-angiography medical patients. Journal of Behavioral
Medicine, 26, 197–209.
Zvolensky, M. J., Feldner, M. T., Eifert, G. H., & Brown, R. A. (2001). Affective style among
smokers: Understanding anxiety sensitivity, emotional reactivity, and distress tolerance
using biological challenge. Addictive Behaviors, 26, 901–915.
Zvolensky, M. J., Feldner, M. T., Leen-Feldner, E. W., Bonn-Miller, M. O., McLeish, A. C.,
& Gregor, K. (2004). Evaluating the role of anxiety sensitivity in smoking
outcome expectancies among regular smokers. Cognitive Therapy and Research, 28,
Zvolensky, M. J., Feldner, M.T., Leen-Feldner, E. W., & McLeish, A. C. (2005). Smoking and
panic attacks, panic disorder, and agoraphobia: A review of the empirical literature.
Clinical Psychology Review, 25, 761–789.
Zvolensky, M. J., Forsyth, J. P, Fuse, T., Feldner, M. T., & Leen-Feldner, E. W. (2002).
Smoking and non-clinical panic attacks: An initial empirical test of panic-relevant
cognitive processes. Cognitive Behaviour Therapy, 31, 170–182.
28 M. J. Zvolensky et al.

Zvolensky, M. J., Kotov, R., Antipova, A. V., & Schmidt, N. B. (2003). Cross cultural
evaluation of smokers risk for panic and anxiety pathology: A test in a Russian epidemio-
logical sample. Behaviour Research and Therapy, 41, 1199–1215.
Zvolensky, M. J., Kotov, R., Bonn-Miller, M. O., Schmidt, N. B., & Antipova, (in press).
Anxiety sensitivity as a moderator of association between smoking status and
panic-related processes in a representative sample of adults. Journal of Psychiatric
Zvolensky, M. J., Lejuez, C. W., Kahler, C. W., & Brown, R. A. (2004). Panic attack history
and smoking cessation: An initial examination. Addictive Behaviors, 29, 825–830.
Zvolensky, M. J., Lejuez, C. W., Kahler, C. W., & Brown, R. A. (2003). Integrating an
interoceptive exposure-based smoking cessation program into the cognitive-behavioral
treatment of panic disorder: Theoretical relevance and clinical demonstration. Cognitive
and Behavioral Practice, 10, 348–358.
Zvolensky, M. J., Sachs-Ericsson, N., Feldner, M. T., Schmidt, B., & Bowman, C. (2006).
Neuroticism moderates the effect of maximum smoking level on lifetime panic disorder:
A test using an epidemiologically defined national sample of smokers. Psychiatry
Research, 141, 321–332.
Zvolensky, M. J., Schmidt, N. B., Antony, M. M., McCabe, R. E., Forsyth, J. P., Feldner, M. T.,
et al. (2005). Evaluating the role of panic disorder in emotional sensitivity processes involved
with smoking. Journal of Anxiety Disorders, 19, 673–686.
Zvolensky, M. J., Schmidt, N. B., Bernstein, A., & Keough, M. E. (2006). Risk-factor research
and prevention programs for anxiety disorders: A translational framework. Behaviour
Research and Therapy, 44, 1219–1239.
Zvolensky, M. J., Schmidt, N. B., & McCreary, B. T. (2003). The impact of smoking on panic
disorder: An initial investigation of a pathoplastic relationship. Journal of Anxiety
Disorders, 17, 447–460.
Zvolensky, M. J., Schmidt, N. B., & Stewart, S. H. (2003). Panic disorder and smoking.
Clinical Psychology: Science and Practice, 10, 29–51.
Zvolensky, M. J., Vujanovic, A. A., Bonn-Miller, M. O., Bernstein, A., Yartz, A. R.,
Gregor, K. L., et al. (2007). Incremental validity of anxiety sensitivity in terms of motivation to
quit, reasons for quitting, and barriers to quitting among community-recruited smokers.
Nicotine and Tobacco Research, 9, 965–975.
Alcohol Use and Anxiety Disorders

Brigitte C. Sabourin and Sherry H. Stewart

The relationship between anxiety and alcohol use is a topic of great theoretical
and practical interest for both scientists interested in the nature and causes of
psychopathology and practitioners working with anxious and/or alcohol abus-
ing clients. Although it has been clearly established that anxiety disorders and
alcohol use disorders are highly ‘‘comorbid’’ or co-occurring conditions (e.g., see
Kushner, Abrams & Borchardt, 2000a for a review), the relationship between the
symptoms or behaviors involved in each disorder (e.g., feelings of anxiety and
levels of alcohol use) has not been as extensively reviewed. This chapter will
review recent empirical evidence linking anxiety and alcohol at both the beha-
vioral and disorder level to determine if similar conclusions can be derived
regarding their relationship from data at both of these levels of enquiry. We
will first briefly describe epidemiological studies linking anxiety disorders and
alcohol use disorders. Then we will examine some of the etiological theories of
the relationship between anxiety and alcohol use and their disorders, with a
review of the empirical evidence supporting each theory. Next, some specific
factors moderating and mediating the relationship between anxiety and alcohol
use will be explored, with an emphasis on individual differences and specific
processes involved in the relationship. A brief discussion of the differences
between factors affecting onset, maintenance, and relapse in the anxiety and
alcohol relationship will follow. The latest empirical evidence and thoughts
about treating both alcohol use and anxiety related problems will also be
reviewed. Finally, we conclude the chapter with some remarks about where the
field stands and directions that future research in this area might profitably take.

Brigitte C. Sabourin
Department of Psychology, Life Sciences Center, Dalhousie University, Halifax, Nova
Scotia, Canada, B3H 4J1, Tel: +902 494 3793, Fax: +902 494 6585

M. J. Zvolensky, J. A. Smits (eds.), Anxiety in Health Behaviors and Physical Illness. 29

Ó Springer 2007
30 B. C. Sabourin, S. H. Stewart

Symptomatic and Syndromal Levels of Enquiry

in the Anxiety – Alcohol Relationship

Anxiety and alcohol use can both be characterized at two different levels: sympto-
matic and syndromal. An association at the former level would entail a clear
relationship, for example, between feelings of anxiety and drinking behavior.
That is, one would expect that higher levels of anxiety would be related to higher
quantities and/or frequency of drinking behavior. In a classic paper, Persons
(1986) describes the advantages of studying psychological phenomena at the
symptomatic level rather than at the diagnostic category (or syndromal) level.
The symptom approach allows for study of important phenomena that may be
ignored by examining only the diagnostic category in question. For example, level
of alcohol consumption is not considered in the diagnosis of alcohol abuse or
dependence according to the fourth edition of the Diagnostic and Statistical
Manual of Mental Disorders (DSM-IV; American Psychiatric Association
[APA], 1994), although there have been some recommendations for incorporating
heavy drinking behaviors in the diagnostic definitions in future editions (Helzer,
Bucholz, & Bierut, 2006). Nonetheless, level of consumption may be an important
risk factor for alcohol problems (Dawson & Archer, 1993) and thus may be of
interest as a ‘‘symptom’’ when considering the anxiety – alcohol relation from the
symptom perspective. Also, the symptom approach recognizes the continuity of
clinical phenomena and behaviors with normal phenomena and behaviors. This is
a crucial point in the study of both drinking behavior and anxiety. For example,
some argue that drinking problems are best viewed as lying on a continuum
ranging from normal, non-problematic social drinking on one end to severe and
pathological alcohol dependence on the other extreme (Sobell et al., 1996).
In addition to the arguments presented above in favor of focusing on symptoms
rather than diagnostic categories, Chilcoat and Breslau’s (1998) discussion of
criteria for establishing causation between anxiety and alcohol abuse also demands
a symptom-focused approach rather than a syndromal- focused approach.
According to Chilcoat and Breslau, one of the criteria for causation includes a
‘‘gradient of effect’’, or dose response relationship between the two phenomena of
interest. That is, as the level of exposure to the causal agent increases, a resulting
increase in the level of the causal outcome should be expected. The gradient of
effect relationship can be studied as it pertains to the relation of any anxiety-related
symptom (e.g., number of panic attacks; severity of cognitive re-experiencing) with
any alcohol-related symptom (e.g., severity of negative consequences resulting
from alcohol use; usual number of alcohol beverages consumed per week).
The relationship between anxiety and alcohol can also be considered at the
syndromal level. At this level, a relationship between alcohol and anxiety would
be demonstrated if a diagnosis of one of the two disorders (i.e., anxiety disorder
or alcohol use disorder) was associated with an increased likelihood of a diag-
nosis of the other disorder. The DSM IV (APA, 1994) distinguishes between two
distinct types of alcohol use disorders: alcohol abuse and alcohol dependence.
Alcohol and Anxiety 31

Alcohol abuse is characterized by ‘‘recurrent and significant adverse conse-

quences related to the repeated use of alcohol’’ (p. 198), whereas alcohol depen-
dence must include ‘‘evidence of tolerance, withdrawal, or compulsive behavior
related to alcohol use’’ (p.214). Alcohol dependence is considered more severe
than alcohol abuse and always overrides the latter diagnosis. As was mentioned
earlier, consumption levels are not considered in the diagnosis of either alcohol
abuse or dependence. On the other hand, for anxiety disorders, both symptom
levels (e.g., repeated panic attacks in the case of panic disorder) and/or negative
consequences of the symptoms (e.g., distress about having another panic attack
in the case of panic disorder) are considered in making a diagnosis.

Epidemiological Findings on the Anxiety – Alcohol Relationship

The relationship between anxiety and alcohol can be described using the concept
of comorbidity, which can be defined as diagnosable, problematic alcohol use and
anxiety symptoms that are both present at some point in a person’s lifetime, but
not necessarily at the same time (Kushner et al., 2000a). Comorbidity rates can be
estimated using either clinical or community samples. Because individuals with
more than one disorder are more likely to seek treatment, clinical samples may in
fact inflate comorbidity estimates (Berkson, 1949). It is believed that community
surveys provide more accurate reflections of the anxiety disorder – alcohol use
disorder relationship. We will review the two most recent large-scale community
surveys, which are representative of the results of these types of surveys.
In this chapter, we will present odds ratios (ORs) to quantify the comorbid
relationship between anxiety disorders and alcohol use disorders. An OR
reflects the odds that individuals will display a second disorder if a first disorder
is present versus if it is not present. An OR of 1.0 reflects a lack of relationship,
with higher ORs reflecting more significant relationships between two disor-
ders. ORs of less than 1.0 reflect a decreased probability of having the second
disorder given the presence of the first disorder.
The National Comorbidity Survey (NCS: e.g., Kessler et al., 1996) reported
12-month ORs for individuals with alcohol dependence and alcohol abuse of
also having suffered from panic disorder (1.7 and 0.5 respectively), social
phobia (2.8 and 2.3), generalized anxiety disorder (4.6 and 0.4), posttraumatic
stress disorder (2.2 and 1.5), and specific phobias (2.2 and 1.2). The 12-month
ORs associated with alcohol dependence are significant for all anxiety disorders
with the exception of panic disorder (although the lifetime OR of alcohol
dependence is significant for panic disorder). On the other hand, although
having any anxiety disorder leads to a significant OR of developing alcohol
abuse, the only specific anxiety disorder with a significant OR is social phobia.
More recently, the National Epidemiologic Survey on Alcohol and Related
Conditions (NESARC; e.g., Grant et al., 2004) reported 12-month ORs for
individuals with alcohol dependence and abuse to also display any anxiety
32 B. C. Sabourin, S. H. Stewart

disorder (2.6 and 1.1, respectively), panic disorder with agoraphobia (3.6 and
1.4) and without agoraphobia (3.4 and 0.8), social phobia (2.5 and 0.9), a
specific phobia (2.2 and 1.1), and generalized anxiety disorder (3.1 and 0.9).1
The NCS (Kessler et al., 1997) also examined sex differences in comorbidity
between alcohol use disorders and anxiety disorders. Anxiety disordered women
and men do not differ significantly in their risk of developing alcohol dependence;
however, women with social phobia, simple phobias or post traumatic stress disorder
hive higher ORs of abusing alcohol than men with these anxiety disorders. Similarly,
the NESARC (Smith et al., 2006) reported ORs of alcohol use disorders and anxiety
disorders by race/ethnicity. Across all races/ethnic groups, there were significant ORs
of any anxiety disorder with alcohol dependence but not with alcohol abuse. How-
ever, the pattern of comorbidity across specific anxiety disorders reveals significant
racial/ethnic effects. For Whites and Blacks, the ORs for alcohol dependence and
anxiety disorders were significant across almost all anxiety disorders. The only
exception was that the OR for panic disorder with agoraphobia was significant for
alcohol abuse but not for alcohol dependence among Blacks. On the other hand, for
Native Americans and Hispanics, only a few of the anxiety disorders were signifi-
cantly associated with alcohol dependence and none with alcohol abuse.
Two general conclusions can be made from the data reported across these
large-scale community surveys. First, the relationship between anxiety disorders
and alcohol dependence appears to be much stronger than between anxiety
disorders and alcohol abuse, with the ORs for dependence much more likely to
be significant than those for abuse. In other words, having a comorbid anxiety
disorder increases the chances of displaying the more severe type of alcohol use
disorder (dependence) moreso than the less severe type (abuse), suggesting a
gradient of effect relationship. . Second, the relationship between alcohol use
disorders and anxiety disorders differs between sexes and racial/ethnic groups.
Although alcohol use is generally more common among men than women (Paa-
vola, Vartiainen, & Haukkala, 2004), alcohol abuse and anxiety disorders are
more closely related for women than for men. Furthermore, it appears that for
Whites and Blacks, anxiety and alcohol dependence are more closely associated
than for Native Americans and Hispanics. The following sections cover etiologi-
cal models of the relationship between alcohol and anxiety, their maintenance,
and their relapse, in an attempt to explicate the high level of comorbidity between
anxiety disorders and alcohol use disorders observed in the epidemiologic surveys.

Etiological Models of the Relationship

In order to gain a better understanding of the relationship between anxiety and

alcohol use, it is important to explore the mechanisms that may affect the
etiology of the co-occurrence of symptoms of both of their disorders. There

The NESARC did not specify significance of reported 12-month ORs.
Alcohol and Anxiety 33

are three main hypotheses that have been put forward, with some evidence
supporting each hypothesis. First, there is some evidence that the associations
between anxiety and alcohol arise from common underlying variables, such as
common genetic or environmental factors, that cause both anxiety symptoms
and problematic alcohol use. Second, some believe that certain aspects of
problematic alcohol use, such as repeated experiences with alcohol withdrawal,
cause anxiety symptoms and ultimately an anxiety disorder. Finally, others
argue that anxiety symptoms cause alcohol misuse, culminating in an alcohol
use disorder. Evidence examining these three hypotheses is presented below.

Common Underlying Variables

There is some evidence, provided by two main lines of research, to support the
hypothesis that certain common, underlying factors are causing both anxiety
symptoms and problematic alcohol use (Kushner et al., 2000a). First, family
and twin studies have provided some evidence of possible common genetic
contributions to the correlation between anxiety symptoms and alcohol con-
sumption (e.g., Tambs, Harris, & Magnus, 1997). Family and twin studies have
also examined the heritability of common underlying personality traits asso-
ciated with both anxiety and alcohol use disorders. For example, several cross-
sectional and longitudinal studies have linked the highly heritable personality
trait of neuroticism (Jang, Livesley, & Vernon, 1996) with anxiety and its
disorders (e.g., Jorm et al., 2000; Weinstock & Whisman, 2006). Neuroticism
has also been linked to alcohol use disorders (Cox, 1987). Another personality
dimension that is closely related to neuroticism, negative emotionality, has been
associated with alcohol use disorders (Swendsen, Conway, Rounsaville, &
Merikangas, 2002). In a study by Swendsen et al. (2002) examining heritability
of negative emotionality, non-alcoholic individuals with alcoholic relatives did
not differ significantly on scores of negative emotionality than those without
alcoholic relatives. If negative emotionality were a heritable risk factor, non-
alcoholic individuals with alcoholic relatives would score higher on this trait
than those without alcoholic relatives. These findings suggest that negative
emotionality may indeed be an individual risk factor rather than a heritable
risk factor for alcohol use disorders. The inconsistent results between these two
related personality traits (i.e. both associated with negative emotional states)
demonstrates that common heritability for some of the personality traits rele-
vant to the anxiety – alcohol relationship is still in need of further investigation.
Conversely, another personality risk factor for both anxiety disorders and
alcohol use disorders, anxiety sensitivity (i.e. fear of anxiety; Stewart & Kushner,
2001), does have a strong heritable component that accounts for nearly half of
the variance in scores on anxiety sensitivity measures (Stein, Lang, & Livesley,
1999). Therefore, it appears that some, but not necessarily all, underlying
personality risk factors associated with both alcohol use disorders and anxiety
disorders have a shared heritable component.
34 B. C. Sabourin, S. H. Stewart

Second, results from some prospective studies suggest a possible common ‘‘third
variable’’ contribution to the alcohol – anxiety relationship. For example, Zimmer-
man et al. (2003) found that remitted panic disorder and social phobia were as
important as current panic/social phobia diagnoses in predicting future alcohol
outcomes. That is, even individuals who were not currently experiencing sufficient
symptoms to receive any anxiety disorder diagnosis were at higher risk of developing
alcohol problems if they had ever been diagnosed with either panic or social phobia in
the past. These findings can be interpreted to suggest that a third underlying factor
(such as a common personality vulnerability or genetic predisposition) was driving
both the alcohol problem and the past or current anxiety disorder.
A 21-year longitudinal study (Goodwin, Fergusson, & Horwood, 2004)
found that once other factors were controlled (i.e., prior substance dependence,
concurrent major depression, and affiliations with deviant peers), the ability of
anxiety disorders to predict the development of alcohol dependence was no
longer significant. The study points to a number of possible third variables
including prior substance dependence which could contribute both to the
development of anxiety disorder (see Norton, Norton, Cox, & Belik, in press)
and of alcohol dependence (e.g., alcohol is consumed in larger quantities when
combined with other substances; Barrett, Darredeau, & Pihl, 2006). Unfortu-
nately the study did not test which of these factors was most important in
explaining the link between anxiety and alcohol dependence.
Some factors that have emerged as possible contributors to the increased
vulnerability of developing comorbid anxiety and alcohol use problems include
either common genetic pre-dispositions (e.g., anxiety sensitivity), biological envir-
onment risk factors (e.g., fetal alcohol syndrome), or non-biological environmental
risk factors (e.g., disruptive familial environment; Merikangas, Stevens, & Fenton,
1996). Unfortunately, no research has yet confirmed one or more of these candidate
factors. More research needs to be conducted exploring these additional underlying
mechanisms before one can make conclusions about their influence.

Alcohol Use Causes Anxiety

The second hypothesis dealing with the relationship between anxiety and
alcohol use posits that prolonged drinking is actually a causal factor in anxiety
symptoms and disorders. This alcohol-induced anxiety can occur through
either psychosocial or physiological mechanisms. Psychosocially, it is hypothe-
sized that alcohol may interfere with normal adaptation to stressful stimuli
or that negative consequences produced by problematic drinking (e.g., loss
of job or relational problems) can lead to anxiety symptoms and increased
vulnerability of developing anxiety disorders (Kushner et al., 2000a).
Physiologically, alcohol withdrawal can often produce anxiety symptoms
such as shakiness (see Kushner et al., 2000a) or increased startle, a common
symptom of PTSD (Stewart et al., 1998). In addition, neural adaptation occurs
Alcohol and Anxiety 35

with frequent and excessive alcohol use over time such that repeated alcohol
withdrawals actually sensitize this withdrawal-induced anxiety (Breese, Over-
street, & Knapp, 2005). This has often been referred to as the ‘‘kindling-stress
hypothesis’’; that is, repeated withdrawals from chronic heavy drinking are
thought to worsen, or ‘‘kindle’’ withdrawal-induced anxiety.
A number of studies have also demonstrated increased norepinephrine
activity as well as hyperexcitability of the central nervous system, especially of
limbic structures, during alcohol withdrawal (Kushner et al., 2000a; Marshall,
1997). These are the same neural systems that have been implicated in panic
attacks and panic disorder, providing a possible physiological explanation for
the link between panic disorder and alcohol use disorders (Marshall, 1997).
A final area of research supporting the hypothesis that alcohol problems
cause anxiety involves prospective studies. One such study by Kushner, Sher,
and Erickson (1999), for example, found that a diagnosis of alcohol dependence
at baseline quadrupled the risk of developing an anxiety disorder three to six
years later. Prospective studies have also examined the relationship between
PTSD and alcohol abuse to ascertain whether heavy alcohol use can be a risk
factor for developing PTSD. It has been hypothesized that physiological and
neurochemical changes due to prolonged heavy alcohol use and/or past reliance
on alcohol to deal with life stressors at the expense of developing other coping
mechanisms may increase an individual’s susceptibility of developing PTSD
after a traumatic experience (Brown & Wolfe, 1994; Stewart et al., 1998).
A prospective study by Acierno, Resnick, Kilpatrick, Saunders, and Best
(1999) found that a history of alcohol abuse increased the risk of developing
PTSD in rape victims almost three-fold (OR = 2.65) when compared to the
absence of this factor.

Anxiety Causes Alcohol Use

It has been hypothesized that anxiety symptoms and anxiety disorders promote
alcohol use, as individuals drink to self-medicate their anxiety. The ‘‘self-med-
ication hypothesis’’ (and the related tension reduction hypothesis) as applied to
the understanding of the relationship between anxiety and alcohol posits that
the pharmacological and/or psychological effects of alcohol lead to decreases in
aversive anxiety symptoms, thereby motivating anxious individuals to increase
their quantity and/or frequency of alcohol use via the process of negative
reinforcement (Kushner et al., 2000a). Although the self-medication and ten-
sion reduction hypotheses clearly do not account for all drinking behavior
(Greeley & Oei, 1999), there has been a good deal of empirical evidence to
support these hypotheses as they apply to the understanding of comorbid
anxiety and alcohol use disorders (Kushner et al., 2000a).
Anxiety disordered individuals do in fact self-report using alcohol to manage
their anxiety (Kushner, Abrams, Thuras, & Hanson, 2000b; Thomas, Randall,
36 B. C. Sabourin, S. H. Stewart

& Carrigan, 2003; see also Kushner, et al., 2000a for a review). In addition,
Thomas and colleagues (2003) found that socially anxious individuals not only
reported that they drank to feel more comfortable in social situations, but that
they would actually avoid social situations if alcohol were unavailable.
As introduced earlier in this chapter, one possible criterion for establishing
causation (Chilcoat & Breslau, 1998) is a dose response, or gradient of effect
relationship: if anxiety causes alcohol use, one would expect that higher levels of
anxiety would be associated with higher levels of alcohol use. Studies have
found positive correlations between severity of PTSD arousal symptoms and
severity of alcohol use disorder symptoms (McFall, Mackay, & Donovan, 1992;
Stewart et al., 1998). Because correlation does not determine causation, one
must rely on laboratory-based studies, such as a study by Abrams, Kushner,
Medina, and Voight (2002), for evidence that induction of anxiety symptoms
causes heavier drinking. The study found that participants with social phobia
consumed more alcohol following an anxiety provoking activity (speaking in
front of a group) than a control activity (reading a book), presumably in an
effort to dampen the anxious feelings caused by the anxiety provoking activity.
Prospective research on non-clinical populations also supports a dose-
response relationship between anxiety and alcohol use. In a diary-based study
by Swendsen and colleagues (2000), moderate drinkers documented their daily
drinking and mood states for a one-month period. The study revealed that only
anxious feelings and not sadness or other negative affective states preceded and
predicted increased alcohol consumption. As can be observed above, findings
from correlational, laboratory-based experimental, and diary-based prospec-
tive research conducted with both clinical and non-clinical populations
converge to provide some evidence for a relationship between anxiety sym-
ptoms and alcohol use where anxiety precedes and contributes to increased
alcohol use.
Social anxiety appears to have a more complicated relationship with alcohol
use than do other types of anxiety, however. Specifically, some studies examin-
ing the relationship between social anxiety and alcohol consumption show a
positive relationship, whereas other studies show either no linear relationship or
even a negative relationship (Ham & Hope, 2005; Stewart, Morris, Mellings, &
Komar, 2006; Tran, Haaga, & Chambless, 1997). The negative relationship
between social anxiety and alcohol consumption may exist because socially
anxious individuals actually avoid the types of social situations that involve
drinking because of their social anxiety, thus leading to lower levels of alcohol
consumption (Stewart et al., 2006). Nonetheless, social anxiety has been found
to predict alcohol dependence, as well as problems caused by alcohol (Gilles,
Turk, & Fresco, 2006; Stewart et al., 2006). Thus, social anxiety does appear
related to alcohol-related consequences, even if it does not always predict
increased alcohol use.
Another criterion discussed by Chilcoat and Breslau (1998) as necessary for
causation is temporality. If anxiety causes increased or problematic alcohol use,
then anxiety symptoms should predate alcohol-related problems, and anxiety
Alcohol and Anxiety 37

disorder diagnoses should precede alcohol disorder diagnoses. In the cases

of social phobia, panic disorder, and PTSD diagnoses, the anxiety disorder
usually preceded the onset of the alcohol use disorder in comorbid individuals
(Cox, Norton, Swinson, & Endler, 1990; Kushner, Sher, & Beitman, 1990;
Stewart & Conrod, 2002).Furthermore, a cross-national investigation by
Merikangas et al. (1998) confirmed that anxiety disorders preceded alcohol
use disorders for the majority of participants. Assessing temporality or relative
order of onset of symptoms of alcohol use disorders and anxiety disorders
among comorbid cases can also help elucidate the anxiety – alcohol
causal relationship. For example, a recent study by Bernstein, Zvolensky,
Sachs-Ericsson, Schmidt, and Bonn-Miller (2006) found that the symptom of
panic attacks predated the onset of heavy drinking behaviors for the vast
majority of participants with both panic attacks and heavy drinking in a
community sample. Although many anxiety disorders and their symptoms
predate alcohol use disorders and heavy drinking, this relationship is normally
inverse for individuals with generalized anxiety disorder (GAD). For the
majority of individuals with GAD, alcohol use disorder predated the GAD
(Kushner et al., 1990), suggesting that, for those cases, anxiety could not have
been the cause of the alcohol use disorder. Thus, as can be seen from the
findings above, for most comorbid individuals (excluding a majority of GAD
individuals), the order of onset supports the possibility that anxiety may play a
causal role in the development of alcohol use disorders. And conversely, for
those with GAD, the alcohol use may play a causal role in the development of
the anxiety disorder. However, Chilcoat and Breslau (1998) clarify that the
temporality criterion is necessary but not sufficient for determining a causal
association between disorders.
Finally, the prospective study by Kushner and colleagues (1999) mentioned
earlier in the chapter that found an increased risk of future anxiety disorders for
individuals with alcohol dependence also found the converse. That is, having a
diagnosis of anxiety at baseline increased the risk three- to five-fold for a new
onset of alcohol dependence three to six years later (see also Goodwin et al., 2004).
The study demonstrates that the causal relationship between anxiety and alcohol
is potentially bi-directional in nature. Overall, the evidence presented above does
seem to support the fact that, at least in some instances, anxiety symptoms and
anxiety disorders do precede and possibly promote problematic alcohol use.

Moderating and Mediating Variables in the Anxiety – Alcohol


More recent work has focused on finding specific variables that either moderate
or mediate the causal relationship between anxiety and alcohol use. A mod-
erator variable is a qualitative (e.g., sex) or quantitative (e.g., anxiety level)
variable that affects the direction and/or strength of the relation between two
38 B. C. Sabourin, S. H. Stewart

other variables (Baron & Kenny, 1986). A mediator variable, on the other hand,
explains how or why the relationship between a predictor and given criterion
(e.g., between anxiety and alcohol use) exists. That is, the mediator actually
accounts for the relationship between the two variables (Baron & Kenny).
Alcohol expectancies. A potential moderator variable between anxiety and
problematic alcohol use includes certain ‘‘alcohol outcome expectancies’’ (i.e.,
beliefs about the consequences of drinking alcohol). For anxious individuals
who self-medicate to avoid anxiety, an important aspect of the self-medication
hypothesis involves the notion that self-medicators anticipate anxiety, and that
they expect that alcohol will actually decrease their feelings of anxiety (e.g.,
Kushner, Sher, Wood, & Wood, 1994; Tran et al., 1997). Studies have shown
that tension reduction expectancies predict drinking frequency and quantity in
non-alcoholic drinkers with panic disorder (Kushner et al., 2000b) and comor-
bid problem drinking in women with PTSD (Ullman, Filipas, & Townswend,
2005). These results support the role of tension reduction alcohol expectancies
as a moderator: increased or problematic drinking occurs among anxiety dis-
order patients only when tension reduction alcohol expectancies are present.
Another methodology that has been employed to investigate the role of
alcohol outcome expectancies in the anxiety – alcohol relationship is the experi-
mental manipulation of expectancies via the placebo-controlled design. If
expecting alcohol were to induce a cognitive or placebo-induced anxiety redu-
cing effect among anxious individuals, such an effect would provide additional
evidence for the contribution of alcohol expectancies in explaining the anxiety –
alcohol relationship. The empirical evidence provided thus far has found mixed
results for this placebo anxiolytic effect. Some studies have found that the belief
that one was consuming alcohol, even when one was actually consuming a
placebo, was enough to lower feelings of anxiety among anxiety-disordered
patients (Abrams, Kushner, Lisdahl, Medina, & Voight, 2001; Lehman, Brown,
Palfai, & Barlow 2002). On the other hand, research by MacDonald, Stewart,
Hutson, Rhyno, and Loughlin (2001) conducted with participants high in
anxiety sensitivity did not support a cognitively-mediated tension reduction
effect of alcohol. The researchers actually found a ‘‘reverse placebo’’ effect,
where high AS participants in a placebo condition, who had expectations of
alcohol-induced tension reduction, but did not benefit from alcohol’s physio-
logical tension-reduction properties, appeared to have even higher levels of
anxiety than participants in a control condition where they neither received
nor expected alcohol. Regardless of the direction of the placebo effect, all of
these findings do suggest a role for cognitive expectancy variables in accounting
for the effects of alcohol among anxious individuals.
A number of studies have examined specific aspects of alcohol expectancies
in individuals with social anxiety. For people high in social anxiety, expecting
that alcohol would decrease social anxiety or increase social assertiveness was
associated with both higher self-reported drinking quantities (Tran et al., 1997)
and higher alcohol consumption in a laboratory setting (Kidorf & Lang, 1999).
More general tension reduction expectancies, on the other hand, had no effect
Alcohol and Anxiety 39

on alcohol consumption. In addition, socially related alcohol expectancies have

been associated with higher levels of alcohol dependence in socially anxious
individuals (Ham, Carrigan, Moak, & Randall, 2005).
These studies demonstrate that different expectancies affect alcohol con-
sumption depending on the particular type of anxiety-related psychopathology
involved. For individuals with panic disorder or PTSD, the research until now
appears to indicate that general tension reduction expectancies provide suffi-
cient motivation for increasing alcohol consumption. However, for individuals
with social anxiety, it appears that specific social-related alcohol expectancies,
but not general tension reduction expectancies, tend to motivate these indivi-
duals to drink, resulting in greater risk of developing alcohol dependence. As
has been done with social anxiety, increasingly directed research in other types
of anxiety-related psychopathology is needed to determine more precisely the
kinds of expectancies that help explain each group’s increased risk for alcohol
problems in order to target these more specifically in treatment.
Self-efficacy. Self-efficacy has been defined by Bandura (1977) as the con-
viction that one can successfully execute a behavior required to produce a
certain outcome. Research conducted with socially anxious individuals suggests
that self-efficacy may act as a link between alcohol expectancies, social anxiety
and heavy drinking behaviors. Positive socially related alcohol expectancies
and low self-efficacy to avoid heavy drinking interact with one another in
increasing problematic drinking among socially anxious individuals (Burke &
Stephens, 1997; Gilles, Turk, & Fresco, 2006). Thus, increasing individuals’
self-efficacy in refusing alcohol or avoiding heavy drinking could be incorpo-
rated when developing treatment programs for comorbid individuals, at least in
the case of social phobia comorbidity. More research is needed to determine the
relevance of the self-efficacy construct to the comorbidity of alcohol use dis-
orders with anxiety disorders other than social phobia.
Drinking motives. Problematic drinking can arise because of maladaptive
drinking motives (reasons for drinking) even in the absence of particularly
elevated levels of alcohol consumption (Stewart et al., 2006). For some indivi-
duals who experience fear or anxiety, avoidance behaviors, such as drinking to
self-medicate, become negative reinforcement strategies used to attenuate or
cope with anxious states. Because such avoidance strategies are negatively
reinforcing through their effects in alleviating anxiety, the drinking behavior
continues over time and the individual comes to rely on drinking as a primary
coping strategy. For example, in many cases, social anxiety is not associated
with higher drinking levels (Tran et al., 1997), but it is with problematic reasons
for drinking, such as drinking to cope with negative emotional states, as well
as with greater risks of developing alcohol problems (Thomas, Randall, &
Carrigan, 2003; Stewart et al., 2006). A further problem with coping-related
drinking is that, as an avoidance behavior, it may serve to maintain anxiety by
preventing habituation of the anxiety response.
Research by Cooper (1994; see also Cooper, Russell, Skinner, & Windle,
1992) exploring drinking motives and their relation to problematic drinking
40 B. C. Sabourin, S. H. Stewart

uncovered four factors explaining motivation for drinking. Two of these

motives involve positive reinforcement from drinking: social (i.e. to obtain
social rewards) and enhancement (i.e. to enhance positive mood or well
being), whereas the other two motives involve negative reinforcement from
drinking: coping (i.e. to cope with negative emotions), and conformity (i.e. to
avoid negative social consequences such as social rejection). Only the negative
reinforcement motives of coping and conformity predict alcohol problems after
controlling for quantity and frequency of alcohol use.
A study by Stewart and colleagues (2006) also found that for undergraduate
students, problem-drinking symptoms were positively associated with the nega-
tive reinforcement motives of coping and conformity drinking. The study also
found that coping and conformity drinking motives mediated the relationship
between social anxiety (specifically, fear of negative evaluation) and drinking
problems. That is, individuals with social anxiety experienced drinking pro-
blems because they drank either to cope with negative emotions (i.e. coping
drinking motives), or to avoid negative social consequences (i.e. conformity
drinking motives). In addition, the study found either no association or even a
negative direction association between social anxiety measures and drinking
quantity and frequency. Together, research by Cooper and her colleagues and
Stewart and her colleagues support the notion that drinking to cope with
anxiety or to conform with peer pressure confers additional risk for problem
drinking over and above the risks associated with level of alcohol consumption.
Additional support for coping and conformity drinking motives as impor-
tant factors in the anxiety – alcohol relationship has been provided by studies
conducted with non-clinically anxious participants. For example, Deacon and
Valentiner (2000) found a significant association between scores on the Beck
Anxiety Inventory (BAI, a widely-used measure of anxiety symptoms) and
coping motivated drinking but not social or enhancement motivated drinking.
In addition, Thomas et al. (2003) found that socially anxious individuals more
often reported drinking before and during social situations in an effort to feel
more comfortable (i.e. to cope with their social anxiety) than non-socially
anxious individuals.
Anxiety sensitivity has been associated with greater drinking levels and
argued to be a potential risk factor for alcohol use disorders (Stewart &
Kushner, 2001), as we describe in greater detail in the next section. Results
from a study by Stewart et al. (2001) supported coping and conformity motives
as mediators in the relationship between AS and higher drinking levels. That is,
high AS individuals’ greater drinking behavior was at least partially explained
by high coping and conformity motive scores.
Anxiety sensitivity. Anxiety sensitivity (AS) is an important individual dif-
ference that may mediate the anxiety – alcohol relationship. It is possible that
anxiety disordered patients are at increased risk of alcohol problems because
of their higher levels of AS (i.e., higher levels of fear of anxiety), which, in turn,
promote greater motivation and behaviors (e.g., alcohol use) to escape or avoid
the feared symptoms (Stewart & Kushner, 2001). Thus, consistent with the
Alcohol and Anxiety 41

self-medication hypothesis, it is possible that individuals high in AS use alcohol

for its anxiolytic or arousal dampening effects. In fact, high AS individuals have
been found to prefer alcohol and other ‘‘depressants’’ over ‘‘stimulant’’ type
drugs (DeHaas, Calamari, & Bair, 2002; Norton, Rockman, & Ediger, 1997).
Additionally, elevated levels of AS have been associated with increased drink-
ing behavior, including increased typical weekly drinking frequency, and yearly
excessive drinking frequency (Stewart, Peterson, & Pihl, 1995; Cox & Klinger,
1988; Stewart, Zvolensky, & Eifert, 2001). Moreover, Stewart, Conrod,
Samoluk, Pihl, and Dongier (2000) found evidence for the mediating role of
AS in explaining the association between PTSD symptoms and negative rein-
forcement drinking.
Stewart et al. (2001) also examined the relationship between lower-order
components of AS and drinking behavior. AS consists of three lower-order
factors: AS physical concerns (e.g., worrying that a rapidly beating heart is a
sign of having a heart attack), AS psychological concerns (e.g., worrying that
not being able to keep one’s mind on a task is a sign of going crazy), and AS
social concerns (e.g., worrying about appearing nervous in front of others; see
Zinbarg, Mohlman, & Hong, 1999). After controlling for the other two factors,
AS social concerns emerged as the only significant predictor for weekly drink-
ing frequency and yearly excessive drinking frequency. Thus, it is possible that
AS also plays an important role in the relationship between social anxiety and
drinking problems given the elevation of AS social concerns among those with
social phobia (Zinbarg et al., 1999).
Summary. Although these mediating and moderating variables are typically
studied in isolation, as can be seen above, alcohol expectancies, self-efficacy,
drinking motives, and anxiety sensitivity may interact with each other or inter-
vene with one another in explaining problematic alcohol use in anxious indivi-
duals. For example, people with certain anxiety disorders may be more likely to
drink to cope with their anxiety sensations (coping motives) because of their
fear of these sensations (anxiety sensitivity), thereby increasing their risk for
drinking problems. Additional research exploring the interplay between the
different variables affecting the anxiety-alcohol relationship would provide
important steps toward creating more effective treatments for comorbid

Anxiety, Alcohol Use, and Other Health Behaviors

Alcohol consumption has been shown to be highly related to risky health

behaviors, such as smoking and illicit drug use (Paavola, Vartiainen, &
Haukkala, 2004; Tolstrup et al., 2005). In a longitudinal study (Paavola et al.,
2004), earlier alcohol use was associated with later smoking, and smoking in
adolescence predicted alcohol use in adulthood. Despite this close association
between alcohol use and smoking, combining the two behaviors does not
42 B. C. Sabourin, S. H. Stewart

increase the rate of anxiety disorders above the rate associated with alcohol use
only (Kandel, Huang, & Davies, 2001).
A diagnosis of an anxiety disorder combined with a drug use disorder
constitutes a significant risk factor for developing alcohol dependence (lifetime
OR =5.81; Kessler et al., 1997). This increased risk appears to be even higher
than the risk associated with being diagnosed with an anxiety disorder alone
(i.e. without a drug use disorder; lifetime OR = 1.85). Unfortunately, like many
epidemiological surveys, the National Comorbidity Survey does not break drug
use disorders down by drug type/class, which could further elucidate the drug –
anxiety – alcohol relationship. One possible explanation for this elevated risk
for alcohol dependence among those with comorbid anxiety – drug use disor-
ders is that problematic drug use, through the drugs’ potentially anxiogenic
effects, exacerbates the need to self-medicate with alcohol, resulting in increased
risk for alcohol dependence relative to those with non-comorbid anxiety

Maintenance of Comorbid Anxiety and Alcohol Problems

Empirical findings supporting all three causal hypotheses suggest the possibility
of multiple causal pathways involved in the etiology of comorbid alcohol use
disorders and anxiety disorders. The specific causal pathway involved may vary
across people or across anxiety sub-types. Regardless of the etiology of the
comorbid anxiety symptoms and problematic alcohol use, there have been
countless studies confirming that, once comorbid, anxiety and alcohol use do,
in fact, exert important influences on each other (see Kushner et al., 2000a).
Furthermore, processes involved in the initiation of the comorbidity may differ
from those involved in the maintenance of problematic alcohol use and anxiety.
A feed-forward model has been proposed (Kushner et al., 2000a) in which once
both alcohol use and anxiety are present, each promotes the maintenance or
exacerbation of the other. For example, anxious individuals may resort to
alcohol to decrease feelings of anxiety, which might be an effective strategy in
the short term, providing reinforcement for this pattern. However, alcohol, and
especially withdrawal from alcohol, increases anxiety-like symptoms in the
longer run via physiological mechanisms such as kindling. Alcohol may also
worsen anxiety levels because of the negative familial, social or occupational
consequences of heavy drinking. Individuals will then increase their drinking
behavior in an attempt to alleviate these worsening feelings of anxiety because
drinking has become a learned strategy for dealing with these symptoms,
especially if there is a failure to recognize that the alcohol may actually be
promoting the anxiety in the medium to long term.
Finally, in individuals with PTSD, alcohol that is used to cope with anxiety
may prevent normal ‘‘habituation’’ of the anxiety symptoms following trauma
exposure. On the other hand, for individuals who are not drinking, anxious
Alcohol and Anxiety 43

feelings caused by PTSD may naturally remit with time. Thus, when individuals
drink in an attempt to numb or avoid these feelings, they may be preventing this
natural recovery from taking place, leading to maintenance of the anxiety
symptoms in the long run (Stewart et al., 1998).

Treatment Outcome and Relapse

When individuals who suffer from both anxiety disorders and alcohol use dis-
orders enter treatment for either disorder, their treatment outcome is often nega-
tively affected by their comorbidity. Alcohol use disorders have been found to
predict poorer anxiety disorder treatment outcomes for patients with both PTSD
(Forbes, Creamer, Hawthorne, Allen, & McHugh, 2003), panic disorder with
agoraphobia, social phobia, and generalized anxiety disorder (Bruce et al., 2005).
Comorbid anxiety problems also increase the likelihood of relapse in
treated or abstinent alcoholics (e.g., Driessen et al., 2001; Kushner et al.,
2005; Willinger et al., 2002). For example, if a comorbid PTSD – alcoholic
individual does not know how to cope with flashbacks and nightmares of the
traumatic event in ways other than through drinking, then continued re-
experiencing symptoms can serve as a major risk factor for return to problem
alcohol use following initially effective alcohol abuse treatment. Not all
studies have shown this relationship, however. In one study (LaBounty,
Hatsukami, Morgan, & Nelson, 1992), alcoholics with comorbid panic dis-
order did not differ in their rates of relapse to drinking problems from non
comorbid alcoholics. However, in a recent review, Bradizza et al. (2006) noted
some methodological issues with this paper, including the absence of a valid
and reliable diagnostic measure and the failure to define relapse, limiting the
conclusions that could be drawn from the study. Moreover, despite the
similar relapse rates, the study did find that more comorbid alcoholic and
panic disordered patients reported relapsing to cope with negative emotions
than non comorbid alcoholics. Thus, these observed differences in the relapse
process might be useful for improving treatments for this group.
Another study compared relapse rates for alcoholic individuals with comor-
bid social phobia, or panic disorder with agoraphobia, or agoraphobia without
a history of panic attacks to relapse rates for alcoholics without any comorbid
anxiety disorder (individuals with other anxiety disorders were excluded) fol-
lowing treatment for alcoholism (Marquenie et al., 2006). The results suggested
that the comorbid anxiety disorders did not have a significant impact on either
relapse rates or days to relapse. Nonetheless, some methodological problems
may account for this study’s failure to support higher alcoholism relapse among
treated alcoholic patients with comorbid anxiety disorders. First, the study used
a retrospective design. Participants were contacted for the study an average of
20.3 months (and up to 42 months) after baseline assessment, even though the
majority of participants who relapsed did so within the first few months post
44 B. C. Sabourin, S. H. Stewart

baseline, raising issues concerning potential inaccuracy in the self-reports due to

retrospective memory bias. Also, compared to the non comorbid group, the
comorbid group had less chronic alcoholism and a shorter period between the
initial and follow-up assessments, which could have led to underestimates of the
rate of relapse in the comorbid group. Although the authors stated the failure to
observe group differences in alcoholism relapse persisted when these group
differences were statistically controlled in the analyses, statistical control of
possible confounds is never definitive, (and in fact may even produce biased
parameter estimates) especially when the confounding variables are correlated
with the predictor of interest, again limiting any firm conclusions that can be
drawn from the Marquenie et al. study.
Several studies, on the other hand, have found higher relapse rates in
comorbid patients. Driessen et al. (2001) found that treated alcoholic patients
with comorbid anxiety had 29% higher alcoholism relapse rates than did
alcoholics without comorbid anxiety. Furthermore, a study examining the
relationship between trait anxiety and relapse in a sample of abstinent alcohol
dependent patients found that higher trait anxiety was significantly predictive
of relapse to uncontrolled drinking (Willinger et al., 2002). Finally, in the best-
controlled study to date, Kushner et al. (2005) found that alcoholic patients
with an anxiety disorder (especially those with comorbid panic disorder or
social phobia) were significantly more likely to relapse to problem drinking
(using multiple criteria for drinking relapse) than alcoholic patients without an
anxiety disorder. These finding are particularly convincing given the methodo-
logical soundness of the study. For example, the study used a prospective design
where all participants were contacted between 90 and 120 days after the begin-
ning of treatment, thus increasing reporting accuracy by reducing reliance on
long-term retrospective memory. Also, all participants were given the same
standardized treatment and assessed at a consistent time following treatment.
Taken together, it appears that comorbid alcohol and anxiety may have a
negative effect on treatment outcome and relapse but with some mixed results.
The methodologically superior studies do seem to suggest such a negative effect. In
addition, although a couple of studies failed to show differences in relapse rates,
one of these negative studies did provide some evidence for the self-medication
hypothesis through highlighting the importance of coping with anxiety in explain-
ing relapse to alcohol use among patients with comorbid alcohol and anxiety
disorders. These differences could be useful in developing specific relapse preven-
tion types of treatments (see Marlatt & Donovan, 2005) for comorbid patients.

Treatments of Comorbid Anxiety and Alcohol Use Disorders

Individuals who suffer from both anxiety problems and alcohol use problems
present a special and challenging population with regards to treatment. As was
shown above, this population often suffers worse anxiety and alcohol treatment
Alcohol and Anxiety 45

outcomes than populations experiencing symptoms in only one of the two

domains. Although the study of specific treatments for anxiety – alcohol
comorbidity is still in its infancy, this area is growing and there are now several
promising approaches to treating comorbid anxiety and alcohol use disorders
(Stewart & Conrod, in press).
There have been mixed findings regarding the effects of pharmacological
treatment for anxiety on drinking outcomes with some studies finding improve-
ments in alcohol outcomes and others finding more equivocal results (see review
by Kushner et al., 2000a). One study (Randall et al., 2001) found that treating
individuals with comorbid social phobia and alcohol use disorder with parox-
etine (a selective serotonin reuptake inhibitor) did improve anxiety, but did not
result in significant decreases in drinking frequency and quantity. Relative to
placebo, paroxetine treatment did, on the other hand, lead to improvements on
the Clinical Global Index for alcohol. Other studies have found that successful
treatment of anxiety with buspirone was also associated with a reduction in
alcohol use (Tollefson, Montague-Clouse, & Tollefson, 1992; Kranzler et al.,
1994). These mixed findings are partially consistent with the self-medication
hypothesis, although the paroxetine study does suggest that there is more to the
maintenance of problematic drinking behavior in anxious individuals than just
the self-medication process.
There have also been mixed findings for the effectiveness of cognitive beha-
vioral therapy (CBT) to improve anxiety and problematic drinking symptoms
in comorbid patients. Thevos et al. (2000) found that, for female patients with
comorbid social phobia and alcohol use disorders involved in project MATCH
(the largest treatment-matching trial to date), CBT treatment for alcohol was
more effective in delaying return to drinking than Twelve-Step Facilitation
(TSF). It was hypothesized that TSF, which encourages participation in Alco-
holics Anonymous (AA), a group-based treatment modality that heavily
involves public speaking as patients share their experiences, may be too intimi-
dating for women with social phobia. Subsequently, Randall, Thomas, &
Thevos, (2001) examined whether conducting parallel CBT treatments aimed
at decreasing social anxiety and at addressing problematic drinking behaviors
would have additional benefits for comorbid patients compared to treatment of
the alcohol disorder alone. For patients receiving the parallel treatments, the
sessions consisted of CBT treatment for alcohol followed immediately by CBT
for social anxiety (i.e. the two treatments are offered simultaneously, but
independently of each other). Surprisingly, they found that patients who parti-
cipated in the parallel treatment had worse drinking outcomes, as assessed by
drinking quantity and frequency measures, than did patients who participated
in the alcohol only treatment. There are several possible explanations for these
unexpected findings. It is possible that clients in the parallel treatment group
engaged in more social situations as a consequence of their social phobia
treatment, resulting in more opportunities to drink. Additional research needs
to be conducted that would include other types of outcome measures that are
not specifically linked to frequency or quantity of drinking. As was mentioned
46 B. C. Sabourin, S. H. Stewart

earlier in the chapter, coping drinking motives and problematic consequences of

drinking are useful therapy targets for comorbid individuals, particularly in the
case of social phobia. It is also possible that the lack of integration of the two
treatments or the excessive demands of combining two already intensive treat-
ments may have affected results. The parallel treatment did in fact lead to
somewhat higher drop out rates than the alcohol treatment alone, suggesting
that the parallel treatment may have been too much for comorbid patients to
handle (Conrod & Stewart, 2005).
On the other hand, another study (Bowen, D’Arcy, Keegan, & Senthilselvan,
2000) found that parallel CBT for panic disorder and standard alcohol treat-
ment in a group of comorbid panic disordered and alcoholic patients did not
result in significantly different treatment outcomes than standard alcohol treat-
ment alone. Both treatments resulted in significant decreases in anxiety and in
drinking behaviors. The authors noted that the relaxation training and stress
management components of the standard alcohol treatment might have limited
the ability to distinguish between treatments as these components may have
been useful in targeting the comorbid anxiety.
A recent randomly-controlled study conducted by Schade and colleagues
(2005) compared standard alcohol treatment alone to standard alcohol treat-
ment with anxiety treatment consisting of CBT plus optional fluvoxamine (a
selective serotonin re-uptake inhibitor) treatment (again, a parallel approach)
in patients with a primary diagnosis of alcohol dependence and a comorbid
diagnosis of panic disorder, agoraphobia, or social phobia. There were no
differences in alcohol outcome measures between both groups of patients.
The additional anxiety treatment did, on the other hand, improve anxiety
symptoms. It can be speculated that improved anxiety scores are significant
for this population, as decreased anxiety may serve as a protective factor for
longer-term outcomes. The study examined outcome results 32 weeks after
initial assessment, but did not look at longer- term outcomes (1 year or later)
in these patients. Future studies should examine longer-term treatment out-
comes in comorbid anxiety and alcohol patients.
Few studies, however, have reported outcomes for truly integrated treat-
ments. Integrated treatment models recognize the complex relationship between
anxiety disorders and alcohol use disorders and their possible mutual mainte-
nance (Zahradnik & Stewart, in press). Furthermore, their aim is to create a
hybrid of the treatments that work best for each disorder separately, and also
include in the treatment strategy an understanding of the reciprocal influences
each disorder has on the other (Zahradnik & Stewart). Integrated treatments
have been developed and tested for certain anxiety disorder – substance use
disorder combinations, but until now only one study (Kushner et al., 2006) has
investigated an integrated treatment focusing on comorbidity with alcohol use
disorders in particular. The treatment integrated CBT for panic disorder with
content focusing on the interaction between alcohol use and panic symptoms.
The integrated treatment was provided on top of treatment as usual (TAU) for
the alcohol use disorder and compared to a group who received only the TAU.
Alcohol and Anxiety 47

The trial was conducted on a sample of comorbid panic disorder – alcoholic

patients, with promising results. The group receiving the integrated treatment
showed better anxiety and alcohol outcomes than the TAU alcohol only treat-
ment group.
It is hoped that integrated treatments will provide a more effective strategy in
treating comorbid patients. Integrated treatments appear to be the most recom-
mended by ‘‘expert opinion’’; however few of the recommendations are sup-
ported by randomized controlled trials, or even quasi-experimental designs
(Watkins, Hunter, Burnman, Pincus, & Nicholson, 2005). More research
needs to be conducted to develop and test integrated treatment strategies and
to compare them against parallel or sequential approaches.

Prevention and/or Early Intervention

Another treatment approach receiving recent research attention involves tar-

geting the vulnerability factors (e.g., AS) associated with problematic drinking
in cases with, or at risk for, comorbid anxiety disorders (Conrod, Stewart,
Comeau, & MacLean, 2006, Watt et al., 2006). A study by Watt et al. found
promising results for brief CBT targeted at reducing AS in a sample of uni-
versity women. Three 50-minute CBT sessions led to a decrease in the propor-
tion of women with high AS who engaged in negative consequence drinking
(based on elevated scores on the Rutgers Alcohol Problem Index; RAPI; White
& Labouvie, 1989), as well as a decrease in conformity-motivated drinking (a
high-risk drinking motive; Cooper et al., 1992; Cooper, 1994) and in emotional
relief alcohol expectancies (i.e. a positive alcohol expectancy similar to tension
reduction expectancies described earlier). The treatment also significantly
reduced AS levels (Watt et al., 2006). Another similar AS-focused brief CBT
delayed drinking onset in young at-risk adolescents (mean age = 14; Conrod,
Castellanos, & Mackie, in press). Later drinking onset has been shown to be a
factor protecting against the development of later alcohol use problems (Grant,
Stinson, & Harford, 2001). The intervention also reduced panic attacks in high
AS adolescents (Castellanos & Conrod, 2006). Together, these results suggest
that brief CBT focused on reduction or management of AS may be a useful
strategy for prevention of or early intervention with anxiety – alcohol use
disorder comorbidity.


Although there is ample evidence supporting the existence of a strong relation-

ship between anxiety and its disorders, and alcohol use and its disorders, much
remains unknown regarding the nature of the relationship. More research
examining particular circumstances in which anxious individuals are more
48 B. C. Sabourin, S. H. Stewart

likely to self-medicate needs to be conducted, with a focus on potential differ-

ences in high-risk drinking situations across the various anxiety disorders. In
addition, there has been little research about protective factors that can
decrease anxious individuals’ needs to self-medicate their anxiety symptoms.
For example, a few studies have explored the role of self-efficacy in heavy
drinking among individuals with anxiety disorders (Burke & Stephens, 1997;
Gilles et al., 2006). Initial results suggest that it might be beneficial for treat-
ments to incorporate specific strategies to increase anxious individuals’ feelings
of self-efficacy, especially about avoiding heavy drinking in particular high risk
situations (e.g., those involving anxiety). Increased knowledge about this and
other potential protective factors may be useful when designing specific treat-
ments for comorbid populations (Burke & Stephens, 1997).
It has also been found that alcohol expectancies, drinking motives, and
anxiety sensitivity all moderate or mediate the relationship between anxiety
and alcohol use/abuse. In addition, these three variables have been found to
interact with or intervene with each other when explaining reasons and circum-
stances for increased and problematic alcohol use in anxious individuals (e.g.,
Stewart et al., 2001). More research exploring the interplay between these and
other moderating and mediating variables would provide deeper and more
complete understanding of the precise mechanisms through which anxiety
and alcohol use affect one another.
Finally, although research on treating and preventing comorbid anxiety
disorders and alcohol use disorders is still in relatively early stages, promising
approaches are emerging in this growing clinical research area. For example, in
prevention/early intervention, recent studies have shown promising results for
targeting personality risk factors directly (e.g., AS) in attempts to reduce both
emergent problematic drinking (Conrod et al., 2006; Watt et al., 2006) and
emerging anxiety disorder symptoms (Castellanos & Conrod, 2006). Recent
research aimed at developing truly integrated treatments for comorbid patients
that consider the interplay between anxiety symptoms and drinking behaviors
(Kushner et al., 2006) has also provided positive preliminary results. Treatment
research efforts should continue exploring ways to address the factors discussed
in this chapter (e.g., moderating and/or mediating factors) linking anxiety and
problematic alcohol use to improve the efficacy of treatments we have available
for comorbid patients.


Abrams, K., Kushner, M., Medina, K. L., & Voight, A. (2002). Self-administration of alcohol
before and after a public speaking challenge by individuals with social phobia. Psychology
of Addictive Behaviors, 16, 121–128.
Abrams, K., Kushner, M., Medina, K. L., & Voight, A. (2001). The pharmacologic and
expectancy effects of alcohol on social anxiety in individuals with social phobia. Drug and
Alcohol Dependence, 64, 219–231.
Alcohol and Anxiety 49

Acierno, R., Resnick, H., Kilpatrick, D. G., Saunders, B., & Best, C. L. (1999). Risk factors
for rape, physical assault, and posttraumatic stress disorder in women: Examination of
differential multivariate relationships. Journal of Anxiety Disorders, 13, 541–563.
American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Dis-
orders, 4th Edition (DSM-IV). Washington, DC: Author.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psycholo-
gical Review, 84, 191–215.
Baron, R. M. & Kenny, D. A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 1173–1182.
Barrett, S. P., Darredeau, C., & Pihl, R. O. (2006). Patterns of simultaneous polysubstance use
in drug using university students. Human Psychopharmocology: Clinical and Experimental,
21, 255–263.
Berkson, J. (1949). Limitations of the application of four-fold tables to hospital data.
Biological Bulletin, 2, 47–53.
Bernstein, A., Zvolensky, M. J., Sachs-Ericsson, N., Schmidt, N. B., & Bonn-Miller, M. O. (2006).
Associations between age of onset and lifetime history of panic attacks and alcohol use, abuse,
and dependence in a representative sample. Comprehensive Psychiatry, 47, 342–349.
Bowen, R. C., D’Arcy, C., Keegan, D., & Senthilselvan, A. (2000). A controlled trial of
cognitive behavioral treatment of panic in alcoholic inpatients with comorbid panic
disorder. Addictive Behvaviors, 25, 593–597.
Bradizza, C. M., Stasiewicz, P. R., & Paas, N. D. (2006). Relapse to alcohol and drug use
among individuals diagnosed with co-occurring mental health and substance use disor-
ders: A review. Clinical Psychology Review, 26, 162–178.
Breese, G. R., Overstreet, D. H., & Knapp, D. J. (2005). Conceptual framework for the
etiology of alcoholism: A ‘‘kindling’’/stress hypothesis. Psychopharmacology, 178, 367–380.
Brown, P. J. & Wolfe, J. (1994). Substance abuse and post-traumatic stress disorder comor-
bidity. Drug and Alcohol Dependence, 35, 51–59.
Bruce, S. E., Yonkers, K. A., Otto, M. W., Eisen, J. L., Weisberg, R. B., Pagano, M., et al.
(2005). Influence of psychiatric comorbidity on recovery and recurrence in generalized
anxiety disorder, social phobia, and panic disorder: A 12-year prospective study. American
Journal of Psychiatry, 162, 1179–1187.
Burke, R. S. & Stephens, R. S. (1997). Effect of anxious affect on drinking self-efficacy in
college students. Psychology of Addictive Behaviors, 11, 65–75.
Castellanos, N. & Conrod, P. (2006). Brief interventions targeting personality risk factors for
adolescent substance misuse reduce depression, panic and risk-taking behaviours. Journal
of Mental Health, 15, 645–658.
Chilcoat, H. D., & Breslau, N. (1998). Investigations of causal pathways between PTSD and
drug use disorders. Addictive Behaviors, 23, 827–840.
Conrod, P., Castellanos, N., & Mackie, C. (in press). Personality-targeted interventions
delay the growth of adolescent drinking and binge drinking. Journal of Child Psychology
and Psychiatry.
Conrod, P. J., & Stewart, S. H. (2005). A critical look at dual-focused cognitive-behavioral
treatments of comorbid substance use and psychiatric disorders: Strengths, limitations,
and future directions. Journal of Cognitive Psychotherapy 19,, 261–284.
Conrod, P. J., Stewart, S. H., Comeau, N., & Maclean, A. M. (2006). Efficacy of cognitive-
behavioral interventions targeting personality risk factors for youth alcohol misuse.
Journal of Clinical Child and Adolescent Psychology, 35, 550–563.
Cooper, M. L. (1994). Motivations for alcohol use among adolescents: Development and
validation of a four-factor model. Psychological Assessment, 6, 117–128.
Cooper, M. L., Russell, M., Skinner, J. B., & Windle, M. (1992). Development and
validation of a three-dimensional measure of drinking motives. Psychological Assessment,
4, 123–132.
50 B. C. Sabourin, S. H. Stewart

Cox, B. J., Norton, G. R., Swinson, R. P., & Endler, N. S. (1990). Substance abuse and panic-
related anxiety: A critical review. Behavior Research and Therapy, 28, 385–393.
Cox, W. M. (1987). Personality theory and research. In H. Blane and K.Leonard (Eds.),
Psychological Theories of Drinking and Alcoholism (pp. 55–89). New York: Guilford Press.
Cox, W. M., & Klinger, E. (1988). A motivational model of alcohol use. Journal of Abnormal
Psychology, 97, 168–180.
Dawson, D. A., & Archer, L. D. (1993). Relative frequency of heavy drinking and the risk of
alcohol dependence. Addiction 88,, 1509–1518.
Deacon, B. J., & Valentiner, D. P. (2000). Substance use and non-clinical panic attacks in a
young adult sample. Journal of Substance Abuse, 11, 7–15.
DeHaas, R. A. B., Calamari, J. E., & Bair, J. P. (2002). Anxiety sensitivity and the situational
antecedents to drug and alcohol use: An evaluation of anxiety patients with substance use
disorders. Cognitive Therapy and Research, 26, 335–353.
Driessen, M., Meier, S., Hill, A., Wetterling, T., Lange, W., & Junghanns, K. (2001). The
course of anxiety, depression and drinking behaviors after completed detoxification in
alcoholics with and without comorbid anxiety and depressive disorders. Alcohol and
Alcoholism, 36, 249–255.
Forbes, D., Creamer, M., Hawthorne, G., Allen, N., & McHugh, T. (2003). Comorbidity as a
predictor of symptom change after treatment in combat-related posttraumatic stress
disorder. Journal of Nervous and Mental Disease, 191, 93–99.
Gilles, D. M., Turk, C. L., & Fresco, D. M. (2006). Social anxiety, alcohol expectancies, and
self-efficacy as predictors of heavy drinking in college students. Addictive Behaviors, 31,
Goodwin, R. D., Fergusson, D. M., & Horwood, L. J. (2004). Association between anxiety
disorders and substance use disorders among young persons: Results of a 21-year long-
itudinal study. Journal of Psychiatric Research, 38, 295–304.
Goodwin, R. D., Lieb, R., Hoefler, M., Pfister, H., Bittner, A., Beesdo, K., et al. (2004). Panic
attack as a risk factor for severe psychopathology. American Journal of Psychiatry, 161,
Grant, B. F., Stinson, F. S., Dawson, D. A., Chou, P., Dufour, M. C., Compton, W., et al.
(2004). Prevalence and co-occurrence of substance use disorders and independent mood
and anxiety disorders. Archives of General Psychiatry, 61, 807–816.
Grant, B. F., Stinson, F. S., & Harford, T. C. (2001). Age of onset of alcohol use and DSM-IV
alcohol abuse and dependence: A 12 year follow-up. Journal of Substance Abuse, 13,
Greeley, J., & Oei, T. (1999). Alcohol and tension reduction. In K. Leonard and H. Blane
(Eds.), Psychological Theories of Drinking and Alcoholism (2nd ed.,pp. 14–53). New York:
Guilford Press.
Ham, L. S., Carrigan, M. H., Moak, D. H., & Randall, C. L. (2005). Social anxiety and
specificity of positive alcohol expectancies: Preliminary findings. Journal of Psychopathol-
ogy and Behavioral Assessment, 27, 115–121.
Ham, L. S., & Hope, D. A. (2005). Incorporating social anxiety into a model of college student
problematic drinking. Addictive Behaviors, 30, 127–150.
Helzer, J. E., Bucholz, K. K., & Bierut, L. J. (2006). Should DSM-V include dimensional
diagnostic criteria for alcohol use disorders? Alcoholism: Clinical and Experimental
Research, 30, 303–310.
Jang, K. L., Livesley, W. J., & Vernon, P. A. (1996). Heritability of the big five personality
dimensions and their facets: A twin study. Journal of Personality, 64, 577–591.
Jorm, A. F., Christensen, H., Henderson, A. S., Jacomb, P. A., Korten, A. E., & Rodgers, B.
(2000). Predicting anxiety and depression from personality: Is there a synergistic effect of
neuroticism and extraversion? Journal of Abnormal Psychology, 109, 145–149.
Kandel D. B., Huang F. Y., Davies M. (2001). Comorbidity between patterns of substance use
dependence and psychiatric syndromes. Drug and Alcohol Dependence, 64, 233–41.
Alcohol and Anxiety 51

Kessler, R. C., Crum, R. M., Warner, L. A., Nelson, C. B., Schulenberg, J., & Anthony, J. C.
(1997). Lifetime co-occurrence of DSM-III-R alcohol abuse and dependence with other
psychiatric disorders in the National Comorbidity Survey. Archives of General
Psychiatry, 54, 313–321.
Kessler, R. C., Nelson, C. B., McGonagle, K. A., Edlund, M. J., Frank, R. G., & Leaf, P. J.
(1996). The epidemiology of co-occurring addictive and mental disorders. American
Journal of Orthopsychiatry, 66, 17–31.
Kidorf, M., & Lang, A. R. (1999). Effects of social anxiety and alcohol expectancies on stress-
induced drinking. Psychology of Addictive Behaviors, 13, 134–142.
Kranzler, H. R., Burleson, J. A., Del Boca, F. K., Babor, T. F., Korner, P., Brown, J., et al.
(1994). Buspirone treatment of anxious alcoholics: A placebo-controlled trial. Archives of
General Psychiatry, 51(9), 720–731.
Kushner, M. G., Abrams, K., & Borchardt, C. (2000a). The relationship between anxiety
disorders and alcohol use disorders: A review of major perspectives and findings. Clinical
Psychology Review, 20, 149–171.
Kushner, M. G., Abrams, K., Thuras, P., & Hanson, K. L. (2000b). Individual differences
predictive of drinking to manage anxiety among non-problem drinkers with panic dis-
order. Alcoholism: Clinical and Experimental Research, 24, 448–458.
Kushner, M. G., Abrams, K., Thuras, P., Hanson, K. L., Brekke, M., & Sletten, S. (2005).
Follow-up study of anxiety disorder and alcohol dependence in comorbid alcoholism
treatment patients. Alcoholism: Clinical and Experimental Research, 29, 1432–1443.
Kushner, M. G., Donahue, C., & Sletten, S. (2006). Cognitive behavioral treatment of
comorbid anxiety disorder in alcoholism treatment patients: Presentation of a prototype
program and future directions. Journal of Mental Health, 15, 697–707.
Kushner, M. G., Sher, K. J., & Beitman, B. D. (1990). The relation between alcohol problems
and the anxiety disorders. American Journal of Psychiatry, 147, 685–695.
Kushner, M. G., Sher, K. J., & Erickson, D. J. (1999). Prospective analysis of the relation
between DSM-III anxiety disorders and alcohol use disorders. American Journal of Psy-
chiatry, 156, 723–732.
Kushner, M. G., Sher, K. J., Wood, M. D., & Wood, P. K. (1994). Anxiety and drinking
behavior: Moderating effects of tension-reduction alcohol outcome expectancies. Alco-
holism: Clinical and Experimental Research, 18, 852–860.
LaBounty, L. P., Hatsukami, D., Morgan, S. F., & Nelson, L. (1992). Relapse among
alcoholics with phobic and panic symptoms. Addictive Behaviors, 17, 9–15.
Lehman, C. L., Brown, T. A., Palfai, T., & Barlow, D. H. (2002). The effects of alcohol
outcome expectancy on a carbon-dioxide challenge in patients with panic disorder.
Behavior Therapy, 33, 447–463.
MacDonald, A. B., Stewart, S. H., Hutson, R., Rhyno, E., & Loughlin, H. L. (2001). The roles
of alcohol and alcohol expectancy in the dampening of responses to hyperventilation
among high anxiety sensitive young adults. Addictive Behaviors, 26, 827–840.
Marlatt, G. A., & Donovan, D.M. (Eds.). (2005). Maintenance strategies in the treatment of
addictive behaviors (2nd ed.). New York: Guilford Press.
Marquenie, L. A., Schade, A., Van Balkom, A. J., Koeter, M., Frenken, S.,Van Den Brink,
W., et al. (2006). Comorbid phobic disorders do not influence outcome of alcohol depen-
dence treatment: Results of a naturalistic follow-up study. Alcohol and Alcoholism, 41,
Marshall, J. R. (1997). Alcohol and substance abuse in panic disorder. Journal of Clinical
Psychiatry, 58(suppl 2), 46–49.
McFall, M. E., Mackay, P. W., & Donovan, D. M. (1992). Combat-related posttraumatic
stress disorder and severity of substance abuse in Vietnam veterans. Journal of Studies on
Alcohol, 53, 357–363.
Merikangas, K. R., Mehta, R. L., Molnar, B. E., Walters, E. E., Swendsen, J. D., Aguilar, S.,
et al. (1998). Comorbidity of substance use disorders with mood and anxiety disorders:
52 B. C. Sabourin, S. H. Stewart

Results of the international consortium in psychiatric epidemiology. Addictive Behaviors,

23, 893–907.
Merikangas, K. A., Stevens, D., & Fenton, B. (1996). Comorbidity of alcoholism and anxiety
disorders. Alcohol Health and Research, 20, 100–105.
Norton, G. R., Norton, P. J., Cox, B. J., & Belik, S. (in press). Panic spectrum disorders and
substance use. In S. H. Stewart & P. J. Conrod (Eds.), The vicious cycle: Theoretical and
treatment issues in comorbid anxiety and substance use disorders. New York: Springer.
Norton, G. R., Rockman, G. E., & Ediger, J. (1997). Anxiety sensitivity and drug abuse in
individuals seeking treatment for substance abuse. Behaviour Research and Therapy, 35,
Paavola, M., Vartiainen, E., & Haukkala, A. (2004). Smoking, alcohol use, and physical
activity: A 13-year longitudinal study ranging from adolescence into adulthood. Journal of
Adolescent Health, 35, 238–244.
Persons, J. B. (1986). The advantage of studying psychological phenomena rather than
psychiatric diagnoses. American Psychologist, 41, 1252–1260.
Randall, C. L., Johnson, M. R., Thevos, A. K., Sonne, S. C., Thomas, S. E., Willard, S. L.,
et al. (2001). Paroxetine for social anxiety and alcohol use in dual-diagnosed patients.
Depression and Anxiety, 14, 255–262.
Randall, C. L., Thomas, S., & Thevos, A. K. (2001). Concurrent alcoholism and social anxiety
disorder: A first step toward developing effective treatments. Alcoholism: Clinical and
Experimental Research, 25, 210–220.
Schade, A., Marquenie, L. A., Balkom, A. J., Koeter, M. W. J., de Ceurs, E., van den Brink,
W., et al. (2005). The effectiveness of anxiety treatment on alcohol-dependent patients with
a comorbid phobic disorder: A randomized controlled trial. Alcoholism: Clinical and
Experimental Research, 29, 794–800.
Smith, S. M., Stinson, F. S., Dawson, D. A., Goldstein, R., Huang, B., & Grant, B. F. (2006).
Race/ethnic differences in the prevalence and co-occurrence of substance use disorders and
independent mood and anxiety disorders: Results from the National Epidemiologic Sur-
vey on Alcohol and Related Conditions. Psychological Medicine, 36, 987–998.
Sobell, L. C., Cunningham, J. A., Sobell, M. B., Agrawal, S., Gavin, D. R., Leo, G. I., et al.
(1996). Fostering self-change among problem drinkers: A proactive community interven-
tion. Addictive Behaviors, 21, 817–833.
Stein, M. B., Lang, K. L., & Livesley, W. J. (1999). Heritability of anxiety sensitivity: A twin
study. American Journal of Psychiatry, 156, 246–251.
Stewart, S. H., & Conrod, P. J. (Eds.). (in press). The vicious cycle: Theoretical and treatment
issues in co-morbid anxiety and substance abuse disorders. New York: Springer.
Stewart, S. H., & Conrod, P. J. (2002). Psychosocial models of functional associations
between posttraumatic stress disorder and substance use disorder. In P. C. Ouimette &
P. J. Brown (Eds.), Trauma and substance abuse: causes, consequences, and treatment of
comorbid disorders (pp. 29–55). Washington, DC: American Psychological Association.
Stewart, S. H., Conrod, P. J., Samoluk, S. B., Pihl, R. O., & Dongier, M. (2000). Posttrau-
matic stress disorder symptoms and situation-specific drinking in women substance abu-
sers. Alcoholism Treatment Quarterly, 18, 31–47.
Stewart, S. H., & Kushner, M. G. (2001). Introduction to the special issue on ‘‘Anxiety
sensitivity and addictive behaviors’’. Addictive Behaviors, 26, 775–785,
Stewart, S. H., Morris, E., Mellings, T., & Komar, J. A. (2006) Relations of social anxiety
variables to drinking motives, drinking quantity and frequency, and alcohol-related
problems in undergraduates. Journal of Mental Health, 15, 671–682.
Stewart, S. H., Peterson, J. B., & Pihl, R. O. (1995). Anxiety sensitivity and self-reported
alcohol consumption rates in university women. Journal of Anxiety Disorders, 9, 283–292.
Stewart, S. H., Pihl, R. O., Conrod, P. J., & Dongier, M. (1998). Functional associations
among trauma, PTSD, and substance related disorders. Addictive Behaviors, 23,
Alcohol and Anxiety 53

Stewart, S. H., Zvolensky, M. J., & Eifert, G. H. (2001). Negative-reinforcement drinking

motives mediate the relation between anxiety sensitivity and increased drinking behavior.
Personality and Individual Differences, 31, 157–171.
Swendsen, J. D., Conway, K. P., Rounsaville, B. J., & Merikangas, K. R. (2002). Are
personality traits familial risk factors for substance use disorders? Results from a con-
trolled family study. American Journal of Psychiatry, 159, 1760–1766.
Swendsen, J. D., Tennen, H., Carney, M. A., Affleck, G., Wilard, A., & Hromi, A. (2000).
Mood and alcohol consumption: An experience sampling test of the self-medication
hypothesis. Journal of Abnormal Psychology, 109, 198–204.
Tambs, K., Harris, J. R., & Magnus, P. (1997). Genetic and environmental contributions to
the correlation between alcohol consumption and symptoms of anxiety and depression:
Results from a bivariate analysis of Norwegian twin data. Behavior Genetics, 27,
Thevos, A. K., Roberts, J. S., Thomas, S. E., & Randall, C. L. (2000). Cognitive behavioral
therapy delays relapse in female socially phobic alcoholics. Addictive Behaviors, 25,
Thomas, S. E., Randall, C. L., & Carrigan, M. H. (2003). Drinking to cope in socially anxious
individuals: A controlled study. Alcoholism: Clinical and Experimental Research, 27,
Tollefson, G. D., Montague-Clouse, J., & Tollefson, S. L. (1992). Treatment of comorbid
generalized anxiety in a recently detoxified alcoholic population with a selective seroto-
nergic drug (buspirone). Journal of Clinical Psychopharmacology, 12, 19–26.
Tolstrup, J. S., Heitmann, B. L., Tjonnelands, A. M., Overvad, O. K., Sorensen, T. I. A., &
Gronbaeki, M. N. (2005). The relation between drinking pattern and body mass index and
waist and hip circumference. International Journal of Obesity, 29, 490–497.
Tran, G. Q., Haaga, D. A. F., & Chambless, D. L., (1997). Expecting that alcohol use will
reduce social anxiety moderates the relation between social anxiety and alcohol consump-
tion. Cognitive Therapy and Research, 21, 535–553.
Ullman, S. E., Filipas, H. H., & Townswend, S. M. (2005). Trauma exposure, posttraumatic
stress disorder and problem drinking in sexual assault survivors. Journal of Studies on
Alcohol, 65, 610–619.
Watkins, K. E., Hunter, S. B., Burnman, S. B., Pincus, H. A., & Nicholson, G. (2005). Review
of treatment recommendations for persons with a co-occurring affective or anxiety and
substance use disorder. Psychiatric Services, 56, 913–926.
Watt, M., Stewart, S., Birch, C., & Bernier, D. (2006). Brief CBT for high anxiety sensitivity
decreases drinking problems, relief alcohol outcome expectancies, and conformity drink-
ing motives. Journal of Mental Health, 15, 683–695.
Weinstock, L. M., & Whisman, M. A. (2006). Neuroticism as a common feature of the
depressive and anxiety disorders: A test of the revised integrative hierarchical model in a
national sample. Journal of Abnormal Psychology, 115, 68–74.
White, H. R., & Labouvie, E. W. (1989). Towards the assessment of adolescent problem
drinking. Journal of Studies on Alcohol, 50, 30–37.
Willinger, U., Lenzinger, E., Hornik, K., Fischer, G., Schonbeck, G., Aschauer, H. N., et al.
(2002). Anxiety as a predictor of relapse in detoxified alcohol-dependent patients. Alcohol
and Alcoholism, 37, 609–612.
Zahradnik, M., & Stewart, S. (in press). Anxiety disorders and substance use disorder
comorbidity: Epidemiology, theories of interrelation, and recent treatment approaches.
In M. Antony & M. Stein (Eds.), Handbook of anxiety and the anxiety disorders. Oxford,
UK: Oxford University Press.
Zimmermann, P. Wittchen, H. U., Hofler, M., Pfister, H., Kessler, R. C. & Lieb, R. (2003).
Primary anxiety disorders and the development of subsequent alcohol use disorders:
A 4-year community study of adolescents and young adults. Psychological Medicine,
33, 1211–1222.
54 B. C. Sabourin, S. H. Stewart

Zinbarg, R. E., Mohlman, J., & Hong, N. N. (1999). Dimensions of anxiety sensitivity.
In S. Taylor (Ed.), Anxiety sensitivity: Theory, research, and treatment of the fear
of anxiety. (pp. 83–114). Mahwah, NJ: Lawrence Erlbaum.
Illicit Drug Use Across the Anxiety Disorders
Prevalence, Underlying Mechanisms, and Treatment

Matthew T. Tull, David E. Baruch, Michelle S. Duplinsky, and C. W. Lejuez

An increasing number of studies are beginning to recognize the heightened risk

for the development of substance use disorders among individuals with an anxiety
disorder diagnosis, and the possible interplay between these disorders is receiving
more attention in theoretical, clinical, and empirical spheres (e.g., Goodwin et al.,
2002; Morissette, Tull, Gulliver, Kamholz, & Zimering, 2007; Stewart, 1996;
Zvolensky & Schmidt, 2004; Zvolensky, Schmidt, & Stewart, 2003). As the
majority of this attention has been focused on nicotine or alcohol use, the goal
of this chapter is to review the extant work specific to illicit drug use across the
anxiety disorders. In keeping with this focus, we will consider only illicit drugs
and refer interested readers to chapters 4 and 5 in this volume for reviews specific
to nicotine and alcohol. We begin by providing a brief list of the drugs that would
fall under the larger category of illicit drugs, as well as prevalence rates of illicit
drug use within the general population. We then move to discuss prevalence rates
of illicit drug use within the anxiety disorders and among individuals with
anxiety-disorder relevant symptoms (e.g., non-clinical panic attacks). Of note,
this chapter will focus primarily on illicit drug use within the anxiety disorders as
opposed to anxiety disorder diagnoses among individuals who use illicit drugs.
Next, we present theories and data pertaining to the temporal progression of illicit
drug use and anxiety disorders. The chapter then concludes with a brief review of
treatments that specifically address comorbid anxiety and illicit drug use.

Illicit Drugs

The term illicit drug is used here in reference to any substance that is illegal,
with the distinction of illegal being specific to the United States.1 Such drugs,
as recognized by the Diagnostic and Statistical Manual of Mental Disorders,

Matthew T. Tull
Department of Psychology, University of Maryland, College Park, MD 20742, Tel: 301-405-
3281, FAX: 310-405-3223
Due to space limitations we do not consider legal drugs used in an illegal manner (e.g., pain
medications used without a prescription).

M. J. Zvolensky, J. A. Smits (eds.), Anxiety in Health Behaviors and Physical Illness. 55

Ó Springer 2008
56 M. T. Tull et al.

4th Edition (DSM-IV; American Psychiatric Association [APA], 1994),

include opioids (e.g., heroin, morphine), cocaine and its variants (e.g.,
crack), amphetamines (e.g., crystal methamphetamine), hallucinogens (e.g.,
LSD, PCP), inhalants, and marijuana. Drug use diagnoses include abuse and
dependence. Abuse refers to a pattern of drug use leading to significant
functional impairment, as evidenced by one of the following within a
12 month period: 1) recurrent use leading to failures to fulfill major obliga-
tions at work, school, or home; 2) recurrent use in situations where it may be
physically hazardous to do so; 3) legal problems resulting from use; or 4)
continued use even when the drug use results in social or psychological
problems (APA, 1994). Dependence refers to a history of drug use marked
by: 1) substance abuse; 2) continued use despite problems associated with that
use; 3) drug tolerance; and 4) withdrawal symptoms (APA, 1994). Specific
reference to use, abuse, and dependence often differ across studies, which can
limit more broad conclusions across studies. Therefore, throughout this
chapter we refer more broadly to use and will refer specifically to abuse and
dependence when appropriate.
In regard to rates of illicit drug use within the general population, the
National Comorbidity Survey (NCS; Kessler et al., 1994), a nationally repre-
sentative household survey of 8,098 adults aged 15–54 in the United States,
found that 4.4% (5.4% of men and 3.5% of women) of respondents met
criteria for lifetime drug abuse and 7.5% (9.2% of men and 5.9% of women)
met criteria for lifetime drug dependence. In regard to 12-month prevalence,
0.8% (1.3% of men and 0.3% of women) met criteria for 12-month drug
abuse and 2.8% (3.8% of men and 1.9% of women) met criteria for 12-month
drug dependence. Data from the NCS Replication (NCS-R; Kessler, Chiu,
Demler, & Walters, 2005), a nationally representative household survey of
adult English-speaking individuals in the United States, provides more recent
12-month prevalence rates. Specifically, the NCS-R found that 1.4% and
0.4% of 5,692 respondents met criteria for drug abuse or dependence (respec-
tively) in the past 12 months (prevalence rates as a function of gender were
not provided).
A limitation of the majority of epidemiological surveys is the examination
of prevalence rates across drugs, with limited specificity to particular drugs.
As a notable exception, the Epidemiological Catchment Area (ECA) study, a
representative household survey of 20,291 adults in the United States,
provides 6-month and lifetime prevalence rates for drug use disorders, as
well as rates for specific illicit drug use disorders (rates are collapsed across
abuse or dependence)(Reiger, et al., 1990). In particular, 2% and 6.1% of the
sample met criteria for a 6-month or lifetime drug use disorder respectively.
In regard to specific drug use disorders, prevalence rates were provided
for marijuana dependence-abuse (1.2% 6-month, 4.3% lifetime), cocaine
dependence-abuse (0.0% 6-month, 0.2% lifetime), opiate dependence-abuse
(0.1% 6-month, 0.7% lifetime), amphetamine dependence-abuse (0.2%
Drug Use and the Anxiety Disorders 57

6-month, 1.7% lifetime), and hallucinogen dependence-abuse (0.0% 6-month,

0.3% lifetime) (Reiger et al., 1990).
The National Epidemiological Survey on Alcohol and Related Conditions
(NESARC; Conway, Compton, Stinson, & Grant, 2006), a representative
household and group quarters survey of 43,093 adults in the United States,
provides a more recent estimate of lifetime prevalence rates of drug use dis-
orders, as well as rates for specific illicit drug use disorders, than the ECA.
Within their sample (which over-sampled for minorities and young adults
between 18–24 years of age), 7.74% (10.6% for men and 5.1% for women)
met criteria for lifetime drug abuse and 2.59% (3.3% for men and 2.0% for
women) for lifetime drug dependence. In regard to specific illicit drug use
disorders: 1.08% (1.6% men, 0.6% women) and 0.34% (0.4% men, 0.3%
women) met criteria for lifetime opioid abuse and dependence respectively,
1.40% (1.9% men, 0.9% women) and 0.60% (0.6% men, 0.6% women) met
criteria for lifetime amphetamine abuse and dependence respectively, 1.45%
(2.1% men, 0.9% women) and 0.24% (0.4% men, 0.1% women) met criteria for
lifetime hallucinogen abuse and dependence respectively, 1.83% (2.7% men,
1.0% women) and 0.98% (1.2% men, 0.7% women) met criteria for lifetime
cocaine abuse and dependence respectively, 7.16% (10% men, 4.5% women)
and 1.30% (1.7% men, 0.9% women) met criteria for lifetime marijuana abuse
and dependence respectively, and 0.30% (0.5% men, 0.1% women) met criteria
for lifetime inhalant/solvent abuse (Conway et al., 2006).
In terms of racial/ethnic differences in drug use disorders, Smith et al. (2006)
found in the NESARC that, in general, Native Americans evidenced the highest
12 month prevalence rates of drug abuse or dependence as compared to all other
racial/ethnic groups. Rates of drug use disorders across racial/ethnic groups
were as follows: White (1.93%), Black/African-American (2.39%, significantly
different from rates for White participants), Native-American (4.91%, signifi-
cantly different from rates for White and Black/African-American partici-
pants), Asian/Asian-American (1.39%, significantly different from rates for
Black/African-Americans and Native-Americans), and Latino (1.74%, signifi-
cantly different from rates for Black/African-Americans and Native-
Not surprisingly, illicit drug use is associated with a variety of health
compromising behaviors including risky sexual behavior (Brook et al.,
2004; Loxley, Bevan, & Carruthers, 1998), licit substance use such as smoking
(Zvolensky, Bonn-Miller et al., 2006) and alcohol abuse or dependence
(Stinson et al., 2005), and poor self-care including poor dietary (Benedict,
Evans, & Calder, 1999; Morabia et al., 1989) and sleep patterns (Pace-Schott
et al., 2005). Combined with an anxiety disorder (which may develop inde-
pendently of, prior to, or as a result of a drug use disorder), the interactive
effects of a co-morbid drug use and anxiety disorder may place an individual
at risk for more severe and extensive health consequences than those with
either disorder alone.
58 M. T. Tull et al.

Illicit Drug Use in the Anxiety Disorders

A number of studies have demonstrated that prevalence rates of anxiety dis-

order diagnoses among illicit drug users are generally higher than what has been
found among individuals from the general population (e.g., Kessler, Berglund,
Demler, Jin, & Walters, 2005; Grant et al., 2004). For example, Kessler, et al.
(1996), in examining the 1-year prevalence of anxiety disorders among 5,877
individuals with a substance use disorder from the NCS, found that 24% of
individuals with drug use without dependence also met criteria for some anxiety
disorder and 45.5% of individuals with an illicit drug use disorder met criteria
for an anxiety disorder. Beyond evidence of anxiety disorders among substance
users, a growing literature focuses on illicit drug use among those with an
anxiety disorder.

General Relationships Between Anxiety Disorders

and Illicit Drug Use

Reiger et al. (1990), in analyzing data from the ECA study, found that
individuals with any anxiety disorder (as determined by DSM-III criteria)
were 2.5 times as likely to also exhibit a comorbid drug use disorder (Odds
Ratio [OR] = 2.4 for drug dependence and OR = 2.3 for drug abuse).
Specifically, 11.9% of participants with any anxiety disorder diagnosis also
met criteria for a drug use disorder, and this rate was significantly greater than
that found among participants without an anxiety disorder diagnosis. Grant
et al. (2004) found slightly higher ORs in examining data from the NESARC.
Individuals with any anxiety disorder were 4.58 times as likely to also exhibit a
comorbid drug use disorder (OR = 2.15 for drug abuse and OR = 2.43 for drug
In another large scale study of comorbidity, Merikangas et al. (1998) ana-
lyzed data from six international epidemiological study sites (Germany,
Mexico, Netherlands, Ontario, and two sites in the United States). They
found that, across all sites, approximately 24.9% of individuals with any life-
time DSM-III-R anxiety disorder diagnosis also exhibited lifetime drug use
(e.g., marijuana, opioids, stimulants, sedatives, or inhalants, but not alcohol),
35.8% exhibited lifetime drug problems (met at least one DSM-III-R abuse
criteria for any drug), and 45.8% met criteria for lifetime DSM-III-R drug
Lopez, Turner, and Saavedra (2005) analyzed data from 1,747 young adults
(92% aged 19–21) from Miami-Dade county (Florida) public schools, and
examined drug dependence among individuals with pure anxiety disorder diag-
noses (i.e., the occurrence of one or more anxiety disorders that is not accom-
panied by any additional psychiatric disorder) or an anxiety disorder that
Drug Use and the Anxiety Disorders 59

was comorbid with some other psychiatric disorder (depression, dysthymia,

conduct disorder, or attention deficit/hyperactivity disorder) (DSM-IV diag-
nostic criteria was used for disorder classification). Of the 4.9% with a pure
anxiety disorder diagnosis (75.3% met diagnostic criteria for posttraumatic
stress disorder [PTSD], 16.5% for social anxiety disorder [SAD], 5.9% for
panic disorder [PD], and 3.5% for generalized anxiety disorder [GAD]),
13.9% also exhibited drug dependence. Further, more men (20.8%) than
women (10.9%) with a pure anxiety disorder diagnosis also met criteria for
drug dependence. Among the 10.2% with a comorbid anxiety disorder diag-
nosis (76.4% met criteria for PTSD, 18% for PD, 15.7% for SAD, 12.4% for
GAD, and in terms of non-anxiety disorder comorbid psychiatric diagnoses,
62.9% met criteria for depression/dysthymia, 53.5% for conduct disorder,
36.5% for antisocial personality disorder, and 28.1% for attention deficit
hyperactivity disorder), 29.2% met criteria for drug dependence. Again, more
men (33.3%) than women (26.9%) with a comorbid anxiety disorder diagnosis
also met criteria for drug dependence.
Studies have also been conducted among adolescent populations and pro-
duced similar findings regarding high prevalence of drug use disorders among
adolescents with anxiety disorder diagnoses. For example, Swadi and Bobier
(2003) examined 62 adolescent patients (Mage=16.35) from an inpatient mental
health facility in New Zealand. Of the 16% of these patients who had a
diagnosable anxiety disorder, 63% also exhibited a drug use disorder (primarily
marijuana, stimulants, and hallucinogens). Goodwin, Fergusson, and
Horwood (2004) examined DSM-IV anxiety disorder and illicit drug depen-
dence comorbidity among 1265 16–18 year-olds from the New Zealand
Christchurch Health and Development Study (CHDS). Of the 175 participants
who met criteria for an anxiety disorder at the age of 16–18, 12% also met
criteria for illicit drug dependence, compared to 3.4% among those without an
anxiety disorder diagnosis. Participants were then assessed again at age 18–21.
Of the 130 who met criteria for an anxiety disorder within this age range, only
9.2% met criteria for illicit drug dependence. Interestingly, this rate was not
significantly different from rates obtained among participants without an anxi-
ety disorder diagnosis (7%).

Relationships Between Specific Anxiety Disorders

and Illicit Drug Use

In addition to examining general relationships between anxiety and drug use

disorders, studies have also focused their attention on examining the relation-
ship between specific anxiety disorders or anxiety disorder-related symptoms
(e.g., non-clinical panic attacks) and the use of specific drugs. Although these
studies are limited, more focused attention on the relationship between specific
anxiety disorder diagnoses and drugs of abuse may further our understanding
60 M. T. Tull et al.

of the functional relationship between disorders. That is, depending upon

the type of anxiety disorder present, an individual may be at risk for the
development of a particular drug use disorder. We will first present data from
the NESARC that provides an overview of 12-month and lifetime prevalence
rates of various illicit drug use disorders among individuals with specific anxiety
disorder diagnoses. We will then direct more focused attention on additional
research that examines illicit drug use within specific anxiety disorders.
The National Epidemiological Survey on Alcohol and Related Conditions.
Conway et al. (2006) have recently presented data from the NESARC
on lifetime prevalence rates of illicit drug use disorders across the DSM-IV
anxiety disorders (with the exception of PTSD and obsessive-compulsive dis-
order [OCD]). In doing so, Conway et al. (2006) have provided the most
comprehensive overview of anxiety disorder-illicit drug use disorder comorbid-
ity to date2. Among respondents with PD with agoraphobia, 6.2% met criteria
for a lifetime opioid use disorder, 9.2% for amphetamine use disorder, 8.1% for
hallucinogen use disorder, 25.6% for marijuana use disorder, 11.5% for cocaine
use disorder, and 0.8% for inhalant/solvent abuse. Among respondents with
PD without agoraphobia, 4.9% met criteria for a lifetime opioid use disorder,
5.8% for amphetamine use disorder, 4.3% for hallucinogen use disorder, 17.7%
for marijuana use disorder, 7.6% for cocaine use disorder, 1.1% for inhalant/
solvent abuse. Of respondents with SAD, 3.6% met lifetime criteria for an
opioid use disorder, 5.5% for amphetamine use disorder, 4.1% for hallucinogen
use disorder, 17.8% for marijuana use disorder, 6.3% for cocaine use disorder,
and 0.7% for inhalant/solvent abuse. In regard to specific phobia, 2.8% of
respondents with a lifetime diagnosis of specific phobia also met lifetime criteria
for an opioid use disorder, 4.9% for amphetamine use disorder, 3.9% for
hallucinogen use disorder, 14.9% for marijuana use disorder, 4.8% for cocaine
use disorder, and 0.5% for inhalant/solvent abuse. Finally, among respondents
with GAD, 3.7% also met lifetime criteria for an opioid use disorder, 6.2% for
amphetamine use disorder, 4.6% for hallucinogen use disorder, 18.5%
for marijuana use disorder, 7.1% for cocaine use disorder, and 0.7% for
inhalant/solvent abuse.
Conway et al. (2006) also examined gender differences in the ORs of illicit
drug use disorders across the anxiety disorders. Few differences were observed.
However, women with SAD (OR = 4.4) were significantly more likely than
men with SAD (OR = 2.3) to exhibit comorbid lifetime opioid use disorder. In
addition, men with PD with agoraphobia (OR = 8.7) were significantly more
likely than women with PD with agoraphobia (OR = 6.5) to exhibit comorbid
lifetime cocaine dependence.
Grant et al. (2004) also provide 12-month anxiety disorder-illicit drug use
disorder co-morbidity prevalence data from the NESARC, although

Due to space limitations, only data pertaining to drug use disorders in general will be
presented. Readers interested in data pertaining to rates of drug abuse or dependence
separately across the anxiety disorder diagnoses are referred to Conway et al. (2006).
Drug Use and the Anxiety Disorders 61

prevalence rates pertaining to specific illicit drug use disorders were not exam-
ined. Among respondents who met 12-month criteria for a DSM-IV anxiety
disorder of PD with agoraphobia, 10.58% also exhibited a 12-month drug use
disorder (4.65% drug abuse and 5.94% drug dependence). Rates of 12-month
drug use disorders for other anxiety disorders were: 6.32% (2.17% drug abuse
and 4.16% drug dependence) for PD without agoraphobia, 5.52% (2.59% drug
abuse and 2.94% drug dependence) for SAD, 4.08% (2.13% drug abuse, 1.95%
drug dependence) for specific phobia, and 8.06% (2.82% drug abuse, 5.24%
drug dependence) for GAD.
Finally, data from the NESARC has been used to examine differences in
racial/ethnic background across comorbid anxiety and drug use disorders.
Specific to abuse, Smith et al. (2006) found significant 12-month associations
between drug abuse and all anxiety disorders (with the exception of PD without
agoraphobia). No other significant 12-month associations between drug abuse
and anxiety disorders were found for any other racial/ethnic group, with the one
exception of a significant 12-month association between drug abuse and specific
phobia among Native American respondents. A greater number of significant
associations were found when examining the relationship between drug depen-
dence and specific anxiety disorders across racial/ethnic groups. Significant
associations were found for drug dependence and all anxiety disorders exam-
ined (PD with and without agoraphobia, SAD, specific phobia, GAD) among
White respondents. Significant associations were found for drug dependence
and all anxiety disorders with the exception of specific phobia for Black/African
American respondents. Among Native American respondents, significant asso-
ciations were found for PD without agoraphobia and SAD. Asian/Asian-
American respondents demonstrated significant associations between drug
dependence and PD without agoraphobia, SAD, and GAD, and Latino respon-
dents exhibited significant associations between drug dependence and PD with
agoraphobia, SAD, specific phobia, and GAD.
As mentioned previously, the NESARC is the first to provide comprehensive
data on the co-occurrence of specific anxiety disorders and illicit drug use
disorders. In general, though, limited data is available that examines drug use
within specific anxiety disorders. The one exception, however, may be PTSD, as
a number of studies have examined the co-occurrence of PTSD and illicit drug
use (for a review, see Chilcoat & Menard, 2003). We now move to a review of
this literature, followed by a review of the literature pertaining to drug use
within other anxiety disorders besides PTSD.
Posttraumatic stress disorder. Data from the ECA found that compared to
men and women without a diagnosis of PTSD, men with PTSD were 5 times
more likely to also exhibit a drug use disorder and women with PTSD were
1.4 times as likely to exhibit a drug use disorder (overall odds ratio of 2.2; Helzer
et al., 1987). In the NCS, Kessler, Sonnega, Bromet, Hughes, and Nelson (1995)
found that compared to men and women without a diagnosis of PTSD, men
with PTSD were approximately 2.97 times as likely and women 4.46 times as
likely to exhibit a drug use disorder. Giaconia and colleagues (1995, 2000)
62 M. T. Tull et al.

collected data from 384 18-year-olds as part of The Early Adulthood Research
Project (EARP) and found that compared to individuals without a history
of traumatic exposure, individuals with a lifetime diagnosis of PTSD were
8.8 times as likely to also meet criteria for a lifetime drug dependence diagnosis
and 14.14 times as likely to meet criteria for past year drug dependence.
Calhoun et al. (2000) assessed drug use among a sample of 341 veterans with
PTSD seeking treatment for PTSD. Opiate and marijuana use was reported by
the largest number of patients (23% and 20% respectively). Benzodiazepine use
was reported by 11% of patients, cocaine use by 8% of the patients, barbitu-
rates by 5% of the patients, amphetamines by 3% of the patients, psilocybin by
3% of the patients, LSD by 1% of the patients, and PCP by 1% of the patients.
Other studies have produced similar findings in regard to the use of specific
drugs. For example, Saxon et al. (2001) found that incarcerated veterans with
PTSD were more likely to report a greater degree of cocaine and heroin use as
compared to individuals without a PTSD diagnosis. Tarrier and Sommerfield
(2003) assessed 120 civilians seeking treatment for chronic PTSD on their drug
use histories. Of the 120 participants, 17 used marijuana, 6 used sedatives, 4
used stimulants, 2 used a psychedelic, and 1 used cocaine. Further, in regard to
rates of drug use disorders among younger individuals with PTSD, Kilpatrick,
et al. (2000) collected data from a national sample of 4,023 adolescents (between
the ages of 12 and 17). They found that PTSD was significantly associated with
illicit drug use, including marijuana abuse and dependence and ‘‘harder’’ drug
abuse and dependence (e.g., stimulants).
Panic disorder and panic disorder-related symptoms. Moving beyond PTSD, a
number of studies have examined illicit drug use among individuals with PD or
PD-related symptoms, such as the experience of non-clinical panic attacks.
Biederman et al. (2005) examined the rates of comorbid disorders among 23
individuals with PD and found that 8% exhibited a comorbid psychoactive
substance use disorder. Among individuals with PD and major depression, 21%
exhibited a comorbid psychoactive substance use disorder. Further, Zvolensky,
Bernstein et al. (2006) examined lifetime associations between marijuana use,
abuse, and dependence and panic attacks in a representative sample of 4,745
individuals. They found a positive association between lifetime panic attack
occurrence and marijuana use, even when controlling for the effect of polysub-
stance use, alcohol abuse, and demographic variables (e.g., age, gender, etc.).
Bonn-Miller, Bernstein, Sachs-Ericsson, Schmidt, and Zvolensky (2007) also
examined the relationship between psychedelic (e.g., PCP, LSD, mescaline,
peyote, psilocybin, DMT) use, abuse, and dependence and lifetime panic attack
history within this same sample. Psychedelic abuse and dependence (although
not use) were significantly associated with a heightened risk for the experience
of lifetime panic attacks, controlling for demographic variables, polysubstance
use, alcohol abuse, and a history of major depressive disorder.
Deacon and Valentiner (2000) examined the relationship between panic
attacks and substance use within a sample of 279 college students. Students
with non-clinical panic attacks (n = 25) were significantly more likely than
Drug Use and the Anxiety Disorders 63

students without panic (n = 222) to report the use of drugs, such as sedatives
(not alcohol), cocaine, and stimulants. Further, among non-clinical panickers,
sedative use was not found to be related to distress about panic attacks, panic
attack frequency, the occurrence of unexpected attacks, or general anxiety or
depression symptoms. Valentiner, Mounts, and Deacon (2004) also investi-
gated the relationship between panic attacks and illicit drug use in a sample of
399 incoming college freshman. Similar to Deacon and Valentiner (2000), they
found that non-clinical panickers (n = 47), compared to non-panickers
(n = 290), were significantly more likely to report lifetime use of sedatives,
stimulants, and opioids, but not tobacco, alcohol, cocaine, or hallucinogens. In
regard to specific rates of drug use among panickers, 12.8% reported lifetime
use of sedatives (not alcohol), 6.4% reported lifetime use of cocaine, 55.3%
reported lifetime use of marijuana, 34% reported lifetime use of stimulants,
21.3% reported lifetime use of opioids, and 23.4% reported lifetime use of
hallucinogens. Additional analyses were conducted by Valentiner et al. (2004)
to determine whether the relationships between panic and substance use dif-
fered as a function of gender or racial/ethnic background. Gender only served
as a moderator in the relationship between panic and cocaine use. In particular,
male panickers were significantly more likely to use cocaine than males without
panic. In addition, among non-clinical panickers, substance use was not due to
the number of panic attacks in the past year, panic attack symptom severity,
and the experience of unexpected panic attacks.
Other anxiety disorders. There is a dearth of studies on the relationship
between specific drug use disorders with SAD, GAD, OCD, and specific
phobia. More research is needed, especially given extant evidence that these
anxiety disorder diagnoses may be associated with heightened risk for the
development of illicit drug use disorders. For example, using data from the
ECA Study, Reiger et al. (1990) found that 18.4% of respondents with lifetime
OCD also exhibited a lifetime drug use disorder (11% for drug dependence and
7.4% for drug abuse). Further, this rate was significantly greater than what was
found among individuals without a diagnosis of OCD.
In regard to SAD, Kessler et al. (1996), in examining data from the NCS,
found that respondents with SAD were significantly at greater risk to exhibit a
12-month co-occurrence of drug dependence (OR = 3.2) and lifetime drug
dependence (OR = 2.6). Kessler et al. (1996) also provide data on GAD and
specific phobia. Respondents with GAD were at significantly greater risk to
exhibit lifetime drug dependence (OR = 3.8). Respondents with specific phobia
were at significantly greater risk to exhibit co-occurring 12-month (OR = 1.8)
or lifetime (OR = 2.5) drug dependence. Fewer studies have specifically exam-
ined the relationship between specific drug use and SAD; however, of note,
several studies have also found evidence of an association between SAD and
marijuana use (Buckner, Mallott, Schmidt, & Taylor, 2006; Buckner, Schmidt,
Bobadilla, & Taylor, 2006; Lindquist, Lindsay, & White, 1979; Lynskey et al.,
64 M. T. Tull et al.

Anxiety and illicit drug use disorder comorbidity in the context of a mood
disorder. It is important to recognize that anxiety disorders are also likely to co-
occur with other disorders, especially mood disorders. Therefore, it will be
important for future studies to begin to examine the impact of mood-anxiety
disorder comorbidity on drug use behavior. Speaking to this potential impact of
mood-anxiety disorder comorbidity on drug use, Goodwin et al. (2002) exam-
ined the relationship between anxiety and drug use disorders among a sample of
130 individuals with a severe affective disorder (recurrent major depression,
bipolar disorder) to examine whether the presence of a comorbid anxiety
disorder diagnoses increases risk for the presence of a drug use disorder. They
found that, in general, among individuals with a severe affective disorder,
having an anxiety disorder was associated with a significantly increased risk
of a cocaine, sedative, stimulant, or opioid use disorder. In examining specific
anxiety disorders, they found that a) the experience of panic attacks increased
the likelihood of having a cocaine, sedative, stimulant, or opioid use disorder; b)
the presence of an OCD diagnosis increase the likelihood of stimulant use
disorder; c) a diagnosis of SAD increased risk for a sedative use disorder; and
d) and a specific phobia diagnosis was associated with increased risk for a
cocaine, sedative, stimulant, and opioid use disorder.

Temporal Order of Anxiety Disorders and Illicit Drug Use

Many of the studies discussed above are cross-sectional in nature, and

therefore, it is impossible to determine the temporal progression of the
disorders. In regard to the specific relationship between anxiety and drug
use disorders, three temporal paths are possible. First, the use of specific
drugs may increase the risk for the onset of an anxiety disorder. Second,
illicit drug use may follow the development of an anxiety disorder diagnosis,
consistent with a self-medication model of illicit drug use. Finally, there may
be an underlying third factor that increases the risk for both anxiety dis-
orders and illicit drug use. That is, there may be a common underlying
mechanism for the comorbid development of these disorders (Goodwin
et al., 2002).3

It is of note that a fourth possibility also exists. Specifically, anxiety disorder and drug use
disorder diagnoses may develop through a completely independent process. Goldenberg et al.
(1995) examined 181 participants in the Harvard Anxiety Research Project who had a history
of substance use disorders and anxiety disorders in order to test hypotheses pertaining to the
chronology of substance use-anxiety disorder comorbidity. They found, among individuals
with a primary anxiety disorder diagnosis, the anxiety disorder diagnosis was present for, on
average, 11.6 years before the onset of the substance use disorder. Further, substance use
tended to occur a mean of 9.6 years before the onset of a secondary anxiety disorder diagnosis.
Given the amount of time between the onset of the diagnoses, the authors concluded that there
is not an etiological connection between the disorders, but instead, the onset of these disorders
are guided by independent processes.
Drug Use and the Anxiety Disorders 65

Illicit Drug Use Precedes Anxiety Disorder Onset

In regard to the first hypothesis, Zvolensky, Bernstein et al. (2006) found that a
lifetime history of marijuana dependence was significantly associated with an
increased risk for panic attacks, even when controlling for polysubstance use,
alcohol abuse, and relevant demographic variables. Further, Cottler, Compton,
Mager, Spitznagel and Janca (1992) found evidence for the high-risk hypothesis
of comorbid PTSD-substance use disorder development. This hypothesis essen-
tially states that drug use increases the likelihood of traumatic exposure (due to
the high risk behaviors drug use is often associated with), thereby increasing
risk for PTSD. Using data from the St. Louis Epidemiologic Catchment Area
study, Cottler et al. (1992) examined the order of onset of PTSD and drug use
within the 2,663 respondents. Cottler et al. (1992) found that drug use tended to
precede the development of PTSD. However, it is important to note, that rates
of PTSD in this sample were low (1.35% overall).
In another study, Compton, Cottler, Phelps, Abdallah, and Spitznagel
(2000) examined the temporal relationship of drug dependence and DSM-III-
R psychiatric diagnoses among 425 individuals in drug treatment. In the major-
ity of cases (90%), drug dependence was found to follow the onset of a phobic
disorder (the authors did not indicate whether this referred to SAD, specific
phobia, or both). However, among individuals with a diagnosis of GAD, in
65% of the cases, the onset of the anxiety disorder followed the onset of the drug
dependence, suggesting that, at least with GAD, the diagnosis may be second-
ary to drug dependence.

Illicit Drug Use Follows Anxiety Disorder Onset

Studies have also provided support for the hypothesis that illicit drug use
follows an anxiety disorder diagnosis. This hypothesis has most extensively
been studied in PTSD. For example, Chilcoat and Breslau (1998) found that a
PTSD diagnosis greatly increased the risk for the subsequent development of a
SUD; however, traumatic exposure not resulting in PTSD did not have this
effect, suggesting that the relationship between traumatic exposure and sub-
stance use is unique to those individuals who develop PTSD. They also found
that substance abuse/dependence did not increase risk for traumatic exposure
or PTSD (thus not providing support for the high-risk hypothesis). Consistent
with a self-medication model of illicit drug use where drug use develops in an
attempt to alleviate anxiety disorder symptoms, it has been demonstrated that
exposure to trauma cues among PTSD-SUD patients is associated with sub-
stance craving, suggesting that the experience of PTSD-related symptoms may
increase motivation to use substances in an attempt to relieve those symptoms
(Coffey et al., 2002; Saladin et al., 2003).
66 M. T. Tull et al.

Additional support for the hypothesis that the presence of an anxiety dis-
order increases risk for illicit drug use comes from findings of Lopez et al. (2005)
who found that, within their sample of 1747 young adults, the mean age of onset
for an anxiety disorder diagnosis was consistently lower than that for substance
dependence. In addition, an anxiety disorder diagnosis tended to occur before
the onset of substance dependence 80% of the time.

Common Third Variables that may Underlie Both Illicit Drug Use
and Anxiety Disorders

Much less research has examined the third hypothesis of third variables that
may function as a common underlying mechanism for both anxiety disorders
and illicit drug use, especially in regard to the identification of common neuro-
biological and psychological mechanisms for the relationship between anxiety
and drug use disorders. Two variables that are worthy of strong consideration
are neurobiology and individual difference variables, which we will review
below. First, however, it is also important to note that another important
third variable for future consideration is gene by environment interactions.
Although no current research has directly examined shared genetic transmis-
sion of anxiety and illicit drug use disorders, this is likely an important area for
further pursuit. Specifically, the interaction between life stress and a poly-
morphism in the regulatory region of the serotonin transporter gene may be a
particularly promising area of study. Indeed, although the preponderance of the
work in this area has focused on the development of depression, some emerging
research has separately identified the role of this interaction in the development
of anxiety (Kendler, 1996) and illicit drug use (see Kreek, Nielsen, Butelman, &
LaForge, 2005).
Neurobiology. One aspect of neurobiology that may be potentially relevant
to the development and maintenance of both anxiety disorders and illicit drug
use involves the hypothalamic-pituitary-adrenal (HPA) axis, which controls the
secretion of hormones for the pituitary and adrenal cortex. The HPA axis plays
a central role in mediating the body’s response to stress and anxiety and is
extremely sensitive to inputs from the limbic system and prefrontal cortex, two
brain areas that are important in modulating reinforcement and motivational
processes. The activity of the HPA axis is reflected in changes in serum or
salivary cortisol levels (de Kloet & Reul, 1987), and has been found to account
for differences in stress reactivity to physical and psychological laboratory
challenges (Kaye et al., 2004; Wetherell et al., 2006). Neurobiological models
of drug addiction hypothesize that dysregulated HPA axis functioning contri-
butes to a state of chronic deviation of the regulatory system from its normal
operating level, resulting in increased reinforcing effects of illicit drugs (Koob &
Le Moal, 2001). More specifically, the HPA and brain stem stress circuits are
hypothesized to be recruited in feed forward loops during response to stress,
Drug Use and the Anxiety Disorders 67

such that the corticotrophin releasing factor (CRF) in the extended amygdala
drives norepinephrine systems in the pons-medulla of the brain stem, which in
turn drives CRF in the extended amygdala. The extended amygdala is thus
hypothesized to play a key role in regulating the HPA axis (i.e., stress response)
and subsequent recruitment of negative reinforcement behavior (Koob &
Le Moal, 2006), which may be especially relevant if substances are used for
self-medication of anxiety symptoms. As much of this work is focused more on
the relationship between stress and illicit drug use, as opposed to anxiety
disorders specifically, it will be important for additional work to target the
relationship between anxiety disorders and illicit drug use specifically. It is
important to note, though, that dysregulation of the HPA axis has been
found to play a role in the anxiety disorders (for a review, see Risbrough &
Stein, 2006) and especially in PD (e.g., Erhardt et al., 2006) and PTSD (e.g., de
Kloet et al., 2006; Yehuda, 2001) – two anxiety disorders that have been found
to frequently co-occur with illicit drug use, suggesting that dysregulated func-
tioning of the HPA axis may be common pathway at least for PD or PTSD and
illicit drug use. However, further research is needed to test this hypothesis.
Anxiety sensitivity. Anxiety sensitivity (AS) is an individual difference vari-
able representing the tendency to fear anxiety-related sensations (e.g., increased
heart rate, shortness of breath) due to beliefs that they will have negative
somatic, cognitive, or social consequences (Reiss, 1991). Reiss (1991) originally
proposed AS as a predisposing personality factor for the pathogenesis of the
anxiety disorders, and research supports this, as demonstrated through the
finding of elevated levels of AS across the anxiety disorders, with the exception
of specific phobia (see Cox, Borger, & Enns, 1999). Although AS was originally
proposed as a vulnerability factor for the anxiety disorders in general, its role as
a specific vulnerability factor for PD has been examined most extensively.
Studies consistently find higher levels of AS among individuals with PD,
compared to individuals with other anxiety disorders (with the exception of
PTSD) and healthy controls (see Cox et al., 1999). Further, AS has been
predictive of the later development of spontaneous panic attacks (e.g., Schmidt,
Lerew, & Jackson, 1997, 1999), as well as fearful responding to biological
challenge tasks, such as hyperventilation and CO2 inhalation (Donnell &
McNally, 1990; Harrington, Schmidt, & Telch, 1996; Rapee & Medoro, 1994;
Schmidt & Telch, 1994; Telch, Silverman, & Schmidt, 1996; see also Zvolensky
& Eifert, 2001, for a review). Similarly, successful treatment of PD has been
found to be associated with corresponding reductions in AS following a
cognitive behavioral group for PD (Telch et al., 1993) and individual cognitive
behavior therapy (CBT) for anxiety medication discontinuation (Bruce,
Spiegel, Gregg, & Nuzzarello, 1995).
Research exploring AS as an underlying vulnerability for psychopathology
in general has also been conducted, producing evidence that AS may underlie
other psychiatric conditions as well, including depression (e.g., Otto, Pollack,
Fava, Uccello, & Rosenbaum, 1995; Taylor, Koch, Woody, & McLean, 1996;
Tull & Gratz, in press; Tull, Gratz, & Lacroce, 2006), borderline personality
68 M. T. Tull et al.

disorder (Gratz, Tull, & Gunderson, in press), and certain types of drug use
patterns (see Lejuez, Paulson, Daughters, Bornovalova, & Zvolensky, 2006;
Otto, Safren, & Pollack, 2004; Stewart & Kushner, 2001; Zvolensky & Schmidt,
2004). In regard to AS and drug use in particular, across the substance use
disorders, McNally (1996) predicted that those individuals high in AS should
specifically be at risk for drugs that include anxiolytic or depressive psycho-
pharmacological effects as opposed to those with arousal properties (i.e.,
stimulants). However, research examining the role of AS in drug use is in its
infancy, and across substances, alcohol and tobacco use have been most exten-
sively studied (see chapters in this book; for reviews, see also Stewart, Samoluk,
& MacDonald, 1999; Zvolensky et al., 2003). Few studies have been conducted
that specifically examine the relationship between AS and other drug use.
However, there is some evidence to suggest that elevated levels of AS may be
associated with certain drug classes (besides tobacco and alcohol).
Consistent with the relationship between arousal dampening drugs and AS,
individuals reporting both chronic back pain and high AS report greater use of
analgesic medications (Asmundson & Norton, 1995), and heightened AS has
been found to be associated with elevated use of anxiety medications in general
(Telch, Lucas, & Nelson, 1989). Bruce et al. (1995) examined successful benzo-
diazepine (Alprazolam) medication discontinuation in individuals receiving
treatment as usual or the same procedure together with CBT treatment.
While CBT significantly decreased anxiety and depression compared to control
at 6-month follow-up, across both groups successful drug discontinuation was
predicted by AS reduction from baseline to post-treatment. While McNally’s
prediction has been generally supported regarding arousal dampening effects
(but see Forsyth, Parker, & Finlay, 2003), the prediction that individuals with
high AS should avoid arousal related drugs have not been confirmed to date.
No differences have been found in preferences for illicit stimulants (e.g., cocaine
or amphetamines) between high and low AS individuals (Norton et al., 1997;
DeHaas, Calamari, Bair, & Martin, 2001; Forsyth et al., 2003) or for other
stimulants such as caffeine (Stewart, Karp, Pihl, & Peterson, 1997). These
findings suggest that while AS may act as a risk factor for drugs with arousal
dampening effects, it does not act as a prophylactic against arousal producing
Research investigating the role of AS in marijuana use has provided mixed
results. Whereas marijuana use has been reported to correlate with high AS in
adolescents (Comeau, Stewart, & Loba, 2001), the opposite was found among
adults in which low AS correlated with marijuana use (Norton et al., 1997;
Stewart et al., 1997). In a recent investigation into adult tobacco users, mar-
ijuana users high in AS were at increased risk for more severe anxiety-related
symptoms (Zvolensky, Bonn-Miller et al., 2006), controlling for the effects of
cigarettes per day, alcohol use, and negative affectivity. In addition, AS has
been found to predict the severity of marijuana withdrawal symptoms among
young adult marijuana smokers, controlling for frequency of past 30-day
marijuana use, number of cigarettes smoked per day, alcohol consumption,
Drug Use and the Anxiety Disorders 69

anxious arousal, and anhedonic depressive symptoms. These conflicting results

indicate that future research is needed to better clarify the interactions between
AS, marijuana, and other variables such as age and co-occurrence with other
addictive substances. This research will aid in determining potential confound-
ing factors, as well as facilitate the identification of more complex relationships
between factors associated with vulnerability for illicit drug use and anxiety
disorders. Further, given evidence that AS may be associated with the severity
of marijuana withdrawal symptoms (Bonn-Miller, Zvolensky, Marshall, &
Bernstein, 2007), it will be important for future research to examine whether
AS may function to maintain drug use (due to fears of withdrawal symptoms)
and/or increase risk for relapse for certain drugs during periods of attempted
Limited research has also been conducted that investigates the relationship
between AS and heroin use. Lejuez et al. (2006) investigated AS among pri-
marily African-American inner city substance users receiving treatment in a
residential treatment facility. Specifically, measures of AS were collected among
heroin users, crack/cocaine users, users of both heroin and crack/cocaine, and
those that used neither. It was found that even when controlling for demo-
graphic variables, depressive symptoms, and the use of other drugs (e.g.,
alcohol, marijuana), primary heroin users evidenced higher levels of AS than
all other groups, suggesting a unique relationship between heroin use and AS.
Elevated AS has also been found to be a risk factor for worse treatment
outcomes among heroin users in residential substance use treatment. Specifi-
cally, Lejuez, et al. (2007) examined AS as a predictor of residential substance
use treatment drop-out among heroin and crack/cocaine users. Heightened
levels of AS were found to predict residential substance use treatment drop-
out among heroin users but not crack/cocaine users.
Prospective studies are needed to determine the specific role AS may play in
the concurrent development of both anxiety and drug use disorders, as well as to
identify the particular mechanisms through which AS may lead to these dis-
orders. In regard to the pathogenesis of an anxiety disorder such as PD,
heightened AS may be associated with attempts to avoid anxiety-related phy-
sical symptoms, thereby preventing functional exposure to the feared stimulus
and maintaining anxious responding. Likewise, in regard to the pathogenesis of
drug use, motivational models acknowledge the influence of positive conse-
quences factoring into substance use yet posit a special role for the removal or
avoidance of negative internal invents especially among chronic users (Cooper,
1994; Simons, Correia, Carey, & Borsari, 1998; Simons, Gaher, Correia,
Hansen, & Christopher, 2005; Tate, Pomerleau, & Pomerleau, 1994). Baker,
Piper, McCarthy, Majeskie, and Fiore (2004) comprehensively reviewed the
basic experimental learning literature and concluded that negative reinforce-
ment remains the primary mechanism behind addiction. Of utmost interest to
the role of AS in addiction, Baker et al. (2004) argue that addiction is main-
tained by the avoidance and or escape of negative affect and other associated
interoceptive cues (i.e. cognitive processes) that signal withdrawal symptoms.
70 M. T. Tull et al.

This is consistent with Hayes and colleagues proposal to reconceptualize psy-

chopathology, including substance use, as a behavior that serves an experien-
tially avoidant function.
Experiential avoidance (defined as attempts to alter the form or frequency of
internal experience such as thoughts, emotions, and bodily sensations; Hayes,
Wilson, Gifford, Follette, & Strosahl, 1996) may be a particularly useful con-
struct to examine in attempting to understand the pathway through which
heightened AS leads to both the pathogenesis of anxiety disorders and illicit
drug use. Experiential avoidance has been found to be associated with AS (e.g.,
Zvolensky & Forsyth, 2002), as well as a variety of anxiety disorder diagnoses
and anxiety disorder-related symptoms (for reviews, see Hayes, Luoma, Bond,
Masuda, & Lillis, 2006; Salters-Pedneault, Tull, & Roemer, 2004), as well as
substance use behaviors (Hayes et al., 1996; Stewart, Zvolensky, & Eifert,
Impulsivity (delay discounting). Impulsivity, defined as delay discounting, may
also provide a useful framework for considering the concurrent development of
anxiety disorders and illicit drug use. Traditionally, delay discounting has served
as a behavioral operationalization for understanding the construct of impulsivity,
which was developed following concern regarding the prevailing trait conceptua-
lizations of the construct. Simply put, a delay discounting perspective considers
impulsive behavior to be the choice of a smaller, immediate reinforcer instead of a
larger, delayed reinforcer (Ainslie, 1975; Rachlin & Green, 1972). Taking this
approach, studies have shown that compared to non-users, illicit drug users are
more likely to select smaller amounts of immediate money instead of larger
amounts that are delayed (e.g., $5 today over $20 in 30 days; Coffey, Gudleski,
Saladin, & Brady, 2003; Kirby, Petry, & Bickel, 1999). In this way, this procedure
provides an analogue for real world behavior of illicit drug users where more high
impact yet delayed long-term goals (e.g., career development, relationship build-
ing) are sacrificed in favor of immediate gain in the form of drug use.
Complimenting this approach to reinforcers, delay discounting also may be
applied in its reciprocal form to aversive stimuli. In this case impulsivity
involves the selection of a larger, delayed aversive stimulus over a smaller, yet
immediate aversive stimulus. Said differently, this describes the tendency to put
off experiencing an aversive event, even though this delay will result in an
exacerbation of the negative consequences. A theoretical account of the rele-
vance of this reciprocal approach to delayed discounting for SAD was provided
by McNeil, Lejuez, and Sorrell (2001), describing the tendency of individuals
with SAD to avoid potentially uncomfortable social interactions despite acute
awareness that this will result in considerably larger long-term consequences.
This approach is easily applied to other anxiety disorders as well. For example
in the case of PD with agoraphobia, an individual may choose to only leave
home with a ‘‘safe’’ person due to fear that of what will happen if alone in public
during a panic attack, thereby resulting in the more severe yet delayed con-
sequences associated with a loss of independence in the service of avoiding
smaller yet immediate discomfort associated with leaving home alone.
Drug Use and the Anxiety Disorders 71

Combining both the reinforcer and reciprocal aversive conceptualizations of

delay discounting provides a unique approach for understanding anxiety
disorder-illicit drug use comorbidity. This may be particularly powerful in
individuals where drug use is not only positively reinforcing (drug induced
euphoria, social interaction) but also highly negatively reinforcing in regard to
anxiety (the removal of overwhelming anxiety). It is also of note that although
discounting is typically considered in terms of delay as discussed here, it is quite
difficult to have a delay without also adding a degree of uncertainty. In this way,
the extent to which delayed positive reinforcers and delayed aversives also are
not perceived as guaranteed, their impact on keeping an individual from making
an impulsive choice is further diminished. For example, while benefits of sobri-
ety participating in exposure-based anxiety treatments seem strikingly powerful
(good health, job security, trust, family), the delay to attaining the rewards are
often long and relatively uncertain (will trust ever be fully returned? Will
employers give you the benefit of the doubt if something goes missing? Will
the anxiety ever really go away?), while pain (anxiety, withdrawal) remains
immediate and certain. Thus, delayed discounting provides a framework to
conceptualize the myriad of environmental factors influencing preference
between two alternatives and suggests intervention implications (i.e., increasing
the immediacy and certainty of alternative behaviors).

Treatment for Anxiety and Drug Use Disorder Comorbidity

Given the high rates of comorbidity between anxiety disorders and substance
use disorders, as well as suggestions that comorbid anxiety and substance use
disorders may be better characterized as a single unique disorder (Hien, Cohen,
Miele, Litt, & Capstick, 2004; Morissette et al., 2007), specialized treatments
designed to specifically target this comorbidity are beginning to be developed.
Of those treatments that are available, the majority are focused on the comor-
bidity between PTSD and substance use disorders.
Seeking Safety. Najavits (2002) developed Seeking Safety to specifically
target comorbid PTSD and substance use disorders. Seeking Safety is a
24-session cognitive behavioral group therapy protocol treatment that teaches
individuals with this comorbid symptom presentation a variety of cognitive,
behavioral, and interpersonal skills particularly applicable to individuals with
both PTSD and substance use difficulties, all of which are designed with the
idea that safety is the top priority in recovery from each disorder (Najavits,
Weiss, & Liese, 1996). That is, coping skills are focused on maintaining absti-
nence, reducing self-destructive and high-risk behavior, and establishing sup-
port. Seeking Safety has been found to be effective, with patients exhibiting
significant reductions in substance use behavior, trauma-related symptoms,
suicide risk, suicidal thoughts, depression, and thoughts about substance use,
and improvements in social adjustment, family functioning, and problem
72 M. T. Tull et al.

solving (e.g., Hien et al., 2004; Najavits et al., 1996; Zlotnick, Najavits, Rohse-
now, & Johnson, 2003).
Concurrent Treatment of PTSD and Cocaine Dependence. Another treatment
specifically designed for individuals with comorbid PTSD and substance use
disorders is Back, Dansky, Carroll, Foa, and Brady’s (2001) Concurrent Treat-
ment of PTSD and Cocaine Dependence (CTPCD). This treatment consists of
16 individual 90-minute sessions. The treatment was designed by integrating
previously validated cognitive behavioral treatments for substance dependence
(Carroll, 1998) and PTSD (Foa, Rothbaum, Riggs, & Murdock, 1991; Foa &
Rothbaum, 1998). CTPCD involves psychoeducation on the link between
PTSD and cocaine dependence, coping skills training, relapse prevention skills,
and cognitive restructuring. Further, patients undergo in-vivo and imaginal
exposure in order to address their PTSD symptoms. In an initial examination of
CTPCD, Brady, Dansky, Back, Foa, and Carroll (2001) found that individuals
who completed treatment evidenced significant reductions in depressive symp-
toms, PTSD symptoms, and cocaine use severity.
Anxiety Sensitivity Treatment for Heroin Users. Targeting AS, as opposed to
any specific disorder, we (Tull, Schulzinger, Schmidt, Zvolensky, & Lejuez, 2007)
recently developed an acceptance-based behavioral treatment (the Anxiety Sensi-
tivity Treatment for Heroin Users; AST-H) meant to be used in conjunction with
standard substance abuse treatment. Our treatment was designed to have specific
relevance for heightened heroin users with heightened AS. Previous research
(Lejuez et al., 2006) has demonstrated that heightened AS may increase risk for
substance use treatment drop-out among heroin users. In particular, a tendency to
fear of and unwillingness to have anxiety-related sensations may prompt indivi-
duals to attempt to avoid these sensations through the use of heroin. Therefore, we
developed a six session adjunctive treatment where individuals engage in inter-
ceptive exposure exercises in order to facilitate acceptance of, and tolerance for,
aversive internal sensations, with the goal of preventing the use of heroin for self-
medication of these sensations. In an initial examination of this treatment’s
effectiveness, the AST-H was found to result in reductions in AS, heroin cravings,
avoidance behavior, and emotion dysregulation (Tull et al., 2007). Improvements
were maintained when measured over one month post-treatment. Current efforts
are underway to replicate this finding within a randomized controlled trial.


Recent years have seen a rise in studies examining the co-occurrence of illicit
drug use and anxiety disorder diagnoses, and these studies provide convincing
evidence that illicit drug use is prevalent among individuals with anxiety dis-
order diagnoses. Further, the impact of this co-occurrence on psychological
functioning and physical health is clear. Yet, although studies have begun to
address the mechanisms underlying this co-occurrence, all anxiety disorders
Drug Use and the Anxiety Disorders 73

and forms of illicit drug use have not been examined equally. Therefore, future
research is needed to better understand the functional relationship between all
anxiety disorders and forms of illicit drug use, with the goal of informing the
development of novel and targeted interventions for this comorbidity, as well as
prevention efforts.


Ainslie, G. (1975). Specious reward: A behavioral theory of impulsiveness and impulse

control. Psychological Bulletin, 82, 463–496.
American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders
(4th ed.). Washington, DC: Author.
Asmundson, G. J., & Norton, G. R. (1995). Anxiety sensitivity in patients with physically
unexplained chronic back pain: A preliminary report. Behaviour Research and Therapy, 33,
Back, S. E., Dansky, B. S., Carroll, K. M., Foa, E. B., & Brady, K. T. (2001). Exposure
therapy in the treatment of PTSD among cocaine-dependent individuals: Description of
procedures. Journal of Substance Abuse Treatment, 21, 35–45.
Baker, T. B., Piper, M. E., McCarthy, D. E., Majeskie, M. R., & Fiore, M. C. (2004).
Addiction motivation reformulated: An affective processing model of negative reinforce-
ment. Psychological Review, 111, 33–51.
Benedict, J., Evans, W., & Calder, J. C. (1999). An exploratory study of recreational drug use
and nutrition-related behaviors and attitudes among adolescents. Journal of Drug Educa-
tion, 29, 139–155.
Biederman, J., Petty, C., Faraone, S. V., Hirshfeld-Becker, D. R., Henin, A., Pollack, M. H.,
et al. (2005). Patterns of comorbidity in panic disorder and major depression: Findings
from a nonreferred sample. Depression and Anxiety, 21, 55–60.
Bonn-Miller, M. O., Bernstein, A., Sachs-Ericsson, N., Schmidt, N. B., & Zvolensky, M. J.
(2007). Associations between psychedelic use, abuse, and dependence and lifetime panic
attack history in a representative sample. Journal of Anxiety Disorders, 21, 730–741.
Bonn-Miller, M. O., Zvolensky, M. J., Marshall, E. C., & Bernstein, A. (2007). Incremental
validity of anxiety sensitivity in relation to marijuana withdrawal symptoms. Addictive
Behaviors, 32, 1843–1851.
Brady, K. T., Dansky, B. S., Back, S. E., Foa, E. B., & Carroll, K. M. (2001). Exposure
therapy in the treatment of PTSD among cocaine-dependent individuals: Preliminary
findings. Journal of Substance Abuse Treatment, 21, 47–54.
Brook, J. S., Adams, R. W., Balka, E. B., Whiteman, M., Zhang, C., & Sugerman, R. (2004).
Illicit drug use and risky sexual behavior among African American and Puerto Rican
urban adolescents: The longitudinal links. Journal of Genetic Psychology, 165, 203–220.
Bruce, T. J., Spiegel, D. A., Gregg, S. F., & Nuzzarello, A. (1995). Predictors of alprazolam
discontinuation with and without cognitive behavior therapy in panic disorder. American
Journal of Psychiatry, 152, 1156–1160.
Buckner, J. D., Mallott, M. A., Schmidt, N. B., & Taylor, J. (2006). Peer influence and gender
differences in problematic cannabis use among individuals with social anxiety. Journal of
Anxiety Disorders, 20, 1087–1102.
Buckner, J. D., Schmidt, N. B., Bobadilla, L., & Taylor J. (2006). Social anxiety and proble-
matic cannabis use: Evaluating the moderating role of stress reactivity and perceived
coping. Behaviour Research and Therapy, 44, 1007–1015.
Calhoun, P. S., Sampson, W. S., Bosworth, H. B., Feldman, M. E,. Kirby, A. C.,
Hertzberg, M. A. et al. (2000). Drug use and validity of substance use self-reports in
74 M. T. Tull et al.

veterans seeking help for posttraumatic stress disorder. Journal of Consulting and Clinical
Psychology, 68, 923–927.
Carroll, K. M. (1998). A cognitive-behavioral approach: Treating cocaine addiction. (pp.
98–4308). Rockville, MD: National Institute on Drug Abuse, NIH Publication.
Chilcoat, H. D., & Breslau, N. (1998). Posttraumatic stress disorder and drug disorders.
Archives of General Psychiatry, 55, 913–917.
Chilcoat, H. D., & Menard, C. (2003). Epidemiological investigations: Comorbidity of post-
traumatic stress disorder and substance use disorder. In P. Ouimette and P. J. Brown
(Eds.), Trauma and Substance Abuse: Causes, Consequences, and Treatment of Comorbid
Disorders (pp. 9–28). Washington, DC: American Psychological Association.
Coffey, S. F., Gudleski, G. D., Saladin, M. E., & Brady, K. T. (2003). Impulsivity and rapid
discounting of delayed hypothetical rewards in cocaine-dependent individuals. Experi-
mental & Clinical Psychopharmacology, 11, 18–25.
Coffey, S. F., Saladin, M. E., Drobes, D. J., Brady, K. T., Dansky, B. S., & Kilpatrick, D. G.
(2002). Trauma and substance cue reactivity in individuals with comorbid posttraumatic
stress disorder and cocaine or alcohol dependence. Drug and Alcohol Dependence, 65,
Comeau, N., Stewart, S. H., & Loba, P. (2001). The relations of trait anxiety, anxiety
sensitivity, and sensation seeking to adolescents’ motivations for alcohol, cigarette, and
marijuana use. Addictive Behaviors, 26, 803–825.
Compton, W. M., Cottler, L. B., Phelps, D. L., Abdallah, A. B., & Spitznagel, E. L. (2000).
Psychiatric disorders among drug dependent subjects: Are they primary or secondary? The
American Journal on Addictions, 9, 126–134.
Conway, K. P., Compton, W., Stinson, F. S., & Grant, B. F. (2006). Lifetime comorbidity of
DSM-IV mood and anxiety disorders and specific drug use disorders: Results from the
National Epidemiologic Survey on Alcohol and Related Conditions. Journal of Clinical
Psychiatry, 67, 247–257.
Cooper, M. L. (1994). Motivations for alcohol use among adolescents: Development and
validation of a four-factor model. Psychological Assessment, 6, 117–128.
Cottler, L. B., Compton, W. M., Mager, D., Spitznagel, E. L., & Janca, A. (1992). Posttrau-
matic stress disorder among substance users from the general population. American
Journal of Psychiatry, 149, 664–670.
Cox, B. J., Borger, S. C., & Enns, M. W. (1999). Anxiety sensitivity and emotional disorders:
Psychometric studies and their theoretical implications. In S. Taylor (Ed.), Anxiety sensi-
tivity: Theory, research, and treatment of the fear of anxiety (pp. 115–148). Mahwah, NJ:
Lawrence Erlbaum.
de Kloet, E. R., & Reul, J. M. (1987). Feedback action and tonic influence of corticosteroids
on brain function: A concept arising from the heterogeneity of brain receptor systems.
Psychoneuroendocrinology, 12, 83–105.
de Kloet, C. S., Vermetten, E. Geuze, E., Kavelaars, A., Heijnen, C. J., & Westenberg, H. G. M.
(2006). Assessment of HPA-Axis functioning in posttraumatic stress disorder: Pharmaco-
logical and non-pharmacological challenge tests, a review. Journal of Psychiatric Research,
40, 550–567.
Deacon, B. J., & Valentiner, D. P. (2000). Substance use and non-clinical panic attacks in a
young adult sample. Journal of Substance abuse, 11, 7–15.
DeHaas, R. A., Calamari, J. E., Bair, J. P., & Martin, E. D. (2001). Anxiety sensitivity and
drug or alcohol use in individuals with anxiety and substance use disorders. Addictive
Behaviors, 26, 787–801.
Donnell, C. D., & McNally, R. J. (1990). Anxiety sensitivity and panic attacks in a nonclinical
population. Behavior Research and Therapy, 28, 83–85.
Erhardt, A., Ising, M., Unschuld, P. G., Kern, N., Lucae, S., Pütz, B. et al. (2006). Regulation
of the hypothalamic-pituitary-adrenocortical system in patients with panic disorder.
Neuropsychopharmacology, 31, 2515–2522.
Drug Use and the Anxiety Disorders 75

Foa, E. B., & Rothbaum, B. O. (1998). Treating the Trauma of Rape: Cognitive-behavioral
Therapy for PTSD. New York: Guilford Press.
Foa, E. B., Rothbaum, B. O., Riggs, D. S., & Murdock, T. B. (1991). Treatment of posttrau-
matic stress disorder in rape victims: A comparison between cognitive-behavioral proce-
dures and counseling. Journal of Consulting and Clinical Psycholoy, 59, 715–723.
Forsyth, J. P., Parker, J., & Finlay, C. G. (2003). Anxiety sensitivity, controllability, and
experiential avoidance and their relation to drug of choice and addiction severity in a
residential sample of substance abusing veterans. Addictive Behaviors, 28, 851–870.
Giaconia, R. M., Reinherz, H. Z., Hauf, A. C., Paradis, A. D., Wasserman, M. S., & Langer
hammer, D. M. (2000). Comorbidity of substance use and posttraumatic stress disorders in a
community sample of adolescents. American Journal of Orthopsychiatry, 70, 253–262.
Giaconia, R. M., Reinherz, H. Z., Silverman, A. B., Pakiz, B., Frost, A. K., & Cohen, E.
(1995). Traumas and posttraumatic stress disorder in a community population of older
adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 34,
Goldenberg, I. M., Mueller, T., Fierman, E. J., Gordon, A., Pratt, L., Cox, K. et al. (1995).
Specificity of substance use in anxiety-disordered subjects. Comprehensive Psychiatry, 36,
Goodwin, R. D., Fergusson, D. M., & Horwood, L. J. (2004). Association between anxiety
disorders and substance use disorders among young persons: Results of a 21-year long-
itudinal study. Journal of Psychiatric Research, 38, 295–304.
Goodwin, R. D., Stayner, D. A., Chinman, M. J., Wu, P., Kraemer Tebes, J., & Davidson, L.
(2002). The relationship between anxiety and substance use disorders among individuals
with severe affective disorders. Comprehensive Psychiatry, 43, 245–252.
Grant, B. F., Stinson, F. S., Dawson, D. A., Chou, S. P., Dufour, M. C., Compton, W., et al.
(2004). Prevalence and co-occurrence of substance use disorders and independent mood
and anxiety disorders. Archives of General Psychiatry, 61, 807–816.
Gratz, K. L., Tull, M. T., & Gunderson, J. G. (in press). The relationship between anxiety
sensitivity and borderline personality disorder: The role of experiential avoidance. Journal
of Psychiatric Research.
Harrington, P. J., Schmidt, N. B., & Telch, M. J. (1996). Prospective evaluation of panic
potentiation following 35% CO2 challenge in nonclinical subjects. American Journal of
Psychiatry, 153, 823–825.
Hayes, S. C., Luoma, J. B., Bond, F. W., Masuda, A., & Lillis, J. (2006). Acceptance and
commitment therapy: Model, processes and outcomes. Behaviour Research and Therapy,
44, 1–25.
Hayes, S. C., Wilson, K. G., Gifford, E. V., Follette, V. M., & Strosahl, K. D. (1996).
Emotional avoidance and behavioral disorders: A functional dimensional approach to
diagnosis and treatment. Journal of Consulting and Clinical Psychology, 64, 1152–1168.
Helzer, J. E., Robins, L. N., & McEvoy, L. (1987). Post-traumatic stress disorder in the
general population: Findings of the Epidemiological Catchment Area Survey. New
England Journal of Medicine, 317, 1630–1634.
Hien, D. A., Cohen, L. R., Miele, G. M., Litt, L. C., & Capstick, C. (2004). Promising
treatments for women with comorbid PTSD and substance use disorders. American
Journal of Psychiatry, 161, 1426–1432.
Kaye, J., Buchanan, F., Kendrick, A., Johnson, P., Lowry, C., Bailey, J. et al. (2004). Acute
carbon dioxide exposure in healthy adults: Evaluation of a novel means of investigating
the stress response. Journal of Neuroendocrinology, 16, 256–264.
Kendler, K. S. (1996). Major depression and generalised anxiety disorder. Same genes, (partly)
different environments–revisited. British Journal of Psychiatry, 168 (Suppl. 30), 68–75.
Kessler, R. C., Berglund, P., Demler, O., Jin, R., & Walters, E. E. (2005). Lifetime prevalence
and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey
Replication. Archives of General Psychiatry, 62, 593–602.
76 M. T. Tull et al.

Kessler, R. C., Chiu, W. T., Demler, O., & Walters, E. E. (2005). Prevalence, severity, and
comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Repli-
cation. Archives of General Psychiatry, 62, 617–627.
Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S., et al.
(1994). Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the
United States: Results from the National Comorbidity Survey. Archives of General
Psychiatry, 51, 8–19.
Kessler, R. C., Nelson, C. B., McGonagle, K. A., Edlund, M. J., Frank, R. G., & Leaf, P. J.
(1996). The epidemiology of co-occurring addictive and mental disorders: Implications for
prevention and service utilization. American Journal of Orthopsychiatry, 66, 17–31.
Kessler, R. C., Sonnega, A., Bromet, E., Hughes, M., & Nelson, C. B. (1995). Posttraumatic
stress disorder in the National Comorbidity Survey. Archives of General Psychiatry, 52,
Kilpatrick, D. G., Acierno, R., Saunders, B., Resnick, H. S., Best, C. L., & Schnurr, P. P.
(2000). Risk factors for adolescent substance abuse and dependence: Data from a national
sample. Journal of Consulting and Clinical Psychology, 68, 19–30.
Kirby, K. N., Petry, N. M., & Bickel, W. K. (1999). Heroin addicts have higher discount rates
for delayed rewards than non-drug-using controls. Journal of Experimental Psychology:
General, 128, 78–87.
Kreek, M., Nielsen, D., Butelman, E., & LaForge, K. (2005). Genetic influences on impulsiv-
ity, risk taking, stress responsivity and vulnerability to drug abuse and addiction. Nature
Neuroscience, 8, 1450–1457.
Koob, G. F., & Le Moal, M. (2001). Drug addiction, dysregulation of reward, and allostasis.
Neuropsychopharmacology, 24, 97–129.
Lejuez, C. W., Paulson, A., Daughters, S. B., Bornovalova, M. A., & Zvolensky, M. J. (2006).
The association between heroin use and anxiety sensitivity among inner-city individuals in
residential drug use treatment. Behaviour Research and Therapy, 44, 667–677.
Lejuez, C. W., Zvolenksy, M. J., Daughters, S. B., Bornovalova, M. A., Paulson, A., Tull, M. T.
et al. (2007). Anxiety sensitivity: A unique predictor of drop-out among inner-city heroin and
crack/cocaine users in residential substance use treatment. Manuscript submitted for
Lindquist, C. U., Lindsay, J. S., & White, G. D. (1979). Assessment of assertiveness in drug
abusers. Journal of Clinical Psychology, 35, 676–679.
Lopez, B., Turner, R. J., & Saavedra, L. M. (2005). Anxiety and risk for substance dependence
among late adolescents/young adults. Journal of Anxiety Disorders, 19, 275–294.
Loxley, W., Bevan, J., & Carruthers, S. (1998). Sex, gender, drugs, and risk: The Australian
study of HIV and injecting drug use. International Journal of Drug Policy, 9, 255–262.
Lynskey, M. T., Heath, A. C., Nelson, E. C., Bucholz, K. K., Madden, P. A. F., Slutske, W. S.
et al. (2002). Genetic and environmental contributions to cannabis dependence in a
national young adult twin sample. Psychological Medicine, 32, 195–207.
McNally, R. J. (1996). Anxiety sensitivity is distinct from trait anxiety. In R. M. Rapee (Ed.),
Current controversies in the anxiety disorders (pp. 214–227). New York, NY: Guilford.
McNeil, D. W., Lejuez, C. W., & Sorrell, J. T. (2001). Behavioral theories of social phobia:
Contributions of basic behavioral principles. In S. G. Hofmann & P. M. DiBartolo (Eds.),
From social anxiety to social phobia: Multiple perspectives (pp. 235–253). Boston, MA:
Allyn & Bacon.
Merikangas, K. R., Mehta, R. L., Molnar, B. E., Walters, E. E., Swendsen, J. D.,
Aguilar-Gaziola, S., et al. (1998). Comorbidity of substance use disorders with mood and
anxiety disorders: Results of the International Consortium in Psychiatric Epidemiology.
Addictive Behaviors, 23, 893–907.
Morabia, A., Fabre, J., Chee, E., Zeger, S, Orsat, E., & Robert, A. (1989). Diet and opiate
addiction: A quantitative assessment of the diet of non-institutionalized opiate addicts.
British Journal of Addiction, 84, 173–180.
Drug Use and the Anxiety Disorders 77

Morissette, S. B., Tull, M. T., Gulliver, S. B., Kamholz, B. W., & Zimering, R. T. (2007).
Anxiety, anxiety disorders, tobacco use, and nicotine: A critical review of interrelation-
ships. Psychological Bulletin, 133, 245–272.
Najavits, L. M. (2002). Seeking safety: A treatment manual for PTSD and substance abuse.
New York, NY: Guilford Press.
Najavits, L. M., Weiss, R. D., & Liese, B. S. (1996). Group cognitive-behavioral therapy for
women with PTSD and substance use disorder. Journal of Substance Abuse Treatment, 13,
Norton, G. R., Rockman, G. E., Ediger, J., Pepe, C., Goldberg, S., Cox, B. J., & Asmundson,
G. J. G. (1997). Anxiety sensitivity and drug choice in individuals seeking treatment for
substance abuse. Behaviour Research and Therapy, 35, 859–862.
Otto, M. W., Pollack, M. H., Fava, M., Uccello, R., & Rosenbaum, J. F. (1995). Elevated
Anxiety Sensitivity Index scores in patients with major depression: correlates and changes
with antidepressant treatment. Journal of Anxiety Disorders, 9, 117–123.
Otto, M. W., Safren, S. A., & Pollack, M. H. (2004). Internal cue exposure and the treatment
of substance use disorders: Lessons from the treatment of panic disorder. Journal of
Anxiety Disorders, 18, 69–87.
Pace-Schott, E. F., Stickgold, R., Muzur, A., Wigren, P. E., Ward, A. S., Hart, C. L. et al.
(2005). Sleep quality deteriorates over a binge-abstinence cycle in chronic smoked cocaine
users. Psychopharmacology, 179, 873–883.
Rachlin, H., & Green, L. (1972). Commitment, choice, and self-control. Journal of the
Experimental Analysis of Behavior, 17, 15–22.
Rapee, R., & Medoro, L. (1994). Fear of physical sensations and trait anxiety as mediators of
the response to hyperventilation in nonclinical subjects. Journal of Abnormal Psychology,
4, 693–699.
Reiger, D. A., Farmer, M. E., Rae, D. S., Locke, B. Z., Keith, S. J., Judd, L. L., et al. (1990).
Comorbidity of mental disorders with alcohol and other drug abuse. Journal of the
American Medical Association, 21, 2511–2518.
Reiss, S. (1991). Expectancy theory of fear, anxiety, and panic. Clinical Psychology Review,
11, 141–153.
Risbrough, V. B., & Stein, M. B. (2006). Role of corticotropin releasing factor in
anxiety disorders: A translational research perspective. Hormones and Behavior, 50,
Saladin, M. E., Drobes, D. J., Coffey, S. F., Dansky, B. S., Brady, K. T., & Kilpatrick, D. G.
(2003). PTSD symptom severity as a predictor of cue-elicited drug craving in victims of
violent crime. Addictive Behaviors, 28, 1611–1629.
Salters-Pedneault, K., Tull, M. T., & Roemer, L. (2004). The role of avoidance of
emotional material in the anxiety disorders. Applied and Preventive Psychology, 11,
Saxon, A. J., Davis, T. M., Sloan, K. L., McKnight, K. M., McFall, M. E., & Kivlahan, D. R.
(2001). Trauma symptoms of posttraumatic stress disorder and associated problems
among incarcerated veterans. Psychiatric Services, 52, 959–964.
Schmidt, N. B., Lerew, D. R., & Jackson, R. J. (1997). The role of anxiety sensitivity in the
pathogenesis of panic: Prospective evaluation of spontaneous panic attacks during acute
stress. Journal of Abnormal Psychology, 106, 355–364.
Schmidt, N. B., Lerew, D. R., & Jackson, R. J. (1999). Prospective evaluation of anxiety
sensitivity in the pathogenesis of panic: Replication and extension. Journal of Abnormal
Psychology, 108, 532–537.
Schmidt, N. B., & Telch, M. J. (1994). Role of fear of fear and safety information in
moderating the effects of voluntary hyperventilation. Behavior Therapy, 25, 197–208.
Simons, J. S., Correia, C. J., Carey, K. B., & Borsari, B. E. (1998). Validating a five-factor
marijuana motives measure: Relations with use, problems, and alcohol motives. Journal of
Counseling Psychology, 45, 265–273.
78 M. T. Tull et al.

Simons, J. S., Gaher, R. M., Correia, C. J., Hansen, C. L., & Christopher, M. S. (2005). An
affective-motivational model of marijuana and alcohol problems among college students.
Psychology of Addictive Behaviors, 19, 326–334.
Smith, S. M., Stinson, F. S., Dawson, D. A., Goldstein, R., Huang, B., & Grant, B. F.
(2006). Race/ethnic differences in the prevalence and co-occurrence of substance use
disorders and independent mood and anxiety disorders: Results from the National
Epidemiological Survey on Alcohol and Related Conditions. Psychological Medicine,
36, 987–998.
Stewart, S. H. (1996). Alcohol abuse in individuals exposed to trauma: A critical review.
Psychological Bulletin, 120, 83–112.
Stewart, S. H., Karp, J., Pihl, R. O., & Peterson, R. A. (1997). Anxiety sensitivity and self-
reported reasons for drug use. Journal of Substance Use, 9, 223–240.
Stewart, S. H., & Kushner, M. G. (2001). Introduction to the special issue on anxiety
sensitivity and addictive behaviors. Addictive Behaviors, 26, 775–785.
Stewart, S. H., Samoluk, S. B., & MacDonald, A. B. (1999). Anxiety sensitivity and substance
use and abuse. In S. Taylor (Ed.), Anxiety sensitivity: Theory, research, and treatment of the
fear of anxiety (pp. 287–319). Mahwah, NJ: Lawrence Erlbaum.
Stewart, S. H., Zvolensky, M. J., & Eifert, G. H. (2002). The relations of anxiety sensitivity,
experiential avoidance, and alexithymic coping to young adults’ motivations for drinking.
Behavior Modification, 26, 274–296.
Stinson, F. S., Grant, B. F., Dawson, D. A., Ruan, W. J., Huang, B., & Saha, T. (2005).
Comorbidity between DSM-IV alcohol and specific drug use disorders in the United
States: Results from the National Epidemiologic Survey on Alcohol and Related Condi-
tions. Drug and Alcohol Dependence, 80, 105–116.
Swadi, H., & Bobier, C. (2003). Substance use disorder comorbidity among inpatient youths
with psychiatric disorder. Australian and New Zealand Journal of Psychiatry, 37, 294–298.
Tarrier, N., & Sommerfield, C. (2003). Alcohol and substance use in civilian chronic PTSD
patients seeking psychological treatment. Journal of Substance Use, 8, 197–204.
Tate, J. C., Pomerleau, C. S., & Pomerleau, O. F. (1994) Pharmacological and non-pharma-
cological smoking motives: a replication and extension. Addiction, 89, 321–330.
Taylor, S., Koch, W. J., Woody, S., & McLean, P. (1996). Anxiety sensitivity and depression:
how are they related? Journal of Abnormal Psychology, 105, 474–479.
Telch, M. J., Lucas, J. A., & Nelson, P. (1989). Nonclinical panic in college students: An
investigation of prevalence and symptomatology. Journal of Abnormal Psychology, 98,
Telch, M. J., Lucas, J. A., Schmidt, N. B., Hanna, H. H., Jaimez, T. L., & Lucas, R. (1993).
Group cognitive-behavioral treatment of panic disorder. Behaviour Research and Therapy,
31, 279–287.
Telch, M. J., Silverman, A., & Schmidt, N. B. (1996). Effects of anxiety sensitivity and
perceived control on emotional responding to caffeine challenge. Journal of Anxiety
Disorders, 10, 21–35.
Tull, M. T., & Gratz, K. L. (in press). Further investigation of the relationship between
anxiety sensitivity and depression: The role of experiential avoidance and difficulties
engaging in goal-directed behavior when distressed. Journal of Anxiety Disorders.
Tull, M. T., Gratz, K. L., & Lacroce, D. M. (2006). The role of anxiety sensitivity and lack of
emotional approach coping in depressive symptom severity among a nonclinical sample of
uncued panickers. Cognitive Behaviour Therapy, 35, 74–87.
Tull, M. T., Schulzinger, D., Schmidt, N. B., Zvolensky, M. J., & Lejuez, C. W. (2007).
Development and initial examination of a brief intervention for heightened anxiety sensi-
tivity among heroin users. Behavior Modification, 31, 1–23.
Valentiner, D., Mounts, N., & Deacon, B. (2004). Panic attacks, depression and anxiety
symptoms, and substance use behaviors during late adolescence. Journal of Anxiety
Disorders, 18, 573–585.
Drug Use and the Anxiety Disorders 79

Wetherell, M. A., Crown, A. L., Lightman, S. L., Miles, J. N. V., Kaye, J., & Vedhara, K.
(2006). The four-dimensional stress test: Psychological, sympathetic-adrenal-medullary,
parasympathetic and hypothalamic-pituitary-adrenal responses following inhalation of
35% CO2. Psychoneuroendocrinology, 31, 736–747.
Yehuda, R. (2001). Biology of posttraumatic stress disorder. Journal of Clinical Psychiatry,
62, 41–46.
Zlotnick, C., Najavits, L. M., Rohsenow, D. J., & Johnson, D. M. (2003). A cognitive-
behavioral treatment for incarcerated women with substance abuse disorder and post-
traumatic stress disorder: Findings from a pilot study. Journal of Substance Abuse Treat-
ment, 25, 99–105.
Zvolensky, M. J., Bernstein, A., Sachs-Ericsson, N., Schmidt, N. B., Buckner, J. D., &
Bonn-Miller, M. O. (2006). Lifetime associations between cannabis, use, abuse, and
dependence and panic attacks in a representative sample. Journal of Psychiatric Research,
40, 477–486.
Zvolensky, M. J., Bonn-Miller, M. O., Bernstein, A., McLeish, A. C., Feldner, M. T., &
Leen-Feldner, E. W. (2006). Anxiety sensitivity interacts with marijuana use in the pre-
diction of anxiety symptoms and panic-related catastrophic thinking among daily tobacco
users. Behaviour Research and Therapy, 44, 907–924.
Zvolensky, M. J., & Eifert, G. H. (2001). A review of psychological factors/processes affecting
anxious responding during voluntary hyperventilation and inhalations of carbon dioxide-
enriched air. Clinical Psychology Review, 21, 375–400.
Zvolensky, M. J., & Forsyth, J. P. (2002). Anxiety sensitivity dimensions in the prediction of
body vigilance and emotional avoidance. Cognitive Therapy and Research, 26, 449–460.
Zvolensky, M. J., & Schmidt, N. B. (2004). Anxiety and substance use disorders: Introduction
to the special series. Journal of Anxiety Disorders, 18, 1–6.
Zvolensky, M. J., Schmidt, N. B., & Stewart, S. H. (2003). Panic disorder and smoking.
Clinical Psychology: Science and Practice, 10, 29–51.
The Promise of Exercise Interventions
for the Anxiety Disorders

Jasper A. J. Smits, Angela C. Berry, Mark B. Powers, Tracy L. Greer, and

Michael W. Otto


There is consistent evidence for the role of exercise in increasing longevity (Lee &
Paffenbarger, 2000; Lee, Hsieh, & Paffenbarger, 1995) and reducing risk for
coronary disease (Berlin & Colditz, 1990; Kohl, 2001), stroke (Wannamethee &
Shaper, 1992), diabetes (Chipkin, Klugh, & Chasan-Taber, 2001), obesity (Ross &
Janssen, 2001) and various cancers (Lee et al., 1995; Lee, Paffenbarger, & Hsieh,
1991; Lee, Paffenbarger, & Hsieh, 1992; Lee, Sesso, & Paffenbarger, 1999).
Evidence from a variety of sources also suggests that physical activity benefits
mental health (see for review Biddle, 2000; Stathopoulou, Powers, Berry, Smits, &
Otto, 2006). Several large cross-sectional population studies have demonstrated
that physical activity is associated with fewer symptoms of anxiety and depression
(Stephens, 1988), lower levels of stress, anger, and cynical distrust (Hassmen,
Koivula, & Uutela, 2000) as well as better social functioning and vitality among
persons with anxiety and substance use disorders (Schmitz, Kruse, & Kugler,
2004). Prospective studies have further shown that physical activity is associated
with a decreased risk of developing depression (Camacho, Roberts, Lazarus,
Kaplan, & Cohen, 1991; Paffenbarger, Lee, & Leung, 1994) even after controlling
for age, social economic status and educational level (Farmer, Locke, Mosciki,
Larson, & Radloff, 1998). Lastly, there is a growing body of research demonstrat-
ing the efficacy of exercise interventions for mental health (Stathopoulou et al.,
If effective for the prevention and treatment of anxiety disorders, physical
activity interventions would have a significant public health impact. In this
chapter, we review the literature on the relationship between physical activity
and anxiety. Specifically, we discuss findings from epidemiological, basic and
clinical studies, and consider potential mechanisms by which physical activity

Jasper A. J. Smits
Southern Methodist University, Department of Psychology, 6424 Hilltop Lane, Dallas, TX.
Tel: 214-768-4125, Fax: 214-768-0821

M. J. Zvolensky, J. A. Smits (eds.), Anxiety in Health Behaviors and Physical Illness. 81

Ó Springer 2008
82 J. A. Smits et al.

may confer protective or anxiolytic effects. We conclude this chapter with

suggested avenues for further research in this area.

Physical Activity Defined

Caspersen, Powell, and Christenson (1985) defined physical activity as ‘‘any

bodily movement produced by skeletal muscle that results in energy expendi-
ture.’’ As such, physical activity encompasses a variety of activities,
including occupational work, leisure activities and exercise, which is defined
as a physical activity that is planned for fitness or health purposes (Caspersen
et al., 1985). Exercise can be further sub-divided into aerobic or anaerobic
exercise. Aerobic exercise includes activities such as walking, swimming,
cycling and running, all of which require maintenance of increased heart
rate. The American College of Sports Medicine (ACSM, 2005) defines aerobic
exercise as a physical activity that lasts longer than three minutes during
which glycogen is consumed with oxygen. In contrast, anaerobic exercise,
which includes activities such as weight training and sprinting, lasts less
than three minutes during which glycogen is consumed without oxygen
(ACSM, 2005).
Physical fitness has been defined as a set of attributes related to the ability to
engage in physical activity (Caspersen et al., 1985). It is a multidimensional
construct (President’s Council on Physical Fitness, 2000) that includes
skill-related (e.g., agility, balance, speed, reaction time), health-related
(e.g., cardiorespiratory endurance, muscular endurance, muscular strength,
body composition, and flexibility) and physiologic components (e.g., glucose
tolerance, blood lipid and cholesterol profiles, fatness, and bone mass). The
latter two components have been more directly linked with physical health and
evidence suggests that they can be more easily improved, relative to the other
component of physical fitness, by engaging in regular physical activity
(ACSM, 2006).
Quantifying the level of physical activity involves the determination of
energy expenditure, which is typically expressed in kilocalories, or the
amount of heat required to increase the temperature of 1 kg of water by
18C. Energy expenditure is best measured by a method called ‘‘doubly labeled
water’’ (Schoeller, 1988). This procedure involves several steps, starting with
the administration of water labeled with stable isotopes of hydrogen and
oxygen. Participants are then subjected to regular urine sampling to track the
loss of these isotopes over a 7- to 14-day period. This procedure allows for an
estimation of daily carbon dioxide production, which can be converted to
total daily energy expenditure in kilocalories. However, instead of determin-
ing energy expenditure, most investigations of the relationship between
physical activity or exercise and anxiety have relied on less expensive and
more practical methods such as self-report questionnaires and interviews.
Promise of Exercise Interventions for the Anxiety Disorders 83

These instruments typically assess several dimensions of physical activity

including type, intensity, frequency, and duration, thereby allowing research-
ers to estimate the total expended kilocalories during a given time
point (e.g., week, month). Such instruments are also often used merely to
categorize participants’ physical activity habits (e.g., engaging in regular
exercise versus not engaging in regular exercise, active versus non-active,
sedentary versus non-sedentary).
Research examining the link between anxiety pathology and fitness has
mostly focused on cardiorespiratory fitness, which is operationalized as the
volume of oxygen consumed while exercising at maximum capacity
(i.e., VO2max). VO2max can be expressed in absolute values (liters/minute) or
relative values (ml/kg/min). The values tend to be higher among men compared
to women and decline with age. According to the ACSM (2006), mean absolute
values for men over age 20 range from 44.2 to 34.6 ml/kg/min, whereas mean
absolute values for women in the same age range are between 37.8 to 26.7 ml/
VO2max can be measured directly using open-circuit spirometry, or indir-
ectly by means of maximal or submaximal exercise tests. Since open-circuit
spirometry and maximal exercise testing require specialized settings, submax-
imal exercise tests are more commonly employed. One of the most widely
used submaximal exercise tests is the Astrand-Ryhming ergometer test
(Astrand & Ryhming, 1954). During this 6-minute protocol, participants
peddle at 50 rotations per minute at a work rate (i.e., watts) that is deter-
mined by the participant’s gender and fitness status. The objective is to reach
a steady-state heart rate between 125 and 170 beats per minute. A nomogram
is then used to estimate VO2max based on the average heart rate during
minutes 5 and 6.
Because it provides an index of cardiorespiratory fitness, VO2max can be
used to guide the prescription of exercise intensity. Specifically, the intensity of
exercise (e.g., very light, light, moderate, vigorous, very hard, maximal) can be

Table 1 Physical activity intensity

Intensity VO2R (%) HRR (%) HR Max (%) RPE
Very Light <20 <50 9
Light 20–39 50–63 11
Moderate 40–59 64–76 13
Hard 60–84 77–93 15
Very Hard 85 94 17
Maximal 100 100 20
Note: VO2R is oxygen reuptake reserve (i.e., the difference between resting
VO2 and maximal VO2); HRR is heart rate reserve (i.e., difference between
resting heart rate and maximum heart rate, where maximum heart rate is
220-age); HR Max is maximum heart rate; RPE is rating of perceived
exertion using a scale from 6 to 20 (Borg, 1998). Table is adapted from
ACSM guidelines for exercise testing and prescription (2006)
84 J. A. Smits et al.

anchored to VO2R (i.e., the difference between resting VO2 and maximal VO2)
values. As can be seen in Table 1, intensity of exercise prescriptions can also be
guided by a person’s heart rate reserve, maximal heart rate, or subjective ratings
of perceived exertion (i.e., RPE; Borg, 1998).

Physical Activity in the United States

Data from the 2002–2004 National Health Interview Survey (NHIS), which
included a representative sample of 93,222 U.S. adults, indicated that 62% of
adults engaged in at least in some leisure-time physical activity over the past year
(i.e., light, moderate or vigorous activity episodes that last at least 10 minutes;
Adams & Schoenborn, 2006). Physical activity status further varies as a function
of several demographic factors. For example, physical activity is significantly
more prevalent among men compared to women (64% versus 60.2%) and among
White and Asian adults (63.7% and 62.1%) compared to Black and Hispanic
adults (51.3% and 47.6%). Likewise, physical activity decreases with age and
increases with education and family income (Adams & Schoenborn, 2006).

Developmental Stages of Adopting Physical Activity

Understanding contributors to the adoption and maintenance of physical

activity is critical to the implementation of physical activity interventions
(see for review Marcus et al., 2000). The Transtheoretical Model (TTM) or
‘‘Stages of Change’’ (Prochaska & DiClemente, 1983) has been applied to many
health-related activities, including physical activity, to describe the process
of adopting a new behavior. The model describes five stages as follows:
1) precontemplation, in which there is no intent to change behavior within
6 months; 2) contemplation, in which there is intent to change behavior within
6 months; 3) preparation, in which there are small or inconsistent changes in a
behavior; 4) action, in which there is active involvement in a behavior for less
than 6 months’ duration; and 5) maintenance, in which a behavioral change is
sustained for 6 months or more. Utilization of the stages in intervention
management frequently involves assessment of individuals’ readiness to change,
evaluation and contributors to stage transition, and stage-targeted strategies
that target behaviors and issues specific to that stage. For example, in the
PACEþ study (Calfas et al., 2002), behaviors such as targeting a specific
exercise behavior (e.g., setting a goal to increase moderate or vigorous physical
activity) was associated with a greater likelihood of moving to the next stage of
change. Self-efficacy has also been shown to be an important factor in transition
to subsequent stages (Plotnikoff, Hotz, Birkett, & Courneya, 2001). Marshall
et al. (2003) reported that a stage-targeted physical activity intervention was
associated with significantly greater increases in amount of post-baseline
Promise of Exercise Interventions for the Anxiety Disorders 85

physical activity (78 minutes) compared to a control group receiving no inter-

vention (12 minutes) after 2 months, although this effect was attenuated by
6 months. In the Project Active study, Dunn and colleagues (1999) showed that
interventions based on the Stages of Change model yield comparable benefits to
physical activity when utilized as a lifestyle or structured intervention, further
supporting the incorporation of the model in intervention strategies.
Indeed, research examining the TTM model has led to the development of
multi-component approaches to adoption of and adherence to exercise. When
assessing the feasibility of a treatment, it is important to consider
behavioral factors that may limit the likelihood of treatment compliance.
Multi-component interventions have been shown to be more effective than
single component interventions because no one strategy is effective for all
individuals (Ockene, Hayman, Pasternak, Schron, & Dunbar-Jacob, 2002;
Roter et al., 1998). Multi-component strategies include behaviors such as
developing a specific action plan for activity and targeting barriers to comple-
tion of the activity to increase the likelihood of exercise behaviors. This is
particularly true with psychiatric populations, where psychological aspects of
the disease process may increase barriers to exercise or create difficulty in
moving on to the next stage of behavioral change.

Population-based Studies of the Relationship Between

Physical Activity and Anxiety

There have been several large-scale studies providing evidence for the negative
association between physical activity and anxiety. For example, Stephens
(1988) collapsed data from four national health surveys conducted in the
United States and Canada between 1971 and 1981 (e.g., the National Health
and Nutrition Examination Study, Canada Health Survey, National Survey of
Personal Health Practices and Consequences, and the Canada Fitness Survey).
This procedure yielded a total sample size of approximately 55,000 persons
between ages 10 and 74. Both mental health and level of leisure time exercise
were assessed by self-report. The findings supported the hypothesis that physi-
cal activity is associated with good mental health, including fewer symptoms of
depression and anxiety, even after controlling for education and physical health
status. Interestingly, the strength of this relationship varied as a function of age
and sex, where a stronger link was observed among women and among those
over age 40. Recently, Schmitz and colleagues (Schmitz et al., 2004)
extended these findings by showing a link between physical activity and
enhanced psychological and health-related well-being among persons suffering
from DSM-IV anxiety disorders. They used data from the German National
Health Interview and Examination Survey (GHS) which comprised a represen-
tative sample of approximately 7,000 persons between ages 18 and 79.
Diagnoses (12-month prevalence) were determined by a combination of
86 J. A. Smits et al.

self-report and interviewer measures, while physical activity level and emotional
well-being were assessed by self-report. Of those suffering from anxiety
disorders (n = 573), 63% indicated that they were inactive. These inactive
individuals reported significantly lower quality of life across a number of health
and mental health domains relative to anxiety disorder sufferers who reported
regular exercise. Further, the associations remained significant after controlling
for theoretically-relevant variables, but they did not depend on age or gender.
In addition to evidence for the association between physical activity and general
emotional well-being, there is also survey data indicating that physical inactivity is
related to clinical levels of anxiety. Using data from the National Comorbidity
Survey (NCS), which comprised a probability sample of approximately 6,000
individuals between ages 15 and 54, Goodwin (2003) estimated the 12-month
prevalence of DSM diagnoses among persons who indicated that they exercised
‘‘regularly’’ (60.3%) and compared the rate to those exercised occasionally, rarely
or never. Findings revealed a dose-response relationship between physical activity
and the likelihood of having an anxiety disorder. Specifically, physical activity was
associated with a significantly decreased likelihood of agoraphobia, panic attacks,
generalized anxiety disorder (GAD), specific phobia, and social phobia, even after
controlling for comorbid physical illness and demographic variables. Interestingly,
the relationship between physical activity and GAD was no longer significant after
additionally adjusting for comorbid mental disorders, suggesting that physical
inactivity may not be directly linked with GAD.

Clinical Studies of the Relationship Between

Physical Fitness and Anxiety

The linkage between phobic anxiety and cardiovascular disease and mortality
(Coryell, Noyes, & House, 1986; Coryell, Noyes, & Clancy, 1982; Gomez-
Caminero, Blumentals, Russo, Brown, & Castilla-Puentes, 2005; Kawachi
et al., 1994; Kawachi, Sparrow, Vokonas, & Weiss, 1994; Kawachi, Sparrow,
Vokonas, & Weiss, 1995; Weissman et al., 1990; see also White et al. in Chapter
11 in this volume) has prompted research on physical fitness among individuals
suffering from pathological anxiety, and particularly panic disorder. There
have been several studies that have employed standardized exercise testing
protocols to determine cardiorespiratory fitness among individuals with panic
disorder (Broocks et al., 1997; Gaffney, Fenton, Lane, & Lake, 1988; Martinsen,
Raglin, Hoffart, & Friis, 1998; Meyer, Broocks, Bandelow, Hillmer-Vogel, &
Ruther, 1998; Schmidt, Lerew, Santiago, Trakowski, & Staab, 2000; Taylor et
al., 1987). For example, Martinsen and colleagues (1998) subjected 35 patients to
a submaximal bicycle ergometer test and found that their VO2max was on
average at 82% (1.9 liters/min) of the expected value. Similarly, Brooks and
colleagues (1997) reported that, relative to a group of non-psychiatric partici-
pants (n = 24), individuals with panic disorder (n = 38) showed significantly
Promise of Exercise Interventions for the Anxiety Disorders 87

reduced VO2max. Specifically, the mean relative VO2max values for panic
disorder patients was 31.0 ml/kg/min (SD = 9.1) versus 37.6 ml/kg/min
(SD = 6.3) for non-psychiatric controls. In a follow-up study, the authors
(Meyer et al., 1998) showed that a 10-session exercise training program was
associated with significant improvements in VO2max values.
To clarify the link between panic disorder and fitness, Schmidt and colleagues
(Schmidt et al., 2000) designed a study to determine whether the relatively poor
cardiorespiratory fitness estimates among individuals with panic disorder may
be biased by elevated levels of anxiety symptoms. To this end, they randomly
assigned 27 panic disorder and 27 matched healthy control participants to a
submaximal bicycle ergometer test with or without heart rate feedback. This
study yielded several important findings. First, panic disorder participants
showed reduced cardiorespiratory fitness relative to non-psychiatric partici-
pants even after controlling for anxiety responses during the testing protocol
(26.4 ml/kg/min (SD = 11.4) versus 34.11 ml/kg/min (SD = 9.8), respectively).
Second, although anxiety sensitivity (i.e., fear of anxiety-related sensations), a
cognitive disposition that has been implicated in the onset and maintenance of
panic disorder (McNally, 2002), was not directly associated with poorer fitness,
it appeared to moderate the effects of diagnostic status and heart rate feedback
on cardiorespiratory fitness. Specifically, cardiorespiratory fitness was particu-
larly poor among panic disorder suffers with elevated levels of anxiety sensitiv-
ity. Similarly, heart rate feedback during testing was only associated with poorer
fitness among individuals who reported elevated levels of anxiety sensitivity.
Collectively, these findings suggest that cardiovascular conditioning is impaired
among individuals with panic disorder, and emphasize the importance of consid-
ering anxiety sensitivity in investigations of the relationship between physical
activity and anxiety. Indeed, support for the linkage between anxiety sensitivity
and physical activity is growing. Recently, McWilliams and Asmundson (2001)
observed an inverse relationship between exercise frequency and anxiety sensitivity
in a non-clinical sample. Likewise, Smits and Zvolensky (2006) reported that
anxiety sensitivity mediated the relationship between physical inactivity and mea-
sures of panic severity in a community sample of individuals with panic disorder.
Further research regarding the linkage between exercise and anxiety sensitivity will
be discussed in the next section on the effects of exercise interventions on anxiety.

Reductions in Anxiety with Exercise

Although the studies discussed above support the hypothesis that physical
activity plays a role in the etiology, maintenance and treatment of anxiety
pathology, the cross-sectional nature of many of these investigations leave
open several alternative explanations for the observed relationships. In this
section, we will discuss a number of studies that have experimentally manipu-
lated physical activity, providing some evidence for the causal effects of physical
activity on anxiety.
88 J. A. Smits et al.

The Effects of Acute Exercise on Anxiety

Evaluation of the beneficial effects of single episodes of exercise on anxiety

has been the focus of numerous quantitative and qualitative reviews
(e.g., Ekkekakis & Petruzzello, 1999; Petruzzello, Landers, Hatfield, Kubitz,
& Salazar, 1991; Schlicht, 1994). For example, Petruzzello and associates (1991)
conducted a meta-analysis of 119 studies that examined the pre- to post-exercise
reductions in self-reported state anxiety. The mean controlled effect size was
d = 0.23, although randomized controlled studies yielded smaller effect sizes
(d = 0.16). In a more recent review, Ekkekakis and Petruzzello (1999) aptly
pointed out that many of the studies examining the effect of acute exercise on
anxiety lack the methodological rigor (e.g., adequate sample sizes and control
conditions) necessary to significantly advance knowledge regarding the dose-
response relationship. They did, however, observe some robust findings. First,
the duration of exercise session is not predictive of anxiety reduction beyond the
effects explained by the degree of exercise intensity. Second, the degree of
benefit is linked to the degree of fitness; in non-clinical samples, demanding
exercise tasks result in greater anxiety reductions among high-fit individuals
versus low-fit individuals (Ekkekakis et al., 1999). In addition to fitness level,
the participant’s baseline anxiety level appears to be predictive of the degree of
benefit. Long and Stavel (1995) reported that effect sizes for studies that
employed high-anxious samples were significantly larger than those that
employed low anxious samples (d = 0.51 and 0.28, respectively). One implica-
tion of this finding is that the degree of benefit from exercise may be much more
profound in clinical samples.
In addition to reducing state anxiety, single exercise sessions also appear to
buffer the effects of social stressor tasks. Using a counterbalanced design,
Rajeski and colleagues (Rejeski, Thompson, Brubaker, & Miller, 1992) asked
48 low-to-moderate physically fit women to exercise or rest prior to completing
two stressor tasks. The experimental sessions began with either 40 minutes of
cycling at 70% heart rate reserve or 40 minutes of rest after which participants
rested 30 minutes. Following this 30 minute period, participants completed a
Stroop mental challenge and a 3-minute impromptu speech on a controversial
topic (e.g., abortion, role of women in society). Relative to rest, exercise was
associated with less blood pressure reactivity following psychosocial stress, and
reduced anticipatory anxiety prior to speech task.
Similar findings have been observed in studies using biological challenge
procedures. Biological provocations such as caffeine ingestion, inhalation
of carbon dioxide-enriched air, and lactate infusion have been shown to
reliably increase bodily and subjective symptoms of anxiety and panic in
healthy controls (Zvolensky & Eifert, 2001). A series of recent studies (Esquivel,
Schruers, Kuipers, & Griez, 2002; Strohle et al., 2005; Youngstedt, O’Connor,
Crabbe, & Dishman, 1998) has demonstrated that emotional responding to
these challenges is significantly attenuated when preceded by a single brief
Promise of Exercise Interventions for the Anxiety Disorders 89

(e.g., 12–30 minutes) intense exercise session (e.g., 70% of max HR; 60–70%
VO2max, >6mm of blood lactate). Together, these findings suggest that single
exercise sessions result in anxiolysis in non-clinical participants.

Parameters of Exercise Interventions

Petruzzello and colleagues (1991) reviewed 62 studies of exercise interventions

that were at least four weeks in length. Most studies involved non-clinical
(n = 48) or non-clinical high anxious (n = 11) individuals. Less than half of
the studies (n = 25) employed a randomized controlled design; the other
37 studies were non-experimental or quasi-experimental in nature. Comparison
groups involved waitlist control, relaxation or motivational control conditions
and anxiety was generally measured using the trait version of the State-Trait
Anxiety Inventory (STAI-T; Spielberger, Gorsuch, Lushene, Vagg, & Jacobs,
1983). The average controlled effect size for anxiety reduction with exercise
training was (d = 0.34). However, post-hoc analyses revealed that the effect size
significantly varied as a function of characteristics of study design and the
exercise intervention. Specifically, randomized controlled studies yielded larger
effects sizes (d = 0.54). Moreover, length of training was significantly
associated with the degree of improvement in anxiety; interventions with a
minimum length of 10 weeks were associated with greater effect sizes
(d =0.36 to 0.90) compared to interventions with a duration of less than
10 weeks (d = 0.14 to 0.17). Similarly, only interventions with session durations
between 21 and 30 minutes or greater than 40 minutes yielded significant
effect sizes.

Exercise Training Programs for Clinical Anxiety

Despite the wealth of studies devoted to the application of exercise interven-

tions to stress and non-clinical anxiety, to our knowledge, there are only two
randomized controlled clinical trials of exercise interventions for
individuals meeting criteria for anxiety disorders. This limited evidence stands
in stark contrast to studies of exercise interventions for major depression. For
depression, exercise interventions are associated with reliable benefit.
For example, a recent meta-analysis of 11 controlled studies of the efficacy
of exercise interventions for depression yielded a controlled effect size of
(d = 1.42) (Stathopoulou et al., 2006). For anxiety disorders, the available
evidence for the efficacy of exercise interventions is limited to the treatment of
panic disorder.
Panic disorder is a particularly apt disorder for the application of exercise
interventions (see for review Smits et al., 2007) given that fears of somatic
sensations of anxiety (i.e., anxiety sensitivity) are central to the etiology and
90 J. A. Smits et al.

maintenance of the disorder (McNally, 2002; Smits, Powers, Cho, & Telch,
2004). Cognitive-behavioral treatments for panic disorder target anxiety sensi-
tivity by prescribing exposure to these feared sensations (i.e., interoceptive
exposure) in conjunction with cognitive interventions to help patients eliminate
catastrophic interpretations of these symptoms (Margraf, Barlow, Clark, &
Telch, 1993). Accordingly, in addition to the general benefits of exercise on
anxiety reduction, as documented for nonclinical samples, exercise also has the
potential to provide patients with exposure to feared sensations of bodily
arousal. Indeed, aerobic exercise induces many of the somatic sensations
(e.g., heart racing, rapid breathing, and sweating) that have shown to elicit
increased anxiety reactions in panic disorder patients (Rief & Hermanutz,
1996). The notion that exercise can serve as an interoceptive exposure proce-
dure is also consistent with early reports that exercise can be anxiogenic for
these patients. For example, Cameron and Hudson (1986) observed significant
increases in subjective anxiety during submaximal exercise testing in 31% of
patients with panic disorder, but only in 7% of patient and non-patient con-
trols. Similarly, two more recent studies reported that panic disorder patients
are more likely to prematurely terminate submaximal exercise testing compared
to non-clinical controls (Schmidt et al., 2000; Stein, Tancer, & Uhde, 1992).
Consistent with previous evidence that repeated exposure to biological pro-
vocation procedures is associated with clinical benefits for panic disorder
sufferers (van den Hout, van der Molen, Griez, Lousberg, & Nansen, 1987),
Broman-Fulks, Berman, Rabian and Webster (2004) developed a brief
exercise intervention to target anxiety sensitivity. The intervention consisted
of six 20-minute high-intensity (60–90% maximal heart rate) sessions on a
treadmill. This intervention was compared to a similar protocol low-intensity
(below 60% of maximal heart rate) level aerobic exercise program. Neither
group received a specific treatment rationale or cognitive interventions. Results
revealed that the high intensity intervention was associated with significantly
greater reductions in anxiety sensitivity compared to the low intensity interven-
tion (pre- to posttreatment raw score mean reductions of 9.14 and 2.88, respec-
tively). Preliminary results from our own laboratories suggest that preceding
the interventions with a rationale emphasizing the importance of exposure to
reducing anxiety sensitivity coupled with coaching participants during and
following exercise sessions (i.e., ‘‘what are you learning from this?’’)
may enhance the effect of this exercise intervention on anxiety sensitivity
(Smits et al., 2007).
In addition to the direct application of exercise as an interoceptive exposure
procedure to reduce anxiety sensitivity, exercise may also have general anxio-
lytic properties for patients with panic disorder. Using a counterbalanced
design, O’Connor (2005) asked 10 women with panic disorder to complete a
maximal treadmill exercise test, 25 minutes of submaximal treadmill walking
(i.e., 65% of VO2 max), and 25 minutes of seated rest during three consecutive
sessions. Self-report of state anxiety was assessed 5 and 15 minutes before and
after each session. Consistent with hypothesis, patients reported significant
Promise of Exercise Interventions for the Anxiety Disorders 91

reductions in anxiety after both maximal and submaximal exercise testing.

Importantly, these reductions were evident within 5 minutes following exercise.
A much broader study of the efficacy of aerobic exercise for panic disorder
was conducted by Broocks et al. (1998). These investigators examined the
efficacy of aerobic exercise relative to clomipramine and pill placebo in a
sample of patients with panic disorder (N = 46). Patients randomized to
clomipramine were prescribed the medication following evidence-based guide-
lines. Patients in the aerobic exercise condition were medication-free and
underwent a 10-week endurance training program. Specifically, patients were
asked to find a four-mile route (forest or park) that was easily accessible from
their home, and complete this entire route at least three times a week, where
walking was allowed during the first six weeks, and running was expected
during the last four weeks. Patients also met with a trainer once each week to
run together. Exercise led to significantly more benefit than placebo treatment
at week 10, and clomipramine yielded significantly greater effects compared to
placebo after four weeks. At posttreatment, both active treatments outper-
formed the placebo condition and were equally effective in reducing anxiety.
Additionally, clomipramine yielded greater changes in global improvement
ratings compared to aerobic exercise. In discussing the magnitude of the effects
observed for exercise in their study, Broocks et al. (1998) hypothesized that
additional cognitive interventions would have enhanced the benefits among
participants in the exercise condition. Specifically, therapists could have
assisted patients in preparing to reappraise some of the feared consequences
of exercise-induced sensations. This type of preparation could have potentially
prevented avoidance of more intense exercise that was evident in a subset of the
patients (Broocks et al., 1998).
The findings reported by Broocks and colleagues (1998) comport well with
an earlier report (Martinsen, Hoffart, & Solberg, 1989). Martinsen et al. (1998)
randomly assigned in-patient participants with panic disorder with agorapho-
bia (n = 56), social phobia (n = 13), or generalized anxiety disorder (n = 10) to
either an aerobic or non-aerobic exercise treatment program. Both programs
were conducted in group format, were eight weeks in length, and involved three
weekly one-hour sessions, of which thirty minutes were devoted to exercise. The
aerobic program involved walking or running at 70% of maximal aerobic
capacity, whereas the intensity of the anaerobic program, which comprised
muscular strength training, was unspecified. The study attrition rate was 11%
and involved only panic disorder patients (n= 9) in the aerobic exercise condi-
tion. Although aerobic exercise was associated with greater improvements in
physical fitness, both conditions showed comparable significant pre- to post
treatment improvements on interviewer and self-report measures of anxiety.
Based on these results, the authors concluded that exercise programs are
effective for reducing pathological anxiety and that this effect cannot be
accounted for by enhanced cardiorespiratory fitness.
Initial feasibility data also supports the application of exercise to obsessive-
compulsive disorder (OCD). Brown and colleagues (Brown et al., 2007) recently
92 J. A. Smits et al.

completed an open trial of an exercise intervention with 15 patients who

were also on a stable dose of cognitive-behavioral treatment, pharmacotherapy,
or their combination. The exercise intervention involved a combination of
supervised and home-based aerobic exercise at 55 to 69% of age predicted
maximal heart rates. Over the course of the 12-week protocol, participants
progressed from 20-minute to 40-minute sessions three to four times a week.
Prior to each supervised group session, participants participated in a 30-minute
meeting with a clinical psychologist and an exercise physiologist to discuss
topics related to compliance with the intervention program (e.g., benefits
of exercise, goal setting, identifying and overcoming barriers to exercise).
Compliance was further stimulated by providing monetary remuneration for
adherence. The Cohen’s d effect size for reductions in Y-BOCS scores was
d = 1.69 from pre- to posttreatment, and d = 1.11 for pretreatment to
6-month follow-up. Moreover, clinically meaningful changes were observed
for 69% and 50% of patients at posttreatment and 6-month follow up,
The initial successes of the application of exercise interventions to the treat-
ment of clinical anxiety problems encourage a more in-depth study of the
feasibility and efficacy of exercise protocols for patients with anxiety disorders.
In addition to large-scale randomized controlled trials, these efforts should
include investigations of the mechanism underlying the effects of exercise on

Mechanisms of Exercise Anxiolysis

There is little research on the mechanism by which exercise exerts its ameliora-
tive effects on anxiety. In this section, we will discuss several proposed
physiological and psychological mechanisms of exercise anxiolysis for which
there is some preliminary supporting evidence.

Central Neurotransmitter Function

Several animal studies have demonstrated that physical activity results in

alterations in many neural systems that are also presumed to underlie the
reductions in depression and anxiety with pharmacological treatments. For
example, increased release of serotonin has been observed in animals both
during (Wilson & Marsden, 1996) and following treadmill running (Dunn &
Dishman, 1991; Meeusen & De Meirleir, 1995). Based upon other animal
studies showing that physical activity leads to increased serotonin metabolism
(Broocks, Schweiger, & Pirke, 1991; Chaouloff, 1997), Broocks and colleagues
(Broocks et al., 1999; Broocks et al., 2001; Broocks et al., 2003) posited that
physical activity may result in down-regulation of postsynaptic serotonin
Promise of Exercise Interventions for the Anxiety Disorders 93

receptors, and specifically the 5-HT2C receptors. In a first study to test this
hypothesis, they compared marathon runners to sedentary controls on their
responses to meta-chlorophenylpiperazine (m-CPP), a 5-HT agonist that pro-
duces anxiogenic symptoms via 5-HT2C receptors (Broocks et al., 1999).
Marathon runners showed a diminished cortisol response to m-CPP, providing
evidence consistent with the idea that chronic exercise results in a reduced
hormonal reaction to m-CPP mediated by postsynaptic 5-HT2C receptors. In
a second study, the authors (Broocks et al., 2001) found that untrained
participants show a similar blunted cortisol response to m-CPP following a
10-week aerobic exercise program of moderate intensity. The authors
concluded that these data collectively suggest that the anxiolytic effects of
exercise may be mediated by the downregulation of 5-HT2C receptors.
The efficacy of benzodiazepines for reducing anxiety forms the basis of
research examining the potential of exercise on -aminobutyric acid (GABA)
function. Studies suggest that measuring the amount of time a rat spends in an
open field (open field locomotion) with or without treadmill exercise is a useful
animal model for studying the anxiolytic effects of exercise (Dishman,
Armstrong, Delp, Graham, & Dunn, 1988; Morgan, Olson, & Pedersen, 1982;
Royce, 1977). Injections of GABA into the nucleus accumbens septi (NAS)
reduces open field locomotion and injections of a GABA antagonist into
the NAS increases locomotion in rats (Jones & Mogenson, 1980a; Jones &
Mogenson, 1980b; Jones, Mogenson, & Wu, 1981). Since exercise also increases
open field locomotion in rats (Tharp & Carson, 1975); (Weber & Lee, 1968),
Dishman and colleagues (1996) have posited that exercise may reduce anxiety
by the downregulation of GABAa receptor density in the corpus striatum.
Consistent with this hypothesis, they found that voluntary exercise in rats
increased open field locomotion with a corresponding GABAa downregulation
(Dishman et al., 1996). Human studies are needed to test the hypothesis that
exercise anxiolysis in humans can be accounted for by GABAa downregulation
or other changes in central neurotransmitter function.

Sleep Restoration

Improved sleep as a proposed mechanism of change first emerged with the

delayed onset hypothesis, or the observation that antidepressant drugs rapidly
change plasma levels and sleep parameters (48–72 hours), and that correspond-
ing mood and anxiety changes in the weeks following (Chen, 1979; Ehlers,
Havstad, & Kupfer, 1996; Hyttel, 1994; Kupfer et al., 1994; Shipley et al.,
1984). There is also evidence that pharmacologic treatment of sleep symptoms
improves depression outcomes (Fava et al., 2006). Moreover, several studies
have now shown that exercise training programs are associated with significant
improvements in self-reported sleep quality (King, Oman, Brassington, Bliwise,
& Haskell, 1997; Singh, Clements, & Fiatarone, 1997). Interestingly, some
94 J. A. Smits et al.

findings converge to suggest that changes in slow wave sleep (SWS), which
occurs during restorative stages 3 and 4 of the sleep cycle, may be particularly
relevant to the anxiolysis following exercise. Specifically, reduced SWS is ubi-
quitous among anxiety disorder sufferers (Arriaga & Paiva, 1990; Arriaga et al.,
1996; Bourdet & Goldenberg, 1994; Fuller, Waters, Binks, & Anderson, 1997)
and exercise appears to enhance SWS (Horne, 1981; Shapiro, Griesel, Bartel, &
Jooste, 1975). There is some controversy with respect to the mechanism under-
lying the effects of exercise on SWS. Some studies suggest that it may be the
elevation in body temperature created by exercise that may be responsible for
sleep effects (Atkinson & Davenne, 2006; Horne & Moore, 1985; Horne &
Shackell, 1987; Horne & Staff, 1983). However, other studies show that low
intensity activity without a corresponding increase in body temperature can
also increase SWS (Naylor et al., 2000). Also, well-controlled studies show that
the anxiolytic effects of acute exercise are not solely due to body temperature
(Youngstedt, Dishman, Cureton, & Peacock, 1993).

Cognitive Refocusing and Self-efficacy

There is some evidence suggesting that exercise may exerts its effect on anxiety
by enhancing perceived coping ability. Steptoe and colleagues (Moses, Steptoe,
Mathews, & Edwards, 1989) observed parallel decreases in perceived coping
and anxiety in anxious individuals who initiated an exercise program. Similarly,
Bodin and Martinsen (2004) found that exercise that targeted self-efficacy
(e.g., 45 minutes of martial arts) corresponded with significantly greater
improvements in positive affect and state anxiety compared to exercise that
did not target self-efficacy (e.g., 45 minutes of stationary bike exercise). As an
alternative to enhanced self-efficacy, some have suggested that physical exercise
may only serve as distraction from ruminations, worries, and anxiety (Bahrke &
Morgan, 1978; Leith, 1994). Interestingly, Goode and Roth (1993) found that it
is not distraction per se but the content of the distraction techniques in which
people engage that is associated with changes in emotional well-being. They
found that that runners who focused on nonassociative thoughts (those not
related to exercising) showed less fatigue and in some cases decreases in tension
and anxiety, compared to runners who focused on associative thoughts
(monitoring the body and the exercise itself).


Recently, Stathopoulou and colleagues (2006) proposed that, among other

effects, exercise may exert positive effects on mental health by teaching persis-
tence in the presence of negative somatic or emotional states. Indeed, the
modification of some of the self-perpetuating patterns in affective disorder
Promise of Exercise Interventions for the Anxiety Disorders 95

(e.g., social withdrawal and inaction in response to feelings of depression, and

avoidance in response to feelings of anxiety), by facilitation of adaptive pursuit
of goals regardless of the presence of aversive thoughts or emotions, is increas-
ingly being discussed as a general feature of adaptive change in therapy
(Barlow, Allen, & Choate, 2004). Accordingly, by training persistence with
exercise, despite the presence of the physical and emotional symptoms of
exertion, exercise interventions may have more general effects on returning
participants to adaptive activity.
As we discussed above, this persistence in the presence of symptoms may
take on special meaning in the case of panic disorder. We discussed exercise,
with its induced array of symptoms that are similar to many of the symptoms of
anxious arousal, as a form of interoceptive exposure – exposure to the somatic
symptoms of anxiety under controlled circumstances. As suggested by the
research to date, inclusion of specific preparation and cognitive coaching for
the symptoms to be induced both mimics typical interoceptive exposure proce-
dures used in cognitive behavior therapy (Smits et al., 2007) and appears to lead
to more optimal outcomes for exercise interventions (Smits et al., 2005; Broocks
et al., 1998).

Conclusions and Future Directions

The application of exercise interventions to clinical anxiety is in the early stages

of development, but the few trials completed to date, as well as the wealth of
evidence for the effects of exercise on non-clinical anxiety, encourage further
study. Although most work to date has been completed with persons suffering
from panic disorder, preliminary findings with respect to the mechanisms of
exercise anxiolysis suggest that exercise of may also offer significant benefits for
those suffering from other forms of anxiety psychopathology. This hypothesis
awaits testing in large-scale randomized controlled trials.
The development of exercise interventions for the anxiety disorders will
likely benefit from studies focused on identifying intervention parameters cri-
tical to efficacy of exercise. There is evidence from the depression literature
suggesting that high intensity exercise provides superior results compared to
lower intensity exercise. For example, Dunn and associates (Dunn, Trivedi,
Kampert, Clark, & Chambliss, 2005) reported that the public health recom-
mended dose of aerobic exercise (total energy expenditure of 17.5 kcal/kg/week)
was more effective in reducing depression relative to low dose aerobic exercise
(total energy expenditure 7.0 kcal/kg/week) or flexibility exercise. A public
health dose of exercise has also yielded reductions in depressive symptom
severity and improvements in quality of life in a preliminary study of exercise
augmentation to antidepressant treatment (Trivedi, Greer, Grannemann,
Chambliss et al., 2006a) and is currently being investigated in the context of a
larger randomized controlled trial (Trivedi, Greer, Grannemann, Church et al.,
96 J. A. Smits et al.

2006b). Other elements of exercise dose that warrant further study are intensity
and duration.
In addition to exercise dose, it is important to determine the relative impor-
tance of exercise type. There is currently a paucity of available data on the
effects of anaerobic exercise (resistance training) for anxiety. Findings
from the depression literature indicate that resistance training may be equally
effective compared to aerobic activity in reducing symptoms of depression
(e.g., Doyne et al., 1987; Martinsen et al., 1989). Support for the use of
anaerobic exercise for depression was further strengthened by a recent study
completed by Singh and colleagues (Singh et al., 2005). They demonstrated that
high intensity progressive resistance training (PRT) (80% maximum load) was
more effective in treating depression than lower intensity PRT (20% maximum
load). Perhaps more important was the finding that the effects of PRT on
depression reduction were accounted for by expectancy in the low intensity
condition but not in the high intensity condition, suggesting that anaerobic
exercise, when prescribed at the higher doses, offers more than a placebo effect.
An important implication of these findings is that resistance training may be an
alternative for patients for whom aerobic activity may be inappropriate or for
those who do not have the initial motivation for aerobic activity.
When examining the utility of new interventions, it is important consider
issues related to effectiveness in addition to efficacy. It remains to be seen how
exercise is best integrated within provider networks. Within specialty care,
exercise interventions may emerge as another adjunctive clinical tool that has
the advantage of providing a broad spectrum of health benefits in addition to its
benefits on mental health (Stathopoulou et al., 2006). In primary care, exercise
may emerge as a more fundamental intervention that can be prescribed by the
primary care physician or provider team. Specifically, exercise has a potential
for targeting many mental and physical health problems simultaneously. In this
application, exercise is likely to bring with it all the challenges of any health
promotion intervention. Adherence to exercise recommendations has been low
in the United States (Schoenborn, Adams, Barnes, Vickerie, & Schiller, 2004);
although there is some evidence suggesting that exercise for mental health
benefits may fare better than when prescribed for improving physical health.
The use of exercise interventions for anxiety disorders has one clear advantage
over exercise for general health promotion. Unlike exercise interventions that
are prescribed for the prevention of health problems, exercise interventions for
treating anxiety disorders are linked with immediate benefits (i.e., acute bouts
of exercise are associated with significant changes in anxiety). As reported by
Christensen-Szalanski and Northcraft (1985), adherence to recommended
health behavior changes is greater when there is direct symptom reduction
that is linked with the behavior change. There also appears to be wide
acceptance of nontraditional treatment strategies for mental health among
individuals suffering from mood and anxiety disorders (Kessler et al., 2001).
Indeed, exercise interventions have the potential of avoiding the social stigma
associated with other mental health interventions (Sirey et al., 2001). Although
Promise of Exercise Interventions for the Anxiety Disorders 97

these data are encouraging, systematic research on issues related to the

dissemination and adoption of exercise interventions for the anxiety disorders
is needed.
From a public health perspective, examining the potential role for physical
activity interventions in the prevention of anxiety pathology may be as impor-
tant as determining its effectiveness for the treatment of anxiety disorders.
Certainly, growing evidence for the efficacy of exercise for reducing anxiety
sensitivity indicate promise for exercise interventions for preventing
panic-related pathology (Schmidt, Lerew, & Jackson, 1997; Schmidt, Lerew,
& Jackson, 1999; Wilson & Hayward, 2005). However, there is a need for
studies that clarify the role of physical (in)activity in the development of anxiety
pathology. When testing the direction of this relationship, it is important to
consider the potential interplay between physical activity and other established
risk factors of anxiety pathology as well as the potential moderating effects of
gender (Stephens, 1988). In this context, it may also be important to assess other
health factors (e.g., fitness, nutrition, substance use, sleep) that have shown to
be linked to physical activity (Aaron, Dearwater, Anderson, & Olsen, 1995).
Similarly, the importance of investigating the impact of exercise on psychoso-
cial function and quality of life should be underscored. Impaired psychosocial
function is associated with increased risk of recurrence of anxiety disorders
(Rodriguez, Bruce, Pagano, & Keller, 2005), and these symptoms must there-
fore be addressed in addition to primary symptoms in order to fully manage the
anxiety disorders.
Finally, we have limited the discussion in this chapter to a unidirectional
model. It should be noted, however, that the evidence to date may point to a
more complex relationship between physical activity and anxiety and panic
psychopathology. Of particular importance to public health would be the
investigation of the potential impact of anxiety and mood problems on the
adoption and maintenance of physical activity.


Aaron, D., Dearwater, S., Anderson, R., & Olsen, T. (1995). Physical activity and the
initiation of high-risk health behaviors in adolescents. Medicine & Science in Sports &
Exercise, 27(12), 1639–1645.
Adams, P. F., & Schoenborn, C. A. (2006). Health behaviors of adults: United States,
2002–04. National Center for Health Statistics. Vital Health Statistics, 10, 20, 1–140.
American College of Sports Medicine. (2005). ACSM’s guidelines for exercise testing and
prescription (6th ed). Philadelphia, PA: Lippincott Williams, & Wilkins.
American College of Sports Medicine. (2006). ACSM’s Advanced exercise physiology.
Baltimore: Lippincott Williams & Wilkins.
Arriaga, F., & Paiva, T. (1990). Clinical and EEG sleep changes in primary dysthymia and
generalized anxiety: A comparison with normal controls. Neuropsychobiology, 24(3),
98 J. A. Smits et al.

Arriaga, F., Paiva, T., Matos-Pires, A., Cavaglia, F., Lara, E., & Bastos, L. (1996). The sleep
of non-depressed patients with panic disorder: a comparison with normal controls. Acta
Psychiatrica Scandanavia, 93(3), 191–194.
Astrand, P. O., & Ryhming, I. (1954). A nomogram for calculation of aerobic capacity
(physical fitness) from pulse rate during sub-maximal work. Journal of Applied Physiology,
7(2), 218–221.
Atkinson, G., & Davenne, D. (2006). Relationships between sleep, physical activity and
human health. Physiology & Behavior, 90, 229–235.
Bahrke, M. S., & Morgan, W. P. (1978). Anxiety reduction following exercise and meditation.
Cognitive Therapy and Research, 2, 323–333.
Barlow, D. H., Allen, L. B., & Choate, M. L. (2004). Toward a unified treatment for
emotional disorders. Behavior Therapy, 35, 205–230.
Berlin, J. A., & Colditz, G. A. (1990). A meta-analysis of physical activity in the prevention of
coronary heart disease. American Journal of Epidemiology, 132(4), 612–628.
Biddle, S. J. H. (2000). Emotion, mood and physical activity. In: S. J. H. Biddle, K. R. Fox, &
S. H. Boutcher (Eds.), Physical activity and psychological well-being. London, UK:
Bodin, T., & Martinsen, E. W. (2004). Mood and self-efficacy during acute exercise in clinical
depression. A randomized, controlled study. Journal of Sport and Exercise Psychology, 26,
Borg, G. (1998). Borg’s perceived exertion and pain scales. Champaign, IL: Human
Bourdet, C., & Goldenberg, F. (1994). Insomnia in anxiety: sleep EEG changes. Journal of
Psychosomatic Research, 38(Suppl 1), 93–104.
Broman-Fulks, J. J., Berman, M. E., Rabian, B. A., & Webster, M. J. (2004). Effects of
aerobic exercise on anxiety sensitivity. Behaviour Research & Therapy, 42(2), 125–136.
Broocks, A., Bandelow, B., Pekrun, G., George, A., Meyer, T., Bartmann, U., et al. (1998).
Comparison of aerobic exercise, clomipramine, and placebo in the treatment of panic
disorder. American Journal of Psychiatry, 155(5), 603–609.
Broocks, A., Meyer, T. F., Bandelow, B., George, A., Bartmann, U., Ruther, E. et al. (1997).
Exercise avoidance and impaired endurance capacity in patients with panic disorder.
Neuropsyschobiology, 36, 182–187.
Broocks, A., Meyer, T., George, A., Hillmer-Vogel, U., Meyer, D., Bandelow, B., et al.
(1999). Decreased neuroendocrine responses to meta-chlorophenylpiperazine (m-CPP)
but normal responses to ipsapirone in marathon runners. Neuropsychopharmacology,
20(2), 150–161.
Broocks, A., Meyer, T., Gleiter, C. H., Hillmer-Vogel, U., George, A., Bartmann, U., et al.
(2001). Effect of aerobic exercise on behavioral and neuroendocrine responses to meta-
chlorophenylpiperazine and to ipsapirone in untrained healthy subjects. Psychopharma-
cology (Berl), 155(3), 234–141.
Broocks, A., Meyer, T., Opitz, M., Bartmann, U., Hillmer-Vogel, U., George, A., et al.
(2003). 5-HT1A responsivity in patients with panic disorder before and after treatment
with aerobic exercise, clomipramine or placebo. European Neuropsychopharmacology,
13(3), 153–164.
Broocks, A., Meyer, T. F., Bandelow, B., George, A., Bartmann, U., Ruther, E., et al. (1997).
Exercise avoidance and impaired endurance capacity in patients with panic disorder.
Neuropsychobiology, 36(4), 182–187.
Broocks, A., Schweiger, U., & Pirke, K. M. (1991). The influence of semistarvation-induced
hyperactivity on hypothalamic serotonin metabolism. Physiology & Behavior, 50(2),
Brown, R. A., Abrantes, A. M., Strong, D. R., Mancebo, M. C., Menard, J., Rasmussen, S. A.,
et al. (2007). A pilot study of moderate-intensity aerobic exercise for obsessive compulsive
disorder. The Journal of Nervous and Mental Disease, 19, 514–520.
Promise of Exercise Interventions for the Anxiety Disorders 99

Calfas, K. J., Sallis, J. F., Zabinski, M. F., Wilfley, D. E., Rupp, J., Prochaska, J. J., et al.
(2002). Preliminary evaluation of a multicomponent program for nutrition and
physical activity change in primary care: PACEþ for adults. Preventative Medicine,
34(2), 153–61.
Camacho, T. C., Roberts, R. E., Lazarus, N. B., Kaplan, G. A., & Cohen, R. D. (1991).
Physical activity and depression: evidence from the Alameda County Study. American
Journal of Epidemiology, 134(2), 220–231.
Cameron, O. G., & Hudson, C. J. (1986). Influence of exercise on anxiety level in patients with
anxiety disorders. Psychosomatics, 27(10), 720–723.
Caspersen, C. J., Powell, K. E., & Christenson, G. M. (1985). Physical activity, exercise, and
physical fitness: definitions and distinctions for health-related research. Public Health
Reports, 100(2), 126–131.
Chaouloff, F. (1997). Effects of acute physical exercise on central serotonergic systems.
Medicine & Science in Sports & Exercise, 29(1), 58–62.
Chen, C. N. (1979). Sleep, depression and antidepressants. British Journal of Psychiatry, 135,
Chipkin, S. R., Klugh, S. A., & Chasan-Taber, L. (2001). Exercise and diabetes. Cardiology
Clinics, 19(3), 489–505.
Christensen-Szalanski, J. J., & Northcraft, G. B. (1985). Patient compliance behavior: the
effects of time on patients’ values of treatment regimens. Social Sciences & Medicine, 21(3),
Coryell, W., Noyes, R., & Clancy, J. (1982). Excess mortality in panic disorder: A comparison
with primary unipolar depression. Archives of General Psychiatry 39, 701–703.
Coryell, W., Noyes, R., & House, J. D. (1986). Mortality among outpatients with anxiety
disorders. American Journal of Psychiatry, 143, 508–510
Dishman, R. K., Armstrong, R. B., Delp, M. D., Graham, R. E., & Dunn, A. L. (1988).
Open-field behavior is not related to treadmill performance in exercising rats. Physiology &
Behavior, 43(5), 541–546.
Dishman, R. K., Dunn, A. L., Youngstedt, S. D., Davis, J. M., Burgess, M. L., Wilson, S. P.,
et al. (1996). Increased open field locomotion and decreased striatal GABAa binding after
activity wheel running. Physiology & Behavior, 60(3), 699–705.
Doyne, E. J., Ossip-Klein, D. J., Bowman, E. D., Osborn, K. M., McDougall-Wilson, I. B., &
Neimeyer, R. A. (1987). Running versus weight lifting in the treatment of depression.
Journal of Consulting and Clinical Psychology, 55(5), 748–754.
Dunn, A. L., & Dishman, R. K. (1991). Exercise and the neurobiology of depression. Exercise
and Sport Sciences Reviews, 19, 41–98.
Dunn, A. L., Marcus, B. H., Kampert, J. B., Garcia, M. E., Kohl, H. W., 3rd, & Blair, S. N.
(1999). Comparison of lifestyle and structured interventions to increase physical activity
and cardiorespiratory fitness: a randomized trial. JAMA, 281(4), 327–34.
Dunn, A. L., Trivedi, M. H., Kampert, J. B., Clark, C. G., & Chambliss, H. O. (2005).
Exercise treatment for depression: efficacy and dose response. American Journal of
Preventive Medicine, 28(1), 1–8.
Ehlers, C. L., Havstad, J. W., & Kupfer, D. J. (1996). Estimation of the time course of
slow-wave sleep over the night in depressed patients: effects of clomipramine and clinical
response. Biological Psychiatry, 39(3), 171–181.
Ekkekakis, P., & Petruzzello, S. J. (1999). Acute aerobic exercise and affect: current status,
problems and prospects regarding dose-response. Sports Medicine, 28(5), 337–374.
Ekkekakis, P., Hall, E. E., & Petruzzello, S. J. (1999). Measuring state anxiety in the context of
acute exercise using the State Anxiety Inventory: An attempt to resolve the brouhaha.
Journal of Sport & Exercise Psychology, 21, 205–229.
Esquivel, G., Schruers, K., Kuipers, H., & Griez, E. (2002). The effects of acute exercise and
high lactate levels on 35% CO2 challenge in healthy volunteers. Acta Psychiatrica
Scandavia, 106(5), 394–397.
100 J. A. Smits et al.

Farmer, M. E., Locke, B. Z., Mosciki, E. K. Dannenberg A. L., Larson, D. B., & Radloff, L. S.
(1998). Physical activity and depressive symptoms: the NHANES I epidemiologic follow-
up study. American Journal of Epidemiology, 128, 1340–1351.
Fava, M., McCall, W., Krystal, A., Wessel, T., Rubens, R., Caron, J., et al. (2006). Eszopi-
clone Co-Administered With Fluoxetine in Patients With Insomnia Coexisting With
Major Depressive Disorder. Biological Psychiatry, 59(11), 1052–1060.
Fuller, K. H., Waters, W. F., Binks, P. G., & Anderson, T. (1997). Generalized anxiety and
sleep architecture: a polysomnographic investigation. Sleep, 20(5), 370–376.
Gaffney, F. A., Fenton, B. J., Lane, L. D., & Lake, C. R. (1988). Hemodynamic,
ventilatory, and biochemical responses of panic patients and normal controls with
sodium lactate infusion and spontaneous panic attacks. Archives of General Psychiatry,
45(1), 53–60.
Gomez-Caminero, A., Blumentals, W. A., Russo, L. J., Brown, R. R., & Castilla-Puentes, R.
(2005). Does panic disorder increase the risk of coronary heart disease? A cohort study of a
national managed care database. Psychosomatic Medicine, 67(5), 688–691.
Goode, K. T., & Roth, D. L. (1993). Factor analysis of cognitions during running: Associa-
tion with mood change. Journal of Sport and Exercise Psychology, 15, 375–389.
Goodwin, R. D. (2003). Association between physical activity and mental disorders among
adults in the United States. Preventive Medicine, 36(6), 698–703.
Hassmen, P., Koivula, N., & Uutela, A. (2000). Physical exercise and psychological well-being:
a population study in Finland. Preventive Medicine, 30(1), 17–25.
Horne, J. A. (1981). The effects of exercise upon sleep: a critical review. Biological Psychology,
12(4), 241–290.
Horne, J. A., & Moore, V. J. (1985). Sleep EEG effects of exercise with and without additional
body cooling. Electroencephalography and clinical Neurophysiology, 60(1), 33–38.
Horne, J. A., & Shackell, B. S. (1987). Slow wave sleep elevations after body heating:
proximity to sleep and effects of aspirin. Sleep, 10(4), 383–392.
Horne, J. A., & Staff, L. H. (1983). Exercise and sleep: body-heating effects. Sleep, 6(1), 36–46.
Hyttel, J. (1994). Pharmacological characterization of selective serotonin reuptake inhibitors
(SSRIs). International Clinical Psychopharmacology, 9( Suppl 1), 19–26.
Jones, D. L., & Mogenson, G. J. (1980a). Nucleus accumbens to globus pallidus GABA
projection subserving ambulatory activity. American Journal of Physiology, 238(1),
Jones, D. L., & Mogenson, G. J. (1980b). Nucleus accumbens to globus pallidus GABA
projection: electrophysiological and iontophoretic investigations. Brain Research, 188(1),
Jones, D. L., Mogenson, G. J., & Wu, M. (1981). Injections of dopaminergic, cholinergic,
serotoninergic and GABAergic drugs into the nucleus accumbens: effects on locomotor
activity in the rat. Neuropharmacology, 20(1), 29–37.
Kawachi, I., Colditz, G. A., Ascherio, A., Rimm, E. B., Giovannucci, E., Stampfer, M. J.,
et al. (1994). Prospective study of phobic anxiety and risk of coronary heart disease in men.
Circulation, 89(5), 1992–1997.
Kawachi, I., Sparrow, D., Vokonas, P. S., & Weiss, S. T. (1994). Symptoms of anxiety and risk
of coronary heart disease. The Normative Aging Study. Circulation, 90(5), 2225–2229.
Kawachi, I., Sparrow, D., Vokonas, P. S., & Weiss, S. T. (1995). Decreased heart rate
variability in men with phobic anxiety (data from the Normative Aging Study). American
Journal of Cardiology, 75(14), 882–885.
Kessler, R. C., Soukup, J., Davis, R. B., Foster, D. F., Wilkey, S. A., Van Rompay, M. I.,
et al. (2001). The use of complementary and alternative therapies to treat anxiety and
depression in the United States. American Journal of Psychiatry, 158(2), 289–294.
King, A. C., Oman, R. F., Brassington, G. S., Bliwise, D. L., & Haskell, W. L. (1997).
Moderate-intensity exercise and self-rated quality of sleep in older adults. A randomized
controlled trial. JAMA, 277(1), 32–37.
Promise of Exercise Interventions for the Anxiety Disorders 101

Kohl, H. W., 3rd. (2001). Physical activity and cardiovascular disease: evidence for a dose
response. Medicine & Science in Sports & Exercise, 33(6 Suppl), S472–S483; discussion
Kupfer, D. J., Ehlers, C. L., Frank, E., Grochocinski, V. J., McEachran, A. B., & Buhari, A.
(1994). Persistent effects of antidepressants: EEG sleep studies in depressed patients during
maintenance treatment. Biological Psychiatry, 35(10), 781–793.
Lee, I. M., Hsieh, C. C., & Paffenbarger, R. S., Jr. (1995). Exercise intensity and longevity in
men. The Harvard Alumni Health Study. JAMA, 273(15), 1179–1184.
Lee, I. M., & Paffenbarger, R. S., Jr. (2000). Associations of light, moderate, and vigorous
intensity physical activity with longevity. The Harvard Alumni Health Study. American
Journal of Epidemiology, 151(3), 293–299.
Lee, I. M., Paffenbarger, R. S., Jr, & Hsieh, C. (1991). Physical activity and risk of developing
colorectal cancer among college alumni. Journal of the National Cancer Institute, 83(18),
Lee, I. M., Paffenbarger, R. S., Jr, & Hsieh, C. C. (1992). Physical activity and risk of
prostatic cancer among college alumni. American Journal of Epidemiology, 135(2),
Lee, I. M., Sesso, H. D., & Paffenbarger, R. S., Jr. (1999). Physical activity and risk of lung
cancer. International Journal of Epidemiology, 28(4), 620–625.
Leith, L. M. (1994). Foundations of exercise and mental health. Morgantown, WV: Fitness
Information Technology.
Long, B. C., & van Stavel, R. (1995). Effects of exercise training on anxiety: A meta-analysis.
Journal of Applied Sport Psychology, 7(2), 167–189.
Marcus, B. H., Dubbert, P. M., Forsyth, L. H., McKenzie, T. L., Stone, E. J., Dunn, A. L.,
et al. (2000). Physical activity behavior change: issues in adoption and maintenance.
Health Psychology, 19(1 Suppl), 32–41.
Margraf, J., Barlow, D. H., Clark, D. M., & Telch, M. J. (1993). Psychological treatment of
panic: work in progress on outcome, active ingredients, and follow-up. Behaviour Research
& Therapy, 31(1), 1–8.
Marshall, A. L., Baumann, A. E., Own, N., Booth, M. L., Crawford, D., & Marcus, B. H.
(2003). Population-based randomized controlled trial of a stage-targeted physical activity
intervention. Annals of Behavioral Medicine, 25, 194–202.
Martinsen, E. W., Hoffart, A., & Solberg, O. (1989). Comparing aerobic with nonaerobic
forms of exercise in the treatment of clinical depression: a randomized trial. Comprehensive
Psychiatry, 30(4), 324–331.
Martinsen, E. W., Raglin, J. S., Hoffart, A., & Friis, S. (1998). Tolerance to intensive
exercise and high levels of lactate in panic disorder. Journal of Anxiety Disorders, 12(4),
McNally, R. J. (2002). Anxiety sensitivity and panic disorder. Biological Psychiatry, 52(10),
McWilliams, L. A., & Asmundson, G. J. (2001). Is there a negative association between
anxiety sensitivity and arousal-increasing substances and activities? Journal of Anxiety
Disorders, 15(3), 161–170.
Meeusen, R., & De Meirleir, K. (1995). Exercise and brain neurotransmission. Sports
Medicine, 20(3), 160–188.
Meyer, T., Broocks, A., Bandelow, B., Hillmer-Vogel, U., & Ruther, E. (1998). Endurance
training in panic patients: spiroergometric and clinical effects. International Journal of
Sports Medicine, 19(7), 496–502.
Morgan, W. P., Olson, E. B., Jr, & Pedersen, N. P. (1982). A rat model of psychopathology for
use in exercise science. Medicine & Science in Sports & Exercise, 14(1), 91–100.
Moses, J., Steptoe, A., Mathews, A., & Edwards, S. (1989). The effects of exercise training on
mental well-being in the normal population: A controlled trial. Journal of Psychosomatic
Research , 33(1), 47–61.
102 J. A. Smits et al.

Naylor, E., Penev, P. D., Orbeta, L., Janssen, I., Ortiz, R., Colecchia, E. F., et al. (2000). Daily
social and physical activity increases slow-wave sleep and daytime neuropsychological
performance in the elderly. Sleep, 23(1), 87–95.
O’Connor, P. J. (2005). State anxiety is reduced after maximal and submaximal exercise
among people with panic disorder. International Journal of Sport and Exercise Psychology,
3(4), 501–508.
Ockene, I. S., Hayman, L. L., Pasternak, R. C., Schron, E., & Dunbar-Jacob, J. (2002).
Task force #4–adherence issues and behavior changes: achieving a long-term solution.
33rd Bethesda Conference. Journal of the American College of Cardiology, 40(4),
Paffenbarger, R. S., Jr, Lee, I. M., & Leung, R. (1994). Physical activity and personal
characteristics associated with depression and suicide in American college men. Acta
Psychiatrica Scandanavia, 377, 16–22.
Petruzzello, S. J., Landers, D. M., Hatfield, B. D., Kubitz, K. A., & Salazar, W. (1991).
A meta-analysis on the anxiety-reducing effects of acute and chronic exercise. Outcomes
and mechanisms. Sports Medicine, 11(3), 143–182.
Plotnikoff, R. C., Hotz, S. B., Birkett, N. J., & Courneya, K. S. (2001). Exercise and the
transtheoretical model: a longitudinal test of a population sample. Preventative Medicine,
33(5), 441–452.
Prochaska, J. O., & DiClemente, C. C. (1983). Stages and processes of self-change of smoking:
toward an integrative model of change. Journal of Consulting and Clinical Psychology,
51(3), 390–395.
Rejeski, W. J., Thompson, A., Brubaker, P. H., & Miller, H. S. (1992). Acute exercise:
buffering psychosocial stress responses in women. Health Psychology, 11(6), 355–362.
Rief, W., & Hermanutz, M. (1996). Responses to activation and rest in patients with panic
disorder and major depression. British Journal of Clinical Psychology, 35( Pt 4), 605–616.
Rodriguez, B. F., Bruce, S. E., Pagano, M. E., & Keller, M. B. (2005). Relationships among
psychosocial functioning, diagnostic comorbidity, and the recurrence of generalized anxi-
ety disorder, panic disorder, and major depression. Journal of Anxiety Disorders, 19(7),
Ross, R., & Janssen, I. (2001). Physical activity, total and regional obesity: dose-response
considerations. Medicine & Science in Sports & Exercise, 33(6 Suppl), S521–527
Roter, D. L., Hall, J. A., Merisca, R., Nordstrom, B., Cretin, D., & Svarstad, B. (1998).
Effectiveness of interventions to improve patient compliance: a meta-analysis. Medical
Care, 36(8), 1138–61.
Royce, J. R. (1977). On the construct validity of open-field measures. Psychological Bulletin,
84(6), 1098–1106.
Schlicht, W. (1994). Does physical exercise reduce anxious emotions? A meta-analysis.
Anxiety, Stress & Coping: An International Journal, 6(4), 275–288.
Schmidt, N. B., Lerew, D. R., & Jackson, R. J. (1997). The role of anxiety sensitivity in the
pathogenesis of panic: prospective evaluation of spontaneous panic attacks during acute
stress. Journal of Abnormal Psychology, 106(3), 355–364.
Schmidt, N. B., Lerew, D. R., & Jackson, R. J. (1999). Prospective evaluation of anxiety
sensitivity in the pathogenesis of panic: replication and extension. Journal of Abnormal
Psychology, 108(3), 532–537.
Schmidt, N. B., Lerew, D. R., Santiago, H., Trakowski, J. H., & Staab, J. P. (2000). Effects of
heart-rate feedback on estimated cardiovascular fitness in patients with panic disorder.
Depression and Anxiety, 12(2), 59–66.
Schmitz, N., Kruse, J., & Kugler, J. (2004). The association between physical exercises and
health-related quality of life in subjects with mental disorders: results from a
cross-sectional survey. Preventive Medicine, 39, 1200–1207.
Schoeller, D. A. (1988). Measurement of energy expenditure in free-living humans by using
doubly labeled water. Journal of Nutrition, 118(11), 1278–1289.
Promise of Exercise Interventions for the Anxiety Disorders 103

Schoenborn, C. A., Adams, P. F., Barnes, P. M., Vickerie, J. L., & Schiller, J. S. (2004). Health
behaviors of adults: United States, 1999–2001. Vital Health Statistics 10, (219), 1–79.
Shapiro, C. M., Griesel, R. D., Bartel, P. R., & Jooste, P. L. (1975). Sleep patterns afted
graded exercise. Journal of Applied Physiology, 39(2), 187–190.
Shipley, J. E., Kupfer, D. J., Dealy, R. S., Griffin, S. J., Coble, P. A., McEachran, A. B.,
et al. (1984). Differential effects of amitriptyline and of zimelidine on the sleep electro-
encephalogram of depressed patients. Clinical Pharmacology and Therapeutics, 36(2),
Singh, N. A., Clements, K. M., & Fiatarone, M. A. (1997). A randomized controlled trial of
the effect of exercise on sleep. Sleep, 20(2), 95–101.
Singh, N. A., Stavrinos, T. M., Scarbek, Y., Galambos, G., Liber, C., & Fiatarone Singh, M. A.
(2005). A randomized controlled trial of high versus low intensity weight training versus
general practitioner care for clinical depression in older adults. The Journals of Gerontology.
Series A, Biological Sciences and Medical Sciences, 60(6), 768–776.
Sirey, J. A., Bruce, M. L., Alexopoulos, G. S., Perlick, D. A., Friedman, S. J., & Meyers, B. S.
(2001). Stigma as a barrier to recovery: Perceived stigma and patient-rated severity of
illness as predictors of antidepressant drug adherence. Psychiatric Services, 52(12),
Smits, J. A. J., Berry, A. C., Rosenfield, D., Powers, M. B., Behar, E., & Otto, M. W.
(In Press). Reducing anxiety sensitivity with exercise. Depression and Anxiety.
Smits, J. A. J., Powers, M. B., Berry, A. C., & Otto, M. W. (2007). Translating
empirically-supported strategies into accessible interventions: The potential utility of
exercise for the treatment of panic disorder. Cognitive and Behavioral Practice, doi:
Smits, J. A., Powers, M. B., Cho, Y., & Telch, M. J. (2004). Mechanism of change in
cognitive-behavioral treatment of panic disorder: evidence for the fear of fear mediational
hypothesis. Journal of Consulting and Clinical Psychology, 72(4), 646–652.
Smits, J. A., & Zvolensky, M. J. (2006). Emotional vulnerability as a function of physical
activity among individuals with panic disorder. Depression and Anxiety, 23(2), 102–106.
Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual
for the State Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press.
Stathopoulou, G., Powers, M. B., Berry, A. C., Smits, J. A. J., & Otto, M. W. (2006). Exercise
Interventions for Mental Health: A Quantitative and Qualitative Review. Clinical
Psychology: Science and Practice, 13(2), 179–193.
Stein, M. B., Tancer, M. E., & Uhde, T. W. (1992). Heart rate and plasma norepinephrine
responsivity to orthostatic challenge in anxiety disorders. Comparison of patients with
panic disorder and social phobia and normal control subjects. Archives of General
Psychiatry, 49(4), 311–317.
Stephens, T. (1988). Physical activity and mental health in the United States and Canada:
evidence from four population surveys. Preventive Medicine, 17(1), 35–47.
Strohle, A., Feller, C., Onken, M., Godemann, F., Heinz, A., & Dimeo, F. (2005). The acute
antipanic activity of aerobic exercise. American Journal of Psychiatry, 162(12), 2376–2378.
Taylor, C. B., King, R., Ehlers, A., Margraf, J., Clark, D., Hayward, C., et al. (1987).
Treadmill exercise test and ambulatory measures in panic attacks. American Journal of
Cardiology, 60(18), 48J–52J.
Tharp, G. D., & Carson, W. H. (1975). Emotionality changes in rats following chronic
exercise. Medicine & Science in Sports, 7(2), 123–126.
Trivedi, M. H., Greer, T. L., Grannemann, B. D., Chambliss, H. O., Jordan, A. N. (2006a).
Exercise as an augmentation strategy for treatment of major depression. Journal of
Psychiatric Practice, 12, 205–213.
Trivedi, M. H., Greer, T. L., Grannemann, B. D., Church, T. S., Galper, D. I., Sunderajan, P.,
et al. (2006b). TREAD: TReatment with Exercise Augmentation for Depression: study
rationale and design. Clinical Trials, 3(3), 291–305.
104 J. A. Smits et al.

US President’s Council for Physical Fitness and Sports. (2000). Definitions of health, fitness
and physical activity. Research Digest, 3, 1–8.
Van den Hout, M. A., van der Molen, G. M., Griez, E., Lousberg, H., & Nansen, A. (1987).
Reduction of CO2-induced anxiety in patients with panic attacks after repeated CO2
exposure. American Journal of Psychiatry, 144(6), 788–791.
Wannamethee, G., & Sharper, A. G. (1992). Physical activity and stroke in British middle
aged men. British Medical Journal, 304, 597–601.
Weber, J. C., & Lee, R. A. (1968). Effects of differing prepuberty exercise programs on the
emotionality of male albino rats. Research Quarterly, 39, 748–751.
Wilson, K. A., & Hayward, C. (2005). A prospective evaluation of agoraphobia and depres-
sion symptoms following panic attacks in a community sample of adolescents. Journal of
Anxiety Disorders, 19, 87–103.
Wilson, W. M., & Marsden, C. A. (1996). In vivo measurement of extracellular serotonin in
the ventral hippocampus during treadmill running. Behavioural Pharmacology, 7,
Youngstedt, S. D., Dishman, R. K., Cureton, K. J., & Peacock, L. J. (1993). Does body
temperature mediate anxiolytic effects of acute exercise? Journal of Applied Physiology, 74,
Youngstedt, S. D., O’Connor, P. J., Crabbe, J. B., & Dishman, R. K. (1998). Acute exercise
reduces caffeine-induced anxiogensis. Medicine & Science in Sports & Exercise, 30,
Zvolensky, M. J., & Eifert, G.H. (2001). A review of psychological factors/processes affecting
anxious responding during voluntary hyperventilation and inhalations of carbon dioxide-
enriched air. Clinical Psychology Review, 21, 375 –400.
Anxiety and Insomnia
Theoretical Relationship and Future Research

Thomas W. Uhde and Bernadette M. Cortese


The major focus of this chapter is to identify future research directions for
advancing knowledge about two phenomena, anxiety and insomnia, which
are highly prevalent, co-existent problems in humans. Despite this universal
observation across different cultures, the medical field has a poor under-
standing of the pathophysiology and causal relationships between anxiety
and insomnia. The physiological and neurobiological relationships between
anxiety and insomnia remain one of medicine’s mysteries. Given the promi-
nence and relevance of these co-occurring complaints among individuals
seeking medical care, the lack of research is somewhat surprising and may
be a by-product of conceptualizing anxiety and insomnia as simply non-
specific manifestations of most systemic diseases. Such attributions impede
the development of new treatments that can improve appreciably the health
and well-being of not only patients with primary insomnia and anxiety dis-
orders but also individuals with cancer, metabolic, and other multi-system
medical diseases.
The goal of this chapter is to encourage clinicians and investigators to think
about anxiety-insomnia from different theoretical, even highly speculative
perspectives, with the hope that new areas of research will be undertaken by
the research community. We understand that some ideas regarding future
research directions are not necessarily logical ‘‘next step’’ studies, but rather,
areas of high-risk investigation that in their own right (and in the opinion of the
authors) may lead to an improved, understanding of the anxiety-insomnia

Thomas W. Uhde
Department of Psychiatry and Behavioral Sciences, Medical University of South
Carolina (MUSC), 67 President Street, 5 South, Charleston, SC

M. J. Zvolensky, J. A. Smits (eds.), Anxiety in Health Behaviors and Physical Illness. 105
Ó Springer 2008
106 T. W. Uhde, B. M. Cortese


Anxiety and insomnia are multidimensional phenomena that largely rely on

impairment criteria to achieve status as a disorder. There are two primary classi-
fication systems used by clinicians for the diagnosis of anxiety and sleep disorders:
Diagnostic and Statistical Manual (DSM-IV) and Tenth Revision of the Interna-
tional Classification of Diseases (ICD-10). While the Diagnostic and Statistical
Manual subdivides the anxiety disorders into a number of subtypes, the essential
feature of the anxiety disorders is impairment in work or social function or
ongoing distress as a result of the condition. For the purposes of this chapter,
unless otherwise indicated by reference to a specific anxiety disorder (e.g. panic
disorder), we are referring to the overall category of the anxiety disorders.
For the purposes of this chapter (and consistent with the conceptualization of
most sleep experts), insomnia refers to the subjective experience of having restless
and un-refreshing sleep, including difficulty falling asleep, multiple nocturnal
awakenings, or early morning waking. Later in this chapter, we make specific
distinctions between the subjective experience of insomnia and absolute sleep
restriction (i.e. insufficient amount of objective EEG sleep). The rationale for
underscoring this point is that there may be a large gap between subjective
experiences of sleep duration versus objective criteria of the hours of sleep per
night or sleep efficiency (i.e. the percent of time in bed when the person is actually
sleeping). In fact, a characteristic of many individuals suffering from insomnia
may be the lack of objective evidence of profound restricted sleep by either
polysomnography or multiple sleep latency testing (MSLT), despite the percep-
tion of having had little or no sleep. Thus, for the purposes of this chapter,
insomnia refers to a subjective experience of poor quality of sleep, which may or
may not be associated with an actual decrease in the normal amounts of sleep.

Theoretical Models

The high prevalence of co-morbid insomnia in primary anxiety disorders and

co-morbid anxiety symptoms in primary insomnia suggest an important under-
lying relationship between these clinical entities. However, using cross-sectional
methods, it is also known that anxiety symptoms, including the cognitive
process of worry, and insomnia are not 100% co-existent in all phases of illness
in either patients with primary insomnia or any primary anxiety disorder. These
observations can be explained on the basis of three theoretical constructs: a)
anxiety (or a diagnostic subtype) and primary insomnia represent a single,
spectrum disorder with a common diathesis (and predictable evolution of
symptoms) (see Fig. 1); b) anxiety and insomnia are separate and distinct
pathological conditions, each of which cause, permit, or promote the down-
stream development of secondary complications (Fig. 2) and c) anxiety and
insomnia are each caused by another critical, independent factor (Fig. 3). Each
Anxiety and Insomnia 107

Fig. 1 In the single spectrum disorder model, anxiety (or a diagnostic subtype) and insomnia
represent the same disorder with a common diathesis and predictable evolution of symptoms.
In this model, the common underlying pathological process (i.e., neurobiological abnorm-
ality) leads to different symptoms or components (e.g. anxiety or insomnia) of the illness

Fig. 2 In the different disorders with secondary complications model, anxiety and insomnia are
separate and distinct pathological conditions with different underlying neurobiological
abnormalities. Early stages of the disorder may reflect this distinction, while later stages
show increasingly similar symptoms (e.g. anxiety and insomnia) due to the downstream
development of secondary complications
108 T. W. Uhde, B. M. Cortese

Fig. 3 In the third factor

model, anxiety and insom-
nia are each caused by a cri-
tical, independent factor(s)
that separately influences
each disorder. These third or
‘‘other’’ factors might be
tal, endogenous/neurobiolo-
gical or familial/genetic in
nature, or possibly involve
contributions from any or
all of these elements

of these theoretical models might explain high co-morbid rates of insomnia and
anxiety observed in patients seeking treatment for an anxiety disorder or
primary insomnia.

What criteria would support a single (spectrum) versus different

disorders versus third factor models for anxiety and insomnia?

A. Single-Spectrum Disorder: In a single-spectrum disorder model, there is

really just one underlying neurobiological disorder that is characterized by
modest variations in clinical presentation. If anxiety (or one of its diagnostic
subtypes) and insomnia are the same basic disorder, one should be able to
document identical pathological mechanisms or neurobiological abnormal-
ities in each of the two clinical syndromes. Thus, either the anxiety or an
anxiety diagnostic subtype (e.g. panic disorder) and insomnia syndromes
would share common neurobiological abnormalities. A medical analog of a
single-spectrum disorder might be multiple sclerosis, wherein a common
underlying pathological process (e.g. demyelization) leads to different symp-
toms at different phases of illness (Fig. 1).
In terms of investigating whether two apparently different clinical
syndromes (i.e. anxiety and insomnia) actually represent a single disease
entity, one would study and expect to find similar ages of onset of illness,
gender distributions, type of symptoms, course of illness, and response to
treatment for the two candidate conditions. One would also predict that
cultural, education, and related expectancy biases might largely contribute to
whether people with the ‘‘same’’ disorder identified themselves as having a
‘‘sleep’’ versus ‘‘anxiety’’ problem. The single-spectrum disorder theory,
therefore, suggests that ‘‘insomnia’’ and ‘‘anxiety’’ (or one or more diagnostic
subtypes of anxiety) are the same single-spectrum disorder and that
non-biological, socio-cultural factors largely determine whether patients
self-identify themselves as having an anxiety or sleep disorder.
Anxiety and Insomnia 109

B. Different Disorders: In contrast to a single-spectrum disorder, distinctly

different disorders would have different underlying abnormalities and be
highly likely to reflect dissimilarities in onset and course of illness, medical
complications and treatment response. However, depending on the under-
lying diathesis or etiology of illness, one might anticipate downstream com-
plications that result in overlapping clinical symptoms. If these are truly
different disease entities, the initial presentation and associated clinical
features would be different. Nonetheless, two different diseases could be
associated with increasingly similar symptoms during later stages of illness
insofar as secondary complications might recruit common neuro-anatomical
substrates or neurobiological systems. Thus, either anxiety or a diagnostic
anxiety subtype and insomnia might each increase the likelihood of devel-
oping secondary complications, which, when the initial and secondary com-
plications are taken together, produce similar clinical profiles. A medical
disease analog might be hypertension leading to impaired renal function
which would share many of the same symptoms as primary renal disease
associated with secondary hypertension (Fig. 2).
To explain high rates of co-morbid anxiety and insomnia wherein each
(i.e. anxiety and insomnia) are fundamentally different disorders would
suggest that we could identify two syndromes: a) patients presenting with
initial insomnia followed by the development of pathological anxiety (or one
of its diagnostic subtypes) and b) patients that present initially with anxiety
or an anxiety disorder subtypes which is later associated with insomnia.
Furthermore, one would predict different (particularly early in the course
of illness) but overlapping (especially in later stages of illness) neuro-
biological abnormalities in the anxiety-insomnia versus insomnia-anxiety
syndromes. Moreover, one would predict that these two syndromes would
have differential responses to treatment, as well as dissimilarities in age of
onset and other demographic variables.
C. Third Factor Models: High co- morbid rates of anxiety and insomnia could
also be explained on the basis of an unknown third factor or factors that have
a direct but independent impact on anxiety and insomnia. In this circum-
stance, third factor(s) would be necessary or contributory but insufficient
alone to produce anxiety and insomnia. Like the single-spectrum and sepa-
rate diagnoses models, the third factors model could explain high co-rates of
insomnia and anxiety, respectively, in people seeking treatment in mental
health (or an anxiety subspecialty program) versus sleep medicine clinics.
These third or ‘‘other’’ factors might be psychosocial/environmental, endo-
genous/neurobiological or familial/genetic in nature, or possibly involve
contributions from any or all of these elements (Fig. 3).
An example of an exogenous-external factor that has an impact on two
separate medical but co-morbidly associated disorders is ultraviolet light/sun
exposure. Data suggest a higher than expected association between patients
with subacute cutaneous lupus erythematosus (SCLE) and polymorphous
light eruption (PLE) (Millard, Lewis et al., 2001; Millard, Kondeatis et al.,
110 T. W. Uhde, B. M. Cortese

2001). Although thought to be different medical conditions, both are

made more evident or severe by sun exposure. Still other third factors
(e.g. genetic polymorphisms) (Millard, Kondeatis et al., 2001) may contribute
to the co-morbid associations of SCLE and PLE, as has been similarly
proposed for apolipoprotein E gene polymorphism in patients with high
HDL-cholesterol levels and dermatophystosis (Tursen et al., 2004). Likewise,
hypercholesterolemia can be conceptualized as playing a third factor role in
the co-morbid association of cardiovascular disease and type 2 diabetes.
Such third factors might be hypothesized for insomnia and anxiety dis-
orders, wherein each is a truly separate neurobiological disease entity but
both suffer an exacerbation of severity due to external factors (e.g. sleep
deprivation) or, in fact, might even share a common genetic polymorphism.
If third factor(s) are contributing to the high prevalence of co-morbid
anxiety and insomnia, we should be able to identify differences in course of
illness and treatment responses in direct relation to the degree to which that
outside ‘‘third’’ factor distinctly and separately influences the essential
qualities of anxiety and insomnia. It should be noted that an outside factor
may affect different components (e.g. symptoms, severity) of anxiety versus
insomnia, although the impact of the third factor should have a consistent,
predictable, and identifiable role in explaining one or more common and
critical feature(s) of anxiety and insomnia.

What Evidence Supports a Single (Spectrum) Versus Different

Disorders Versus Third Factor Models for Anxiety and Insomnia?

A. Age: Age of onset could help to distinguish anxiety and insomnia as separate
and distinct disorders if it was found to be considerably different for these
two disorders. Recent data from the World Mental Health Survey (WMH)
describes an early age of onset for some anxiety disorders, with a later onset
for others. For example, phobias and separation anxiety disorder was found
to have a median age of onset in the range of 7–14 years (interquartile range
[IQR] = 4–20 years). The age of onset distributions for panic, generalized
anxiety, and posttraumatic disorders, on the other hand, ranged from 25 to
53 years (IQR = 15–75 years; Kessler et al., 2007). The NCS-R median age
of 11 for the onset for anxiety reported by Kessler et al. (2005) and a 7 year
(IQR = 7 years) median age of onset for a first anxiety disorder reported by
Johnson, Roth, and Breslau (2006) are consistent with the WMH results and
further demonstrate the early age of onset for anxiety.
With respect to insomnia, the evidence supports the view that older
individuals are most at risk (Morphy, Dunn, Lewis, Boardman, & Croft,
2007). In fact, several studies reveal that nearly half of individuals over the
age of 65 report sleeping difficulties and insomnia-like symptoms (Monane,
1992; Foley et al., 1995; Ganguli, Reynolds, & Gilby, 1996).
Anxiety and Insomnia 111

Fewer studies have specifically assessed the age of onset for insomnia. In
one retrospective analysis of adolescents from the ages 13 to 15 years,
Johnson et al. (2006) reported a median age of onset for insomnia at age 11
(IQR = 5 years). Although this study found a relatively young age of onset
for children identified specifically with insomnia, most other studies with
children assess individuals with unspecified sleep problems that includes
insomnia, but could also include numerous other difficulties surrounding
sleep (Paavonen, Solantaus, Almqvist, & Aronen, 2003; Gregory &
O’Connor, 2002; Johnson, Chilcoat, & Breslau, 2000), making it difficult
to draw clear conclusions about age of onset for insomnia.
B. Gender: A second factor that could distinguish anxiety and insomnia as a
single spectrum disorder versus two separate disorders is gender distribution.
With respect to both insomnia and anxiety, data suggest a similar gender
distribution for these two disorders. Anxiety, for example, is nearly twice as
prevalent in women as in men (Kessler et al., 1994, 2005). Although not as
consistent as the anxiety data, most studies report higher prevalence rates for
females than males, with female gender as a strong risk factor for insomnia
(Ford & Kamerow, 1989; Mellinger, Balter, & Uhlenhuth, 1985; Bixler,
Vgontzas, Lin, Vela-Bueno, & Kales, 2002; Zhang & Wing, 2006).
C. Symptoms and Sleep EEG: Insomnia, as a subjective sleep complaint, is
reported by many patients with anxiety disorders. The report of insomnia
is so prevalent among patients with anxiety disorders that it is widely
assumed among clinicians, and even by most anxiety disorder specialists,
that the treatment of core anxiety symptoms in anxiety disorder patients will
be associated with parallel improvement in sleep quality (see discussion).
Some anxiety disorders, however, appear to be more commonly associated
with insomnia and two specific anxiety disorders, panic disorder and gen-
eralized anxiety disorder, deserve special mention insofar as they appear to
share important but different symptom characteristics compared with
patients with primary insomnia.
Patients with generalized anxiety disorder often have subjective com-
plaints that are nearly identical to those reported by patients with primary
insomnia. These include worrying about obtaining good quality or sufficient
amounts of sleep and problems falling or maintaining sleep. Patients with
primary insomnia, essentially by operational criteria, report almost identical
sleep complaints, but, for reasons that remain unclear, experience their
complaints within the context of a sleep problem, rather than within the
context of a mental health problem. Patients with GAD and primary insom-
nia typically report symptoms of physiological hyperarousal and increased
vigilance (‘‘feeling keyed up’’) with intermittent fatigue and disturbances in
concentration and memory. There are also nearly identical polysomno-
graphic findings; sleep architecture and REM measures are normal whereas
both syndromes have EEG evidence of increased sleep latency and distur-
bances in maintaining sleep (for reviews, see Papadimitriou & Linikowski,
2005; Uhde, 2000).
112 T. W. Uhde, B. M. Cortese

Of interest, patients with GAD, when awakened under experimental

conditions after several minutes of EEG-documented stage 2 sleep, report
not having been asleep or experiencing any drowsiness whatsoever. These
observations, combined with our ongoing studies in patients with recurrent
sleep paralysis indicate that some individuals find it nearly impossible to
distinguish between awake and sleep states, including REM-stage sleep. We
have reported that patients with recurrent sleep paralysis, as well as some
patients with GAD and PD (especially those with nocturnal panic attacks)
find it difficult to separate at the experiential level the difference between
sleep versus wakeful states (Uhde et al., 2006). Thus, the study of sleep in
patients with primary insomnia, as well as patients with selective anxiety
disorders and recurrent sleep paralysis may be of interest to neuroscientists
investigating the neurobiological and theoretical basis of consciousness. In
fact, the impressive degree of self-awareness reported by patients with recur-
rent sleep paralysis during sleep has been described by our research team as
‘‘consciousness intruding upon REM stage sleep’’ or as ‘‘sleep consciousness’’
(Uhde et al., 2006).
Taken together, these observations suggest that GAD and panic disorder,
particularly patients with nocturnal panic attacks, share many symptomatic
and sleep EEG characteristics with chronic insomnia sufferers. On the other
hand, certain other anxiety disorders such as social phobia (Brown, Black, &
Uhde, 1994) can be easily distinguished from chronic insomnia on a number
of clinical and physiological criteria (for review, see Uhde, 2000), largely
based on the absence of either impressive clinical or EEG findings.
D. Course: Anxiety and insomnia are, in many cases, both disabling and
chronic conditions that put a significant number of individuals with these
disorders at risk for developing secondary complications. The high
co-morbidity between anxiety and insomnia suggests a strong relationship
between the two but information on the temporal relationship is lacking.
Some research on insomnia and anxiety has assessed the longitudinal course
of illness in relation to each other. The findings, however, are limited and
inconsistent, with important questions remaining as to whether insomnia
typically precedes anxiety or if insomnia more often develops subsequent to
1. Insomnia, with secondary anxiety.
Some findings suggest that anxiety symptoms/disorders can develop from
primary insomnia. For example, longitudinal data from the National
Institute of Mental Health Epidemiologic Catchment Area Study (ECA)
revealed that adults with uncomplicated insomnia (i.e., defined as insomnia
without the lifetime presence of a psychiatric disorder) at baseline were
5 times more likely to experience a panic attack in addition to being at a
significant increased risk for developing panic disorder by a follow-up inter-
view, one year later, compared to individuals without baseline insomnia or a
psychiatric disorder (Weissman, Greenwald, Niño-Murcia, & Dement,
Anxiety and Insomnia 113

1997). Breslau, Roth, Rosenthal, and Andreski (1996) also found that a
history of insomnia increased the risk for developing anxiety compared to
individuals with no history of insomnia (OR=1.97, 95% CI 1.08–3.60).
2. Anxiety, with secondary insomnia.
Other evidence supports the course of illness progression from primary
anxiety to secondary insomnia. Ohayon and Roth (2003) retrospectively
assessed the temporal relationship between insomnia and anxiety in a
large, multinational European, general population study. In this study,
current severe insomnia was the strongest predictor of a past psychiatric
history (OR=5.8, 95% CI 2.4–14.0). Further evidence demonstrating an
illness progression from anxiety to insomnia included the finding that the
onset of insomnia preceded the development of anxiety in only 18% of the
cases, while the onset of anxiety preceded the insomnia in more than 43%
of the cases. In another retrospective analysis of a community-based
sample of adolescents from the ages 13 to 15 years, Johnson et al. (2006)
sought to assess the directionality of association between insomnia
and psychiatric disorders including anxiety and depression. This study
reported a high prevalence rate of insomnia in individuals with a
history of anxiety that varied between 24% and 43% depending on the
specific anxiety disorder. Additionally in 73% of the individuals that
were co-morbid for anxiety and insomnia, the anxiety disorder preceded
the onset of the insomnia. Risk associated with either anxiety or insomnia
by the prior onset of the other disorder was also assessed to evaluate
directionality. This analysis revealed that a prior anxiety disorder increased
the risk of subsequent insomnia more than 3 fold, but that prior insomnia
was unrelated to the later development of an anxiety disorder.
E. Neuroanatomy & Pharmacology: Any attempt to examine the relationship
between anxiety and insomnia should consider the neuroanatomical sub-
strates and related neurotransmitter-neuromodulatory receptor systems that
mediate wake versus sleep states, keeping in mind that ‘‘alarm’’ mechanisms
are likely to influence both wakefulness and sleep. This is based on observa-
tions that states of fear (Uhde 2000) can take place during sleep and wakeful-
ness. Neither insomnia nor anxiety, therefore, necessarily reflect
disturbances in those neurobiological systems that mediate either wakeful-
ness or sleep per se. In fact, neuroreceptor-neurotransmitters that mediate
alarm functions might represent a third factor that contributes to the high
co-morbidity of chronic insomnia in many anxiety disorders (see Fig. 3).
Keeping these theoretical constructs in mind, we briefly review the neuro-
anatomy of wakefulness-promotion and sleep-promotion in relation to those
neurotransmitter and neuropharmacological systems most commonly impli-
cated in anxiety and insomnia.
1. Wakefulness-Promotion versus Sleep-Promotion
Wakefulness is mediated in part by midbrain and pontine-acetyl choli-
nergic, pontine locus ceruleus-noradrenergic, reticular formation system
114 T. W. Uhde, B. M. Cortese

(superior)-sertonergic, and posterior hypothalamus (tuberomammillary

nucleus)-histaminergic systems whereas sleep-promotion is linked to
anterior hypothalamus (ventrolateral preoptic nucleus) and reticular
formation (caudal)-GABAergic systems.
2. Anxiety and Insomnia
Many of the substrates implicated in sleep- or wakefulness-promotion
also have been implicated in either anxiety or insomnia or both. The
noradrenergic, GABAergic and serotonergic systems have been particu-
larly associated in the pathophysiology or treatment of both anxiety and
a) Noradrenergic:
Anatomy: The pontine nucleus locus ceruleus (LC) plays a key role in
arousal and vigilance in animals and disturbances in this nucleus or
its neuronal projections such as the amygdala and related limbic
substrates, and nucleus accumbens have been implicated in panic
disorder, GAD, PTSD and insomnia (Abelson, Khan, Liberzon, &
Young, 2007; Alttoa, Eller, Herm, Rinken, & Harro, 2007; Charney &
Redmond, 1983; Charney et al., 1990; DeViva, Zayfert, & Mellman,
2004; Sullivan, Coplan, Kent, & Gorman, 1999; Uhde & Singareddy,
2002). The locus ceruleus likely mediates its wakefulness-promoting
effects, and possibly insomnia, via a number of actions including direct
activation of the cortex and pedunculopontine tegmental nucleus and
inhibitory inputs to the ventral lateral preoptic nucleus (VLPO), a
galanin-related structure implicated in insomnia associated with
aging (Gaus, Strecker, Tate, Parker, & Saper, 2002).
Drug Action: Clonidine, an 2 adrenergic agonist, has short-term
but not long-term anti-anxiety actions (Uhde et al., 1989), which
theoretically parallels the inhibition and then tolerance to clonidine’s
direct application via iontophoresis onto the LC in animals. Likewise,
it is likely that the sedating and anti-anxiety effects of other 2 adre-
nergic agonists such as dexmedetomidine and lofexidine are mediated
via locus ceruleus inhibition. Similarly, the 1 adrenergic antagonist,
prasozin, has been recently reported to have beneficial effects in the
treatment of the hyper-arousal, anxiety, and sleep-related problems in
PTSD, including insomnia and nightmares (Raskind et al., 2003).
b) GABAergic:
Anatomy: The neurotransmitter pathway most widely implicated in
both anxiety and insomnia is the GABAergic-benzodiazepine-chloride
(GBC) system, with the ionotropic GABAA receptor that regulates
chloride channels being particularly relevant for the biological func-
tions of alarm, arousal and sedation.
The role of the GABAergic system in animal fear is well-established
(Zorumski & Isenberg, 1991). Alterations in GABAergic receptor
function also are linked to arousal states, involving the full-range of
Anxiety and Insomnia 115

sedation-drowsiness to alertness-hypervigilence and alarm states in

humans. However, there is no known specific or final common pathway
disturbance known to be associated with insomnia, anxiety or any
anxiety disorder.
Extrapolating from observations largely obtained in panic disorder
patients using flumazenil, a pure benzodiazepine-receptor antagonist,
there are few consistent data to support the idea that pathological
anxiety states are directly mediated by either an excess or deficiency,
respectively, of an endogenous benzodiazepine inverse agonist or pure
agonist (Coupland, Bell, Potokar, Dorkins, & Nutt, 2000; Nutt et al.,
1990). It remains possible, however, that disturbances in benzodiaze-
pine receptor distribution, location or sensitivity or secondary
GABAergic influences could play a role in the pathophysiology of
anxiety or insomnia (Kalueff & Nutt, 1997; Kaschka, Feistel, &
Ebert, 1995; Malizia et al., 1998; Mohler, Fritschy, & Rudolph, 2002;
Mohler, Fritschy, Vogt, Crestani, & Rudolph, 2005). Consistent with
this notion are prevalent decreases in flumazenil and iomazenil binding
throughout the brains of panic and generalized anxiety disorder
patients, respectively, as well as decreases in the fear-associated and
anxiety-linked orbitofrontal cortex, amygdala and hippocampus
regions (Malizia et al., 1998, Tiihonen et al., 1997).
Of interest, Buhr and coworkers (2002) recently reported a mutation
in the b3 subunit of the GABAA receptor in a person with chronic
insomnia from a family with sleep problems. Unfortunately, there was
no comprehensive information provided, or available, on the specific
qualities of insomnia or anxiety within the individual or family. None-
theless, these observations, combined with separate lines of evidence in
rodents and humans suggest that the GABAA receptors, perhaps
particularly 1, 2, 3, and b3 subunits, may play a partial role in
the pathophysiology and treatment of insomnia and co-morbid anxi-
ety symptoms (Laposky, Homanics, Baile, & Mendelson, 2001; Moh-
ler et al., 2002, 2005; Rowlett, Cook, Duke, & Platt, 2005; Rush, 1998).
It is theoretically possible that some co-morbid anxiety-insomnia syn-
dromes have particular relevance to a specific GABAA subunit(s),
although developing such a specific animal co-morbidity model will
be a major challenge.
Drug Action: Many drugs with direct or indirect actions at the
GABAergic-benzodiazepine-chloride (GBC) receptor complex
have short-term sedating, sleep-promoting and anti-anxiety profiles.
Barbiturates, alcohol, benzodiazepines, and anesthetic agents, which
act at different sites of the GBC receptor complex, have both sedating
and subjective anxiety-reducing effects in humans, although the beha-
vioral effects of barbiturates, alcohol, and anesthetic agents in patients
who meet contemporary DSM-IV diagnostic criteria for anxiety or
sleep disorders have not been thoroughly examined. Propofol, an
116 T. W. Uhde, B. M. Cortese

anesthetic agent with GABAA receptor actions, is often reported by

subjects undergoing colonoscopy to produce the equivalent of a ‘‘full-
night’s restful’’ and restorative sleep (author observations).
Traditional hypnotic benzodiazepine drugs such as flurazepam,
estazolam, temazepam, triazolam and quazepam, which have FDA
indications for the short-term treatment of insomnia, produce anxio-
lysis but are generally considered impractical for such off-label use due
to short half-life pharmacokinetics. Benzodiazepine hypnotics admi-
nistered to patients with insomnia as target symptoms have been
shown to improve insomnia as well as reduce co-morbid daytime
anxiety (Fontaine, Beaudry, Le Morvan, Beauclair, & Chouinard,
1990) and improve a sense of physical well-being (Roth, Walsh, Krys-
tal, Wessel, & Roehrs, 2005). It is widely appreciated among clinicians
that benzodiazepine compounds and so-called Z drugs reduce anxiety
and subjective insomnia. These observations are consistent with the
notion that anxiety and insomnia share overlapping neuroanatomical
substrates, at least in terms therapeutic response patterns to drugs that
act at the GBC receptor complex. These observations, however, con-
tribute to the possible misperception that a critical, or even requisite,
ingredient of drug-mediated anxiolysis and especially sleep-promotion
is the induction of drowsiness or sedation. Drug properties of drowsi-
ness and sedation may not only be unnecessary for the effective treat-
ment of anxiety or insomnia or co-morbid anxiety-insomnia
conditions, but, theoretically may actually reduce the overall effective-
ness of sedating drugs in the treatment of selective anxiety disorders,
which require vigilance to monitor threat during sleep (Singareddy,
Uhde, & Commissaris, 2006).
c) Serotonergic:
Anatomy: The reticular formation has long been associated with func-
tions of wakefulness. Moruzzi and Magoun (1949) conducted studies
demonstrating that the transection of the reticular formation above the
pons in the face of intact sensory inputs to higher brain regions are
associated with behavior and EEG patterns consistent with sleep,
whereas lesions of the reticular formation below the pons are asso-
ciated with a reduction in sleep. These classic studies and subsequent
research indicate that different neural networks within and impinging
onto the reticular formation mediate wakefulness and arousal func-
tions, including serotonergic projections and activation of 5HT2A
receptors at the level of the cerebral cortex and possibly hypothalamus.
The serotonergic receptor system is strongly implicated in fearful
animal behaviors and human anxiety disorders, particularly panic and
obsessive-compulsive disorders (for review, see Uhde & Singareddy,
2002). Knock-outs of 5-HT1A receptors are associated with increased
fear behaviors in animal models including, but not limited to, the elevated
plus maze and foot shock (Heisler et al., 1998), forced swim (Ramboz et
Anxiety and Insomnia 117

al., 1998), and open field (Parks, Robinson, Sibille, Shenk, & Toth, 1998)
tests. In humans, fenfluramine, a 5-HT releasing agent, induces both
anxiety and increases in blood cortisol. Of interest, over-activity of the
hypothalamic-pituitary-adrenal (HPA) axis is associated with some types
of severe insomnia (for review, see Roth, Roehrs, & Pies, 2007).
Drug Action: Similar to benzodiazepines, selective serotonin reup-
take inhibitors (SSRI) are widely used by clinicians to treat most of the
anxiety disorders. And, trazodone, a tetracylic SSRI with 5-HT2
antagonist effects, may be the most frequently prescribed medication
for the treatment of insomnia by psychiatrists and primary care phy-
sicians. There are limited data regarding the secondary improvement
in insomnia following the targeted treatment of anxiety (or vice versa).
It is known that older patients with anxiety disorders (60 years of age
with mainly GAD) show decreased change scores on the Pittsburgh
Sleep Quality Index (PSQI), suggesting that quality of sleep improves
after the targeted treatment of anxiety in elderly patients (Blank et al.,
2006). Not all anxiety disorders (e.g. PTSD), however, show conver-
gent and parallel improvements in anxiety symptoms and insomnia (or
sleep efficiency on polysomnography) when treated with SSRI’s;
moreover, responses to some SSRI’s may be either ineffective or
actually induce insomnia in PTSD, people with primary insomnia or
healthy subjects (Davis, Frazier, Williford, & Newell, 2006; Winokur
et al., 2001). Thus, while the SSRI’s play a crucial role in the treatment
of many anxiety disorders, including those wherein sleep problems are
a core feature, the effects of these same agents on the treatment of
insomnia appear to be less predictable.
F. Other ‘‘Third’’ Factors: An important question that remains to be answered is
whether sleep deprivation is a contributing factor to insomnia and/or anxiety.
Contrary to popular belief, there is a lack of good evidence for significant sleep
loss in insomnia. Despite the fact that insufficient sleep is a hallmark feature of
insomnia, there is little data demonstrating actual sleep loss or sleep restriction
in individuals reporting insomnia. Sleep polysomnography (PSG) documents
this lack of relationship between subjective reports of insomnia and objective
measures of poor sleep including reduced sleep time. In general, individuals
with insomnia underestimate sleep time compared to actual sleep time recorded
by PSG. For example, Rosa and Bonnet (2000) reported no relationship
between the subjective experience of insomnia and poor EEG sleep, defined
by increased sleep latency or decreased efficiency. Specifically, chronic insom-
niacs reported significantly worse laboratory sleep compared to controls, while
objective EEG-assessed sleep revealed little difference between the groups.
Means, Edinger, Glenn, and Fins (2003) who also assessed individuals with
insomnia and compared them to normal sleepers corroborate the findings of
Rosa and Bonnet (2000) and go further to describe that insomniacs show a
wide range of sleep misperception, from considerable underestimation to both
118 T. W. Uhde, B. M. Cortese

accurate and overestimation of sleep time. In all, these studies suggest that
insomnia is a complex condition associated with factors that extend beyond
reduced sleep time.
Recent findings from our anxiety, stress and trauma laboratory also
suggest a difference between subjective reports of sleep duration (i.e., sleep
quantity) and subjective experience of sleep quality. We administered the
Pittsburgh Sleep Quality Index (PSQI) to a small group of individuals with
motor vehicle accident-related PTSD. The PSQI is a widely used, valid and
reliable instrument designed to measure sleep disturbance through seven
components including sleep quality, sleep latency, sleep duration, habitual
sleep efficiency, sleep disturbances, use of sleeping medications, and daytime
dysfunction (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). To
achieve a pure score for sleep quality (i.e., score that excludes sleep quantity)
we eliminated the two components of the sleep efficiency domain (i.e., sleep
duration and habitual sleep efficiency) from the overall PSQI score (Cole et
al., 2006). We then correlated this modified score (i.e., sleep quality) with
subjective reports of average sleep duration (i.e., amount of sleep in minutes)
and found no relationship between these two variables (r = 0.20;
F1,26=1.04, p=0.32). Our findings in this small group of anxious individuals
demonstrate an apparent distinction between sleep quantity and quality and
suggest that individuals suffering from anxiety who report poor sleep may
not have a reduction in sleep duration (i.e., restriction/deprivation).
Although limited data exists concerning the effects of sleep deprivation on
insomnia and anxiety, the available evidence suggests that sleep deprivation or
restriction could have opposing effects on insomnia or anxiety, findings that
support a clear distinction between these 2 disorders. On one hand, sleep
deprivation/restriction has shown positive results in people with insomnia,
while no consistent benefits have been shown for sleep deprivation in indivi-
duals diagnosed with anxiety. Specifically, sleep deprivation is associated with
an increase in anxiety symptoms in healthy individuals and has been shown to
worsen anxiety in patients with some but not all anxiety disorders. Roy-Byrne,
Uhde, and Post (1986) assessed depression and anxiety levels after one night of
total sleep deprivation in patients with panic disorder who were not currently
depressed. Individual responses of panic patients to sleep deprivation varied,
with a subset of panic patients (7/12 [58%]) demonstrating a worsening of
anxiety and 4 of the 12 panic patients (33%) experiencing a spontaneous panic
attack the day following the sleep deprivation procedure. Labbate et al. (1997)
also assessed the effects of sleep deprivation on anxiety and reported a
worsening of anxiety symptoms after sleep deprivation in a small sample of
panic (n=5) patients. In this study, 3 of the 5 panic patients experienced at
least one panic attack the morning after sleep deprivation, while none of the
control subjects experienced sleep deprivation-induced panic. With respect to
other anxiety subtypes, one night of sleep deprivation revealed inconsistent,
and for some negative, effects in patients with primary OCD (Joffe & Swinson,
Anxiety and Insomnia 119

1988; Labbate et al., 1997), social phobia (SP) and generalized anxiety dis-
order (GAD; Labbate et al., 1998).
Several studies describe the positive effects of sleep deprivation/
restriction on insomnia and suggest its potential utility in treating this dis-
order. In one study (Stepanski, Zorick, Roehrs, & Roth, 2000), sleepiness
and total sleep time for primary insomniacs was significantly increased,
compared to baseline, during the recovery night following one night of
total sleep deprivation. In addition, post-deprivation sleep measures in the
insomniacs were comparable to the age- and sex-matched normal sleeper


Certainly, the collective data reviewed in the aforementioned sections argue

against the idea that anxiety and insomnia are entirely separate disorders with
distinctly different neurochemical disturbances. However, it is impossible with-
out prospective studies, conducted in at-risk populations, to ascertain whether
these neurobiological findings represent different phases of a common diathesis
(Fig. 1), different disorders with coupled downstream abnormalities (Fig. 2), or
the consequence(s) of unknown ‘‘other’’ endogenous and/or external/exogen-
ous factors (Fig. 3).
Given the prevalence of insomnia reported by patients with anxiety disorders
(Brown & Uhde, 2003; Craske & Tsao, 2005; Hoehn-Saric, 1981; Mellman &
Uhde, 1989, 1990; Saletu et al., 1997; Uhde, 2000), the portfolio of polysomno-
graphy and related sleep research studies in patients with anxiety disorders is
fairly modest in number. The greatest amount of polysomnography and related
sleep research has been conducted in panic disorder (PD), post-traumatic stress
disorder (PTSD), generalized anxiety disorder (GAD) and, to a lesser extent,
obsessive-compulsive disorder (OCD) (Dow, Kelsoe, & Gillin, 1996; Engdahl,
Eberly, Hurwitz, Mahowald, & Blake, 2000; Hurwitz, Mahowald, Kuskowski,
& Engdahl, 1988; Papadimitriou & Linkowski, 2005; Mellman, 1997; Ross
et al., 1994; Sheikh, Woodward, & Leskin, 2003; Uhde et al., 1984; Uhde,
2000). While sleep disturbances are a ‘‘core’’ and distinguishing feature of
PTSD, the anxiety disorders of GAD and PD, especially those patients with
sleep-nocturnal panic attacks (Craske & Tsao, 2005; Cortese & Uhde, 2006;
Mellman & Uhde, 1989), appear to share particularly poignant characteristics
with primary insomnia patients in terms of symptomatology, co-morbidity,
pharmacologic treatment response, and, sleep EEG measures. Because symp-
toms of ‘‘psychic distress’’, worry, difficulty concentrating, feelings of jitteri-
ness, agitation, and muscle tension are almost universally reported in part or
whole by patients with generalized anxiety disorder and primary insomnia, it is
not surprising that GAD was found to be the most prevalent co-morbid anxiety
120 T. W. Uhde, B. M. Cortese

disorder assessed in a large general population study of individuals with insom-

nia complaints (Ohayon, Caulet, & Lemoine, 1998).
Taken together, these observations suggest that any attempt to further
dissect the theoretical models in a prospective manner should give priority to
the systematic investigation of patients with primary insomnia versus general-
ized anxiety disorder. The convergence of symptoms and their robust treatment
response to benzodiazepines are particularly compelling from a clinical per-
spective. If GAD and primary insomnia represent different phases of the same
disorder (Fig. 1) or different conditions whose neurobiology significantly con-
verge over time (Fig. 2), one would predict that at later stages of illness both
GAD and primary insomnia would respond favorably to not only the same
pharmacologic treatments but also the same non-pharmacological interven-
tions. Thus, one might predict that cognitive behavioral treatments specifically
designed to target the cognitive distortions of worry would be useful in both
syndromes. In an important preliminary investigation, Bélanger, Morin,
Langlois and Ladouceur (2004) administered group CBT to GAD patients
based on targeting GAD-related worries (Dugas & Ladouceur, 2000), wherein
sleep complaints were not specifically addressed as part of the treatment. In this
study, 86.5% of the GAD patients reported that they had never experienced
insomnia without worry and the majority reported difficulties maintaining
sleep; indeed, 25% reported suffering from all phases of insomnia (i.e. early,
middle, and late). Additionally, group CBT was associated with a significant
improvement on the Insomnia Severity Index (Morin, 1993).
One cannot help but speculate that cultural or socioeconomic factors might
largely influence whether one self-identifies anxiety-insomnia as a medical
(i.e. sleep) versus mental health problem (i.e. anxiety), which in the case of
GAD and primary insomnia might lead to the conclusion that major difference
in primary insomnia versus GAD patients are mainly related to help-seeking

Future Research

To validate any of the theoretical models, future research must investigate

patients with primary insomnia and anxiety disorders using both cross-sectional
and retrospective (Uhde, Boulenger, Roy-Byrne, Vittone, & Post, 1985) life-
charting as well as prospective methods. Prospective studies in patients whose
initial presentation is anxiety versus insomnia or, in well-defined at-risk popula-
tions, would be particularly desirable. Assessment tools must be developed,
validated and employed in a systematic fashion. The conduct of such studies
will be particularly challenging due to pragmatic considerations, which will
necessitate that such investigations be conducted across different institutional
sites and specialty clinics (i.e. sleep medicine clinics versus anxiety disorder
clinics). Ideally, it is desirable to conduct such studies within department(s) or
Anxiety and Insomnia 121

institution(s), which have research expertise in both sleep and anxiety disorder
research in order to minimize internal and external sources of variation. To
investigate the convergent, co-morbid relationship between anxiety and insom-
nia, priority might be given to the study of patients who a) meet DSM-IV criteria
for GAD plus seek treatment at anxiety disorder clinics versus b) patients meet-
ing ICD-10 criteria for insomnia plus seek treatment from sleep clinics. Other
than these entry criteria, we recommend that there be few, if any, exclusion
criteria in designing such comparator studies. Specifically, we would not include
or exclude on the basis of sleep misperception or, perhaps, even the degree or
duration of subjective insomnia. Such differences or similarities across groups
seeking treatment for anxiety versus insomnia might themselves be markers that
distinguish group differences.
A particular problem for clinicians is the lack of information on the impact
of long-term pharmacological treatments or the comparative efficacy of drug
versus cognitive behavioral interventions. Such studies are time-intensive but
necessary to advance knowledge about the validity of any of the proposed three
theoretical models but also to develop evidence-based treatment packages for
the long-term treatment of anxiety-insomnia syndromes. Even more proble-
matic is the lack of information on the course of anxiety and insomnia after
treatment discontinuation.
It is beyond the scope of this chapter to review and recommend the beha-
vioral, physiological, and neuroimaging studies, which might be conducted to
best characterize the relationship between insomnia and anxiety. Clearly, neu-
roimaging strategies would be useful, although current MRI, MRS and fMRI
imaging strategies have resolution limitations that make it difficult to define with
precision the neuroanatomical substrates and functional circuits underlying
anxiety, insomnia, and mixed anxiety-insomnia syndromes. Nonetheless, the
emerging field of sleep neuroimaging (Nofzinger, 2004) may ultimately provide
tools for understanding fundamental constructs such as ‘‘time perception’’ or
‘‘sleep consciousness’’ (Uhde et al., 2006). Such hypothesized third factors
(Fig. 3) might be amenable to investigation with imaging strategies, which are
not conducive to examination with traditional polysomnography or even
spectral analysis methods (Nofzinger et al., 1999, 2002, 2004). An examination
of the comparator effects of sleep deprivation under laboratory-controlled
conditions would be of interest to better differentiate the behavioral, cognitive,
MSLT and neuroendocrine effects of sleep restriction. Likewise, caffeine is an
ideal chemical model of anxiety (Lin, Uhde, Slate, & McCann, 1997; Uhde,
1995) and has been recently proposed as a tool to study primary insomnia
(Drake, Jefferson, Roehrs, & Roth, 2006). Studying the behavioral (including
sleep perception and polysomnographic measures), neuroendocrine, and phy-
siological effects of caffeine in primary insomnia and GAD would provide useful
information. Disturbances in hypothalamic-pituitary-adrenal axis function
remain a focus of much anxiety (Uhde & Singareddy, 2002) and insomnia
research (Drake, Roehrs, & Roth, 2003; Vgontzas & Chrousos, 2002); yet, the
122 T. W. Uhde, B. M. Cortese

specific origin, maintenance, and neuroanatomical impact of putative glucocor-

ticoid disturbances in the anxiety disorders and insomnia remain unknown.
The most exciting area of future research is the theoretical possibility that
designer drugs might be developed with non-sedating anxiolysis or, even,
cognitive enhancing, pro-vigilance, anxiolytic properties. As reviewed else-
where (Dawson, Collinson, & Atack, 2005; Mohler et al., 2002, 2005), the
development of a compound that binds to 2 GABAA and/or 3 GABAA
sites, which are located primarily in the fear-related substrates (i.e. amygdala)
and cortex, but without binding to the sedation-associated 1 GABAA binding
site, would theoretically have non-sedating anxiolytic properties. Such a drug
would be particularly useful for patients who wish to maintain alertness or who
experience increased anxiety as a result of increasing relaxation or sedation
(e.g. patients with nocturnal panic attacks). Even more speculative, but equally
provocative, is the idea that drugs with hypocretin agonist actions might also be
useful in the treatment of nocturnal panic attacks or patients whose insomnia is
related to a fear of sleeping (Uhde, 2000; Singareddy et al., 2006).


Abelson, J. L., Khan, S., Liberzon, I., & Young, E. A. (2007). HPA axis activity in patients
with panic disorder: review and synthesis of four studies. Depression and Anxiety, 24(1),
Alttoa, A., Eller, M., Herm, L., Rinken, A., & Harro, J. (2007). Amphetamine-induced
locomotion, behavioral sensitization to amphetamine, and striatal D2 receptor function
in rats with high or low spontaneous exploratory activity: differences in the role of locus
coeruleus. Brain Research, 1131(1), 138–148.
Bélanger, L., Morin, C. M., Langlois, F., & Ladouceur, R. (2004). Insomnia and generalized
anxiety disorder: effects of cognitive behavior therapy for gad on insomnia symptoms.
Journal of Anxiety Disorders, 18(4), 561–571.
Blank, S., Lenze, E., Mulsant, B. H., Dew, M. A., Karp, J. F., Shear, M. K., et al. (2006).
Outcomes of Late-Life Anxiety Disorders During 32 Weeks of Citalopram Treatment. The
Journal of Clinical Psychiatry, 67, 468–472.
Bixler, E. O., Vgontzas, A. N., Lin, H. M., Vela-Bueno, A., & Kales, A. (2002). Insomnia in
central Pennsylvania. Journal of Psychosomatic Research, 53(1), 589–592.
Breslau, N., Roth, T., Rosenthal, L., & Andreski, P. (1996). Sleep disturbance and psychiatric
disorders: a longitudinal epidemiological study of young adults. Biological Psychiatry,
39(6), 411–418.
Brown, T., & Uhde, T. W. (2003). Sleep Panic Attacks: A Micro-movement Analysis.
Depression and Anxiety, 18, 214–220.
Brown, T. M., Black, B., & Uhde, T. W. (1994). Sleep architecture in social phobia. Biological
Psychiatry, 35, 420–421.
Buhr, A., Bianchi, M. T., Baur, R., Courtet, P., Pignay, V., Boulenger, J.-P., et al. (2002).
Functional characterization of the new human GABAA receptor mutation b3(R192H).
Human Genetics, 111, 154–160.
Buysse, D. J., Reynolds, C. F., Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The
Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research.
Psychiatry Research, 28(2), 193–213.
Anxiety and Insomnia 123

Charney, D. S., & Redmond, D. E. (1983). Neurobiological mechanisms in human anxiety.

Evidence supporting central noradrenergic hyperactivity. Neuropharmacology, 22(12B),
Charney, D. S., Woods, S. W., Nagy, L. M., Southwick, S. M., Krystal, J. H., & Heninger, G. R.
(1990). Noradrenergic function in panic disorder. The Journal of Clinical Psychiatry,
51(Suppl A), 5–11.
Cole, J. C., Motivala, S. J., Buysse, D. J., Oxman, M. N., Levin, M. J., & Irwin, M. R. (2006).
Validation of a 3-factor scoring model for the Pittsburgh sleep quality index in older
adults. Sleep, 29(1), 112–116,.
Cortese, B. M., & Uhde, T. W. (2006). Immobilization Panic. The American Journal of
Psychiatry, 163, 1453–1454.
Coupland, N. J., Bell, C., Potokar, J. P., Dorkins, E., & Nutt, D. J. (2000). Flumazenil
challenge in social phobia. Depression and Anxiety, 11(1), 27–30.
Craske, M. G., & Tsao, J. C. (2005). Assessment and treatment of nocturnal panic attacks.
Sleep Medicine Reviews, 9(3), 173–184.
Davis, L. L., Frazier, E. C., Williford, R. B., & Newell, J. M. (2006). Long-term
pharmacotherapy for post-traumatic stress disorder. CNS Drugs, 20(6), 465–476.
Dawson, G. R., Collinson, N., & Atack, J. R. (2005). Development of subtype selective
GABAA modulators. CNS Spectr, 10(1), 21–27.
DeViva, J. C., Zayfert, C., & Mellman, T. A. (2004). Factors associated with insomnia among
civilians seeking treatment for PTSD: an exploratory study. Behavioral Sleep Medicine,
2(3), 162–176.
Diagnostic and Statistical Manual of Mental Disorders DSM-IVTM American Psychiatric
Association 2000. (DSM-IV-TR) Diagnostic and statistical manual of mental disorders,
4th edition, text revision. Washington, DC: American Psychiatric Press, Inc.
Dow, B. M., Kelsoe, J. R., Jr., & Gillin, J. C. (1996). Sleep and dreams in Vietnam PTSD and
depression. Biological Psychiatry, 39, 42–50.
Drake, C. L., Roehrs, T., & Roth, T. (2003). Insomnia causes, consequences, and therapeu-
tics: an overview. Depress Anxiety, 18(4), 163–176.
Drake, C. L., Jefferson, C., Roehrs, T., & Roth, T. (2006). Stress-related sleep disturbance and
polysomnographic response to caffeine. Sleep Medicine, 7(7), 567–572.
Dugas, M. J., & Ladouceur, R. (2000). Treatment of GAD: targeting intolerance of uncer-
tainty in two types of worry. Behavior Modification, 24, 635–657.
Engdahl, B. E., Eberly, R. E., Hurwitz, T. D., Mahowald, M. W., & Blake, J. (2000). Sleep in a
community sample of elderly war veterans with and without posttraumatic stress disorder.
Biological Psychiatry, 47, 520–525.
Foley, D. J., Monjan, A. A., Brown, S. L., Simonsick, E. M., Wallace, R. B., & Blazer, D. G.
(1995). Sleep complaints among elderly persons: an epidemiologic study of three commu-
nities. Sleep, 18(6), 425–432.
Fontaine, R., Beaudry, P., Le Morvan, P., Beauclair, L., & Chouinard, G. (1990). Zopiclone
and triazolam in insomnia associated with generalized anxiety disorder: a placebo-
controlled evaluation of efficacy and daytime anxiety. International Clinical Psychophar-
macology, 5(3), 173–83.
Ford, D. E., & Kamerow, D. B. (1989). Epidemiologic study of sleep disturbances and
psychiatric disorders. An opportunity for prevention? JAMA, 262(11), 1479–1484.
Ganguli, M., Reynolds, C. F., & Gilby, J. E. (1996). Prevalence and persistence of sleep
complaints in a rural older community sample: the MoVIES project. Journal of the
American Geriatrics Society, 44(7), 778–784.
Gaus, S. E., Strecker, R. E., Tate, B. A., Parker, R. A., & Saper, C. B. (2002). Ventrolateral
preoptic nucleus contains sleep-active, galaninergic neurons in multiple mammalian spe-
cies. Neuroscience, 115(1), 285–94.
Gregory, A. M., & O’Connor, T. G. (2002). Sleep problems in childhood: a longitudinal study
of developmental change and association with behavioral problems. Journal of the
American Academy of Child and Adolescent Psychiatry, 41(8), 964–971.
124 T. W. Uhde, B. M. Cortese

Heisler, L. K., Chu, H. M., Brennan, T. J., Danao, J. A., Bajwa, P., Parsons, L. H., et al.
(1998). Elevated anxiety and antidepressant-like responses in serotonin 5-HT1A receptor
mutant mice. Proceedings of the National Academy of Sciences of the United States of
America, 95(25), 15049–15054.
Hoehn-Saric, R. (1981). Characteristics of chronic anxiety patients. In: D. F. Klein & J. G. Rabkin
(Eds.), Anxiety: New research and changing concepts. New York: Raven Press.
Hurwitz, T. D., Mahowald, M. S., Kuskowski, M., & Engdahl, B. E. (1988). Polysomno-
graphic sleep is not clinically impaired in Vietnam combat veterans with chronic posttrau-
matic stress disorder. Biol Psychiatry, 44(10), 1066–73.
International Statistical Classification of Diseases and Related Health Problems (ICD-10)
The ICD-10 Classification of Mental and Behavioural Disorders: Clinical descriptions and
diagnostic guidelines. Geneva. World Health Organization, 1992.
Joffe, R. T., & Swinson, R. P. (1988). Total sleep deprivation in patients with obsessive-
compulsive disorder. Acta psychiatrica Scandinavica, 77(4), 483–487.
Johnson, E. O., Chilcoat, H. D., & Breslau, N. (2000). Trouble sleeping and anxiety/depres-
sion in childhood. Psychiatry Research,. 94(2), 93–102.
Johnson, E. O., Roth, T., & Breslau, N. (2006). The association of insomnia with anxiety
disorders and depression: Exploration of the direction of risk. Journal of psychiatric
Research, 40(8), 700–708.
Kalueff, A., & Nutt, D. J. (1997). Role of GABA in memory and anxiety. Depression and
Anxiety, 4, 100–110.
Kaschka, W., Feistel, H., & Ebert, D. (1995). Reduced benzodiazepine receptor binding in
panic disorders measured by iomazenil SPECT. Journal of Psychiatric Research, 29(5),
Kessler, R. C., Amminger, G. P., Aguilar-Gaxiola, S., Alonso, J., Lee, S., & Ustün, T. B.
(2007). Age of onset of mental disorders: a review of recent literature. Current Opinion
Psychiatry, 20(4), 359–364.
Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005).
Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National
Comorbidity Survey Replication. Archives of General Psychiatry, 62(6), 593–602.
Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., & Eshleman, S.
(1994). Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the
United States. Results from the National Comorbidity Survey. Archives of General Psy-
chiatry, 51(1), 8–19.
Labbate, L. A., Johnson, M. R., Lydiard, R. B., Brawman-Mintzer, O., Emmanuel,
N., Crawford M., et al. (1997). Sleep deprivation in panic disorder and obsessive-compul-
sive disorder. Canadian Journal of Psychiatry. Revue Canadienne de Psychiatrie, 42(9),
Labbate, L. A., Johnson, M. R., Lydiard, R. B., Brawman-Mintzer, O., Emmanuel,
N., Crawford, M., et al. (1998). Sleep Deprivation in social phobia and generalized anxiety
disorder. Biological Psychiatry, 43(11), 840–842.
Laposky, A. D., Homanics, G. E., Baile, A., & Mendelson, W. B. (2001). Deletion of the
GABA(A) receptor beta 3 subunit eliminates the hypnotic actions of oleamide in mice.
Neurorepot, 12, 4143–4147.
Lin, A. S-K., Uhde, T. W., Slate, S. O., & McCann, U. D. (1997). Effects of intravenous
caffeine administered to healthy males during sleep. Depression and Anxiety, 5(1),
Malizia, A. L., Cunningham, V. J., Bell, C. J., Liddle, P. F., Jones, T., & Nutt, D. J. (1998).
Decreased brain GABA(A)-benzodiazepine receptor binding in panic disorder: preli-
minary results from a quantitative PET study. Archives of General Psychiatry, 55(8),
Means, M. K., Edinger, J. D., Glenn, D. M., & Fins, A. I. (2003). Accuracy of sleep perceptions
among insomnia sufferers and normal sleepers. Sleep Medicinal, 4(4), 285–296.
Anxiety and Insomnia 125

Mellinger, G. D., Balter, M. B., & Uhlenhuth, E. H. (1985). Insomnia and its treatment:
Prevalence and correlates. Archives of General Psychiatry, 42(3), 225–232.
Mellman, T. A. (1997). Psychobiology of sleep disturbance in posttraumatic stress disorder.
Ann N.Y. Academy of Sciences, 821, 142–149.
Mellman, T. A., & Uhde, T. W. (1990). Patients with frequent sleep panic: Clinical
findings and response to medication treatment. The Journal of clinical psychiatry, 51(12),
Mellman, T. A., & Uhde, T. W. (1989). Sleep panic attacks: New clinical findings and
theoretical implications. The American Journal of Psychiatry, 146, 1204–1207.
Millard, T. P., Lewis, C. M., Khamashta, M. A., Hughes, G. R., Hawk, J. L., & McGregor, J. M.
(2001). Familial clustering of polymorphic light eruption in relatives of patients with lupus
erythematosus: evidence of a shared pathogenesis. The British Journal of Dermatology, 144,
Millard, T. P., Kondeatis, E., Cox, A., Wilson, A. G., Grabczynska, S. A., Carey, B. S., et al.
(2001). A candidate gene analysis of three related photosensitivity disorders: cutaneous
lupus erythematosus, polymorphic light eruption and actinic prurigo. The British Journal
of Dermatology, 145, 229–236.
Mohler, H., Fritschy, J. M., & Rudolph, U. (2002). A new benzodiazepine pharmacology. The
Journal of Pharmacology and Experimental Therapeutics, 300, 2–8.
Mohler, H., Fritschy, J. M., Vogt, K., Crestani, F., & Rudolph, U. (2005). Pathophysiology
and pharmacology of GABA(A) receptors. Handbook of Experimental Pharmacology,
169, 225–247.
Monane, M. (1992). Insomnia in the elderly. The Journal of Clinical Psychiatry, 53 (Suppl),
Morin, C. M. (1993). Insomnia: Psychological assessment and management. New York:
Guilford Press.
Morphy, H., Dunn, K. M., Lewis, M., Boardman, H. F., & Croft, P. R. (2007). Epidemiology
of insomnia: A longitudinal study in a UK population. Sleep, 30(3), 274–280.
Moruzzi, G., & Magoun. H. W. (1949). Brain stem reticular formation and activation of the
EEG. Electroencephalography and Clinical Neurophysiology, 1, 455–473.
Nofzinger, E. A., Buysse, D. J., Miewald, J. M., Meltzer, C. C., Price, J. C., Sembrat, R. C.,
et al. (2002). Human regional cerebral glucose metabolism during non-rapid eye move-
ment sleep in relation to waking. Brain, 125(Pt 5), 1105–15.
Nofzinger, E. A., Nichols, T. E., Meltzer, C. C., Price, J., Steppe, D. A., Miewald, J. M., et al.
(1999). Changes in forebrain function from waking to REM sleep in depression: prelimin-
ary analyses of [18F]FDG PET studies. Psychiatry Research, 91(2), 59–78.
Nofzinger, E. A. (2004). What can neuroimaging findings tell us about sleep disorders? Sleep
Medicinal, 5 (Suppl 1), S16–22.
Nutt, D. J., Glue, P., Lawson, C. (1990). The Neurochemistry of anxiety: an update. Prog
Neuropsychopharmacol Biol Psychiatry 14, 737–752.
Ohayon, M. M., Caulet, M., & Lemoine, P. (1998). Comorbidity of mental and insomnia
disorders in the general population. Comprehensive Psychiatry, 39(4), 185–97.
Ohayon, M. M., & Roth, T. (2003). Place of chronic insomnia in the course of depressive and
anxiety disorders. Journal of Psychiatric Research, 37(1), 9–15.
Paavonen, E. J., Solantaus, T., Almqvist, F., & Aronen, E. T. (2003). Four-year follow-up
study of sleep and psychiatric symptoms in preadolescents: relationship of persistent and
temporary sleep problems to psychiatric symptoms. Journal of Developmental and Beha-
vioral Pediatrics, 24(5), 307–314.
Papadimitriou, G. N., & Linikowski, P. (2005). Sleep disturbance in anxiety disorders. Int
Review of Psychiatry, 17(4), 229–236.
Parks, C. L., Robinson, P. S., Sibille, E., Shenk, T., & Toth, M. (1998). Increased anxiety of
mice lacking the serotonin1A receptor. Proceedings of the National Academy of Sciences of
the United States of America, 95(18), 10734–10739.
126 T. W. Uhde, B. M. Cortese

Ramboz, S., Oosting, R., Amara, D. A., Kung, H. F., Blier, P., Mendelsohn, M., et al. (1998).
Serotonin receptor 1A knockout: an animal model of anxiety-related disorder. Proceed-
ings of the National Academy of Sciences of the United States of America, 95(24),
Raskind, M., Peskind, E., Kanter, E., Petrie, E. C., Radant, A., Thompson, C. E., et al.
(2003). Reduction of nightmares and other PTSD symptoms in combat veterans
by prazosin: a placebo-controlled study. The American Journal of Psychiatry, 160,
Rosa, R. R., & Bonnet, M. H. (2000). Reported chronic insomnia is independent of poor sleep
as measured by electroencephalography. Psychosomatic Medicine, 62(4), 474–482.
Ross, R. J., Ball, W. A., Dinges, D. F., Kribbs, N. B., Morrison, A. R., Silver, S. M., et al.
(1994). Rapid eye movement sleep disturbance in posttraumatic stress disorder. Biological
Psychiatry, 35, 195–202.
Roth, T., Roehrs, T., & Pies, R. (2007). Insomnia: pathophysiology and implications for
treatment. Sleep Medicine Reviews, 11(1), 71–79.
Roth, T., Walsh, J. K., Krystal, A., Wessel, T., & Roehrs, T.A. (2005). An evaluation of the
efficacy and safety of eszopiclone over 12 months in patients with chronic primary
insomnia. Sleep Medicine, 6(6), 487–495.
Roy-Byrne, P. P., Uhde, T. W., & Post, R. M. (1986). Effects of one night’s sleep depriva-
tion on mood and behavior in panic disorder. Patients with panic disorder compared
with depressed patients and normal controls. Archives of General Psychiatry, 43(9),
Rowlett, J. K., Cook, J. M., Duke, A. N., & Platt, D. M. (2005). Selective antagonism of
GABAA receptor subtypes: an in vivo approach to exploring the therapeutic and side
effects of benzodiazepine-type drugs. CNS Spectr, 10, 40–48.
Rush, C. R. (1998). Behavioral pharmacology of zolpidem relative to benzodiazepines: a
review. Pharmacology, Biochemistry, and Behavior, 61, 253–269.
Saletu, B., Saletu-Zyhlarz, G., Anderer, P., Brandstater, N., Frey, R., Gruber, G., et al.
(1997). Nonorganic insomnia in generalized anxiety disorder: controlled studies on
sleep. Awakening and daytime vigilance utilizing polysomnography and EEG mapping.
Neuropsychobiology, 36, 117–129.
Sheikh, J. I., Woodward, S. H., & Leskin, G. A. (2003). Sleep in post-traumatic stress
disorder: convergence and divergence. Depression and Anxiety, 18(4), 187–197.
Singareddy, R., Uhde, T. W., & Commissaris, R. (2006). Differential Effects of Hypocretins
on Noise Alone versus Potentiated Startle Responses. physiology Behavior, 89(5),
Stepanski, E., Zorick, F., Roehrs, T., & Roth T. (2000). Effects of sleep deprivation on
daytime sleepiness in primary insomnia. Sleep, 23(2), 215–219.
Sullivan, G. M., Coplan, J. D., Kent, J. M., & Gorman, J. M. (1999). The noradrenergic
system in pathological anxiety: a focus on panic with relevance to generalized anxiety and
phobias. Biological Psychiatry, 46(9), 1205–1218.
Tiihonen, J., Kuikka, J., Rasanen, P., Lepola, U., Koponen, H., Liuska, A., et al. (1997).
Cerebral benzodiazepine receptor binding and distribution in generalized anxiety disor-
ders: A fractal analysis. Molecular Psychiatry, 6, 464–471.
Tursen, U., Kaya, T. I., Eskandari, G., Bocekli, E., Muslu, N., Camdeviren, H., et al. (2004).
Apolipoprotein E gene polymorphism and serum lipids in patients with superficial fungal
disease. Yonsei Medical Journal, 45, 375–379.
Uhde, T. W. (2000). The anxiety disorders. In: M. H. Kryger, T. Roth, & W. Dement (Eds.),
Principles and practice in sleep medicine (3rd ed., pp. 1123–1139). W. B. Saunders: Phila-
delphia, Pennsylvania.
Uhde, T. W. (1995). Caffeine-induced anxiety: An ideal chemical model of panic disorder: In:
G. Asnis & H. M. van Praag (Eds.), Einstein monograph series in psychiatry (pp. 181–205).
New York: Wiley-Liss.
Anxiety and Insomnia 127

Uhde, T. W., Boulenger, J-P., Roy-Byrne, P. P., Vittone, B. J., & Post, R. M. (1985). Long-
itudinal course of panic disorder: Clinical and biological considerations. Progress Neu-
ropsychopharmacology Biological Psychiatry, 9, 39–51.
Uhde, T., Merritt-Davis, O., Yaroslavsky, Y., Glitz, D., Singareddy, R. K., & Cortese, B. M.
(2006). Sleep paralysis: overlooked fearful arousal. 159th annual meeting of the American
Psychiatric Association, Toronto, Canada, May 20–25, 2006 in syllabus & proceedings
summary, symposium 96, fearful sleep arousals (pp. 250–251), Abstract #96D.
Uhde, T. W., Roy-Byrne, P. P., Gillin, J. C., Mendelson, W. D., Boulenger, J-P., Vittone, B. J.,
et al. (1984). The sleep of patients with panic disorder: A preliminary
report. Psychiatry Research, 12(3), 251–259.
Uhde, T. W., & Singareddy, R. (2002). Biological research in anxiety disorders. In: M. Maj
(Ed.), Psychiatry as a Neuroscience, (pp. 237–285). New York City, New York: John Wiley
and Sons.
Uhde, T. W., Stein, M. B., Siever, L. J., Vittone, B. J., Boulenger, J-P., Klein, E. H., et al.
(1989). Behavioral and physiological effects of short-term and long-term administration of
clonidine in panic disorder. Archives of general psychiatry, 46, 170–177.
Vgontzas, A. N., & Chrousos, G. P. (2002). Sleep, the hypothalamic-pituitary-adrenal axis,
and cytokines: multiple interactions and disturbances in sleep disorders. Endocrinal
Metabolic clinics of North America, 31(1), 15–36.
Weissman, M. M., Greenwald, S., Niño-Murcia, G., & Dement, W. C. (1997). The morbidity
of insomnia uncomplicated by psychiatric disorders. General Hospital Psychiatry, 19(4),
Winokur, A., Gary, K. A., Rodner, S., Rae-Red, C, Fernando, A. T., & Szuba, M. P. (2001).
Depression, sleep physiology, and antidepressant drugs. Depression and Anxiety, 14(1),
Zhang, B., & Wing, Y. K. (2006). Sex differences in insomnia: a meta-analysis. Sleep, 29(1),
Zorumski, C. F., & Isenberg, K. E. (1991). Insights into the structure and function of GABA-
benzodiazepine receptors: Ion channels and psychiatry. The American journal of psychia-
try, 148, 162–173.
Part II
Physical Conditions and Anxiety Disorders
Anxiety Disorders and Physical Illness
Comorbidity: An Overview

Tanya Sala, Brian J. Cox, and Jitender Sareen


There has been a considerable body of research exploring associations between

physical illness and depressive disorders, but similar research examining a
possible association of physical illness with anxiety disorders has lagged behind.
More recently, research into the association of anxiety disorders with physical
illness has also been expanding, prompted in part by large epidemiologic survey
data revealing a high prevalence of anxiety disorders in community samples
(Kessler et al., 2005). Some studies have reported an association with anxiety
disorders as strong as or stronger than that with mood disorders (McWilliams,
Cox, & Enns, 2003; McWilliams, Goodwin, & Cox, 2004; Von Korff et al.,
2005). Some studies examining associations between anxiety disorders and
specific illnesses, including thyroid disease, cancer, diabetes, cardiac disease,
gastrointestinal disease, respiratory disease, and chronic pain, have found levels
of anxiety disorders among patients seeking treatment for medical conditions to
be higher than expected compared to the general population.
Attempts to explore and clarify associations between anxiety disorders and
physical disorders have focused on two main sources of data: clinical samples
and community samples. Within these two broad categories, sample size and
methodology of studies vary considerably. Many studies offer intriguing find-
ings, but are challenging to interpret due to inconsistencies in nosological
terminology, psychometric measurements, and methodology. This overview
chapter is designed to be selective rather than exhaustive, and attempts to

Tanya Sala
Department of Psychiatry, University of Manitoba. PsycHealth Centre, PZ430-771
Bannatyne Avenue, Winnipeg, MB, Canada, R3E 3N4, Tel: 204-787-7078, Fax: 204-787-4879

Acknowledgements Dr. Sareen is supported by the Canadian Institutes of Health Research

New Investigator Award. Dr. Cox is supported by the Canada Research Chair Award. The
authors thank Natalie Mota for her assistance in manuscript preparation.

M. J. Zvolensky, J. A. Smits (eds.), Anxiety in Health Behaviors and Physical Illness. 131
Ó Springer 2008
132 T. Sala et al.

focus on large epidemiologic and clinical studies where available in the existing
literature. However, the relative lack of such studies mandates some considera-
tion of research that falls outside these parameters. The majority of currently
available studies in this area are cross-sectional in nature. To avoid unnecessary
repetition, unless otherwise specified, all studies referred to are cross-sectional.
The chapter is organized into sections for each anxiety disorder, with subsections
by system of disease where warranted by available literature.
For the purposes of this literature review, searches of PubMed and PsycInfo
were performed using the general term anxiety disorder as well as each specific
anxiety disorder (e.g. panic disorder), in combination with the general terms
medical condition, physical illness, and medical illness as well as terms for
individual systems of disease (e.g. cardiovascular). Searches included studies
published from 1985 to 2007. For a discussion of associations between pain and
anxiety disorders, please refer to the chapter by Asmundson et al.

Possible Explanations for Observed Relationships

The cross-sectional design of the majority of existing studies of anxiety psycho-

pathology and physical illness precludes any firm conclusions about the exact
nature of the observed associations. However, if a true association exists, pos-
sible explanations can be understood within three basic types of relationships.
Although intriguing, it must be emphasized that these relationships remain
speculative and are not yet convincingly supported by empirical evidence.
1. Direct causal relationship – The presence of an anxiety disorder may directly
increase the risk of suffering from a physical illness. For example, it has been
postulated that chronically high levels of anxiety may produce changes in
physiological functioning, which may in turn increase risk for physical illness
(e.g. Kubzansky, Koenen, Spiro, Vokonas, & Sparrow, 2007). Conversely,
the presence of a physical illness may directly increase the risk of having an
anxiety disorder, a possibility explored in studies of asthma and anxiety
disorders, especially panic disorder (e.g. Goodwin, Jacobi, & Thefeld,
2003; Goodwin & Pine, 2002). An interesting related concept is the model
of mutual maintenance (Sharp & Harvey, 2001), which refers to the phenom-
enon of physical symptoms and anxiety symptoms each exacerbating and
perpetuating the other.
2. Indirect causal relationship – An anxiety disorder may indirectly increase the
risk of having a physical illness through some intermediate factor, such as
smoking, obesity, physical exercise, or substance abuse. Alternatively, hav-
ing a physical illness could indirectly increase the risk of suffering from an
anxiety disorder, such as when the use of a particular medication has an
adverse effect on anxiety symptoms. In individuals being treated for respira-
tory disease, use of bronchodilators has been suggested as a possible exacer-
bating factor for anxiety symptoms.
Anxiety Disorders and Physical Illness Comorbidity: An Overview 133

3. Shared risk factors – There may be no causal relationship between anxiety

disorders and physical illness. Rather, they may share risk factors, such as
genetic, environmental (e.g. poverty), or personality characteristics.
It is also possible, although unlikely, that the reported associations are not
true associations, but the result of another process. Some studies have found
anxiety disorders to be associated with high utilization of medical services (e.g.
Calhoun, Bosworth, Grambow, Dudley, & Beckham, 2002; Hoge, Terhakopian,
Castro, Messer, & Engel, 2007; Richardson, Elhai, & Pedlar, 2006). Increased
frequency of medical attention, and possibly increased somatic complaints, may
lead to increased detection of physical illness in individuals with anxiety dis-
orders. This explanation may be particularly important to exclude for those
physical illnesses that are typically asymptomatic in the early stages, such as
Type II diabetes and hypertension.
A possible link between anxiety disorders and health behaviors such as
obesity (Simon et al., 2006), physical exercise, and smoking remains under-
studied. Although some studies have adjusted for health behaviors such as
smoking (Sareen, Cox, Clara, & Asmundson, 2005), the majority of studies
have not adjusted for important health behaviors. Given the well-known impact
of lifestyle and health choices on risk for physical illness, this remains an issue
for interpretive caution. However, debate exists about the utility of adjusting
for relevant health behaviors. Some authors argue that if health behaviors are
intermediate factors along a causal pathway, rather than confounding vari-
ables, these adjustments could potentially obscure an indirect causal relation-
ship as described above.

Anxiety Symptoms in General

Sareen et al. (2005) examined the relationship between anxiety disorders and a
range of physical disorders in the US National Comorbidity Survey (NCS), a
large nationally representative dataset. This study found that among respon-
dents with one or more physical disorders, a comorbid anxiety disorder diag-
nosis was associated with an increased likelihood of disability even after
adjusting for severity of pain, comorbid mood, and substance use disorders.
Interestingly, adjusting for smoking did not affect these associations. Using
German Health Survey data, Sareen et al. (2006) again examined these associa-
tions. Advantages of this study included both a large sample size and the fact
that the presence of physical illness was based on physician assessment and not
individual self-report, thus increasing reliability. Unfortunately, this study did
not include PTSD. After adjusting for sociodemographic factors and other
common mental disorders, the presence of an anxiety disorder was significantly
associated with thyroid disease, respiratory disease, gastrointestinal disease,
arthritis, migraine headaches, and allergic conditions. The presence of a comor-
bid anxiety disorder with one or more physical disorders was significantly
134 T. Sala et al.

associated with poor quality of life and disability in this study. Important
clinical implications of these associations exist, including the possibility that
the co-occurrence of physical disorders and anxiety disorders may confer a
more disabling condition.
Method of assessment of anxiety and physical illness varies considerably
between studies. Some have assessed anxiety as a unified entity, rather than
distinguishing amongst anxiety disorders (Anderson et al., 2002; Engum, 2007;
Engum, Bjoro, Mykletun, & Dahl, 2002; Hermanns, Kulzer, Krichbaum,
Kubiak, & Haak, 2005; Kruse, Schmitz, & Thefeld, 2003; Shaban, Fosbury,
Kerr, & Cavan, 2006), adding to the challenge of interpretation and clinical

Posttraumatic Stress Disorder (PTSD)

Editions of the Diagnostic and Statistical Manual of Mental Disorders (DSM)

prior to the DSM-IV explicitly excluded life-threatening illness as a stressor that
could be considered a traumatic precipitant of PTSD. With the publication of
DSM-IV, individuals could receive a diagnosis of PTSD as a direct result of
being traumatized by the experience of a life-threatening illness. This develop-
ment has contributed to interest in the association between PTSD and physical
Much of the early work examining the co-occurrence of PTSD with physical
disorders involved studies of war veterans. Higher scores on measures of post-
traumatic stress symptoms have been linked with increased medical care utiliza-
tion and lower ratings of general health (Calhoun et al., 2002; Hoge et al., 2007;
Richardson et al., 2006). Frayne et al. (2004) reported that female veterans with
PTSD had a greater number of physical illnesses and poorer physical health
functioning than female veterans with either depression alone or neither
Within the general population of the United States, Sareen et al. (2005)
reported that, among DSM-IV anxiety disorders, PTSD had the greatest num-
ber of significant associations with chronic physical disorders, including neu-
rological, cardiovascular, gastrointestinal, metabolic/autoimmune, and bone
or joint conditions. A study of primary care patients found PTSD to be a
stronger predictor of a reported number of medical problems than trauma
history, physical injury, lifestyle factors, or comorbid depression (Weisberg
et al., 2002).
Studies have not been limited to adult populations. A descriptive epidemio-
logic case-control analysis of Medicaid service-use data in the United States
found PTSD to be associated with adverse health outcomes in female children
and adolescents (Seng, Graham-Bermann, Clark, McCarthy, & Ronis, 2005),
and significant associations with PTSD were found across a wide range of
disease categories. The importance of PTSD diagnosis as a predictor of having
Anxiety Disorders and Physical Illness Comorbidity: An Overview 135

a physical illness appeared to increase with age. Studies of different ethnic

groups have also identified PTSD as being significantly associated with cardi-
ovascular disease, even after controlling for major depression and traditional
cardiovascular risk factors including age, sex, education, diabetes, high blood
pressure, and smoking (Sawchuk et al., 2005).
Efforts to understand the biological pathways through which PTSD and
physical illness might be linked have included examination of the hypothala-
mic-pituitary-adrenal and sympathetic-adrenal-medullary axes. Findings of
lower cortisol, higher catecholamine levels, and increased circulating
T-lymphocytes in individuals with PTSD support the possibility of biological
underpinnings (e.g. Yehuda et al., 1995).

Cardiovascular Disease

A link between PTSD or trauma and cardiovascular disease has been suggested
by a number of studies. A recent prospective study of community-dwelling US
military veterans found a significant association between PTSD symptoms at
baseline and subsequent development of coronary heart disease, even after
controlling for depressive symptoms (Kubzansky et al., 2007). A study of
Korean War and World War II veterans reported increased rates of physi-
cian-diagnosed cardiovascular disease among veterans with PTSD (Schnurr,
Spiro, & Paris, 2000). Studies demonstrating increases in basal cardiovascular
activity in PTSD sufferers offer a potential biological pathway (Buckley &
Kaloupek, 2001).
One important possible consequence of PTSD secondary to a medical con-
dition or medical treatment is the potential for lower levels of adherence to
medical treatment. Shemesh et al. (2001, 2004, 2006) examined the hypothesis
that symptoms of MI-related PTSD were linked to nonadherence to medical
follow-up. The authors hypothesized that patients who experience myocardial
infarction as a traumatic event may not take medications as prescribed, with
nonadherence representing part of the avoidance dimension of PTSD. In these
studies, PTSD symptoms were significantly associated with nonadherence to
medical treatment. In support of these findings, others have reported that
alleviation of PTSD symptoms in a series of burn patients could improve
adherence to treatment (Countermanche & Robinow, 1989).

Neurological Disease

A number of studies have sought to determine whether PTSD diagnosis is

associated with psychogenic nonepileptic seizure (PNES), also referred to as
pseudoseizure and psychogenic seizure. Fiszman, Alves-Leon, Nunes,
D’Andrea, and Figueira (2004) recently conducted a review of studies on the
136 T. Sala et al.

prevalence of traumatic events and/or PTSD in patients with PNES. Among the
17 studies included in this review, ten analyzed the PTSD diagnosis in PNES
patients. Although the prevalence rates reported in these ten studies varied
considerably, rates of PTSD were consistently higher than those found in the
general population. However, when compared with control groups who had
epilepsy, only two studies found a significant difference in rates of PTSD.
Limitations acknowledged by the authors include the fact that all 17 studies
were hospital-based, and many had small sample sizes. The authors speculate
that PNES, which has been traditionally thought of as a dissociative phenom-
enon, might be understood as a severe form of PTSD that includes prominent
dissociative features. There is currently too little information to warrant con-
clusions on this premise.

Endocrine/Metabolic/Autoimmune Disease

Among the significant associations with PTSD in the study by Sareen et al.
(2005), metabolic/autoimmune conditions showed the most robust association
(AOR 3.32, CI 1.96-5.62). A number of studies have supported an alteration
in thyroid function in individuals with PTSD, both in chronic combat-related
PTSD (Mason et al., 2004; Wang & Mason, 1999; Wang et al., 1995) and in
survivors of childhood sexual abuse (Friedman, Wang, Jalowiec, McHugo, &
McDonagh-Coyle, 2005). In both studies by Wang et al., alterations in
thyroid function were specifically associated with increased hyperarousal
symptoms. Through promotion of ongoing noradrenergic transmission,
elevations in triiodothyronine (T3) have been proposed by some authors as
a possible underlying biological mechanism perpetuating arousal features in
PTSD (Prange, 1999), although this hypothesis remains speculative at this
Using data from a large national sample of Vietnam veterans, Boscarino
(2004) tested the hypothesis that chronic sufferers of PTSD may be at increased
risk for certain autoimmune diseases, including rheumatoid arthritis, thyroid
disease, insulin-dependent diabetes, and psoriasis. After adjusting for a variety
of factors, including age, alcohol and drug abuse, and history of cigarette
smoking, the authors noted significant associations between chronic PTSD
and all the above-mentioned illnesses. In a large study of US male veterans
with diabetes, Trief, Ouimette, Wade, Shanahan, and Weinstock (2006) noted
significantly higher cholesterol, LDL, weight, and BMI in subjects with PTSD
and depression, compared to depression alone, PTSD alone, or neither.
Goodwin and Davidson (2005), using the NCS dataset, examined the asso-
ciation between self-reported diabetes and PTSD in a community sample of
adults. In this study, self-reported diabetes was found to be significantly asso-
ciated with an increased likelihood of PTSD, but not with an increased like-
lihood of any other mental disorder. This study not only provides evidence for
Anxiety Disorders and Physical Illness Comorbidity: An Overview 137

generalizability of previous findings to adults in the community, it also suggests

some specificity for the association with PTSD compared to other mental


The majority of studies assessing incidence or prevalence of PTSD in indivi-

duals with cancer have involved cross-sectional assessment of individuals with
breast cancer following completion of primary treatment, and have reported
rates ranging from 0 to 32% (e.g. Andrykowski, Cordova, Studts, & Miller,
1998). Several studies (Andrykowski, Cordova, McGrath, Sloan, & Kenady,
2000; Mehnert & Koch, 2007) also included a longitudinal component in the
form of either a 12 or 6-month follow-up assessment for PTSD. Analysis of data
from a large US community sample (Honda & Goodwin, 2004) found no
significant association between cancer and PTSD.
A variety of factors have been reported to be predictive of PTSD in indivi-
duals with cancer, including increased emotional distress following diagnosis,
female gender, a history of negative life stressors, prior psychological distur-
bance, lower socioeconomic status, poor social support, and reduced physical
functioning (e.g. Jacobsen et al., 2002). Some have reported an association
between more severe PTSD symptoms and a more advanced stage of disease
(Jacobsen et al., 1998), while other studies have found no association between
medical variables of the illness and risk of PTSD (Alter et al., 1996; Green et al.,
1998). With respect to the predictive value of any specific medical variable,
studies have been mixed.
In summary, challenges exist in the interpretation of existing PTSD litera-
ture, one of which is the predominance of cross-sectional methodology in
evaluating associations of PTSD with physical illnesses. Research examining
PTSD related to discrete trauma may not generalize to the experience of a
significant physical illness. Clarifying the precise event or events that represent
trauma can be difficult but clinically relevant, as some evidence exists that
prolonged or multiple traumatic events may result in greater severity or chroni-
city of PTSD symptoms. The relative contributions of the diagnosis itself,
treatment, side effects, prognosis, disrupted physical, social and occupational
functioning, and risk of exacerbation or recurrence cannot be examined within
a cross-sectional design.
Diagnostic validity using current DSM-IV criteria also presents special
challenges in this population. Divergent definitions of the stressor may increase
variability in the diagnosis of PTSD in the context of physical illness. The utility
of specific symptoms within DSM-IV criteria for PTSD needs to be assessed in
this population. For example, ‘‘sense of a foreshortened future’’ may be a
realistic concern in individuals with more serious physical illnesses such as
cancer. Arousal symptoms may overlap considerably with side effects of
138 T. Sala et al.

treatment. PTSD also needs to be distinguished from reactive stress to ongoing

medical issues, or a grief reaction when the diagnosis is a terminal or signifi-
cantly disabling illness.
There is good evidence supporting cognitive-behavior therapy (CBT) for
treatment of PTSD. However, solid research supporting CBT for treatment of
PTSD co-occurring with physical illness is lacking. The relative benefits of
cognitive vs. behavioral components of CBT may also vary in this population.

Panic Disorder (PD)

Adding to the complexity of the relationship between panic disorder and

physical illness, both for individuals suffering from these disorders and for the
professionals who provide them with care, is the high degree of overlap between
symptoms of panic and certain physical illnesses, particularly cardiac and
respiratory disease.

Cardiovascular Disease

Given the high level of symptom overlap, it is not surprising that an association
between panic disorder and cardiac disease has been relatively well studied. Of
the 13 symptoms of a panic attack listed in DSM-IV, many could also represent
cardiovascular disease. Many patients with panic disorder present first to the
emergency room requesting medical assessment (Huffman & Pollack, 2003).
Data from the NIMH ECA study suggested that individuals who present with
numerous medically unexplained symptoms have 200 times increased odds of
having panic disorder, compared with 17 for major depression (Simon & Von
Korff, 1991). A strong relationship has been reported between panic attacks and
hypertension diagnosed by physical exam in a primary care sample (Davies et al.,
1999). Similarly, myocardial infarction has been associated in at least two studies
with increased likelihood of new-onset panic attacks. Physiological mechanisms
such as decreased heart rate variability, increased platelet activity, and increased
sympathetic tone have been implicated in the relationship between anxiety and
cardiac disease (Kawachi, Sparrow, Vokonas, & Weiss, 1995).
In a retrospective study of a large random community sample, Weissman,
Markowitz, Ouellette, Greenwald, and Kahn (1990) examined the association
between PD and cardiovascular/cerebrovascular disorders. After adjusting for
a number of demographic factors, participants with PD were at higher risk of
reporting high blood pressure, heart attack, and stroke than a group of respon-
dents with no psychiatric disorder, but when compared with other psychiatric
disorders, only stroke remained significant. One limitation of this study is that
existence of cardiovascular or cerebrovascular disease was based on self-report.
Anxiety Disorders and Physical Illness Comorbidity: An Overview 139

Kawachi et al. (1994) reported on a prospective study of phobic anxiety and

the risk of coronary artery disease (CAD) in a sample of 34,000 male health
professionals. Although not a DSM-IV disorder, phobic anxiety likely most
closely resembles panic symptoms, and these findings are therefore discussed
here. Subjects were initially free of cardiovascular disease at baseline. A sig-
nificant difference was found in the age-adjusted relative risk of fatal coronary
artery disease when groups with the highest and lowest anxiety scores were
compared. The excess risk was confined to sudden death. A more recent pro-
spective study examined the relationship between phobic anxiety and coronary
artery disease among women participating in the Nurses’ Health Study over
12 years of follow-up (Albert, Chae, Rexrode, Manson, & Kawachi, 2005). In
this sample of over 72,000 women with no history of cardiac disease at baseline,
a higher anxiety score was associated with an increased risk of sudden cardiac
death and fatal coronary artery disease, but not of nonfatal MI. After adjusting
for possible intermediaries (hypertension, diabetes, and elevated cholesterol),
only a trend toward increased risk persisted for sudden cardiac death.

Respiratory Disease

Chronic obstructive pulmonary disease (COPD) and asthma are featured most
prominently in the literature, although a smaller number of studies examine an
association between panic disorder and allergic reactions in both children and
adults (Goodwin, 2002; Kovalenko et al., 2001). As with cardiovascular disease,
the high level of symptom overlap presents unique challenges in understanding
repeatedly observed associations. Given the early age of onset of asthma, many
have speculated that it could be a contributor to the development of panic
disorder, although there is evidence supporting a bidirectional influence of the
disorders (Hasler et al., 2005), as well as studies questioning whether both may
be due to a third variable (Goodwin, Fergusson, & Horwood, 2004).
Lifetime rates of respiratory disease have been reported to be as high as 47%
in individuals with panic disorder (Zandbergen et al., 1991). Rates of panic
disorder in individuals with respiratory illness have also been found to be
increased above those reported in general population studies, not only within
clinical samples of adults, but also in community samples and within child and
adolescent populations (Goodwin & Eaton, 2003; Goodwin, Jacobi, et al.,
2003; Goodwin, Olfson, et al., 2003; Goodwin, Pine, & Hoven, 2003). In a
community sample of more than 4000 individuals with current severe asthma,
10% also had panic disorder (Goodwin, Jacobi, et al., 2003).
Using the NCS dataset, Sareen et al. (2005) did not find a relationship
between panic attacks and respiratory disease, as previous studies have reported
(Goodwin & Eaton, 2003; Goodwin, Jacobi, et al., 2003; Goodwin et al., 2003;
Goodwin, Pine, et al., 2003; Ortega, McQuaid, Canino, Goodwin, & Fritz,
2004). A possible explanation suggested by the authors is that the discrepancy
140 T. Sala et al.

may be related to severity. The NCS did not distinguish severity of physical
disorders, but in a German community sample, severe asthma diagnosed by a
physician was associated with panic attacks while non-severe asthma was not
(Goodwin, Jacobi, et al., 2003).
Special considerations arise in the treatment of comorbid asthma and panic
disorder. Treatments for asthma, in particular short-acting b2 agonists, may
exacerbate anxiety symptoms as a side effect of their use, presenting a dilemma
both for the individual and health care provider. If symptoms are attributed to
asthma and a short-acting b2 agonist is used, symptoms may worsen if related to
panic rather than respiratory causes. Some evidence exists that patients with
comorbid panic disorder and asthma may make greater use of short-acting b2
agonists than asthma-only patients, independent of pulmonary function,
asthma medication class, and sociodemographic status (Feldman, Lehrer,
Borson, Hallstrand, & Siddique, 2005).
Several studies have reported an increased risk of panic disorder in indivi-
duals with COPD (e.g. Karajgi, Rifkin, Doddi, & Kolli, 1990). A recent review
by Brenes (2003) noted consistently increased rates of panic disorder and GAD
across studies, but commented on the lack of evidence to guide treatment
decisions in this population.

Neurological Disease

Studies examining associations with neurological illness have focused mainly on

epilepsy or vestibular dysfunction. The majority of those involving epilepsy
have concentrated on the diagnostic challenges in differentiating partial seizure
epilepsy involving pre-ictal or ictal fear from panic attacks or panic disorder,
and available literature is limited mainly to case reports (e.g. Bernik, Corregiari,
& Braun, 2002; Thompson, Duncan, & Smith, 2000). Some authors have
speculated on the existence of a subgroup of panic attacks, in which panic
constitutes symptoms of simple partial seizures with primarily psychic content
(Alvarez-Silva, Alvarez-Rodriguez, Perez-Echeverria, & Alvarez-Silva, 2006).
Patients with partial seizures may have symptoms similar to panic attacks –
prodromal tension, fear, and autonomic disturbances, including changes in
blood pressure, heart rate, and skin color. This category of overlap once
again highlights the importance of treatment providers maintaining an index
of suspicion to rule out physical illness as an etiologic factor in the presentation
of anxiety symptoms. Possible indications for neurological investigations
include atypical symptoms or age at onset, non-response to standard treat-
ments, abnormalities in routine physical exam or laboratory investigations, or
family history.
With respect to vestibular dysfunction, a two-year prospective study exam-
ining the role of cognitions in the development of panic disorder after the
experience of vestibular neuritis (Godemann, Schabowska, Naetebusch,
Anxiety Disorders and Physical Illness Comorbidity: An Overview 141

Heinz, & Strohle, 2006) reported that although fear arising on the first day
of an acute vestibular episode did not predict the development of panic,
fear of vertigo one week after the dysfunction was a significant predictor.
After six weeks, persistent fear of vertigo remained a significant predictor.
Results of a small clinical study suggested that subclinical vestibular dys-
function might contribute to symptoms of panic disorder, (Jacob, Furman,
Durrant, & Turner, 1996). Although the majority of studies assessing
vestibular function in individuals with panic disorder report a high pre-
valence of abnormal results of vestibular testing, no consistent pattern of
vestibular dysfunction has been observed across studies. Similarly, although
results of studies examining psychiatric symptoms in patients with vestibu-
lar dysfunction in general show increased levels of anxiety symptoms,
especially panic, no consistent pattern of association has yet been

Gastrointestinal (GI) Disease

The majority of studies within this system of disease have focused on associa-
tions of anxiety disorders with ‘‘Functional’’ GI symptoms. Functional refers
to symptoms reported by an individual that do not correlate to any tissue
pathology detectable by medical testing. In a nationally representative survey
of over 13,000 individuals in the United States, Lydiard et al. (1994) found
that individuals with panic disorder had the highest rate of unexplained GI
symptoms (7.2%) compared with other diagnostic categories. Respondents
who reported two gastrointestinal symptoms had significantly higher lifetime
prevalence rates for panic disorder and agoraphobia than those who reported
no GI symptoms. In patients with severe irritable bowel syndrome (IBS),
panic disorder has been associated with impaired functioning (Creed et al.,
In a longitudinal study of a representative cohort of Swiss adults by
Hochstrasser and Angst (1996), subjects were assessed for functional GI
symptoms and anxiety and depressive disorders. Cross-sectionally, a signif-
icant relationship was found between functional GI complaints and panic
disorder, subthreshold panic disorder, agoraphobia, social phobia, simple
phobia, and GAD. However, those who reported functional GI complaints
at younger ages were not at increased risk for subsequent development of
an anxiety disorder. The authors concluded that functional gastrointestinal
complaints reflect a non-specific concomitant vegetative disturbance asso-
ciated with anxiety disorders, not a risk factor for the development of a
specific anxiety disorder. Data from a large-scale, nationally representative
sample of Japanese subjects was analyzed for comorbidity of IBS with
panic disorder and agoraphobia (Kumano et al., 2004). Significantly higher
rates of PD and agoraphobia were observed in subjects with IBS than in
142 T. Sala et al.

non-IBS subjects. Although IBS was more prevalent in females, comorbid-

ity did not differ between males and females.

Endocrine/Metabolic Disease

Within this category, thyroid disease and diabetes have received the most
attention from researchers. A recent review of studies examining associations
between thyroid dysfunction and anxiety disorders found significantly elevated
rates for panic disorder in patients with a history of thyroid disease, and
concluded that inquiring about thyroid symptoms and screening for thyroid
disease is warranted in patients with panic disorder (Simon et al., 2002). How-
ever, of the 12 studies included in this review, only six were controlled, and the
findings of studies varied considerably.
Results from larger studies have been mixed. Analyzing a community
sample, Sareen et al. (2005) did not find an association between panic attacks
and self-reported diagnosis of metabolic/endocrine disease. Analysis of a
different community sample also did not find a significant association
between panic disorder/agoraphobia and self-reported thyroid disease
(Patten, Williams, Esposito, & Beck, 2006). However, analysis of a large
clinical sample showed an association between panic disorder and thyroid
disease (Rogers et al., 1994).


A review of studies between 1980 and 1994 examining psychological distur-

bances in cancer patients (van’t Spijker, Trijsburg, & Duivenvoorden, 1997)
found no significant differences from the general population with respect to
anxiety and psychological distress. In fact, in comparison with other medical
patients, cancer patients showed significantly less anxiety in this review. In
contrast, Honda and Goodwin (2004) used NCS data to examine the associa-
tion between cancer and mental disorders, adjusting for sociodemographic
characteristics. Among other reported findings, cancer was associated with
increased risk of agoraphobia, although this finding barely reached statistical
significance. There was no significant association in this study with panic
disorder. The findings of Honda and Goodwin have been criticized on a number
of grounds (Coyne & Palmer, 2005), including the low prevalence of cancer
diagnosis in the NCS (n=45, 0.76%), making it possible for small sampling
biases to have large effects. It has also been suggested that diagnoses made in the
NCS may have been less valid when a significant medical comorbidity was
present, as distinguishing between psychiatric symptoms and physical symp-
toms could pose an even greater challenge for lay interviewers than clinicians.
Anxiety Disorders and Physical Illness Comorbidity: An Overview 143

Generalized Anxiety Disorder (GAD)

Cardiovascular Disease

Barger and Sydeman (2005), using data from a large US community sample of
adults aged 25–74, found that independently of depression, subjects with GAD
were more likely to have coronary artery disease risk factors, as measured by
self-reported smoking status, body mass index, and recent medication use for
hypertension, hypercholesterolemia, and diabetes.

Gastrointestinal Disease

Goodwin and Stein (2002) examined a possible link between GAD and peptic ulcer
disease (PUD) among adults in a community sample. After controlling for differ-
ences in sociodemographic characteristics, psychiatric comorbidity, and physical
comorbidity, the authors found a significantly increased risk of self-reported PUD
in subjects with GAD. A dose-response relationship was also found between
number of GAD symptoms and higher likelihood of PUD. A review of a number
of smaller studies (Lydiard, 2001) reported a consistent link between irritable
bowel syndrome and GAD. In the NCS sample, Sareen et al. (2005) found GAD
to be associated exclusively with gastrointestinal disease after adjusting for socio-
demographic factors, other anxiety disorders, and depression. See the section on
Panic Disorder and gastrointestinal disease for discussion of a relevant study by
Hochstrasser and Angst (1996).

Endocrine/Metabolic/Autoimmune Disease

Studies of this body system have primarily examined associations with either
thyroid disease or diabetes mellitus. In a small study of a clinical sample, Carta
et al. (2005) found that subjects with Hashimoto disease showed higher fre-
quencies of GAD compared to subjects with euthyroid goiter and controls. See
the section on Panic Disorder and endocrine/metabolic disease for discussion of
a relevant review by Simon et al. (2002).
Studies in clinical samples have also supported an association between GAD
and diabetes (Kovacs, Goldston, Obrosky, & Bonar, 1997). However, both studies
of diabetes and studies of thyroid disease typically involve smaller numbers of
individuals and are subject to the sampling bias that accompanies studies of
treatment-seeking populations. In a recent review of anxiety in adults with diabetes
(Grigsby, Anderson, Freedland, Clouse, & Lustman, 2002), the authors concluded
that 14% of diabetic subjects in the included studies also had GAD, well in excess
of the prevalence reported in community surveys (Kessler, Chui, Demler,
144 T. Sala et al.

Merikangas, & Walters 2005). However, of the 18 studies included in this review,
only 5 included a non-diabetic control group, and most of the subjects were drawn
from clinical populations. In contrast, an analysis of general population data
(Sareen et al., 2005) found no significant association between metabolic/endocrine
or autoimmune diseases and GAD.

Obsessive-Compulsive Disorder (OCD)

The bulk of the research exploring a link between obsessive-compulsive dis-

order and physical illness has focused on a possible etiologic link between
Group A b-Hemolytic Streptococcus (GABHS) and the onset and subsequent
exacerbations of OCD. Allen, Leonard, and Swedo (1995) published an early
description of cases in which the onset of obsessive-compulsive symptoms and
tics appeared to follow streptococcal infection. Based on the first 50 cases
described, Swedo et al. (1998) defined five inclusion criteria for this postulated
subgroup of OCD cases, referred to as PANDAS – pediatric autoimmune
neuropsychiatric disorders associated with streptococcus infection: (1) presence
of OCD and/or a tic disorder (meeting DSM-IV criteria); (2) pediatric symptom
onset (between 3 years and the beginning of puberty); (3) episodic course
characterized by abrupt onset of symptoms or dramatic symptom exacerba-
tions; (4) temporal relationship between GABHS infections and symptom
exacerbations; (5) neurological abnormalities (e.g. choreiform movements,
motoric hyperactivity, and/or tics) present during symptom exacerbations.
The postulated biological mechanism for PANDAS is referred to as mole-
cular mimicry, a process by which pathogens attempt to evade host immune
defenses designed to distinguish between host and foreign antigens. In the case
of PANDAS, the mimicry is thought to exist between surface antigens of the
streptococcus bacterium and neuronal antigens in the basal ganglia, resulting in
neuropsychiatric symptoms.
The majority of studies examining PANDAS are limited by small sample size,
and there is a strong bias in the literature towards samples from specialty clinics.
Although a robust association was reported in a large, population-based case-
control study (Mell, Davis, & Owens, 2005), another recent prospective study
found no association between OCD or tic symptom exacerbations and GABHS
infections beyond that expected based on chance alone (Luo et al., 2004).
The presence of anti-basal ganglia antibodies (ABGA) in the peripheral
blood of patients with tics and/or OCD may be an important clue to the
involvement of autoimmunity in these disorders (Dale, Heyman, Giovannoni,
& Church, 2005). However, it is important to note that the presence of ABGA
does not prove autoimmunity as a causal mechanism. Autoantibodies against a
variety of tissue antigens are present in healthy individuals, and therefore, the
presence of these antibodies does not by itself prove pathological significance.
In fact, a recent study did not detect differences in antineuronal antibody
Anxiety Disorders and Physical Illness Comorbidity: An Overview 145

profiles between children with Tourette’s syndrome, PANDAS, or age-matched

controls (Singer, Hong, Yoon, & Williams, 2005).
Efforts to elucidate a possible autoimmune mechanism have included a
number of treatment studies. A pilot study of antibiotic prophylaxis for pre-
vention of symptom exacerbations, analogous to penicillin prophylaxis in
rheumatic fever, revealed no difference in number of infections or obsessive-
compulsive or tic symptom severity between antibiotic and placebo groups
(Garvey et al., 1999). A study of 29 patients fulfilling proposed PANDAS
criteria reported sustained benefits after both intravenous immunoglobulin
and plasma exchange, compared to placebo (Perlmutter et al., 1999). However,
this small study had significant methodologic problems, and its findings have
not been convincingly replicated. Other authors have reported rapid diminu-
tion of OC symptoms in cases fulfilling PANDAS criteria following treatment
with antibiotics (Murphy & Pichichero, 2002), and positive correlations
between streptococcal titers and obsessive-compulsive severity rating changes
in subjects with large symptom changes (Murphy et al., 2004).
Despite a large body of literature in this area, the existence of a PANDAS
subgroup of OCD sufferers remains controversial. There is a clear need for long-
itudinal studies aimed at delineating the precise relationships between GABHS
infections, OC symptom exacerbations, and alterations in immune parameters. It
remains uncertain whether an association, if it truly exists, refers to a specific
subgroup of patients with OCD or tics (PANDAS), to tic disorders and OCD in
general, or to a broader range of diagnostic entities perhaps encompassing eating
disorders and impulse control disorders such as trichotillomania. Based on the
evidence so far, some argue for routine throat culture and antistreptolysin-O titres
in children with an abrupt onset or exacerbation of OCD or tic disorder, although
others dispute this recommendation. In summary, there is ample evidence to
warrant further research in this area, but insufficient data to definitively describe
the PANDAS subgroup or any treatments specific to such a group.

Social Anxiety Disorder

Gastrointestinal Disease

See the section on Panic Disorder and gastrointestinal disease for discussion of
a relevant study by Hochstrasser and Angst (1996).

Endocrine/Metabolic Disease

Results of smaller studies examining an association between social anxiety

disorder and thyroid disease have been mixed, with some authors reporting
significant associations (e.g. Carta et al., 2005), and others finding no increased
prevalence compared to the general population (Simon et al., 2002). Studies
146 T. Sala et al.

using larger community samples have generally supported a significant associa-

tion. Analysis of NCS data by Sareen et al. (2005) revealed a strong association
between social phobia and metabolic/autoimmune disorders. Using data from a
Canadian general population sample, Patten et al. (2006) reported that after
adjusting for sex, age, and other chronic conditions, social phobia remained
associated with thyroid disease.

Autoimmune Disease

Lindal, Thorlacius, Steinsson, and Stefansson (1995), in an analysis of all

known individuals with lupus in Iceland, reported that social phobia was
more common in persons with lupus than in the general population. Reasons
for this association are unclear, although the authors speculated that the
occurrence of disfiguring facial rashes in certain lupus sufferers could be related
to increased social anxiety.

Specific or Simple Phobia

Respiratory Disease

Goodwin, Jacobi, et al. (2003) reported a strong association between physician-

diagnosed severe asthma and specific phobia in a German community sample,
which included assessment by a physician rather than self-report by study
participants. This finding is consistent with analysis of NCS data by Sareen
et al. (2005), which found a significant association between specific phobia and
respiratory disease.

Gastrointestinal Disease

See the section on Panic Disorder and gastrointestinal disease for discussion of
a relevant study by Hochstrasser and Angst (1996).

Endocrine/Metabolic Disease

As part of an examination of the epidemiology of blood-injection-injury phobia

in the Baltimore ECA Follow-up Study, Bienvenu and Eaton (1998) found no
Anxiety Disorders and Physical Illness Comorbidity: An Overview 147

strong, broad general health ramifications of this subtype of specific phobia.

However, individuals with diabetes and blood-injection-injury phobia had
higher than expected rates of macrovascular complications. This finding sup-
ports those from a previous study reporting that phobic symptoms are frequent
in patients with Type I diabetes, and that the intensity of blood and injury fears
negatively correlated with number of blood glucose measurements per day and
metabolic control (Berlin et al., 1997). These findings together suggest the
possibility that decreasing blood-injury fears could lead to improved metabolic


See the section on Panic Disorder and cancer for discussion of relevant studies
by van’t Spijker et al. (1997) and Honda and Goodwin (2004).

Concluding Comments


Examination of the relationship between anxiety disorders and physical illness

is a limited but expanding area of the scientific literature. The prevalence of
both anxiety disorders and many physical illnesses in the population argues
convincingly for the clinical relevance of this area of research. In addition to
informing screening and detection of both mental and physical disorders, early
evidence suggesting an impact on risk of death and disability underscores the
importance of extending our understanding of these relationships. The impact
of this comorbidity on treatment for either condition remains largely unex-
plored. Results of an analysis of primary care patients with panic disorder
(Roy-Byrne et al., 2005) suggest that patients with a higher burden of physical
illness responded equally well to CBT and pharmacotherapy targeted at panic
symptoms, but more research in this area is needed to provide support for these
findings and to explore treatment implications for other anxiety disorders.
Although an impressive number of studies support associations between
anxiety disorders and physical illness, the variety among reported findings is
equally striking. This variety represents the current state of the literature:
sufficient evidence exists to argue convincingly that associations between anxi-
ety disorders and physical illness exist, but the precise nature of the relation-
ships remains elusive, and a number of important methodological concerns
limit conclusions. Which anxiety disorders are associated with which physical
illnesses, and the exact nature of these associations, is a task for future research
148 T. Sala et al.

An essential caveat accompanying any discussion of associations between

mental and physical disorders is the necessity of ruling out a physical illness
masquerading as an anxiety disorder by virtue of overlapping symptomatology.
In order for this very real possibility to be reliably distinguished from comor-
bidity, a thorough history and physical exam is required in any patient present-
ing for the first time with symptoms of an anxiety disorder, in any patient with
markedly altered nature or severity of symptoms, and in any patient whose
symptoms are resistant to standard treatments. Only when any existing physical
illness is diagnosed and optimally treated can associations with mental disor-
ders be accurately and safely explored.

Limitations and Future Directions

Within the currently available literature, a number of methodological issues

limit our understanding of associations between anxiety disorders and physical
illness. The relative lack of large clinical or epidemiological studies, and relative
preponderance of smaller studies, is one of the primary limiting factors. In
addition, much of the available research still relies heavily on self-report in
assessment of physical illness. In a population of individuals shown to be
sensitized to perception of physical symptoms, reliance on self-report is parti-
cularly problematic.
The lack of understanding of the role of other health behaviors, such as
physical exercise and substance use, must also be highlighted as an issue for
interpretive caution. It remains to be discovered whether health behaviors play
a significant role in the observed associations between anxiety disorders and
physical illness, and if so, whether they represent confounding variables or inter-
mediate factors in an indirect causal pathway. The majority of studies in this area
are cross-sectional, precluding any causal inferences. The scarcity of longitudinal
data is a major limitation of existing research, and the need for more prospective,
longitudinal epidemiologic data with objective measures of both physical and
mental health status stands out as a clear mandate for future research efforts.
Despite the limitations in available studies, existing research is compelling and
argues convincingly that further attention to this area is warranted.


Albert, C. M., Chae, C. U., Rexrode, K. M., Manson, J. E., & Kawachi, I. (2005). Phobic
anxiety and risk of coronary heart disease and sudden cardiac death among women.
Circulation, 111, 480–487.
Allen, A. J., Leonard, H. L., & Swedo, S. E. (1995). Case study: A new infection-triffered,
autoimmune subtype of pediatric OCD and Tourette’s syndrome. Journal of the American
Academy of Child & Adolescent Psychiatry, 34, 307–311.
Alter, C. L., Pelvcovitz, D., Axelrod, A., Goldenberg, B., Harris, H., Meyers, B., et al. (1996).
Identification of PTSD in cancer survivors. Psychosomatics, 37, 137–143.
Anxiety Disorders and Physical Illness Comorbidity: An Overview 149

Alvarez-Silva, S., Alvarez-Rodriguez, J., Perez-Echeverria, M. J., & Alvarez-Silva, I. (2006).

Panic and epilepsy. Anxiety Disorders, 20, 353–362.
Anderson, R. J., Grigsby, A. B., Freedland, K. E., de Groot, M., McGill, J. B., Clouse, R. E.,
et al. (2002). Anxiety and poor glycemic control: A meta-analytic review of the literature.
International Journal of Psychiatry In Medicine, 32, 235–247.
Andrykowski, M. A., Cordova, M. J., McGrath, P. C., Sloan, D. A., & Kenady, D. E. (2000).
Stability and change in posttraumatic stress disorder symptoms following breast cancer
treatment: A 1-year follow-up. Psychooncology, 9, 69–78.
Andrykowski, M. A., Cordova, M. J., Studts, J. L., & Miller, T. W. (1998). Posttraumatic
stress disorder after treatment for breast cancer: Prevalence of diagnosis and use of the
PTSD Checklist-Civilian Version (PCL-C) as a screening instrument. Journal of Consult-
ing and Clinical Psychology, 66, 586–590.
Barger, S. D., & Sydeman, S. J. (2005). Does generalized anxiety disorder predict coronary
heart disease risk factors independently of major depressive disorder? Journal of Affective
Disorders, 88, 87–91.
Berlin, I., Bisserbe, J. C., Eiber, R., Balssa, N., Sachon, C., Bosquet, F., et al. (1997). Phobic
symptoms, particularly the fear of blood and injury, are associated with poor glycemic
control in type I diabetic adults. Diabetes Care, 20, 176–178.
Bernik, M. A., Corregiari, F. M., & Braun, I. M. (2002). Panic attacks in the differential
diagnosis and treatment of resistant epilepsy. Depression & Anxiety, 15, 190–192.
Bienvenu, O. J., & Eaton, W. W. (1998). The epidemiology of blood-injection-injury phobia.
Psychological Medicine, 28, 1129–1136.
Boscarino, J. A. (2004). Posttraumatic stress disorder and physical illness. Annals of the New
York Academy of Sciences, 1032, 141–153.
Brenes, G. A. (2003). Anxiety and chronic obstructive pulmonary disease: Prevalence, impact,
and treatment. Psychosomatic Medicine, 65, 963–970.
Buckley, T. C., & Kaloupek, D. G. (2001). A meta-analytic examination of basal cardiovascular
activity in posttraumatic stress disorder. Psychosomatic Medicine, 63, 585–594.
Calhoun, P. S., Bosworth, H. B., Grambow, S. C., Dudley, T. K., & Beckham, J. C. (2002).
Medical service utilization by veterans seeking help for posttraumatic stress disorder.
American Journal of Psychiatry, 159, 2081–2086.
Carta, M. G., Hardoy, M. C., Carpiniello, B., Murru, A., Marci, A. R., Carbone, F.,
et al. (2005). A case control study on psychiatric disorders in Hashimoto disease
and Euthyroid Goitre: Not only depressive but also anxiety disorders are associated
with thyroid autoimmunity. Clinical Practice and Epidemiology in Mental Health, 1,
Countermanche, D. L., & Robinow, O. (1989). Recognition and treatment of the posttraumatic
stress disorder in the burn victim. Journal of Burn Car & Rehabilitation, 10, 247–250.
Coyne, J. C., & Palmer, S. C. (2005). National Comorbidity Survey data concerning cancer
and depression lack credibility. Psychotherapy and Psychosomatics,74, 260–261.
Creed, F., Ratcliffe, J., Fernandes, L., Palmer, S., Rigby, C., Tomenson, B., et al. (2005).
Outcome in severe irritable bowel syndrome with and without accompanying depressive,
panic and neurasthenic disorders. British Journal of Psychiatry, 186, 507–515.
Dale, R. C., Heyman, I., Giovannoni, G., & Church, A. J. (2005). Incidence of anti-brain
antibodies in children with obsessive-compulsive disorder. British Journal of Psychiatry,
187, 314–319.
Davies, S. J. C., Ghahramani, P., Jackson, P. R., Noble, T. W., Hardy, P. G., Hippisley-Cox,
J., et al. (1999). Association of panic disorder and panic attacks with hypertension.
American Journal of Medicine, 107, 310–316.
Engum, A. (2007). The role of depression and anxiety in onset of diabetes in a large popula-
tion-based study. Journal of Psychosomatic Research, 62, 31–38.
150 T. Sala et al.

Engum, A., Bjoro, T., Mykletun, A., & Dahl, A. A. (2002). An association between depres-
sion, anxiety and thyroid function – a clinical fact or an artefact? Acta Psychiatrica
Scandinavica, 106, 27–34.
Feldman, J. M., Lehrer, P. M., Borson, S., Hallstrand, T. S., & Siddique, M. I. (2005). Health
care use and quality of life among patients with asthma and panic disorder. Journal of
Asthma, 42, 179–184.
Fiszman, A., Alves-Leon, S. V., Nunes, R. G., D’Andrea, I., & Figueira, I. (2004). Traumatic
events and posttraumatic stress disorder in patients with psychogenic nonepileptic sei-
zures: A critical review. Epilepsy & Behavior, 5, 818–825.
Frayne, S. M., Seaver, M. R., Loveland, S., Christiansen, C. L., Spiro, A., Parker, V. A., et al.
(2004). Burden of medical illness in women with depression and posttraumatic stress
disorder. Archives of Internal Medicine, 164, 1306–1312.
Friedman, M. J., Wang, S., Jalowiec, J. E., McHugo, G. J., & McDonagh-Coyle, A. (2005).
Thyroid hormone alterations among women with posttraumatic stress disorder due to
childhood sexual abuse. Biological Psychiatry, 57, 1186–92.
Garvey, M. A., Perlmutter, S. J., Allen, A. J., Hamburger, S., Lougee, L., Leonard, H. L.,
et al. (1999). A pilot study of penicillin prophylaxis for neuropsychiatric exacerbations
triggered by streptococcal infections. Biological Psychiatry, 45, 1564–1571.
Godemann, F., Schabowska, A., Naetebusch, B., Heinz, A., & Strohle, A. (2006). The impact
of cognitions on the development of panic and somatoform disorders: A prospective study
in patients with vestibular neuritis. Psychological Medicine, 36, 99–108.
Goodwin, R. D. (2002). Self-reported hay fever and panic attacks in the community. Annals of
Allergy, Asthma & Immunology, 88, 556–559.
Goodwin, R. D., & Davidson, J. R. (2005). Self-reported diabetes and posttraumatic stress
disorder among adults in the community. Preventive Medicine, 40, 570–574.
Goodwin, R. D., & Eaton, W. W. (2003). Asthma and the risk of panic attacks among adults
in the community. Psychological Medicine, 33, 879–885.
Goodwin, R. D., Fergusson, D. M., & Horwood, L. J. (2004). Asthma and depressive and
anxiety disorders among young persons in the community. Psychological Medicine, 34,
Goodwin, R. D., Jacobi, F., & Thefeld, W. (2003). Mental disorders and asthma in the
community. Archives of General Psychiatry, 60, 1125–1130.
Goodwin, R. D., Olfson, M., Shea, S., Lantigua, R. A., Carrasquilo, O., Gameroff, M. J.,
et al. (2003). Asthma and mental disorders in primary care. General Hospital Psychiatry,
25, 479–483.
Goodwin, R. D., & Pine, D. S. (2002). Respiratory disease and panic attacks among adults in
the United States. Chest, 122, 645–650.
Goodwin, R. D., Pine, D. S., & Hoven, C. W. (2003). Asthma and panic attacks among youth
in the community. Journal of Asthma, 40, 139–145.
Goodwin, R. D., & Stein, M. B. (2002). Generalized anxiety disorder and peptic ulcer disease
among adults in the United States. Psychosomatic Medicine, 64, 862–866.
Green, B. L., Rowland, J. H., Krupnick, J. L., Epstein, S. A., Stockton, P., Stern, N. M., et al.
(1998). Prevalence of posttraumatic stress disorder in women with breast cancer. Psycho-
somatics, 39, 102–111.
Grigsby, A. B., Anderson, R. J., Freedland, K. E., Clouse, R. E., & Lustman, P. J. (2002).
Prevalence of anxiety in adults with diabetes: A systematic review. Journal of Psychoso-
matic Research, 53, 1053–1060.
Hasler, G., Gergen, P. J., Kleinbaum, D. G., Ajdacic, V., Gamma, A., Eich, D., et al. (2005).
Asthma and panic in young adults: A 20-year prospective community study. American
Journal of Respiratory and Critical Care Medicine, 171, 1224–1230.
Hermanns, N., Kulzer, B., Krichbaum, M., Kubiak, T., & Haak, T. (2005). Affective and
anxiety disorders in a German sample of diabetic patients: Prevalence, comorbidity and
risk factors. Diabetic Medicine, 22, 293–300.
Anxiety Disorders and Physical Illness Comorbidity: An Overview 151

Hochstrasser, B., Angst, J. (1996). The Zurich Study: XXII. Epidemiology of gastrointestinal
complaints and comorbidity with anxiety and depression. Eur Arch Psychiatry Clin
Neurosci, 246, 261–272.
Hoge, C. W., Terhakopian, A., Castro, C. A., Messer, S. C., & Engel, C. C. (2007). Associa-
tion of posttraumatic stress disorder with somatic symptoms, health care visits, and
absenteeism among Iraq war veterans. American Journal of Psychiatry, 164, 150–153.
Honda, K., & Goodwin, R. D. (2004). Cancer and mental disorders in a national community
sample: Findings from the National Comorbidity Survey. Psychotherapy and Psychoso-
matics, 73, 235–242.
Huffman, J. C., & Pollack, M. H. (2003). Predicting panic disorder among patients with chest
pain: an analysis of the literature. Psychosomatics, 44, 222–236.
Jacob, R. G., Furman, J. M., Durrant, J. D., & Turner, S. M. (1996). Panic, agoraphobia, and
vestibular dysfunction. American Journal of Psychiatry, 153, 503–512.
Jacobsen, P. B., Sadler, I. J., Booth-Jones, M., Soety, E., Weitzner M. A., & Fields, K. K.
(2002). Predictors of posttraumatic stress disorrder symptomatology following bone
marrow transplantation for cancer. Journal of Consulting and Clinical Psychology, 70,
Jacobsen, P. B., Widows, M. R., Hann, D. M., Andrykowski, M. A., Kronish, L. E., & Fields,
K. K. (1998). Posttraumatic stress disorder symptoms after bone marrow transplantation
for breast cancer. Psychosomatic Medicine, 60, 366–371.
Karajgi, B., Rifkin, A., Doddi, S., & Kolli, R. (1990). The prevalence of anxiety disorders in
patients with chronic obstructive pulmonary disease. American Journal of Psychiatry, 147,
Kawachi, I., Colditz, G. A., Ascherio, A., Rimm, E. B., Giovannucci, E., Stampfer, M. J.,
et al. (1994). Prospective study of phobic anxiety and risk of coronary heart disease in men.
Circulation, 89, 1992–1997.
Kawachi, I., Sparrow, D., Vokonas, P.S., & Weiss, T. (1995). Decreased heart rate variability
in men with phobic anxiety (data from the Normative Aging Study). American Journal of
Cardiology, 75, 882–885.
Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E.
(2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in
the National Comorbidity Survey Replication. Archives of General Psychiatry, 62,
Kessler, R. C., Chiu, W. T., Demler, O., Merikangas, K. R., & Walters, E. E. (2005).
Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National
Comorbidity Survey Replication. Archives of General Psychiatry, 62, 617–627.
Kovacs, M., Goldston, D., Obrosky, D. S., & Bonar, L. K. (1997). Psychiatric disorders in
youths with IDDM: Rates and risk factors. Diabetes Care, 20, 36–44.
Kovalenko, P. A., Hoven, C. W., Wu, P., Wicks, J., Mandell, D. J., & Tiet, Q. (2001).
Association between allergy and anxiety disorders in youth. Australian and New Zealand
Journal of Psychiatry, 35, 815–821.
Kruse, J., Schmitz, N., & Thefeld, W. (2003). On the association between diabetes and mental
disorders in a community sample: Results from the German National Health Interview
and Examination Survey. Diabetes Care, 26, 1841–1846.
Kubzansky, L. D., Koenen, K. C., Spiro, A., Vokonas, P. S., & Sparrow, D. (2007). Pro-
spective study of posttraumatic stress disorder symptoms and coronary heart disease in the
Normative Aging Study. Archives of General Psychiatry, 64, 109–116.
Kumano, H., Kaiya, H., Yoshiuchi, K., Yamanaka, G., Sasaki, T., & Kuboki, T. (2004).
Comorbidity of irritable bowel syndrome, panic disorder, and agoraphobia in a Japanese
representative sample. American Journal of Gastroenterology, 99, 370–376.
Lindal, E., Thorlacius, S., Steinsson, K., & Stefansson, J. G. (1995). Psychiatric disorders
among subjects with systemic lupus erythematosus in an unselected population. Scandi-
navian Journal of Rheumatology, 24, 346–351.
152 T. Sala et al.

Luo, F., Leckman, J. F., Katsovich, L., Findley, D., Grantz, H., Tucker, D. M., et al. (2004).
Prospective longitudinal study of children with tic disorders and/or obsessive-compulsive
disorder: Relationship of symptom exacerbations to newly acquired streptococcal infec-
tions. Pediatrics, 113, 578–585.
Lydiard, R. B., Greenwald, S., Weissman, M. M., Johnson, J., Drossman, D. A., &
Bellenger, J. C. (1994). Panic disorder and gastrointestinal symptoms: Findings from the
NIMH Epidemiologic Catchment Area project. American Journal of Psychiatry, 151,
Lydiard, R. B. (2001). Irritable bowel syndrome, anxiety, and depression: What are the links?
Journal of Clinical Psychiatry, 62, 38–45.
Mason, J., Southwick, S., Yehuda, R., Wang, S., Riney, S., Bremner, D., et al. (2004).
Elevation of serum free triiodothyronine, total triiodothyronine, thyroxine-binding glo-
bulin, and total thyroxine levels in combat-related posttraumatic stress disorder. Archives
of General Psychiatry, 51, 629–641.
McWilliams, L. A., Cox, B. J., Enns, M. W. (2003). Mood and anxiety disorders associated
with chronic pain: an examination in a nationally representative sample. Pain, 106,
McWilliams, L. A., Goodwin, R. D., & Cox, B. J. (2004). Depression and anxiety associated
with three pain conditions: results from a nationally representative sample. Pain, 111, 77–83.
Mehnert, A., & Koch, U. (2007). Prevalence of acute and posttraumatic stress disorder and
comorbid mental disorders in breast cancer patients during primary cancer care: a pro-
spective study. Psychooncology, 16, 181–188.
Mell, L. K., Davis, R. L., & Owens, D. (2005). Association between streptococcal infection and
obsessive-compulsive disorder, Tourette’s syndrome, and tic disorder. Pediatrics, 116, 56–60.
Murphy, M. L., & Pichichero, M. E. (2002). Prospective identification and treatment of children
with pediatric autoimmune Neuropsychiatric disorder associated with group A streptococ-
cal infection (PANDAS). Archives of Pediatrics Adolescent Medicine, 156, 356–361.
Murphy, T. K., Sajid, M., Soto, O., Shapira, N., Edge, P., Yang, M., et al. (2004). Detecting
pediatric autoimmune Neuropsychiatric disorders associated with streptococcus in chil-
dren with obsessive-compulsive disorder and tics. Biological Psychiatry, 55, 61–68.
Ortega, A. N., McQuaid, E. L., Canino, G., Goodwin, R. D., & Fritz, G. K. (2004).
Comorbidity of asthma and anxiety and depression in Puerto Rican children. Psychoso-
matics, 45, 93–99.
Patten, S. B., Williams, J. V., Esposito, E., & Beck, C. A. (2006). Self-reported thyroid disease
and mental disorder prevalence in the general population. General Hospital Psychiatry, 28,
Perlmutter, S. J., Leitman, S. F., Garvey, M. A., Hamburger, S., Feldman, E., Leonard, H. L.,
et al. (1999). Therapeutic plasma exchange and intravenous immunoglobulin for obses-
sive-compulsive disorder and tic disorders in childhood. Lancet, 354, 1153–1158.
Prange, A. J. (1999). Editorial comment: Thyroid axis sustaining hypothesis of posttraumatic
stress disorder. Psychosomatic Medicine, 61, 139–140.
Richardson, J. D., Elhai, J. D., & Pedlar, D. J. (2006). Association of PTSD and depression
with medical and specialist care utilization in modern peacekeeping veterans in Canada
with health-related disabilities. Journal of Clinical Psychiatry, 67, 1240–1245.
Rogers, M. P., White, K., Warshaw, M. G., Yonkers, K. A., Rodriguez-Villa, F., Chang, G.,
et al. (1994). Prevalence of medical illness in patients with anxiety disorders. International
Journal of Psychiatry in Medicine, 24, 83–96.
Roy-Byrne, P., Stein, M. B., Russo, J., Craske, M., Katon, W., Sullivan, G., et al. (2005).
Medical illness and response to treatment in primary care panic disorder. General Hospital
Psychiatry, 27, 237–243.
Sareen, J., Cox, B. J., Clara, I., & Asmundson, G. J. G. (2005). The relationship between
anxiety disorders and physical disorders in the U.S. National Comorbidity Survey.
Depression and Anxiety, 21, 193–202.
Anxiety Disorders and Physical Illness Comorbidity: An Overview 153

Sareen, J., Jacobi, F., Cox, B. J., Belik, S., Clara, I., & Stein, M. B. (2006). Disability and poor
quality of life associated with comorbid anxiety disorders and physical conditions.
Archives of Internal Medicine, 166, 2109–2116.
Sawchuk, C. N., Roy-Byrne, P., Goldberg, J., Manson, S., Noonan, C., Beals, J., et al. (2005).
The relationship between posttraumatic stress disorder, depression and cardiovascular
disease in an American Indian tribe. Psychological Medicine, 35, 1785–1794.
Schnurr, P. P., Spiro, A., & Paris, A. (2000). Physician-diagnosed medical disorders in relation
to PTSD symptoms in older male military veterans. Health Psychology, 19, 91–97.
Seng, J. S., Graham-Bermann, S. A., Clark, M. K., McCarthy, A. M., & Ronis, D. L. (2005).
Posttraumatic stress disorder and physical comorbidity among female children and ado-
lescents: Results from service-use data. Pediatrics, 116, 767–776.
Shaban, M. C., Fosbury, J., Kerr, D., & Cavan, D. A. (2006). The prevalence of depression
and anxiety in adults with Type 1 diabetes. Diabetic Medicine, 23, 1381–1384.
Sharp, T. J., & Harvey, A. G. (2001). Chronic pain and posttraumatic stress disorder: mutual
maintenance? Clinical Psychology Review, 21, 857–77.
Shemesh, E., Koren-Michowitz, M., Yehuda, R., Milo-Cotter, O., Murdock, E., Vered, Z.,
et al. (2006). Symptoms of posttraumatic stress disorder in patients who have had a
myocardial infarction. Psychosomatics, 47, 231–239.
Shemesh, E., Rudnick, A., Kaluski, E., Milovanov, O., Salah, A., Alon, D., et al. (2001). A
prospective study of posttraumatic stress symptoms and nonadherence in survivors of a
myocardial infarction (MI). General Hospital Psychiatry, 23, 215–222.
Shemesh, E., Yehuda, R., Milo, O., Dinur, I., Rudnick, A., Vered, Z., et al. (2004). Posttrau-
matic stress, nonadherence, and adverse outcome in survivors of a myocardial infarction.
Psychosomatic Medicine, 66, 521–526.
Simon, G., & Von Korff, M. (1991). Somatization and psychiatric disorder in the
NIMH Epidemiologic Catchment Area study. American Journal of Psychiatry, 148,
Simon, G., Von Korff, M., Saunders, K., Miglioretti, D. L., Crane, P. K., van Belle, G., et al.
(2006). Association between obesity and psychiatric disorders in the US adult population.
Archives of General Psychiatry, 63, 824–830.
Simon, N. M., Blacker, D., Korbly, N. B., Sharma, S. G., Worthington, J. J., Otto, M. W.,
et al. (2002). Hypothyroidism and hyperthyroidism in anxiety disorders revisited: new data
and literature review. Journal of Affective Disorders, 69, 209–217.
Singer, H. S., Hong, J. J., Yoon, D. Y., & Williams, P. N. (2005). Serum autoantibodies do not
differentiate PANDAS and Tourette syndrome from controls. Neurology, 65, 1701–1707.
Swedo, S. E., Leonard, H. L., Garvey, M., Mittleman, B., Allen, A. J., Perlmutter, S., et al.
(1998). Pediatric autoimmune Neuropsychiatric disorders associated with streptococcal
infections: clinical description of the first 50 cases. American Journal of Psychiatry, 155,
Thompson, S. A., Duncan, J. S., Smith, S. J. M. (2000). Partial seizures presenting as panic
attacks. British Medical Journal, 321, 1002–1003.
Trief, P. M., Ouimette, P., Wade, M., Shanahan, P., & Weinstock, R. S. (2006). Posttraumatic
stress disorder and diabetes: Co-morbidity and outcomes in a male veterans sample.
Journal of Behavioral Medicine, 29, 411–418.
van’t Spijker, A., Trijsburg, R. W., & Duivenvoorden, H. J. (1997). Psychological Sequelae of
Cancer Diagnosis: A meta-analytical review of 58 studies after 1980. Psychosomatic
Medicine, 59, 280–293.
Von Korff, M., Crane, P., Lane, M., Miglioretti, D. L., Simon, G., Saunders, K., et al. (2005).
Chronic spinal pain and physical-mental comorbidity in the United States: results from the
national comorbidity survey replication. Pain, 113, 331–339.
Wang, S., Mason, J., Southwick, S., Johnson, D., Lubin, H., & Charney, D. (1995). Relation-
ships between thyroid hormones and symptoms in combat-related posttraumatic stress
disorder. Psychosomatic Medicine, 57, 398–402.
154 T. Sala et al.

Wang, S., & Mason, J. (1999). Elevations of serum T3 levels and their association with
symptoms in World War II veterans with combat-related posttraumatic stress disorder:
replication of findings in Vietnam combat veterans. Psychosomatic Medicine, 61, 131–138.
Weisberg, R. B., Bruce, S. E., Machan, J. T., Kessler, R. C., Culpetter, L., & Keller, M. B.
(2002). Nonpsychiatric illness among primary care patients with trauma histories and
posttraumatic stress disorder. Psychiatric Services, 53, 848–854.
Weissman, M. M., Markowitz, J. S., Ouellette, R., Greenwald, S., & Kahn, J. P. (1990). Panic
disorder and cardiovascular/cerebrovascular problems: results from a community survey.
American Journal of Psychiatry, 147, 1504–1508.
Yehuda, R., Kahana, B., Binder-Brynes, K., Southwick, S. M., Mason, J. W., & Giller, E. L.
(1995). Low urinary cortisol excretion in holocaust survivors with posttraumatic stress
disorder. American Journal of Psychiatry, 152, 982–986.
Zandbergen, J., Bright, M., Pols, H., Fernandez, I., de Loof, C., & Griez, E. J. (1991). Higher
lifetime prevalence of respiratory diseases in panic disorder? American Journal of Psychia-
try, 148, 1583–1585.
The Relation Between Puberty and Adolescent
Anxiety: Theory and Evidence

Ellen W. Leen-Feldner, Laura E. Reardon, Chris Hayward, and Rose C. Smith

The Role of Puberty in Adolescent Anxiety: Theory and Evidence

Anxiety disorders are among the most prevalent forms of psychopathology

affecting youth, with recent estimates ranging from 3% to 18% for particular
disorders (Albano, Chorpita, & Barlow, 1996). These rates are alarming, in
light of the fact that anxiety psychopathology negatively impacts functioning
across multiple domains (e.g., McGee & Stanton, 1990), maintains a chronic
course for a significant proportion of youth affected (Orvaschel, Lewinsohn, &
Seeley, 1995), and increases the risk for other types of disorders (Cole, Peeke,
Martin, Truglio, & Seroczynski, 1998; Goodwin & Hamilton, 2002a, 2002b).
The period of adolescence appears to be a particularly ‘‘high-risk’’ epoch in
terms of the onset and intensification of anxiety problems; for instance, panic
attacks (Macaulay & Kleinknecht, 1989; Warren & Zgourides, 1988), social
phobia (Inderbitzen & Hope, 1995; Liebowitz, Gorman, Fyer, & Klein, 1985)
and obsessive-compulsive disorder (Rasmussen & Eisen, 1990) commonly
emerge during this developmental stage. Furthermore, traumatic event exposure
is common among youth, with estimates indicating 16%–47% of youth ages
12–17 years have been exposed to at least one traumatic event (Giaconia,
Reinherz, Silverman, & Stashwick, 1995; Perkonigg, Kessler, Storz, & Wittchen,
2000; Kilpatrick et al., 2000). Overall, adolescence represents a critical period
during which anxiety vulnerability may be transformed into psychopathology.
Efforts to develop sophisticated and comprehensive etiological models of
adolescent anxiety psychopathology must necessarily consider the role of pub-
erty, conceptualized as the most significant milestone of this period (Hayward &
Sanborn, 2002). In addition, a growing body of literature links various aspects of
puberty to clinical problems (Graber, Lewinsohn, Seeley, & Brooks-Gunn, 1997;
Graber, Seeley, Brooks-Gunn, & Lewinsohn, 2004), including depression

Ellen W. Leen-Feldner
University of Arkansas, Department of Psychology, 216 Memorial Hall, Fayetteville, AR
72701, Tel: 479-575-5329, Fax: 479-575-3219

M. J. Zvolensky, J. A. Smits (eds.), Anxiety in Health Behaviors and Physical Illness. 155
Ó Springer 2008
156 E. W. Leen-Feldner et al.

(Angold, Costello, & Worthman, 1998), eating disorders (Killen, et al., 1992),
and substance use problems (Patton et al., 2004).
With this backdrop, researchers have begun to consider the potential role
that puberty and its associated parameters (e.g., timing) may play in the
development and maintenance of anxiety problems among adolescents. Indeed,
a number of empirical articles have been published on this topic in the past
decade; nonetheless, the literature is still in its relative infancy. Accordingly, the
overarching objective of the current chapter is to stimulate additional research
in this important domain; our specific goals are to 1) discuss methodological
and conceptual issues pertinent to the study of puberty, 2) present an overview
of the empirical literature focused on the association between puberty and
anxiety, and 3) provide some conceptually-driven suggestions for future work
in the area.

Operationalizing and Assessing Puberty: Methodological

and Conceptual Issues

Puberty is a period of profound biopsychosocial development; in a relatively

short time-span (i.e., two to four years), youth experience extensive physical
development, including reaching skeletal maturity (i.e., growth spurt), devel-
oping primary and secondary sexual characteristics (e.g., breast and penis
growth), and attaining reproductive capability (Rogol, Roemmich, & Clark,
2002; Sheehy, Gasser, Molinari, & Largo, 1999). Although the experience can
be exciting and relatively positive for many youth, puberty also is characterized
by potentially undesirable bodily events, such as spontaneous erections,
nocturnal emissions, gynecomastia, as well as onset of menses and irregularity
of initial menstrual cycles (Kipke, 1999). Indeed, some youth report feeling
under-prepared for the events of puberty, potentially contributing to a sense of
‘‘unexpectedness’’ or dyscontrol around these somatic events (Forest, Strange,
Oakley, & The RIPPLE Study Team, 2004; Omar, McElderry, & Zakharia,
2003). In addition, puberty is associated with increased susceptibility to nega-
tive affectivity (Brooks-Gunn, Graber, & Paikoff, 1994; Susman, Dorn, &
Chrousos, 1991), sleep deprivation (Carskadon et al., 2002), conflict in
parent-child relationships (Paikoff & Brooks-Gunn, 1991), and enhanced
emotional lability (Buchanan, Eccles, & Becker, 1992; Spear, 2003).
Collectively, these data suggest puberty may be a difficult experience, at least
for some adolescents. Indeed, puberty may be associated with the onset or
exacerbation of clinically meaningful symptomatology among vulnerable
youth (Caspi, & Moffitt, 1991, 1993). As the field endeavors to construct
developmentally sensitive models of the etiology and maintenance of anxiety
psychopathology among youth, it has become apparent that puberty deserves
theoretical and empirical attention. However, operationalizing and assessing
the complex construct of puberty is a challenging task.
Puberty and Anxiety 157

Because puberty is a multifaceted and dynamic process, there is no

‘‘gold standard’’ for its assessment (Hayward, 2003). When considering the
mechanisms by which puberty may confer risk for psychopathology, research-
ers typically highlight three aspects of puberty: pubertal status, timing, and
hormones, all of which are conceptualized to be in dynamic relation with one
another and the larger social context (Graber, Brooks-Gunn, & Warren, 2006).
A brief overview of each, including assessment approaches, is provided in the
next section. For more comprehensive discussions of methodological
approaches to the study of puberty, the reader is referred to reviews by Graber,
Petersen, and Brooks-Gunn (1996) as well as Dorn, Dahl, Woodward, and Biro
(2006). Finally, it is important to note in this context that, despite a significant
correlation between the two variables, puberty is not redundant with chrono-
logical age. For example, age does not necessarily reflect pubertal status
because the age at which the events of puberty occur vary across individuals
(e.g., female breast development typically begins anywhere from ages 8 to 13;
Tanner, 1962). In addition, a number of studies have shown that during early
adolescence, puberty, and not age, is related to clinical problems such as eating
disorders (Killen et al., 1992) and depression (Angold et al., 1998). Therefore, it
is important to consider how age and puberty, both independently and in
concert, alter risk over time for the development of specific clinical outcomes
(Hayward, 2003).
Pubertal Status. Pubertal status refers to an adolescent’s current level of
morphologic development (Dubas, Graber, & Petersen, 1991) and can be
indexed by asking participants to self-report their development in terms of
specific physical indicators (e.g., height, pubic hair growth, skin changes;
Petersen, Crockett, Richards, & Boxer, 1988) or experience of particular
pubertal events such as spermarche (first emission of spermatozoa) or menarche
(first menstrual period or bleeding). One of the most commonly used indices of
pubertal status is the Tanner Staging System (Tanner, 1962; Morris & Udry,
1980), a five-stage classification system based on maturation of secondary
sexual characteristics (i.e., breasts and pubic hair for females; genitalia and
pubic hair for males) ranging from Tanner stage I (immature) to Tanner stage
V (mature). Tanner stage can be self- or parent-reported (by asking respondents
to select from a series of schematics), or directly evaluated by a practitioner. The
Tanner staging system is a standard assessment of pubertal status in the
literature; however, some limitations of the instrument include 1) inconsistent
reliability between self report and practitioner evaluation, 2) concerns about
the generalizability of the Tanner schematics across ethnic groups, and 3) the
degree to which perceptions of status may affect adolescents’ self ratings of
physical development (Dorn, Dahl et al., 2006).
Pubertal Timing. Pubertal timing refers to the timing of pubertal onset relative to
peers (i.e., early, on-time, or late). Timing can be objectively indexed by classifying
participants on the basis of established standards of physical development (e.g.,
Todd standards of skeletal maturity) or comparing participants’ status on concep-
tually-relevant indicators, such as breast/pubic hair development (Wilson et al.,
158 E. W. Leen-Feldner et al.

1994), menarche/spermarche (Kaltiala-Heino, Marttunen, Rantanen, & Rimpela,

2003), or age at peak height velocity (Dubas et al., 1991) with sample- or epide-
miologically-defined norms. The timing of puberty also can be indexed via sub-
jective self-report (e.g., ‘‘do you think your development is (was) any earlier or later
than most other boys/girls your age?’’ Petersen et al., 1988). Although such reports
evidence modest associations with objective measures (r’s range from 0.28 to 0.56;
Dubas et al., 1991), puberty researchers conceptualize adolescents’ interpretation of
pubertal development as reflecting meaningful variance in participants’ affective
response to pubertal events. In other words, tapping into adolescents’ sense of their
experience of puberty, irrespective of the degree to which it maps onto actual
development, is theoretically important, as it is indicative of the broader psycho-
social context in which the youth is developing. As an illustrative example, Michael
and Eccles (2003) found that objectively indexed early timing (i.e., self-report of
onset of menarche) was not associated with psychological distress (e.g., depressive
symptoms; anger problems) among a sample of 266 African American females ages
13–14 years. However, when perceptions of pubertal timing were evaluated,
females who viewed their maturation as early or late relative to on-time peers
evidenced elevated depressive, eating, anger, and problem behavior symptoms.
Maternal report corroborated these adolescent self report data. These findings
suggest that subjective indices tap conceptually-relevant processes that may be
important in terms of risk for psychopathology during puberty (cf., Silbereisen, &
Kracke, 1997). Indeed, assessing perceptions of timing is recommended in studies
where outcome variables are theorized to vary as a function of individual differ-
ences in the experience of puberty (e.g., perceptions of pubertal timing may be more
relevant if self-perception is related to the outcome variable [self-esteem; anxiety;
body image] whereas objective indices are indicated if research questions are
focused on the effect of actual physical development on a particular outcome
[aggression; muscle strength]; Dorn, Dahl et al., 2006).
Pubertal Hormones. The two primary hormonal processes that drive pub-
ertal development are adrenarche and gonadarche (Fechner, 2002). The first
phase of puberty, adrenarche, commences between ages five and nine years in
girls (approximately one year later in boys; Grumbach, & Styne, 2003).
Activation of the adrenal axis results in the release of the adrenal hormones
including cortisol, dehydroepiandrosterone (DHEA), its sulfate (DHEAS), and
androstendione (4-A; Parker, 1999). The second phase of puberty,
gonadarche, commences around age 9–10 years in females and is characterized
by the release of leutinizing hormone (LH) and follicle stimululating hormone
(FSH) by gonadotropes in the anterior pituitary (Johnson & Everitt, 2000).
These hormones stimulate the enlargement of the gonads, which in turn release
testosterone (T) and estrogen (E)/estradiol (E2) that promote the development
of breasts and genitalia. Hormone concentrations typically are measured via
blood serum or salivary sample assay. Salivary samples are less invasive and
typically more cost effective, with professional assay services offered by
well-respected companies such as Salimetrics, Inc. Moreover, there is
evidence of adequate inter-correlations between the two assessment approaches
Puberty and Anxiety 159

(e.g., r =0.91 [cortisol]; r = 0.86 [DHEAS]; Salimetrics, LLC). However,

salivary assay may not be sensitive enough to detect some hormones
(e.g., estradiol; Shirtcliff, Reavis, Overman, & Granger, 2001) and can be
affected by multiple variables, including contraceptive use, teeth brushing,
medication, as well as recent meals and exercise (e.g., Bhagwager, Hafizi, &
Cohen, 2002; Kirschbaum, Kudielka, Gaab, Schommer, & Hellhammer, 1999).
Therefore, blood serum assessments may be preferable, particularly if
researchers have access to the appropriate personnel and testing services
(e.g., affiliations with a medical school). Regardless of the approach however,
rigorous assessment (e.g., multiple assessments within- and across- days;
Goodyer, Park, & Herbert, 2001) is critical to ensure valid hormone indices
because hormones evidence significant fluctuation (e.g., ultradian; circadian;
monthly rhythms; Cauter, 2001).
Puberty as a ‘‘Risk Factor.’’ A necessary first step toward theoretical and
empirical integration of puberty into models of etiology and maintenance of
anxiety psychopathology is to clarify how the various aspects of puberty fit into
contemporary conceptualizations of risk. Specifically, Helena Kraemer and her
colleagues (Kazdin, Kraemer, Kessler, Kupfer, & Offord, 1997; Kraemer, Stice,
Kazdin, Offord, & Kupfer, 2001; Kraemer et al., 1997) have developed a
standardized nomenclature for operationalizing risk factor processes in which
a risk factor is defined as characteristic, experience, or event that indicates the
relative likelihood of a given outcome among individuals within a population
(e.g., if the entire population is exposed to the variable in question, it is not a risk
factor). The authors suggest that a characteristic can be classified as a risk factor
(i.e., a characteristic that temporally precedes, and is correlated with, an
outcome) or a causal risk factor (i.e., a characteristic that temporally precedes
and, when manipulated, produces systematic change in an outcome). Risk
factors can be further categorized according to whether they are malleable; if
the characteristic changes spontaneously (e.g., age) or can be manipulated
(e.g., skill set) is it classified as a variable risk factor, whereas if the characteristic
is non-modifiable (e.g., gender), it is referred to as a fixed marker.
According to the Kraemer model, puberty per se (i.e., an individual’s
pubertal status; hormonal changes characteristic of puberty) is not a risk
factor; all youth are exposed to puberty and as such it does not designate
sub-populations of youth more or less likely to evidence a given outcome.
A more accurate conceptualization of puberty generally is as a ‘‘critical period,’’
reflecting a high-risk phase for youth who are otherwise vulnerable. Indeed, the
unparalleled physical, cognitive, and psychosocial changes associated with
puberty can be stressful for some adolescents, making it an important time
for screening and intervention. Pubertal timing, on the other hand, is a facet of
puberty that reflects individual differences in an otherwise universal human
experience. Accordingly, pubertal timing more readily lends itself to classifica-
tion within the Kraemer model. Specifically, pubertal timing characterizes
specific sub-groups of youth at risk for a number of psychiatric outcomes
(e.g., early-maturing females evidence increased risk of substance use problems;
160 E. W. Leen-Feldner et al.

Dick, Rose, Viken, & Kaprio, 2000). Although the timing of puberty has
changed in the past two decades (earlier onset), given the current state of our
knowledge, altering the timing of puberty is not currently conceptualized as a
reasonable target of intervention. Therefore, pubertal timing is considered
non-malleable and classified as a fixed marker. Importantly, pubertal timing
is non-specific in nature as it has been linked to an array of clinical sequelae,
including problematic health behaviors (e.g., excessive dieting and
exercise [McCabe & Ricciardelli, 2004]; alcohol, cigarette, and marijuana use
[Lanza & Collins, 2002; Wichstrom, 2001; Weisner & Ittel, 2002]) as well as
elevated risk of clinical symptomatology and diagnoses (e.g., depressive-type
problems [Graber, Seeley, Brooks-Gunn, & Lewinsohn, 2004; Kaltiala-Heino,
Kosunen, & Rimpela, 2003); eating-related pathology [Kaltiala-Heino,
Rimpela, Rissanen, & Rantanen, 2001]).
As highlighted by the current volume, the important link between health
behaviors and anxiety is just beginning to be fully appreciated. As these two
outcomes, anxiety and unhealthy behaviors, both increase at puberty, it will
become increasingly important to untangle both pubertal status and pubertal
timing effects associated with health behaviors in addition to unraveling the
potential interaction between onset of anxiety and initiation of unhealthy
behaviors in the peripubertal period.

Overview of Findings from the Empirical Literature –

Differentiating Pubertal Status and Timing Effects

The available empirical literature focused on the puberty-anxiety association is

briefly reviewed in this section, addressing anxiety in relation to each of the
aspects of puberty discussed above. In light of the marked gender differences in
the process and experience of puberty (Hayward, 2003), findings among males
and females are presented separately. Finally, in light of space limitations, more
detailed descriptions are presented for studies that, relative to others in the
literature, are characterized by greater methodological rigor (e.g., sample size;
prospective assessments) thereby allowing for more confidence in the observed
Pubertal Status and Anxiety. In terms of the main effects of pubertal status,
findings are mixed. Among females, for instance, a significant positive
relation was observed in two large samples of non-clinical youth conducted in
the United States (Hayward et al., 1992) and Australia (Patton et al., 1996).
Specifically, Hayward and colleagues found, after controlling for age, a
two-fold increase in the likelihood of having had a panic attack for each
one-point change in self-reported Tanner stage among 754 ethnically diverse
6th and 7th grade females. Interestingly, none of the females at Tanner stages
I or II reported having had a panic attack. Similarly, in a sample of more than
2,000 adolescents, Patton and colleagues reported a positive association
Puberty and Anxiety 161

between psychiatric symptoms (including anxiety) and pubertal status

(self-reported age at menarche) after controlling for chronological age. Indeed,
females for whom menarche had occurred more than 36 months prior to the
study were more than twice as likely to evidence psychiatric morbidity.
Studies also have demonstrated an association between advancing pubertal
status and trait anxiety among females (Huerta & Brizuela-Gamino, 2002)
but others have found either a weak negative association or no relation at
all (Canals,Marti-Henneberg, Fernandez-Ballart, Cliville, & Domenech, 1992;
Laitinen-Krispijn, van der Ende, & Verhulst, 1999; Susman et al., 1991; Stone &
Barker, 1939).
Among males, Ge, Conger, and Elder (2001) reported a positive association
between ‘‘internalized distress’’ (i.e., summary measure of self-reported depres-
sive, anxious, somatization, and eight additional symptoms) and self-reported
pubertal status among 202 adolescent males between the ages of 12 and 14
assessed annually over a four year period. However, the association did not
persist, suggesting this effect may be short-term in nature. Specifically, pubertal
status indexed in grade 7 predicted internalized distress at grades 8 and 10, but
pubertal status assessed at grades 8, 9, and 10 was unrelated to internalized
distress at any other assessment point. Similarly, Susman and colleagues (1991)
found that physician assessed pubertal status related positively to interview-
assessed depressive and anxiety symptoms 12 months later among 56 males
aged 10 to 15 years (age was entered simultaneously into the model). However,
inconsistent with these two studies, other work has found no association
between pubertal status and anxious symptoms (Patton et al., 1996), a decrease
in anxious-depressed symptoms as a function of advancing pubertal stage
(Laitinen-Krispijn et al., 1999), or mixed findings (i.e., elevated state anxiety
at the beginning and end of puberty; Canals et al., 1992).
One reason that the literature is mixed may be that key constructs that
might influence or moderate the pubertal status-anxiety association have
received limited attention. This relative degree of neglect is unfortunate,
particularly because, as discussed above, puberty per se is a normative
developmental process (Rosenfeld & Nicodemus, 2003). Therefore, the char-
acteristic changes of puberty would only be expected to increase risk for
psychopathology under certain conditions (e.g., among psychologically vul-
nerable adolescents; Buchanan et al., 1992). This conceptualization is in line
with the idea that puberty in and of itself is a ‘‘critical period,’’ or a period of
high risk for youth who are otherwise vulnerable. Specifically, the accentua-
tion hypothesis (Caspi & Moffitt, 1991, 1993) postulates that the changes
associated with puberty amplify pre-existing vulnerability, thereby resulting
in psychopathology among at-risk youth. Consistent with this perspective,
panic-relevant (anxiety sensitivity; Leen-Feldner, Reardon, et al., 2006; Leen-
Feldner, Reardon, & Zvolensky, 2007) and biological (seratonergic dysfunc-
tion; Twitchell, Hanna, Cook, Fitzgerald, & Zucker, 2000), individual differ-
ence vulnerability variables have been found to moderate the association
between advancing pubertal status and anxiety outcomes. For instance,
162 E. W. Leen-Feldner et al.

among 123 psychologically healthy adolescents aged 12–17 years, Leen-Feld-

ner and colleagues (2006) found that more mature youth who also were high
in anxiety sensitivity [AS], a cognitive vulnerability factor reflecting a ten-
dency to fear the consequences of anxiety (Reiss & McNally, 1985), reported
the most fearful reactivity in response to a panic-relevant voluntary hyper-
ventilation challenge. That is, advancing pubertal status was associated with
panic-relevant reactivity only among those youth who were vulnerable to
such reactivity by virtue of elevated cognitive vulnerability (i.e., AS). These
effects were significant above and beyond age, gender, and pre-experimental
anxiety. These data highlight the utility of investigating potential moderators
of the pubertal status-anxiety association. Indeed, a number of conceptually
relevant diatheses may be important in terms of enhancing anxiety vulner-
ability during pubertal development, including behavioral inhibition (Hirsh-
feld et al., 1992), negative affectivity (Craske, 2003), anxiety sensitivity
(Schmidt, Zvolensky, & Maner, 2006), and traumatic event exposure
(Caffo, Forresi, & Lievers, 2005). With the exception of AS, no empirical
work has addressed these questions, and thus they represent important future
directions (please see the Suggestions for Future Research section for addi-
tional recommendations).
Pubertal Hormones and Anxiety. Perhaps due to the complexities of hormone
research described earlier, the literature focused on the link between anxiety and
pubertal hormones is quite limited, precluding integrative conclusions. Prior to
reviewing the available literature, it is worth noting that several studies
conducted with adolescents have investigated the general association between
anxiety and specific hormones that are present during puberty (e.g., cortisol;
Gerra et al., 2000). For instance, in a study of 131 adolescents, Terleph, Klein,
and Roberson-Nay (2006) reported a positive association between mean
cortisol levels before and during a biological challenge procedure and panic
attacks elicited by the challenge, suggesting elevated cortisol may represent a
panic-relevant vulnerability variable. Although these findings are theoretically
interesting in their own right, without an index of pubertal status, it is difficult
to extrapolate their relevance in terms of the puberty-anxiety association
because observed associations could be due to other variables that
are confounded with hormone levels, such as the social consequences of
hormonally-driven secondary sex characteristic development (Slap, Khalid,
Paikoff, & Brooks-Gunn, 1994). Therefore, studies reviewed below included
an index of pubertal status.
In a seminal study, Olweus, Mattsson, Schalling, and Low (1980) reported
no association between blood serum testosterone and both trait and state
anxiety elicited by the blood draw among 58 Swedish males aged 15 to
17 years. In a more fine-grained analysis of the role of testosterone, Granger
and colleagues (2003) found, in a mixed gender study of over 200 youth aged
11–17 years, a negative association between testosterone and concurrent
anxious/depressed symptomatology. In addition, anxious/depressed symptoms
were related to diurnal variation such that those males who exhibited less of a
Puberty and Anxiety 163

decrease in testosterone across the course of the day evidenced elevated

symptomatology. No relation was observed among females. Nottelmann and
colleagues (1987) measured LH, FSH, DHEA, DHEAS, T, E2 4-A, and
internalizing-type symptoms among 108 youth aged 9 to 14 years. The only
associations observed were among males; a positive correlation between 4-A
and an index of obsessive-compulsive symptoms and a negative correlation
between ‘‘emotional tone’’ (combined sadness/anxiety ratings) and a T/E2 ratio.
Finally, in a one-year prospective study of more than 100 youth (ages–years),
Susman and colleagues (1991) measured cortisol in addition to the hormones
indexed in the Nottelmann et al. study. Among females, cortisol and T relate-
positively to concurrent anxious symptoms and emotional tone, respectively.
DHEAS related negatively to concurrent maternal report of internalizing
symptoms. Prospective analyses with females indicated LH (positive associa-
tion) and DHEAS (negative association) predict anxiety symptoms assessed
one year later. Among males, T and E2 (negative associations) as well as 4-A
(positive association) related concurrently to emotional tone. In addition, LH
and cortisol related negatively to maternal report of internalizing symptoms
12 months later; DHEA and 4-A also prospectively predicted anxious symp-
toms (positive associations).
As noted above, it seems premature to draw integrative conclusions about
the role of hormones in potentiating anxiety, although there is some evidence
that T and 4-A are related to anxious symptomatology among males; reliable
associations among females have not been observed across studies. There are
multiple empirical questions to address in this domain (please see the
Suggestions for Future Research section). From a conceptual perspective, it is
useful to highlight the fact that hormones represent a ‘‘piece of the puzzle’’ as
they exhibit reciprocal effects on other hormones, behavior, and pubertal
development (Nelson, 2000). Therefore, the role of pubertal hormones in
anxiety development will likely be understood only if studied within the context
of a larger, biopsychosocial model (Brooks-Gunn et al., 1994; Booth, Johnson,
Granger, Crouter, & McHale, 2003).
Pubertal Timing and Anxiety. Although the literature is not uniform, com-
pared to pubertal status and hormones, pubertal timing evidences the most
consistent and robust associations with anxiety psychopathology, particularly
among females. Specifically, early-maturing females report elevations in
anxious symptoms (Dorn, Hitt, & Rotenstein, 1999; Graber et al., 2006;
Kaltiala-Heino et al., 2003; Silbereisen & Kracke, 1997; Sonis et al., 1985),
psychological distress (including anxiety symptoms; Ge, Conger, & Elder, 1996;
Hayward et al., 1997), panic attacks (Hayward et al., 1997) and anxiety
disorders when prospectively assessed (Graber et al., 2004). As an illustrative
example, in one of the most comprehensive studies completed to date,
Graber and colleagues examined perceptions of pubertal timing in relation
to psychiatric symptoms and interview assessed diagnoses among an
epidemiologically-defined mixed gender sample of 1,709 adolescents aged
14–18 years. Concurrent analyses (adjusted for chronological age) indicated
164 E. W. Leen-Feldner et al.

early maturing females evidenced elevated depressive symptoms as well as

disruptive behavior, eating, and major depressive disorders (but not anxiety
disorders). These associations were significant (with the exception of eating
disorders) at the follow-up assessment (approximately 10 years after the initial
evaluation). In addition, there appeared to be a ‘‘sleeper effect’’ for anxiety;
compared to their on-time counterparts, early maturing females were almost
twice as likely to meet criteria for an anxiety disorder (lifetime). There is no
evidence that late maturation is associated with anxiety among females.
Among males, the pattern is somewhat less consistent; studies have found
early maturation is associated with increased anxiety symptoms (Ge, Brody,
Conger, & Simons, 2006), state anxiety (i.e., a single question on which parti-
cipants self-reported the ease with which they became anxious or distressed;
Kaltiala-Heino et al., 2003), observations of manifest anxiety (i.e., trained
interviewers’ ratings of participants’ anxiety (e.g., tendency to worry) based
on direct observation and interviews with family and teachers, taken yearly
between the ages of 5 and 16; Peskin, 1967), and ‘‘internalized distress’’ (i.e., a
self-report of depression, anxiety, somatization, and other complaints such as
poor appetite, trouble sleeping, and overeating, as measured on the SCL-90-R;
Ge et al., 2001). In contrast, other reports suggest late maturation relates
positively to self-reported internalizing-type symptoms (including anxiety;
Graber et al., 1997) and observer rated social anxiety (Jones & Bayley, 1971).
In the studies conducted by Graber and colleagues discussed above, the follow
up study indicated that the positive association between late maturation and
internalizing symptoms does not persist among males and, in contrast to
females, pubertal timing does not prospectively predict anxiety diagnoses
(although late maturation was associated with current substance use and
lifetime disruptive behavior diagnoses at the follow-up assessment).
There are a number of individual and contextual factors that appear to
influence the association between pubertal timing and anxiety. For instance,
significant interactions have been observed between early maturation and life
stress (Ge et al., 2001), as well as early maturation and pre-pubertal internaliz-
ing symptoms (Hayward et al., 1997) in the prediction of anxious symptoma-
tology. Specifically, early maturing youth who also reported elevated life stress
and internalizing symptoms, relative to all other variable combinations,
evidenced the most anxiety. In a more fine-grained analysis, Ge and colleagues
(2001) examined pubertal timing as indexed by age at menarche in a four-year
longitudinal study of 216 females (7th through 10th grade). These authors
found evidence for the idea that early maturation exacerbates vulnerability to
psychological distress and that this process is influenced by a number of
variables. Specifically, among early maturing females only (cf., on-time
and late developers), early psychological distress relates positively to later
psychological distress, suggesting early maturation increases susceptibility to
continued emotional problems. In addition, early maturing females were more
likely to associate with deviant peers, which in turn predicted later psychologi-
cal distress. Finally, peer group composition moderated the timing-distress
Puberty and Anxiety 165

association such that, compared to on-time and late developers, early maturing
females who also were associated with mixed (as opposed to same-sex) peer
groups in the 7th through 9th grades were more likely to evidence psychological
distress in the 10th grade. This study highlights the complex inter-relations that
characterize the association between pubertal timing and anxiety, and further-
more suggests that negative effects of off-time maturation may not be transient,
but rather persist over time.
Overall, pubertal timing appears to be an important fixed marker in the
development and maintenance of anxiety-type symptoms, particularly among
females. Interestingly, not only does the timing of puberty alter psychiatric risk
but psychological factors may alter the timing of puberty (reviewed in more
detail below). Thus, in theory, pubertal timing is malleable. However, from an
intervention standpoint, understanding how pubertal timing affects outcomes
may be a more feasible goal than attempting to alter the timing of puberty itself.
Therefore, at this stage of research development, we consider pubertal timing to
be non-malleable (i.e., fixed marker). In addition, as noted earlier, the associa-
tion between pubertal timing and psychiatric outcomes is non-specific in
nature. For both these reasons (non-malleability; non-specificity) identification
of mediators and moderators of the pubertal timing-anxiety association
is critical. In particular, it will be important for researchers to consider
gender-specific characteristics that may mediate or moderate the timing-anxiety
association, as the link between timing and anxiety appears to differ across
males and females. For instance, sexual assault exposure (Hayward & Sanborn,
2002) and body image and weight concerns (Richards, Boxer, Petersen, &
Albrecht, 1990) may be particularly important among females, whereas
risk-taking activities (e.g., substance use; Andersson & Magnusson, 1990)
may be important to consider among males.
Temporal Dimensions of the Pubertal Timing-Anxiety Association. As
mentioned above, a complicating facet of the putative link between pubertal
timing and anxiety pertains to the chicken-and-the-egg issue. Specifically, two
pathways are viable. First, the timing of maturation may potentiate anxiety for
some youth (e.g., early maturing females develop anxiety-related problems;
Graber et al., 2004). However, it has also been posited that the direction of
the association may be from contextual ‘‘stress’’ in the early environment to
pubertal timing. Both neurobiological (e.g., changes to the HPA-axis secondary
to childhood stress; Meschke, Johnson, Barber, & Eccles, 2003) and
sociobiological (Belsky, Steinberg, & Draper, 1991) mechanisms have been
theorized. For instance, Belsky and colleagues highlight the utility, from an
evolutionary perspective, of commencing reproductive activities as soon as
possible if an organism is reared in a stressful environment where the viability
of progeny is questionable. Therefore, children reared in a stressful environ-
ment are predicted to experience the hormonal and physical changes of puberty
earlier than youth from stable home environments. There is a relatively large
body of work examining the role of early stress in relation to pubertal timing.
For instance, among females, dysfunctional family background (e.g., parental
166 E. W. Leen-Feldner et al.

divorce; alcoholism; Hulanika, 1999) and childhood sexual abuse (Herman-

Giddens, Sandler, & Friedman, 1988) have been found to predict earlier
menarche. Similarly, familial conflict in the childhood environment has been
shown to predict early spermarche (Kim & Smith, 1998). Conflicting findings
also have been observed, however, with some studies suggesting childhood
stress predicts later maturation (e.g., Malo & Tremblay, 1997) and others
reporting no association at all (see Kim, Smith, & Palermiti, 1997, for a review).
With regard to anxiety symptoms specifically, the literature is similarly
mixed, with some studies reporting that anxious symptomatology in
childhood is associated with earlier menarche (Kim & Smith, 1998) and
spermarche (Kim & Smith, 1999), others suggesting it is linked with later
spermarche (Malo & Tremblay, 1997), and still others indicating it is unrelated
to maturational timing (Graber, Brooks-Gunn, & Warren, 1995). In addition,
at least one study calls attention to the importance of the type of anxiety being
assessed. Specifically, Meschke and colleagues (2003) found, among males,
economic and school anxieties predicted earlier and later puberty, respectively.
These studies represent a useful first step toward specifying the role of early-life
anxiety in affecting pubertal onset. However, the mixed findings, due at least in
part to methodological issues (e.g., reliance on retrospective recall) make it
difficult to draw definitive conclusions about issues of temporal order.
Nevertheless, these data suggest that interventions designed to reduce anxiety
early in life might also ‘‘secondarily’’ alter the timing of puberty. This is a critical
area for future work.
Summary of the Empirical Literature Focused on the Puberty-Anxiety
Association. Although the available research is promising, there is a great deal
of work to be done to fully understand the association between puberty and
anxiety. Much of the early work was characterized by significant methodolo-
gical limitations, including small sample size, assessment of only one aspect
of puberty, one-time hormone assessments, analyses with males and females
combined, and utilization of non-specific measures of anxiety. Indeed, in
drawing conclusions from the extant work, greater weight should be placed
on those studies (e.g., Ge et al., 1996, 2001; Graber et al., 1997, 2004; Hayward
et al., 1992, 1997) where some of these limitations are addressed. Accordingly,
three tentative conclusions can be drawn at this stage of research development.
First, anxiety symptomatology appears to increase as youth traverse puberty;
the more methodologically rigorous studies suggest a positive association
between pubertal status and anxiety that cannot be explained by chronological
age (e.g., Patton et al., 1996). From a conceptual perspective, however, it makes
sense to continue exploring mechanisms that may explain this association, as
puberty itself is not considered to be pathological in its own right (Graber,
2003); examination of individual-difference variables that might be expected to
influence the association is an exemplar of an important next step in this
literature. Second, the hormone literature is too under-developed to draw any
firm conclusions at this point; additional work is needed to cull out the
main, additive, and interactive effects of pubertal hormones on anxiety
Puberty and Anxiety 167

symptomatology. Finally, pubertal timing, particularly early maturation

among females, appears to be a fixed marker for anxiety (e.g., Graber, 2003);
there is less compelling evidence for the effects of early or late maturation on
anxiety among males. As with pubertal status, our understanding of the effects
of off-time maturation will be most comprehensive when other variables, such
as the role of concomitant life stress, are considered. In addition, there is at least
some evidence indicating that early-life anxiety predicts early maturation,
suggesting the association between anxiety and pubertal timing is complex
and likely bi-directional, at least for some youth. Despite the overarching
promise of the emerging literature in this domain, the direction, magnitude,
and nature of the puberty-anxiety association have yet to be fully explicated.
Building on the suggestions for future research proposed in previous sections,
the next section offers additional ideas within the context of the overarching
limitations of the extant work.

Moving this Literature Forward: Conceptually-Driven Suggestions

for Future Research

We have three specific suggestions for future research, aimed at increasing the
methodological and conceptual sophistication of the literature. First, there is a
general need for more comprehensive and integrated assessments of puberty.
For instance, it will be important to replicate and extend findings from the
hormone literature suggesting a main effect of specific hormones on anxious
symptomatology (Granger et al., 2003). Here, methodologically innovative
approaches that involve multiple assessments across the course of puberty
will be critical for addressing questions related to the short-term (e.g., rapidly
escalating hormones; Apter, 1980) and long-term effects of hormonal changes
during puberty. In addition, because hormones affect each other in complex
ways (Johnson & Everitt, 2000) it will be important for researchers to obtain
indices of several different hormones and examine their independent and com-
bined effects on anxiety. Although expansion of the hormone-anxiety literature
would fill an important empirical gap, perhaps a more pressing challenge is to
evaluate the combined (additive and interactive) effects of pubertal status,
timing, and hormones on anxiety, as well as the impact of childhood anxiety
on pubertal maturation. An integrated approach to pubertal assessment is
critical here, as these aspects of puberty are fundamentally inter-related
(Brooks-Gunn et al., 1994). As an illustrative example, consider breast devel-
opment among females. This process is driven by gonadal hormones, which
may have an independent effect on anxious reactivity (Susman et al., 1991).
However, personal and social (e.g., parents; peers) reactions to a female’s
status on this relatively visible pubertal event may also contribute to anxiety
(Brooks-Gunn, 1984), particularly if she is ‘‘off-time’’ with respect to her
contemporaries (Brooks-Gunn & Warren, 1985). Of course, some of the
168 E. W. Leen-Feldner et al.

greatest yield in terms of improving our conceptual understanding of the

puberty-anxiety association will come from prospective assessments of the
inter-relations between the two across the course of puberty, beginning, ideally,
as young as five or six years so that the early stages of puberty are studied
(e.g., awakening of the adrenal glands; Dorn, Dahl et al., 2006) and questions of
the temporal association can begin to be addressed. In addition, advanced
statistical techniques, aimed at delineating the associations among childhood
anxiety, early puberty, and subsequent anxiety, are needed. Overall,
multi-modal, integrated, and longitudinal assessments of puberty will be neces-
sary to cultivate a comprehensive understanding the puberty-anxiety
Related to this first point, the literature would also benefit from more
fine-grained assessments of anxiety. For instance, with some important
exceptions (e.g., Graber et al., 1997, 2004; Hayward et al., 1992, 1997), the
majority of available studies focus on associations among pubertal indices and
non-specific anxiety-relevant outcomes (e.g., anxiety as a part of a larger
internalizing construct). Although this approach yields useful information in
terms of the role of puberty in terms of general risk trajectories (e.g., internaliz-
ing versus externalizing), it does not speak to the aspects of puberty that may
potentiate fearful reactivity to distinct classes of stimuli (e.g., social; bodily;
mental; Craske, 2003) or the development of specific types of anxiety
psychopathology (e.g., social anxiety disorder; post-traumatic stress disorder).
In addition to the conceptual utility of such tests, data pertinent to the role of
puberty in the development of specific anxiety disorders would be useful for the
development of tailored preventative interventions (e.g., Mrazek & Haggerty,
1994; Zvolensky, Schmidt, Bernstein, & Keough, 2006). To this end, investiga-
tors could explore pubertal processes in the context of specific anxiety
outcomes, including symptom checklists that correspond to specific anxiety
disorders (e.g., generalized anxiety disorder subscale of the Screener for Child
Anxiety and Related Disorders; Birmaher et al., 1997) and diagnostic status
across the anxiety disorders as indexed via structured clinical interview. These
data would be extraordinarily useful in clarifying the nature and magnitude of
the puberty-anxiety association and would furthermore position researchers to
examine whether and how pubertal processes affect anxious symptomatology
across disorders (e.g., is puberty more likely to potentiate social anxiety as
compared to obsessive compulsive disorder?). As a complement to this
approach, there may be significant merit in integrating laboratory-based assess-
ments of anxiety processes (e.g., are aspects of puberty related to changes in
social anxiety as indexed via distress elicited by a laboratory-based social
evaluative task?; Ferrell, Beidel, & Turner, 2004). Laboratory-based methodol-
ogy has the advantage of allowing for the evaluation of ‘‘real-time’’ emotional
responding, thereby delimiting recall biases other types of reporting errors
common to affective processing (Zvolensky, Lejuez, Stuart, & Curtin, 2001).
Finally, the extant literature can be described as largely descriptive in nature,
focusing on whether a relation exists between puberty and anxiety. Building on
Puberty and Anxiety 169

this foundation will require systematic testing of the putative ‘‘mechanisms of

action’’ underlying the link between puberty and anxiety. The broad-based
questions that need to be addressed include 1) precisely how does puberty
contribute to anxiety psychopathology (and vice versa)? 2) what individual
difference and social-contextual variables increase the explanatory power of
puberty-anxiety models? and 3) what types of ‘‘process variables’’ mediate the
puberty-anxiety association? Given the multi-faceted nature of both puberty
and anxiety, there are multiple systematic research programs that will converge
to address these questions. Consistent with our suggestion to develop the
literature on the role of puberty in the development and maintenance of specific
anxiety-relevant outcomes, we will utilize panic disorder (PD) as the outcome
of interest to exemplify this issue. However, because pubertal timing is a
non-specific fixed marker, it is our hope that investigators with expertise in
other types of anxiety psychopathology (e.g., post-traumatic stress disorder;
social anxiety disorder) will extrapolate ideas relevant to their own programs of
research from the current discussion.
Panic disorder is a common, severe, and chronic condition associated with
substance use disorders, poor physical health outcomes, social/occupational
disability, and high rates of healthcare utilization (APA, 2000; Greenberg et al.,
1999). Importantly, adolescence is a critical developmental period during which
panic problems first manifest, making it an important stage to study in terms of
panic-related problems (e.g., Ollendick, Mattis, & King, 1994; Von Korff,
Eaton, & Keyl, 1985). The hallmark characteristic of PD is fearful reactivity
to bodily sensations, which is posited to result, at least in part, from interocep-
tive learning (Craske, 2003). Specifically, the repeated pairing of somatic cues
and personal threat/anxiety results in bodily arousal becoming a phobic
stimulus (Bouton, Mineka, & Barlow, 2001). Puberty may be a ‘‘critical period’’
for such interoceptive learning. Specifically, puberty is replete with bodily
oriented changes, some of which, as discussed earlier, may be conceptualized
as unpredictable, uncontrollable, and undesirable bodily events (e.g., sponta-
neous erection; Omar et al., 2003). Importantly, an array of studies using
diverse methodological procedures have demonstrated that uncontrollable
and unpredictable aversive events are associated with heightened anxiety states
and bodily arousal (Maier & Seligman, 1976; Mineka & Kihlstrom, 1978;
Sanderson, Rapee, & Barlow, 1989; Sapolsky, 1990; Sapolsky, Alberts, &
Altman, 1997; Zvolensky, Eifert, & Lejuez, 2001). As youth traverse puberty,
they are increasingly exposed to somatic events that may result in
panic-relevant interoceptive learning. This hypothesis predicts that puberty,
reflective of adolescents’ relative degree of exposure to such learning trials,
will be associated with elevated panic symptomatology. However, because
puberty per se is not a malleable risk factor (all youth experience puberty, but
not all youth develop clinical problems [Graber, 2003]), investigators are tasked
with identifying other variables that may explain the puberty-panic association.
To identify risk factors that are malleable and therefore subject to interven-
tion it is necessary to identify the specific mediators and moderators of the
170 E. W. Leen-Feldner et al.

puberty panic association. Specifically, identification of mediators/moderators

will 1) elucidate the causal chain that links pubertal effects to panic outcomes,
2) point researchers toward modifiable factors that will aid in the prevention of
panic, and 3) increase model specificity (e.g., anxiety-relevant mediators and
moderators such as fear of negative evaluation may improve the prediction of
anxiety problems, whereas depression-relevant mediators/moderators such as
dysfunctional attitudes will likely aid in the prediction of depressive outcomes).
Discussions of some candidate mediators and moderators of the puberty-panic
association are provided below.
Mediators temporally follow and are correlated with a risk characteristic (in
this case, pubertal effects), and attenuate the association between the risk
characteristic and the outcome (i.e., anxiety; Kraemer et al., 2001). One poten-
tial mediator may be the fear-relevant somatic learning that occurs during the
course of puberty (see Fig. 1).This hypothesis requires an assessment of the
posited mediating process. To this end, it would be helpful to design a
psychometrically sound self-report index of fear-relevant experiences with
puberty-related bodily sensations (e.g., ‘‘I was scared when I first started having
menstrual cramps’’). With a valid and reliable index, researchers would be well
positioned to address questions such as whether fear-relevant learning changes
across the course of puberty and whether such learning accounts for observed
associations between puberty and panic. Another class of conceptually inter-
esting mediators, given their effects on neurotransmitter systems hypothesized
to be involved in mood regulation (McEwen, 1991), are the gonadal steroids
associated with puberty. For instance, estrogen modulates the serotonin
neurotransmitter system in a number of ways, including effects on serotonin
synthesis and receptor density (Biegon et al., 1990). Animal research suggests
puberty ushers in changes in the serotonin system reaction to stress, particularly
among females. For instance, following separation, post-pubertal female rhesus
monkeys, compared to males, evidence higher levels of 5-Hydroxyindoleacetic
acid (5-HIAA), a primary metabolite of serotonin and an index of the
organism’s level of serotonin (Higley, Suomi, & Linnoila, 1990). This difference
is not apparent prior to puberty, suggesting gonadal steroids may potentiate the
stress response among pubertal females (Sanborn & Hayward, 2003).


Puberty Panic-Related Problems

(e.g., fear-relevant
learning; gonadal

Fig. 1 Mediators of the puberty-panic association

Puberty and Anxiety 171

It also is important to identify potential moderators of the puberty- panic

association. These variables precede and are uncorrelated with aspects of
puberty (Kraemer et al., 2001). An important potential individual-difference
factor that could moderate the puberty panic association is anxiety sensitivity
(AS), or the fear of anxiety and anxiety-related sensations (Reiss & McNally,
1985), which has been theoretically (Barlow, 2002) and empirically (Hayward,
Killen, Kraemer, & Taylor, 2000) supported as a key cognitive vulnerability
factor involved in the development of anxious responding to bodily sensations –
a hallmark characteristic of panic attacks. From a conceptual perspective, AS
might be expected to amplify the panic-relevant learning that occurs during
puberty. That is, compared to low AS youth, adolescents with elevated AS may
perceive the bodily events that occur during puberty (e.g., early breast
development; menstrual discomfort/accidents; autonomic arousal secondary
to hormonal shifts) as more threatening. Across the course of puberty,
vulnerable youth (i.e., high AS youth) may more readily learn that physical
sensations are aversive and anxiety-relevant, possibly resulting in a panic
attack. In contrast, adolescents low in AS may be less susceptible to
panic-relevant interoceptive learning during puberty because they are less
fearful of bodily sensations (see Fig. 2).
In addition to individual-difference characteristics, there are a number of
panic-relevant social/contextual moderators that would be theoretically
interesting to investigate. For instance, there is emerging evidence for the role
of parenting behavior in increasing the ‘‘threat value’’ of bodily sensations via
direct (e.g., informational transmission), or indirect (e.g., modeling) reinforce-
ment of fearful responding to somatic cues (Craske & Rowe, 1997; Ehlers, 1993;
Muris, Merckelbach, & Meesters, 2001; Watt & Stewart, 2000). Thus, a parent

(e. g., youth HIGH in
anxiety sensitivity)

Presence of Panic-
Puberty Related Problems

Absence of Panic-
Related Problems

(e. g., youth LOW in
anxiety sensitivity)

Fig. 2 Moderators of the puberty-panic association

172 E. W. Leen-Feldner et al.

who communicates fear in the context of bodily arousal (e.g., dizziness

secondary to a childhood fever) may inadvertently facilitate the child’s pairing
of fear and interoceptive experiences. Subsequently, the bodily perturbation
characteristic of puberty (particularly when such experiences are off-time com-
pared to peers) may be experienced as panic-relevant. In this example, the
moderating effect of specific parenting behaviors would be expected to increase
the likelihood of panic-relevant learning during puberty. Another conceptually-
relevant social-contextual moderator is pre-pubertal traumatic event exposure,
such as childhood sexual abuse. A number of studies suggest a link between
significant stress and changes (e.g., hypersecretion of corticotropin-releasing
hormone) to the hypothalamic-pituitary-adrenal (HPA) axis (e.g., Miller,
Chen, & Zhou, 2007). Such changes are thought to directly increase
vulnerability to psychopathology; this outcome may be more likely among
females because of the role estrogen plays in enhancing the HPA stress
response. When this vulnerability is combined with the normative HPA axis
changes that occur during puberty, Sanborn and Hayward (2003) suggest that
impairments in HPA functioning may reach a threshold, thereby putting
trauma-exposed, pubertal females at particular risk for anxiety psychopathol-
ogy (including panic).
Together, although the discussion was focused on panic-related problems,
the third general suggestion is for researchers to complement the available
descriptive research and work toward identifying key mediators and
moderators of the puberty-anxiety association. Understanding the processes
that influence the association between puberty and anxiety will be critical from
a public health perspective. For instance, if research bears out the prediction
that panic-relevant learning mediates the association between puberty and
panic, high risk youth (e.g., early maturing females) could be identified in a
primary care setting and presented with a brief intervention focused on reducing
fear of bodily sensations and encouraging interoceptive exposure to attenuate
the effects of subsequent panic-relevant leaning trials (e.g., latent inhibition;
Bouton et al., 2001). However, at this stage of research development, the key
tasks relate to clarifying the nature of the puberty-anxiety association. To this
end, the role of individual-difference and social-contextual factors, as well
as variables that may mediate the puberty-anxiety relation need to be
systematically and comprehensively addressed, irrespective of what type of
anxiety serves as the outcome variable.


The period of adolescence is characterized by both puberty and an increase in

anxiety-related psychopathology, prompting researchers to examine the
relation between the two. The available literature suggests there is a meaningful
association between puberty and anxiety; for some youth (particularly females),
Puberty and Anxiety 173

the process of puberty appears to enhance anxiety-related symptoms, although

there are a number of important gaps in the research base on which researchers
could focus their investigative efforts, including improved methodological
approaches and more sophisticated and comprehensive conceptual models.
Nonetheless, this is a promising area of research; we hope the suggestions
presented here will stimulate what we conceptualize as important future work
aimed at clarifying the association between puberty and anxiety among youth.


Albano, A. M., Chorpita, B. F., & Barlow, D. H. (1996). Childhood anxiety disorders. In
E. J.Mash & R. A. Barkley (Eds.), Child Psychopathology (pp. 196–241). New York,
NY: Guilford Press.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental
disorders (4th ed. – Text Revision). Washington, DC: Author.
Andersson, T., & Magnusson, D. (1990). Biological maturation in adolescence and the
development of drinking habits and alcohol abuse among young males: A prospective
longitudinal study. Journal of Youth and Adolescence, 19, 33–41.
Angold, A., Costello, E. J., & Worthman, C. M. (1998). Puberty and depression: The roles of
age, pubertal status, and pubertal timing. Psychological Medicine, 28, 661–666.
Apter, D. (1980). Serum steroids and pituitary hormone in female puberty: A partly
longitudinal study. Clinical Endocrinology, 12, 107–120.
Barlow, D. H. (2002). Anxiety and its disorders: The nature and treatment of anxiety and panic
(2nd ed.). New York: The Guilford Press.
Belsky, J., Steinberg, L., & Draper, P. (1991). Childhood experience, interpersonal
development, and reproductive strategy: An evolutionary theory of socialization. Child
Development, 62, 647–670.
Bhagwager, Z., Hafizi, S., & Cohen, P. J. (2002). Acute citalopram administration produces
correlated increase in plasma and salivary cortisol. Psychopharmacology, 163, 119–120.
Biegon, A., Grinspoon, A., Blumenfeld, B., Bleich, A., Apter, A., & Mester, R. (1990).
Increased serotonin 5-HT2 receptor binding on blood platelets of suicidal men.
Psychopharmacology, 100, 165–167.
Birmaher, B., Khetarpal, S., Brent, D., Cully, M., Balach, L., Kaufman, J., et al. (1997). The
Screen for Child Anxiety-Related Disorders (SCARED): Scale construction and
psychometric characteristics. Journal of the American Academy of Child and Adolescent
Psychiatry, 36, 545–553.
Booth, A., Johnson, D. R., Granger, D. A., Crouter, A. C., & McHale, S. (2003). Testosterone
and child and adolescent adjustment: The moderating role of parent-child relationships.
Developmental Psychology, 39, 85–98.
Bouton, M. E., Mineka, S., & Barlow, D. H. (2001). A modern learning theory perspective on
the etiology of panic disorder. Psychological Review, 108, 4–32.
Brooks-Gunn, J. (1984). The psychological significance of different pubertal events to young
girls. Journal of Early Adolescence, 4, 315–327.
Brooks-Gunn J., & Warren, M. (1985). The effects of delayed menarche in different contexts;
dance and non-dance students. Journal of Youth and Adolescence, 14, 285–300.
Brooks-Gunn, J., Graber, J., & Paikoff, R. L. (1994). Studying links between hormones and
negative affect: Models and measures. Journal of Research on Adolescence, 4, 469–486.
Buchanan, C. M., Eccles, J. S., & Becker, J. B. (1992). Are adolescents the victims of raging
hormones? Evidence for activational effects of hormones on moods and behaviors in
adolescence. Psychological Bulletin, 111, 62–107.
174 E. W. Leen-Feldner et al.

Caffo, E., Forresi, B., & Lievers, L. S. (2005). Impact, psychological sequelae and manage-
ment of trauma affecting children and adolescents. Current Opinion in Psychiatry, 18,
Canals, J., Marti-Henneberg, C., Fernandez-Ballart, J., Cliville, R., & Domenech, E. (1992).
Scores on the state-trait anxiety inventory for children in a longitudinal study of pubertal
Spanish youth. Psychological Reports, 71, 503–512.
Carskadon, M. A., Harvey, K., Duke, P., Anders, T. F., Litt, I. F., & Dement, W. C. (2002).
Pubertal changes in daytime sleepiness. Sleep: Journal of Sleep and Sleep Disorders
Research, 25, 525–605.
Caspi, A., & Moffitt, T. E. (1991). Individual differences are accentuated during periods of
social change: The sample case of girls at puberty. Journal of Personality and Social
Psychology, 61, 157–168.
Caspi, A., & Moffitt, T. E. (1993). When do individual differences matter? A paradoxical
theory of personality coherence. Psychological Inquiry, 4, 247–271.
Cauter, E. V. (2001). Endocrine rhythms. In K. L. Becker (Ed.), Principles and practice of
endocrinology and metabolism (3rd ed.). Philadelphia: Lippincott Williams and Wilkins.
Cole, D. A., Peeke, L. G., Martin, J. M., Truglio, R., & Seroczynski, A. D. (1998).
A longitudinal look at the relation between depression and anxiety in children and
adolescents. Journal of Consulting and Clinical Psychology, 66, 451–460.
Craske, M. G. (2003). Origins of phobias and anxiety disorders: Why more women than men?
New York, NY: Elsevier.
Craske, M. G., & Rowe, M. K. (1997). Nocturnal panic. Clinical Psychology: Science &
Practice, 4, 153–174.
Dick, D. M., Rose, R. J., Viken, R. J., & Kaprio, J. (2000). Pubertal timing and substance
use: Associations between and within families across late adolescence. Development
Psychology, 36, 180–189.
Dorn, L. D., Dahl, R. E., Woodward, H. R., & Biro, F. (2006). Defining the boundaries of
early adolescence: A user’s guide to assessing pubertal status and pubertal timing in
research with adolescents. Applied Developmental Science, 10, 30–56.
Dorn, L. D., Hitt, S. F., & Rotenstein, D. (1999). Biopsychological and cognitive differences
in children with premature vs. on-time adrenarche. Archives of Pediatric and Adolescent
Medicine, 153, 137–146.
Dubas, J. S., Graber, J. A., & Petersen, A. C. (1991). A longitudinal investigation of adoles-
cents’ changing perceptions of pubertal timing. Developmental Psychology, 27, 580–586.
Ehlers, A. (1993). Somatic symptoms and panic attacks: A retrospective study of learning
experiences. Behaviour Research and Therapy, 31, 269–278.
Fechner, P. Y. (2002). Gender differences at puberty. Journal of Adolescent Health, 30, 44–48.
Ferrell, C., Beidel, D. & Turner, S. (2004). Assessment and treatment of socially phobic
children: A cross-cultural comparison. Journal of Clinical Child and Adolescent Psychol-
ogy, 31, 69–79.
Forest, S. Strange, V. Oakley, A., & The RIPPLE Study Team. (2004). What do young people
want from sex education? The results of a needs assessment from a peer-led sex education
programme. Culture, Health, & Sexuality, 6, 337–354.
Ge, X., Brody, G. H., Conger, R. D., & Simons, R. L. (2006). Pubertal maturation and
African American children’s internalizing and externalizing symptoms. Journal of Youth
and Adolescence, 35, 531–540.
Ge, X., Conger, R. D., & Elder, G. H. (1996). Coming of age too early: Pubertal influences on
girls’ vulnerability to psychological distress. Child Development, 67, 3386–3400.
Ge, X., Conger, R. D., & Elder, G. H. (2001). The relation between puberty and psychological
distress in adolescent boys. Journal of Research on Adolescence, 11, 49–70.
Gerra, G., Zaimovic, A., Zambelli, U., Timpano, M., Reali, N., Bernasconi, S., et al. (2000).
Neuroendocrine responses to psychological stress in adolescents with anxiety disorder.
Neuropsychobiology, 42, 82–92.
Puberty and Anxiety 175

Giaconia, R. M., Reinherz, H. Z., Silverman, A. B., & Stashwick, C. K. (1995). Traumas and
posttraumatic stress disorder in a community population of older adolescents. Journal of
the American Academy of Child & Adolescent Psychiatry, 34, 1369–1380.
Goodwin, R. D., & Hamilton, S. P. (2002a). Early-onset fearful panic attack: A possible
prodrome of early-onset severe psychopathology. Comprehensive Psychiatry, 42, 22–27.
Goodwin, R. D., & Hamilton S. P. (2002b). The early-onset fearful panic attack as a predictor
of severe psychopathology. Psychiatry Research, 109, 71–79.
Goodyer, I. M., Park, R. J., & Herbert, J. (2001). Psychosocial and endocrine features of
chronic first-episode major depression in 8–16 year olds. Biological Psychiatry, 50,
Graber, J. A. (2003). Puberty in context. In C. Hayward’s (Ed.), Gender differences at puberty
(pp. 307–325). New York: Cambridge University Press.
Graber, J., A., Brooks-Gunn, J., & Warren, M. P. (1995). The antecedents of menarcheal age:
Heredity, family environment, and stressful life events. Child Development, 66, 346–359.
Graber, J. A., Brooks-Gunn, J., & Warren, M. P. (2006). Pubertal effects on adjustment in
girls: Moving from demonstrating effects to identifying pathways. Journal of Youth and
Adolescence, 35, 413–423.
Graber, J. A., Lewinsohn, P. M., Seeley, J. R., & Brooks-Gunn, J. (1997). Is psychopathology
associated with the timing of pubertal development? Journal of the American Academy of
Child and Adolescent Psychiatry, 36, 1768–1776.
Graber, J. A., Petersen, A., & Brooks-Gunn, J. (1996). Pubertal processes: Methods, mea-
sures, and models. In J. A. Graber, J. Brooks-Gunn, & A. Petersen (Eds.), Transitions
Through Adolescence: Interpersonal Domains and Context (pp. 23–53). Mahwah, NJ:
Lawrence Erlbaum Associated.
Graber, J. A., Seeley, J. R., Brooks-Gunn, J., & Lewinsohn, P. M. (2004). Is pubertal timing
associated with psychopathology in young adulthood? Joural of the American Academy of
Child and Adolescent Psychiatry, 43, 718–726.
Granger, D. A., Shirtcliff, E. A., Zahn-Waxler, C., Usher, B., Klimes-Dougan, B., &
Hastings, P. (2003). Salivary testosterone diurnal variation and psychopathology in ado-
lescent males and females: Individual differences and developmental effects. Development
and Psychopathology, 15, 431–449.
Greenberg, P. E., Sisitsky, T., Kessler, R. C., Finkelstein, S. N., Berndt, E. R., Davidson, J. R. T.,
et al. (1999). The economic burden of anxiety disorders in the 1990s. Journal of Clinical
Psychiatry, 3, 1–9.
Grumbach, M. M., & Styne, D. M. (2003). Puberty: Ontogeny, neuroendocrinology, and
disorders. In P. R. Larsen, H. M. Kronenberg, S. Melmed, & K. S. Polonsky (Eds.),
Williams textbook of endocrinology (10th ed., pp. 1115–1286). New York: Elsevier.
Hayward, C. (2003). Gender differences at puberty. New York: Cambridge University Press.
Hayward, C., & Sanborn, K. (2002). Puberty and the emergence of gender differences in
psychopathology. Journal of Adolescent Health, 30S, 49–58.
Hayward, C., Killen, J. D., Hammer, L. D., Litt., I. F., Wilson, D. M., Simmonds, B. et al.
(1992). Pubertal stage and panic attack history in sixth- and seventh-grade girls. American
Journal of Psychiatry, 149, 1239–1243.
Hayward, C., Killen, J. D., Kraemer, H. C., & Taylor, C. B. (2000). Predictors of panic attacks
in adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 39,
Hayward, C., Killen, J. D., Kraemer, H. C., Blair-Greimer, A., Strachowski, D., Cuning, D.,
et al. (1997). Assessment and phenomenology of nonclinical panic attacks in adolescent
girls. Journal of Anxiety Disorders, 11, 17–32.
Herman-Giddens, M. E., Sandler, A. D., & Friedman, N. E. (1988). Sexual precocity in girls:
An association with sexual abuse? American Journal of Diseases of Children, 142, 431–433.
Higley, J. D., Suomi, S. J., & Linnoila, M. (1990). Developmental influences on the serotonergic
system and timidity in the nonhuman primate. In E. F. Coccaro & D. L. Murphy (Eds.),
176 E. W. Leen-Feldner et al.

Serotonin in major psychiatric disorders (pp. 29–46). Washington, DC: American Psychiatric
Hirshfeld, D. R., Rosenbaum, J. F., Biederman, J., Bolduc, E. A., Faraone, S. V., Snideman, N.,
et al. (1992). Stable behavioral inhibition and its association with anxiety disorder. Journal of
the American Academy of Child and Adolescent Psychiatry, 31, 103–111.
Huerta, R., & Brizuela-Gamino, O. L. (2002). Interaction of pubertal status, mood, and
self-esteem in adolescent girls. Journal of Reproductive Medicine, 47, 217–225.
Hulanika, B. (1999). Acceleration of menarcheal age of girls from dysfunctional families.
Journal of Reproductive and Infant Psychology, 17, 119–132.
Inderbitzen, H. M., & Hope, D. (1995). Relationship among adolescent reports of social
anxiety, anxiety, and depressive symptoms. Journal of Anxiety Disorders, 9, 385–396.
Johnson, M., & Everitt, B. (2000). Essential Reproduction (5th ed.). Oxford, England:
Blackwell Scientific.
Jones, M. C., & Bayley, N. (1971). Physical maturing among boys as related to behavior. In
M. C. Jones, N. Bayley, J. W. MacFarlane, & M. P. Honzik (Eds.), The Course of human
development (pp. 252–257). Waltham, MA: Xerox College Publishing.
Kaltiala-Heino, R., Kosunen, E., & Rimpela, M. (2003). Pubertal timing, sexual behaviour
and self-reported depression in middle adolescence. Journal of Adolescence, 26, 531–545.
Kaltiala-Heino, R., Marttunen, M., Rantanen, P., & Rimpela, M. (2003). Early puberty is
associated with mental health problems in middle adolescence. Social Science & Medicine,
57, 1055–1064.
Kaltiala-Heino, R., Rimpela, M., Rissanen, A., & Rantanen, P. (2001). Early puberty and
early sexual activity are associated with bulimic-type eating pathology in middle
adolescence. Journal of Adolescent Health, 28, 346–352.
Kazdin, A. E., Kraemer, H. C., Kessler, R. C., Kupfer, D. J., & Offord, D. R. (1997).
Contributions of risk-factor research to developmental psychopathology. Clinical
Psychology Review, 17, 375–406.
Killen, J. C., Hayward, C., Litt, I. F., Hammer, L. D., Wilson, D. M., Milner, B., et al. (1992).
Is puberty a risk factor for eating disorders? American Journal of Childhood Disorders, 146,
Kilpatrick, D. G., Acierno, R., Saunders, B., Resnick, H. S., Best, C. L., & Schnurr, P. P.
(2000). Risk factors for adolescent substance abuse and dependence: Data from a national
sample. Journal of Consulting and Clinical Psychology, 68, 19–30.
Kim, K., & Smith, P. K. (1998). Childhood stress, behavioural symptoms, and
mother-daughter pubertal development. Journal of Adolescence, 21, 231–240.
Kim, K., & Smith, P. K. (1999). Family relations in early childhood and reproductive
development. Journal of Reproductive and Infant Psychology, 17, 133–148.
Kim, K., Smith, P. K., & Palermiti, A. L. (1997). Conflict in childhood and reproductive
development. Evolution and Human Behavior, 18, 109–142.
Kipke, M. (1999). Adolescent development and the biology of puberty: Summary of a workshop
on new research: Forum on adolescence (Compass Series). Washington, DC: National
Academy Press.
Kirschbaum, C., Kudielka, B. M., Gaab, J., Schommer, N. C., & Hellhammer, D. H. (1999).
Impact of gender, menstrual cycle phase, and oral contraceptives on activity of the
hypothalamic-pituitary-adrenal axis. Psychosomatic Medicine, 61, 154–162.
Kraemer, H. C., Kazdin, A., Offord, D., Kessler, R. C., Jensen, P. S., & Kupfer. D. J. (1997).
Coming to terms with the terms of risk. Archives of General Psychiatry, 54, 337–343.
Kraemer, H. C., Stice, E., Kazdin, A., Offord, D., & Kupfer, D. (2001). How do risk factors
work together? Mediators, moderators, and independent, overlapping, and proxy risk
factors. American Journal of Psychiatry, 158, 848–856.
Laitinen-Krispijn, S., van der Ende, J., & Verhulst, F. C. (1999). The role of pubertal progress
in the development of depression in early adolescence. Journal of Affective Disorders, 54,
Puberty and Anxiety 177

Lanza, S. T., & Collins, L. M. (2002). Pubertal timing and the onset of substance use in
females during early adolescence. Prevention Science, 3, 69–82.
Leen-Feldner, E. W., Reardon, L. E., McKee, L. G., Feldner, M. T., Babson, K, A., &
Zvolensky, M. J. (2006). The interactive role of anxiety sensitivity and pubertal status in
predicting anxious responding to bodily sensations among adolescents. Journal of
Abnormal Child Psychology, 34, 797–810.
Leen-Feldner, E. W., Reardon, L. E., & Zvolensky, M. J. (2007). Pubertal status and
emotional reactivity to a voluntary hyperventilation challenge predicting panic symptoms
and somatic complaints: A laboratory-based multi-informant test. Behavior Modification,
31, 8–31.
Liebowitz, M. R., Gorman, J. M., Fyer, A. J., & Klein, D. F. (1985). Social phobia: Review of
a neglected anxiety disorder. Archives of General Psychiatry, 42, 729–736.
Macaulay, J. L., & Kleinknecht, R. A. (1989). Panic and panic attacks in adolescents. Journal
of Anxiety Disorders, 3, 221–241.
Maier, S. F., & Seligman, M. E. P. (1976). Learned helplessness: Theory and evidence. Journal
of Comparative and Physiological Psychology, 88, 554–564.
Malo, J., & Tremblay, R. E. (1997). The impact of parental alcoholism and maternal social
position on boys’ school adjustment, pubertal maturation and sexual behavior: A test of
two competing hypotheses. Journal of Child Psychology and Psychiatry, 38, 187–197.
McCabe, M. P., & Ricciardelli, L. A. (2004). A longitudinal study of pubertal timing and
extreme body change behaviors among adolescent boys and girls. Adolescence, 39, 145–166.
McEwen, B. S. (1991). Non-genomic and genomic effects of steroids on neural activity. Trends
in Pharmacological Science, 12, 141–147.
McGee, R., & Stanton, W. R. (1990). Parent reports of disability among 13-year olds with DSM-
III disorders. Journal of Child psychology & Psychiatry and Allied Disciplines, 31, 793–801.
Meschke, L. Johnson, P. J., Barber B., & Eccles J. (2003). Psychosocial factors predicting
pubertal onset. In C. Hayward (Ed.), Gender differences at puberty (pp. 217–238).
Cambridge, UK: Cambridge University Press.
Michael, A., & Eccles, J. S. (2003). When coming of age means coming undone: Links between
puberty and psychosocial adjustment among European American and African American
girls. In C. Hayward (Ed.), Gender differences at puberty (pp. 277–303). Cambridge, UK:
Cambridge University Press.
Mineka, S., & Kihlstrom, J. (1978). Unpredictable and uncontrollable aversive events. Journal
of Abnormal Psychology, 87, 256–271.
Miller, G. E., Chen, E., & Zhou, E. S. (2007). If it goes up, must it come down? Chronic stress
and the hypothalamic-pituitary-adrenocortical axis in humans. Psychological Bulletin,
133, 25–45.
Morris, N. M., & Udry, J. R. (1980). Validation of a self-administered instrument to assess
stage of adolescent development. Journal of Youth and Adolescence, 9, 271–280.
Mrazek, P. G., & Haggerty, R. J. (1994). Reducing risk for mental disorders: Frontiers for
preventive intervention research. Washington, DC: National Academy Press.
Muris, P., Merckelbach, H., & Meesters, C. (2001). Learning experiences and anxiety sensitivity
in normal adolescents. Journal of Psychopathology and Behavioral Assessment, 23, 279–283.
Nelson, R. J. (2000). An introduction to behavioral endocrinology. New York: Sinaur.
Nottelmann, E. D., Susman, E. J., Inoff-Germain, B. A., Cutler, G., B., Loriaux, D. L., &
Chrousos, G. P. (1987). Developmental processes in early adolescence: Relationships
between adolescent adjustment problems and chronologic age, pubertal stage, and
puberty-related serum hormone levels. Adolescent Medicine, 110, 473–480.
Ollendick, T. H., Mattis, S. G., & King, N. J. (1994). Panic in children and adolescents:
A review. Journal of Child Psychology and Psychiatry, 35, 113–134.
Olweus, D., Mattsson, A., Schalling, D., & Low, H. (1980). Testosterone, aggression,
physical, and personality dimensions in normal adolescent males. Psychosomatic
Medicine, 42, 253–269.
178 E. W. Leen-Feldner et al.

Omar, H., McElderry, D., & Zakharia, R. (2003). Educating adolescents about puberty: What
are we missing? International Journal of Adolescent Health and Medicine, 15, 79–83.
Orvaschel, H., Lewinsohn, P. M., & Seeley, J. R. (1995). Continuity of psychopathology in a
community sample of adolescents. Journal of the American Academy of Child & Adolescent
Psychiatry, 34, 1525–1535.
Paikoff, R. L., & Brooks-Gunn, J. (1991). Do parent-child relationships change during
puberty? Psychological Bulletin, 110, 47–66.
Parker, C. R. (1999). Dehydroepiandrosterone and dehydroepiandrosterone sulfate produc-
tion in the human adrenal gland during development and aging. Steroids, 64, 640–647.
Patton, G. C., Hibbert, M. E., Carlin, J., Shao, Q., Rosier, M., Caust, J., et al. (1996).
Menarche and the onset of depression and anxiety in Victoria, Australia. Journal of
Epidemiology and Community Health, 50, 661–666.
Patton, G. C., McMorris, B. J., Toumbourou, J. W., Hemphill, S. A., Donath, S., &
Catalano, R. F. (2004). Puberty and onset of substance use and abuse. Pediatrics, 114,
Perkonigg, A., Kessler, R. C., Storz, S., & Wittchen, H. (2000). Traumatic events and
post-traumatic stress disorder in the community: Prevalence, risk factors, and
comorbidity. Acta Psychiatrica Scandinavica, 101, 46–59.
Peskin, H. (1967). Pubertal onset and ego functioning. Journal of Abnormal Psychology, 72, 1–15.
Petersen, A. C., Crockett, L., Richards, M., & Boxer, A. (1988). A self-report measure of
pubertal status: reliability, validity, and initial norms. Journal of Youth and Adolescence,
17, 117–133.
Rasmussen, S. A., & Eisen, J. L. (1990). Epidemiology of obsessive compulsive disorder.
Journal of Clinical Psychiatry, 53, 3–10.
Reiss, S., & McNally, R. J. (1985). Expectancy model of fear. In S. Reiss & R. R. Bootzin
(Eds.), Theoretical issues in behavior therapy (pp.107–121). San Diego: Academic Press.
Richards, M. H., Boxer, A.W., Petersen, A. C., & Albrecht, R. (1990). Relation of weight to
body image in pubertal girls and boys from two communities. Developmental Psychology,
26, 313–321.
Rogol, A. D., Roemmich, J. N., & Clark, P. A. (2002). Growth at puberty. Journal of
Adolescent Health, 31, 192–200.
Rosenfeld, R. G., & Nicodemus, B. C. (2003). The transition from adolescence to adult life:
Physiology of the ‘transition’ phase and its evolutionary basis. Hormone Research, 60,
Sanborn, K., & Hayward, C. (2003). Hormonal changes at puberty and the emergence of
gender differences in internalizing disorders. In C. Hayward (Ed.), Gender differences at
puberty (pp. 307–325). New York: Cambridge University Press.
Sanderson, W., Rapee, R., & Barlow, D. (1989). The influence of an illusion of control on
panic attacks induced via inhalation of 5.5% carbon dioxide-enriched air. Archives of
General Psychiatry, 46, 157–162.
Sapolsky, R. M. (1990). Stress in the wild. Scientific American, 262, 116–123.
Sapolsky, R. M., Alberts, S. C., & Altman, J. (1997). Hypercortisolism associated with social
subordinance or social isolation among wild baboons. Archives of General Psychiatry, 54,
Schmidt, N. B., Zvolensky, M. J., & Maner, J. K. (2006). Anxiety sensitivity: Prospective
prediction of panic attacks and Axis I pathology. Journal of Psychiatric Research, 40,
Sheehy, A., Gasser, T., Molinari, L., & Largo, R. H. (1999). An analysis of variance of the
pubertal and mid-growth spurts for length and width. Annals of Human Biology, 26,
Shirtcliff, E. A., Reavis, R., Overman, W. H., & Granger, D. A. (2001). Measurement of
gonadal hormones in dried blood spots versus serum: Verification of menstrual cycle
phase. Hormones and Behavior, 39, 258–266.
Puberty and Anxiety 179

Silbereisen, R. K., & Kracke, B. (1997). Self-reported maturational timing and adaptation in
adolescence. In J. Schulenberg, J. L. Maggs, & K. Hurrelmann (Eds.), Health risks and
developmental transitions during adolescence (pp. 85–109). New York, NY, US: Cambridge
University Press.
Slap, G. B., Khalid, N., Paikoff, R., & Brooks-Gunn, J. (1994). Evolving self-image, pubertal
manifestations, and pubertal hormones: Preliminary findings in young adolescent girls.
Journal of Adolescent Health, 15, 327–335.
Sonis, W., Comite, F., Blue, J., Pescovitz, O. H., Rahn, C. W., Hench, K. D., et al. (1985).
Behavior problems and social competence in girls with true precocious puberty. Journal of
Pediatrics, 106, 156–160.
Spear, L. P. (2003). Neurodevelopment during adolescence. In D. Cicchetti & E. Walker (Eds.),
Neurodevelopmental mechanisms in psychopathology (pp. 62–83). New York: Cambridge
University Press.
Stone, C. P., & Barker, R. G. (1939). The attitudes and interests of premenarcheal and
postmenarcheal girls. Journal of Genetic Psychology, 54, 27–71.
Susman, E. J., Dorn, L. D., & Chrousos, G. P. (1991) Negative affect and hormone levels in
young adolescents: Concurrent and predictive perspectives. Journal of Youth and
Adolescence, 20, 167–190.
Tanner, J. M. (1962). Growth at adolescence. Thomas: Springfield, IL.
Terleph, T. T., Klein, R. G., & Roberson-Nay, R. (2006). Stress responsivity and HPA axis
activity in juveniles: Results from a home-based CO2 inhalation study. American Journal of
Psychiatry, 163, 738–740.
Twitchell, G. R., Hanna, G. L., Cook, E. H., Fitzgerald, H. E., & Zucker, R. A. (2000).
Serotonergic function, behavioral disinhibition, and negative affect in children of
alcoholics: The moderating effects of puberty. Alcoholism, 24, 972–979.
Von Korff, M. R., Eaton, W. W., & Keyl, P. M. (1985). The epidemiology of panic attacks
and panic disorder. American Journal of Epidemiology, 122, 970–981.
Warren, R., & Zgourides, G. (1998). Panic attacks in high school students: Implications for
prevention and intervention. Phobia Practice and Research Journal, 1, 97–113.
Watt, M. C., & Stewart, S. H. (2000). Anxiety sensitivity mediates the relationships between
childhood learning experiences and elevated hypochondriacal concerns in young adult-
hood. Journal of Psychosomatic Research, 49, 107–118.
Weisner, M., & Ittel, A. (2002). Relations of pubertal timing and depressive symptoms to
substance use in early adolescence. Journal of Early Adolescence, 22, 5–23.
Wichstrom, L. (2001). The impact of pubertal timing on adolescents’ alcohol use. Journal of
Research on Adolescence, 11, 131–150.
Wilson, D. M., Killen, J. D., Hayward, C., Robinson, T. N., Hammer, L. D., Kraemer, H. C.,
et al. (1994). Timing and rate of sexual maturation and the onset of cigarette and alcohol
use among teenage girls. Archives of Pediatric and Adolescent Medicine, 148, 789–795.
Zvolensky, M. J., Eifert, G. H., & Lejuez, C.W. (2001). Offset control during recurrent 20%
carbon dioxide-enriched air induction: Relation to individual difference variables.
Emotion, 1(2), 148–165.
Zvolensky, M. J., Lejuez, C. W., Stuart, G. L., & Curtin, J. J. (2001). Experimental
psychopathology in psychological science. Review of General Psychology, 5, 371–381.
Zvolensky, M. J., Schmidt, N. B., Bernstein, A., & Keough, M. E. (2006). Risk factor research
and prevention programs for anxiety disorders: A translational research framework.
Behaviour Research and Therapy, 44, 1219–1239.
Anxiety, Anxiety Disorders,
and the Menstrual Cycle

Sandra T. Sigmon and Janell G. Schartel

Fluctuations in mood, bodily sensations, and behavior have long been

associated with the premenstrual phase of the menstrual cycle. Research has
found that women with psychological disorders often experience symptom
exacerbation and psychiatric admission rates are increased during the
premenstrual phase (e.g., Hendrick, Altshuler, & Burt, 1996). There are a few
studies focusing on the follicular or intermenstrual phase of the menstrual cycle
but the majority has focused on the premenstrual phase. This chapter covers the
empirical research on the relationship between anxiety, anxiety disorders and
the premenstrual phase of the menstrual cycle; however it is not an exhaustive
review of the literature on menstrual cycle effects. There is a significant
premenstrual literature on depression and other mood states; however this
chapter focuses on anxiety.
The complex relationship between anxiety and menstruation is presented by
(1) identifying the continuum of experience of premenstrual anxiety, (2) addres-
sing issues in assessment and treatment, (3) reviewing the literature on the
premenstrual experience of normal and clinical samples of women, (4) discussing
current models used to explain this relationship, and (5) briefly reviewing how
health behaviors are influenced by the menstrual cycle. Anxiety is discussed both
as a common symptom of menstrual distress and as clinical disorders that are
influenced by the menstrual cycle. Limitations of the current literature and
suggestions for future research are also presented.

The Menstrual Cycle

Phases of the normal menstrual cycle are typically based on a 28-day timeline
that corresponds to changing hormone levels in a woman’s body (e.g., Hillard &
Deitch, 2005). The normal range of women’s cycles varies from 24–35 days. The

Sandra T. Sigmon
University of Maine, 5742 Little Hall, Orono, ME 04469-5742, Tel: 207-581-2049,
Fax: 207-581-6128

M. J. Zvolensky, J. A. Smits (eds.), Anxiety in Health Behaviors and Physical Illness. 181
Ó Springer 2008
182 S. T. Sigmon, J. G. Schartel

uterus, ovary, and pituitary play a large role in menstrual cycle functioning with
the hypothalamus playing a central control function. Research in the menstrual
literature has typically delineated 3 distinctive phases: follicular (days 1–13),
ovulation (days 13–14), and luteal (days 15–28). Days 1–7 represent menses and
days 21–28 are generally considered the premenstrual or late luteal phase of the

Premenstrual Symptoms and Disorders

Premenstrual Symptoms. Menstrual cycle fluctuations in mood and behavior

have been addressed for over 2500 years (e.g., Hippocrates [600 B.C.]
described premenstrual complaints) in Western and non-Western cultures
(e.g., Chrisler & Caplan, 2002; Hylan, Sundell, & Judge, 1999). Epidemiolo-
gical surveys suggest approximately 75% of menstruating women experience
premenstrual symptoms each month (APA, 2000). Approximately 40% of
women experience premenstrual and menstrual symptoms of a mild nature
with only 2–10% reporting severe symptoms (e.g., Logue & Moos, 1986).
Over 150 symptoms have been identified with the premenstrual phase of the
menstrual cycle, the most common domains being affective (e.g., depression,
anxiety, irritability, mood lability), somatic (e.g., bloating, breast tenderness),
and behavioral (e.g., difficulty concentrating, overeating). Symptoms of
anxiety include physiological sensations, emotional and cognitive symptoms,
as well as behavioral symptoms. Typical anxiety symptoms associated with
the premenstrual phase include muscle tension, stomach pain, feeling restless
or irritable, difficulty concentrating, and avoidance behaviors.
Premenstrual Syndrome. There are no universal or uniform definitions for
premenstrual syndrome [PMS] (e.g., Johnson, 2004), but the defining factor is
typically presence of symptoms exclusively during the late luteal phase. To
assist researchers and foster more uniformity in PMS research, the American
College of Obstetricians and Gynecologists (ACOG, 2000) proposed the
following criteria: symptoms must only occur five days prior to menses onset,
remit by day 4 of menses and symptoms must be absent during the follicular
phase (or symptom free for at least one week). These criteria also require
prospective reporting for 2 consecutive cycles to document the pattern. There
are two consistent aspects of PMS examined in the literature; (1) the timing of
the symptoms, and (2) the symptoms must be distressing and interfere with
normal functioning.
Premenstrual Dysphoric Disorder. A controversial debate ensued when Late
Luteal Phase Dysphoric Disorder (LLPDD) was included in the DSM-III-R in
1987 and continued with the inclusion of Premenstrual Dysphoric Disorder
(PMDD) in DSM-IV (APA, 1994). Women with LLPDD reported greater
affective symptoms during the premenstrum than controls; however anxiety
symptoms do not reach the level of severity reported by women diagnosed with
Anxiety and the Menstrual Cycle 183

anxiety disorders such as panic disorder (Veeninga, de Rutter, & Kraaimaat,

1994). Although the estimates for women experiencing PMS symptoms are some-
what high, research indicates that only 3–8% of women may meet diagnostic
criteria for PMDD (e.g., APA, 1994; Halbreich, Borenstein, Pearlstein, & Kahn,
2003). However, subsyndromal levels that may also result in severe impairment are
estimated to occur in 13–18% of women.
Currently, PMDD is diagnosed under ‘‘Depressive Disorder Not Otherwise
Specified’’ and appears under ‘‘Criteria Sets and Axes Provided for Further
Study’’ (APA, 2000). There are four major criteria for a diagnosis of PMDD to
be assigned. First, at least 5 of 11 identified symptoms must meet the same criteria
used for PMS. There are four core affective symptoms (depressed mood, anxiety,
affective lability, and persistent irritability or anger), at least one of which must be
present for diagnosis. This symptom pattern must occur in most menstrual cycles
for at least one year. The second criterion for PMDD requires that the symptoms
cause marked interference with work, school, social activities or relationships.
Third, the symptoms cannot be an exacerbation of another diagnosable disorder.
For example, a diagnosis of PMDD would exclude the many women with panic
disorder who experience an exacerbation of their panic symptoms premenstrually.
And finally, the symptom pattern must be verified by at least two consecutive
cycles of prospective daily symptom ratings for two symptomatic cycles.

Issues and Methods of Assessment

Retrospective versus Prospective Reporting. The research that exists on

pmenstrual cycle effects in general has yielded inconsistent results. One
reason may be that early research on menstrual cycle effects relied almost
exclusively on women’s retrospective self-reports. Women often overestimate
the exacerbation of their anxiety symptoms when asked to recall their symp-
toms retrospectively (e.g., Endicott & Halbreich, 1982; Rubinow, 1985).
Indeed, many of the retrospective studies on anxiety and the menstrual
cycle have found strong cyclic effects in that anxiety symptoms worsen
during the premenstrum (Breier, Charney, & Heninger, 1986; Cameron,
Kuttesch, McPhee, & Curtis, 1988; Williams & Koran, 1997). Retrospective
accounts of the premenstrual experience have been hypothesized to reflect a
reliance on stereotypical and cultural biases regarding the menstrual cycle
(e.g., McFarland, Ross, & DeCourville, 1989; Ruble, 1977).
Prospective assessment of menstrual symptoms has addressed some of the
concerns stemming from retrospective measures, yet has simultaneously
presenting challenges to retrospective results, often finding no significant
fluctuations premenstrually (e.g., Stein, Schmidt, Rubinow, & Uhde, 1989).
Prospective studies that have shown menstrual fluctuations in anxiety disorder
symptomatology suggest much smaller fluctuations than have been found in
retrospective studies (Cameron et al., 1988; Cook et al., 1990). In the discussion
184 S. T. Sigmon, J. G. Schartel

of anxiety that follows, research will be discussed in terms of retrospective

versus prospective methodologies.
Assessment Measures. The gold standard for assessing any symptom across
the menstrual cycle is prospective monitoring. Thus, fluctuations in anxiety
across the cycle can be tracked by having women repeatedly complete any state-
dependent measure of anxiety or visual analogue scales focused for specific
symptoms (e.g., frequency of panic attacks or obsessive-compulsive rituals).
As one of the core symptoms of PMS and PMDD, anxiety and anxiety
fluctuations across the menstrual cycle are often assessed within that context.
There are several empirically supported measures for PMS and PMDD, all of
which include items assessing anxiety (see Table 1). Readers are encouraged to see
Haywood, Slade, and King (2002) for a review of retrospective and prospective
measures of symptoms across the menstrual cycle. On retrospective measures,
women are asked to consider either several recent menstrual cycles or to report on
their typical experience across the menstrual cycle. Prospective measures typically
take the form of a ‘‘daily diary,’’ asking women to indicate if a symptom did or did
not occur or rate on a Likert scale the severity of individual symptoms each day.
Some measures, such as the Premenstrual Tension Rating Scale (Steiner, Haskett,
& Carroll, 1980) include both a self and observer rating form. Researchers and
clinicians need to be aware of the format and the intended purpose of the
assessment, as not all assessment measures are diagnostic in nature.
A major difficulty with requiring prospective analysis over two months is
time and effort to complete the measures. Many women are unable to complete
daily ratings over several months’ time and many of the rating forms are

Table 1 Sample of measures used in assessing symptoms across the menstrual cycle
Measure Purpose (PMS/PMDD) or Prospective
Premenstrual Tension Rating Scale, self Diagnostic and Treatment Both
and observer forms (Steiner et al., 1980) Response – PMS
Daily Rating Form (Endicott et al., 1986) Assess Symptoms – PMS P
Premenstrual Symptom Diary Diagnostic – PMS P
(Thys-Jacobs et al., 1995)
Calendar of Premenstrual Experiences Diagnostic – PMS P
(Mortola et al., 1990)
Menstrual Distress Questionnaire (Moos, Assess Symptoms – Both R
Menstrual Distress Diagnostic – PMS; P
Questionnaire – Today version (Blake et al., Treatment response
Premenstrual Assessment Form (Halbreich Assess Symptoms – Both R
et al., 1982)
Daily Record of Severity of Problems Diagnostic – PMDD P
(Endicott et al., 2006)
Daily Symptom Rating Scale (Taylor, 1979) Assess Symptoms – Both P
Anxiety and the Menstrual Cycle 185

lengthy. Researchers have made efforts to validate short forms of empirically

supported measures with limited success (Haywood et al., 2002). As such, it is
common clinical practice for these forms to be adapted or for clinicians to create
unique versions of these forms to suit the needs of their individual clients.
However, clinicians should take several things into consideration when refor-
matting assessment measures. First, because of validity issues noted above, it is
recommended to rate symptoms on a severity scale rather than in yes/no format
(Freeman, 2003). Symptoms most relevant or most distressing to the client
should be the focus of monitoring. Consolidating information into as few
pages as possible increases compliance, reduces errors and minimizes data.
One common way to do this is to have a monthly page, much like a calendar,
rather than daily pages to record information (Freeman, 2003). Most impor-
tantly, women must make note of the first day of bleeding so that ‘‘menstrual
time’’ can be established and accurate comparisons made between the follicular
and luteal phases.
Co-morbidity versus Premenstrual Exacerbation. With regard to assessment
of menstrual-related symptoms, three options need to be considered: (1) is
anxiety manifesting as a primary symptom of a menstrual problem, (2) are
symptoms of an anxiety disorder worsening premenstrually, or (3) are
symptoms consistent with a comorbid menstrual problem and anxiety disorder.
Differentiating between comorbidity and premenstrual exacerbation (PME)
may be one of the most difficult challenges facing researchers and clinicians.
Anxiety disorders and menstrual difficulties are highly comorbid (see below).
Specifically, Kim and colleagues (2004) found that panic disorder occurs in
25% of PMDD samples, social phobia in 19–23%, OCD 11–13% and GAD in
4–38% across studies. Moreover, many women with anxiety disorders report
premenstrual exacerbation (PME) of their symptoms. Previous research has
found that women with Generalized Anxiety Disorder (GAD), Panic Disorder
(PD), and simple phobia (SP) report premenstrual worsening of their anxiety
symptoms (e.g., Busch, Costa, Whitehead, & Heller, 1988). Moreover, 45–50%
of psychiatric admissions for women occur during the premenstrual phase (e.g.,
Targum, Caputo, & Ball, 1991).
Distinguishing PME from comorbidity presents challenges for women with
PMS. Stout and colleagues (1986) found that in a sample of 223 women seeking
treatment for PMS, 81% of the women had at least one psychiatric diagnosis
lifetime with 65% meeting criteria for phobia and 16% meeting criteria for
OCD. Hsiao, Hsiao, and Liu (2004) examined the prevalence of PMS and
premenstrual exacerbation (PME) in 200 Chinese women and found that
78% of the women with GAD and 68 % of the women with PD met criteria
for PMS. In addition, 52% of the women with GAD and 36% of the women
with PD reported PME of their anxiety symptoms.
The distinction between comorbidity and PME might stem from controversy
surrounding the diagnostic validity of PMS and PMDD. Some researchers
argue that PMDD is a distinct disorder with irritability and mood lability as
defining features (e.g., Landén & Eriksson, 2003) rather than a subtype of
186 S. T. Sigmon, J. G. Schartel

depression or anxiety. Hartlage and Gehlert (2001) suggest the problem is that
symptoms of PMDD overlap significantly with other affective disorders and
argue for phase specificity in diagnosing PMDD (i.e., symptoms occur solely
during the premenstrum). They propose that PME can be distinguished from
PMDD if clinicians consider one symptom at a time. This leaves unresolved the
possibility of some symptoms of an anxiety disorder occurring across the entire
cycle while others do not. How many of the symptoms of PMDD must be
distinct from the symptoms of the ongoing anxiety disorder remains unclear.
The criteria of five symptoms has also been challenged as being arbitrary and
too restrictive to accurately assess the impact of symptoms on women’s
functioning (Smith et al., 2003; Freeman, 2003). Yonkers (1997) suggests that
premenstrual complaints can vary in severity, degree of comorbidity with other
psychological disorders and level of impairment. For example, a woman may
only demonstrate two out of five symptoms, but they may be severe enough to
cause significant interference with her day to day functioning.
Researchers have also questioned the validity of defining and measuring the
‘‘absence’’ of a symptom. The requirement of symptom absence is unique to
PMS/PMDD, and there are currently no good assessment measures that
reliably measure this criterion. Furthermore, most criteria require symptoms
be absent for the duration of the follicular phase. This requirement has been
challenged as being unrealistic due to the natural fluctuations in mood and
occurrence of life events over the course of the cycle (Freeman, 2003; Smith
et al., 2003). For example, Bailey and Cohen (1999) assessed the number of
women seeking treatment for premenstrual symptoms who experienced symp-
toms across the menstrual cycle. Over 60% of their sample met criteria for either
a mood or anxiety disorder or both, indicating they experienced symptoms
diagnostic of PMS across all phases of the cycle. Instead, researchers like
Freeman (2003) suggest we would be better served to look at changes in
symptom severity or overall frequency (i.e., degree of change). Although
suggestions have been made (e.g., Smith et al., 2003), no conclusion has yet
been reached regarding what would represent a clinically significant change
from one menstrual cycle phase to the next phase.
There is a significant amount of variability both between and within indivi-
dual women’s symptoms from cycle to cycle. Bloch, Schmidt, and Rubinow
(1997) looked at the stability of premenstrual symptoms in 16 women with PMS
(defined as 30% increase in symptom frequency during the luteal phase) for 3 or
more cycles. Results indicated that symptom presentation from cycle to cycle
varied considerably, with anxiety (83%), irritability (85%), and mood lability
(77%) exhibiting the most stability across the 3 cycles. Freeman (2003) also
notes that the timing of symptom appearance/remission varies between women.
Some women experience PME for only a few days of the premenstrual phase at
either the beginning or end and some symptoms are experienced on nonconse-
cutive days. Thus, the literature suggests that in well-defined PMS samples,
symptoms persist across time and can cause significant impairment in women’s
Anxiety and the Menstrual Cycle 187

lives even if two consecutive symptomatic cycles are not established during the
diagnostic period.

Research on Premenstrual Anxiety

Normal Samples. Normal samples of women have been studied extensively to

ascertain the level of distress associated with the premenstrual phase. It should
be noted that although most studies indicate a normal sample, typically no
assessment has been made of pre-existing levels of trait anxiety or anxiety
disorders. Studies utilizing retrospective methodology have demonstrated
increases in negative affect (depressed mood and anxiety in particular) and
increases in severity of physical symptoms during the premenstrual phase
compared to the intermenstrual phase of the cycle (e.g., Boyle & Grant,
1992). Some prospective studies of menstrual cycle fluctuations in affective
and somatic symptoms in normal populations have shown premenstrual effects
(e.g., Gallant, Hamilton, Popiel, Morokoff, & Chakraborty, 1991) whereas
others have failed to find increased symptoms during the premenstrual phase
(e.g., Slade, 1984). When normal women report more anxiety during the
premenstrual phase, the level of anxiety appeared to be more comparable to
women who are mildly stressed, not to a clinical sample (Golub, 1976). This
equivocal pattern was not clarified whether or not participants were informed
of the true purpose of the study (e.g., Gallant et al., 1991).
Although the majority of individuals who seek treatment for PMS are
typically in their mid 20’s to mid 30’s, Hillard and Deitch (2005) found that
75% of older adolescents who visited a health care provider perceived a
problem with menstruation. However, Golub and Harrington (1981) found
no significant differences between premenstrual and menstrual physical or
mood symptoms in 158 15–16 year old females. Moreover, they did not
significantly differ from comparisons to same-age males during the same time
period. In a prospective study (Layton, 1989), adolescent females reported
elevated rates of state anxiety during the premenstrual phase compared to the
menstrual phase. Although the findings for premenstrual worsening of mood
symptoms vary in adolescent samples, it appears that premenstrual complaints
are common across different age samples.
Large community surveys have also been conducted to determine if women
experience greater distress during the premenstrual phase of the menstrual cycle
(e.g., Strine, Chapman, & Ahluwalia, 2005). As part of the 2002 National
Health Survey, over 11,000 women from ages 18–55 completed a computer-
assisted personal interview. Approximately 19% of the sample indicated men-
strual difficulties over the past year with white non-Hispanic women reporting
more difficulties than Hispanic women and 16% reported interference from
menstrual problems. In addition, women who reported menstrual problems
were more likely to report psychological distress including anxiety, sleep
188 S. T. Sigmon, J. G. Schartel

difficulties, and negative health behaviors (e.g., smoking, drinking, obesity).

Unfortunately, no associations between distress and cycle phase were assessed.
A growing area of research on menstrual cycle effects examines physiological
differences in normal women during phases of the menstrual cycle. Researchers
have found phase differences in hypothalamic pituitary axis (HPA) functioning,
specifically related to cortisol and DHEA (i.e., dehydroepiandrosterone) levels
(Symonds, Gallagher, Thompson, & Young, 2004). Cortical activity has also
been found to fluctuate across menstrual cycle phase in samples of normal
women (Asso & Braier, 1982; Protopopescu et al., 2005). It has been suggested
that varied levels of estrogen and progesterone across the menstrual cycle may
affect neuronal circuitry (Protopopescu et al., 2005). Asso and Braier (1982)
found increased skin conductance during the premenstrual phase compared to
the intermenstrual phase in a normal sample of women. It is clear that more
research needs to be conducted regarding normal menstrual cycle psychophy-
siology in order to inform how such processes may be impaired in women with
anxiety disorders.
Premenstrual Anxiety in PMS Samples. Similar to the findings in normal
samples, there are mixed results in women self-reporting PMS with retrospec-
tive and prospective methods. Women with PMS retrospectively report higher
levels of state (e.g., Haskett, Steiner, & Carroll, 1984; Mira, Vizzard, &
Abraham, 1985) and trait anxiety (e.g., Giannini, Price, Loiselle, & Giannini,
1985; Watts et al., 1980) during the premenstrum. However, prospective
self-reports of premenstrual changes have yielded inconsistent results (e.g.,
Sampson & Jenner, 1977; Taylor, 1979). In a sample of 69 undergraduate and
community women, Kessel (2000) found women who experienced significant
premenstrual changes (>30%) reported higher levels of anxiety compared to
women who did not report severe changes. Christensen and Oei (1989) found
that women with PMS prospectively reported more anxiety, depression, and
automatic negative thoughts than controls during the premenstrual phase.
Magos, Brincat and Studd (1986) found that retrospective and prospective
reports of women seeking treatment for PMS were similar, with 61–85%
reporting symptoms consistent with PMS. In contrast, Hardie (1997) found
that none of the retrospective reports of PMS from their sample of 101 women
were confirmed by 2 prospectively monitored cycles. Fluctuations in mood were
better predicted by interpersonal relationship quality and stress level than cycle
phase. Thus, anxiety fluctuations across the menstrual cycle in PMS popula-
tions are just as inconclusive as those in normal samples.
Other research has focused on assessing attributions for PMS symptoms (e.g.,
Veeninga & Kraaimaat, 1995). For example, women with PMS were more likely
than controls to attribute symptoms to their cycle when in the premenstrual
phase than when they were in the intermenstrual phase. Women with PMS also
reported more menstrual symptoms both premenstrually and during the inter-
menstrual phase (Veeninga & Kraaimaat, 1995). However, a large number of the
women with PMS also met criteria for an anxiety disorder and thus this study
Anxiety and the Menstrual Cycle 189

could not distinguish between women with PMS alone or women with PMS who
also met criteria for an anxiety disorder.
Cross-cultural research in Western societies seems to parallel the results
obtained in the United States. A study on women in the United States, United
Kingdom, and France (N = 1045) revealed considerable impairment at home,
in social relationships, and work. Across countries, more than 60% of women
reported increases in anxiety, fears and tension premenstrually, and less than
25% had sought treatment for their symptoms (Hylan et al., 1999).
Premenstrual Dysphoric Disorder. Researchers have estimated that the impact
and burden associated with PMDD to be similar to that of other affective
disorders (e.g., Halbreich et al., 2003). Wittchen, Becker, Lieb, and Krause
(2002) contend that PMDD can persist for the duration of a woman’s reproduc-
tive years. In their research, depression was the first ranked symptom (91%) and
anxiety was noted by 67% of the sample, with 47% of women with PMDD
meeting criteria for a comorbid anxiety disorder. Several studies have found
that comorbid anxiety disorders may be found in as many as 59% of women
diagnosed with PMDD (Fava et al., 1992; Pearlstein, Frank, Rivera-Tovar,
Thoft, et al., 1990). Research has also indicated that women with PMDD report
more stressors and experience more distress premenstrually than intermenstrually
compared to controls (e.g., Fontana & Badawy, 1997) and also report more state
anxiety than controls (e.g., Christensen & Oei, 1989).

Menstrual Cycle Effects on Anxiety and Anxiety Disorders

Anxiety Sensitivity. Anxiety sensitivity (AS; i.e., fear of symptoms of anxiety

and its consequences; Reiss & McNally 1985) has been well-studied as a
vulnerability to anxiety disorders, panic disorder in particular (e.g., Schmidt,
Lerew, & Trakowski, 1997; Weems, Hayward, Killen, & Taylor, 2002) and
gender differences in anxiety sensitivity may account for some of the gender
differential in anxiety disorder prevalence rates. In a series of studies, Sigmon
and colleagues have investigated the relationship between AS, self-reports of
menstrual symptoms, gender roles, somatic beliefs, and psychophysiological
reactions to anxiety stimuli across the menstrual cycle (Sigmon, Dorhofer,
Rohan, & Boulard, 2000; Sigmon, Fink, Rohan, & Hotovy, 1996; Sigmon,
Rohan, Boulard, Dorhofer, & Whitcomb, 2000).
In the first study (Sigmon et al., 1996), women high or low in anxiety sensi-
tivity were assessed during either the intermenstrual or premenstrual phase.
Results indicated that women high in AS reported more severe premenstrual
symptoms, more state and trait anxiety, and exhibited greater skin conductance
responses to auditory anxiety-provoking stimuli than women low in AS, regard-
less of current cycle phase. Researchers proposed that women high in anxiety
sensitivity are more likely to focus on internal stimuli, have a higher nonspecific
arousal level, and negatively interpret normal bodily sensations. In addition to
190 S. T. Sigmon, J. G. Schartel

confirming greater premenstrual distress reported by women high in AS, two

subsequent studies (Sigmon, Rohan et al., 2000; Sigmon, Dorhofer, et al., 2000)
suggested that women high in AS may demonstrate more gender specificity
(endorsing more feminine and less masculine characteristics), illness attitudes,
and may be more vigilant to bodily sensations than women low in AS.
The above results led Sigmon and colleagues to propose the menstrual reac-
tivity hypothesis (Sigmon, Dorhofer, et al., 2000) which states that certain women
(e.g., high in AS, women with asthma) are more likely to report greater menstrual
distress due to accurate reports of symptoms, expectations about the meanings of
symptoms, hypervigilance to bodily sensations, and increased psychophysiologi-
cal responding in response to anxiety stimuli. The menstrual cycle may represent a
cyclical stressor that provides increased opportunities for self-focus, symptom
attribution and interpretation. The authors suggest that researchers should rou-
tinely assess for AS and for gender-related variables that could add to a more
complete account of the relationship between the menstrual cycle and anxiety.
Although existing research on menstrual effects and anxiety sensitivity has
focused on AS as a continuous variable, recent research has found support for a
latent structure that is taxonic (i.e., categorical in nature; e.g., Schmidt, Kotov,
Lerew, Joiner, & Ialongo, 2005; Bernstein et al., 2005). In general, the taxon
accounts for more variance than variables such as body sensations and body
vigilance and possesses incremental validity in predicting future panic attacks.
Bernstein and colleagues (2005) suggest that taxon criteria may differ across
gender and population, pointing to the need for more research to determine if
AS has a latent taxonic structure for women experiencing PME of an anxiety
disorder and women for whom anxiety is the core symptom of a menstrual
disorder. It is possible that taxon identification may be useful in identifying
women who are vulnerable to menstrual cycle influences on anxiety and
formulating successful treatment strategies.

Premenstrual Symptoms and Anxiety Disorders

Anxiety represents a prominent premenstrual symptom for some women and

comorbidity between premenstrual and anxiety disorders is high (e.g., Fava
et al., 1992), however the course of specific anxiety disorders across the
menstrual cycle is less clear. Few studies have investigated if and how specific
anxiety disorders are affected by the menstrual cycle and that which does exist
is complicated by issues surrounding retrospective and prospective recording.
Panic Disorder. Although evidence suggests that panic disorder symptoms can
be exacerbated premenstrually (e.g., Klein, 1993; Cameron et al., 1988), retro-
spective and prospective self-reports of menstrual symptoms are inconsistent.
Given the similarity between certain premenstrual symptoms (e.g., muscle tension,
hot flashes, increases in breathing rates, anxiety) and symptoms of panic, the
ability to distinguish between the exacerbation of panic symptoms or the
Anxiety and the Menstrual Cycle 191

experience of premenstrual symptoms is complicated, regardless of the methodol-

ogy used. Research on PME of panic symptoms assumes that women are able to
accurately distinguish the causes of their symptoms, and symptom misattribution
has been cited as a contributor to inconsistent findings as well as a significant
component of panic disorder (Stein, Schmidt, Rubinow, & Uhde, 1989).
Retrospective assessments suggest that a large proportion of women experi-
ence an increase in panic symptoms premenstrually. For example, Breier and
colleagues (1986) found that more than half of women reported PME of their
symptoms. Women in other studies have reported increases in severity of
symptoms (Cameron et al., 1988), overall anxiety, frequency of panic attacks
and phobic avoidance (Cook et al., 1990).
Prospective studies have yielded mixed results with respect to PME of panic
disorder. Cameron et al. (1988) found that women reported all symptoms as more
severe retrospectively, but only muscle tension was statistically significant
prospectively. Similarly, other researchers have failed to find premenstrual
increases in overall levels of anxiety (e.g., Stein et al., 1989; Cook et al., 1990)
or frequency of panic attacks (Cook et al., 1990). Women have also rated
symptoms as more severe during the premenstrual phase. For example, Cook
and colleagues (1990) found that approximately half of their sample prospectively
rated one out of two menstrual cycles with a greater than 30% increase in panic
symptom severity. In contrast, Kaspi, Otto, Pollack, Eppinger and Rosenbaum,
(1994) found that panic attack frequency increased premenstrually whereas panic
severity, avoidance, and anticipatory anxiety did not. Half of the participants in
this study experienced twice as many panic attacks during the premenstrual phase
compared to only 5 women who retrospectively reported symptom worsening.
Unsubstantiated retrospective reports of premenstrual worsening only occurred
in four out of twelve participants.
Several reasons for these inconsistencies have been suggested. The experience
of negative life events, fatigue, or other stressors may affect women’s experience
of anxiety from one cycle to the next and assuming consistency across
consecutive cycles may be inappropriate (Cook et al., 1990). The influence of
beliefs about menstruation, social expectations or social roles may also play a
part in women’s symptom expression (Sigmon, Dorhofer, et al., 2000). Memory
biases (Cameron et al., 1988) and symptom misattribution during different
phases have also been suggested. Natural fluctuations in anxiety might be
mistakenly attributed to the menstrual cycle because it is a salient anchor
point for women (Stein et al., 1989).
Obsessive-Compulsive Disorder. Retrospective studies have suggested a
significant portion of women with OCD experience premenstrual worsening
of their symptoms (Labad et al. 2005; Williams & Koran, 1997). Interestingly,
many of the women reporting premenstrual worsening also reported significant
premenstrual mood symptoms indicative of PMS or PMDD (Williams &
Koran, 1997). No studies utilizing daily prospective symptom monitoring
have been conducted; however Vulink, Denys, Bus, and Westenberg (2006)
assessed women at three time points concordant with menstrual, premenstrual,
192 S. T. Sigmon, J. G. Schartel

and intermenstrual phases of their cycles. Results indicated that half of the
women reported PME of OCD symptoms.
Generalized Anxiety Disorder. Only one study has investigated menstrual cycle
relationships with GAD (McLeod, Hoehn-Saric, Foster, & Hipsley, 1993) and
PTSD (Perkonigg, Yonkers, Pfister, Lieb, & Wittchen, 2004). McLeod and
colleagues (1993) recruited women with GAD with and without comorbid PMS.
Results indicated that women who had both GAD and PMS reported significant
PME of anxiety symptoms during the luteal phase. Unfortunately, assessment
points were not consistent between groups, making comparisons difficult.
PTSD. Perkonigg and colleagues (2004) followed women with PTSD over a
span of 42 months, assessing for predictors of PMDD at three time points. The
best predictor of PMDD diagnosis was subthreshold PMDD symptoms at the
initial interview. However, the experience of traumatic events and anxiety
disorder presence both significantly predicted subsequently meeting diagnostic
criteria for PMDD. PMDD symptoms were only assessed retrospectively and
interviews were conducted without respect to menstrual cycle phase; however
their results do suggest that trauma and anxiety may impact women’s experi-
ence of symptoms across the menstrual cycle.
Although results across different disorders, populations, and methods of
reporting have been inconsistent, there is nonetheless some evidence to suggest
that for some women, menstrual cycle phase may have a significant impact on
their experience, or at least the reporting of symptoms. Symptom patterns may
also vary from cycle to cycle and may be related to life events, stress, fatigue,
interpersonal or relationship problems. Many authors have suggested that this
heterogeneity both between and among women may interfere with accurate
diagnosis and interpretation of effect sizes in treatment studies (Vulink et al.,
2006). Given recent advances in statistical methods for prospective data and in
general understanding of anxiety disorders, more research investigating
menstrual cycle effects on women’s experience of anxiety is warranted.

Models of Anxiety-Premenstrual Cycle Phase Interaction

Feminist Views. Chrisler and Caplan (2002) provide an excellent overview of

the evolution of scientific thought regarding PMS. The authors discuss the
implications surrounding the medicalization of the menstrual cycle and
the promulgation of ‘‘the belief that the (menstrual) cycle itself is a problem to
be solved’’ (p. 283). Ussher (2003) also discusses hegemonic truths that
need to be carefully scrutinized: ‘‘PMDD as a thing that can be objectively
defined and measured; PMDD is pathology to be eradicated; PMDD is caused
and treated by one factor; PMDD as a bodily phenomenon, and PMDD causes
women’s symptoms’’ (p. 131). In a similar vein, Clare (1983) argues that it may
be unrealistic to assume that women would be symptom free during any given
menstrual cycle given that the symptoms that are most reported are ubiquitous
Anxiety and the Menstrual Cycle 193

to humans. To study any aspect of the menstrual cycle without the contexts of
stress and lifestyle will result in incomplete assumptions (e.g., Koeske, 1983).
Thus, feminist views about the menstrual cycle mirror similar concerns that
have been raised in other approaches.
Biological Explanations. Reproductive hormones play a significant role in
mood symptom fluctuation across the menstrual cycle; however, the exact
mechanisms involved are not well understood (e.g., Hendrick, Altshuler, &
Burt, 1996). Progesterone metabolites and estrogen levels typically decrease
during the premenstrual phase and remain low through the early follicular
phase. These hormones modulate neurotransmitter levels (e.g., serotonin,
GABA, dopamine, norepinephrine) that are hypothesized to play an important
role in the development and maintenance of psychological disorders, in
particular anxiety and depression (e.g., Dubrovsky, 2006; Le Mellédo &
Baker, 2002). This area of research is relatively new and is hampered by the
use of small samples, retrospective methodology, and lack of precision in
determining menstrual cycle phase.
Biological explanations of PMS typically begin with a discussion of hormonal
control of the menstrual cycle. The development of premenstrual symptoms is
purported to be linked to the rapid decrease in progesterone during the late luteal
phase (e.g., Poromaa, Smith, & Gulinello, 2003). However women with and
without PMS often do not show differences in absolute levels of progesterone
across the menstrual cycle (e.g., Rubinow et al., 1988; Hammarbäck, Damber, &
Bäckström, 1989). Thus, researchers have proposed that there may be differ-
ences in sensitivity to the changing hormone levels (e.g., Poromaa et al., 2003;
Halbreich, 2003). In addition, because progesterone metabolites enhance GABA
(i.e., gamma-aminobutyric acid) activity (e.g., Olsen & Sapp, 1995) and GABA
has been implicated in anxiety symptoms (e.g., Ninan et al. 1982), researchers
have proposed models of complex hormonal interactions that may link
premenstrual symptoms and anxiety disorders (e.g., Facchinetti, Tarabusi, &
Nappi, 1998). However, more research is needed to determine the mechanisms
affecting sensitivity to these interactions (Roy-Byrne, Cowley, Greenblatt,
Shader, & Hommer, 1990).
Research has also demonstrated several biological links between menstrual
disorders and anxiety disorders. Women diagnosed with PMS and LLPDD
(i.e., Late Luteal Phase Dysphoric Disorder) report increased in panic
symptoms, negative interpretations of sensations, and state anxiety in response
to a lactate challenge (e.g., Facchinetti et al., 1998; Kent et al., 2001). Further,
approximately 60–70% of women with panic disorder and with LLPDD experi-
ence panic attack in response to lactate infusion and inhaled carbon dioxide.
Panicogenic responding to challenge tasks was previously thought to be specific
to panic disorder, and the similar rates of responding may suggest a shared
biological vulnerability and account for the high levels of anxiety reported by
LLPDD patients (for a review see Vickers & McNally, 2004). Other evidence to
support a link between PMS and anxiety disorders comes from the responsive-
ness to alprazolam in women with PMS (e.g., Facchinetti et al., 1998). Although
194 S. T. Sigmon, J. G. Schartel

these findings are preliminary, they do suggest intriguing links between anxiety
symptoms and PMS. Biological links between PMDD and anxiety have also
been explored. For example, symptoms disappear or decrease in women diag-
nosed with PMDD when they do not ovulate, when ovaries are removed, or
when they take medications that inhibits ovulation (e.g., Steiner, 2000).
Biopsychosocial Models. More complex models of menstrual distress incor-
porating social and psychological processes are surfacing as it has become clear
that biological explanations alone cannot suffice (e.g., Anson, 1999). A woman’s
experience of premenstrual symptoms is influenced by socialization practices and
environmental contextual factors (e.g., Anson, 1999). Menstrual attitudes and
beliefs that develop from interactions with peers, family, and the media (e.g.,
Aubuchon & Calhoun, 1985; Woods, Mitchell, & Lentz, 1995) may play a
significant role in the reporting and experience of premenstrual symptoms.
Reports of premenstrual complaints vary depending on retrospective reports
(e.g., Marván & Cortés-Imiestra, 2001; Rapkin, Chang, & Reading, 1988),
number and severity of daily and life stressors (e.g., Fontana & Palfai, 1994),
use of maladaptive coping strategies (Sigmon, Whitcomb-Smith, Rohan, &
Kendrew, 2004), complex role conflicts (e.g., Ross & Steiner, 2003), and history
of sexual abuse (e.g., Ross & Steiner, 2003). These represent but a few of the
factors proposed to play a contributory role in the experience and reporting of
premenstrual difficulties.
Researchers have also examined the role that the perception of stress and
interference with functioning might play in the reporting of increases in anxiety
during the premenstrual phase. Davydov and colleagues (2004) assessed anxiety
symptoms during the luteal and follicular phases in nurses, who reported higher
levels of anxiety on working days during the luteal phase than during days off
during the luteal phase or during the follicular phase. Studies examining
menstrual interference in college samples (e.g., Brooks-Gunn & Ruble, 1986)
have found that perceived interference, menstrual pain, and premenstrual pain
were the best predictors of emotional distress. The menstrual cycle itself has
been conceptualized as a stressor (e.g., Logue & Moos, 1986). Elliott and
Harkins (1992) found that for normal women menstrual and premenstrual
pain and perception of interference were the strongest predictors of emotional
distress. According to the authors, results provided support that how
individuals appraise or perceive interference from a particular condition or
stressor (e.g., menstrual cycle) may represent an important component in stress
and coping processes.

Treatment Approaches for Premenstrual Anxiety

There are several treatment options available for women who experience
significant levels of anxiety either as a result of PME or as a core symptom of
PMS or PMDD. Efficacious treatments include lifestyle changes, medications,
Anxiety and the Menstrual Cycle 195

hormone supplements, and psychotherapy. Researchers have suggested that

when there are questions of comorbidity, patients should be treated first for the
primary anxiety disorder, as many premenstrual symptoms will resolve with its
effective treatment and additional treatment for residual premenstrual
symptoms can then be provided (e.g., Steiner & Born, 2000).
Lifestyle changes, stress management and/or dietary supplements (e.g.,
Clayton, 2003; Steiner & Born, 2000) are often recommended as the first
step in treatment. Establishing and maintaining a regular sleep-wake cycle
(Clayton, 2003) and regular exercise may reduce the severity and impact of
both physical symptoms and anxiety during the premenstrum (e.g., Kirkby &
Linder, 1998; Hightower, 1997). Studies have found particular positive effects
of exercise on anxiety disorders such as OCD (Lancer, 2005) and PTSD
(Manger, 2005).
Various forms of psychotherapy have been found to be helpful in
alleviating menstrual difficulties and may have implications for anxiety symp-
toms. Although random controlled trials on psychotherapeutic interventions
for menstrual-related problems are few, the existing research is promising.
Relaxation training has been shown to be superior to a distracting activity
such as leisure reading and to symptom monitoring (Goodale, Domar, &
Benson, 1990) and coping skills groups have demonstrated superior efficacy
to relaxation and hormone therapy (Morse, 1999). Cognitive-behavior
therapy for PMS has demonstrated effectiveness in both group and individual
format (Morse, 1999; Slade, 1984) when compared to placebo treatments
(Kirkby, 1994) and wait list control (Blake, Salkovskis, Gath, Day, & Garrod,
1998). Educational groups have demonstrated effectiveness when compared to
CBT groups (Christensen & Oei, 1995).
As with most forms of CBT, it is critical that a strong, collaborative
therapeutic alliance between patient and therapist be established. Most
women who experience menstrual-related problems typically adhere to medical
explanations for their problems (Hunter, Ussher, Cariss, Browne, Jelley, &
Katz, 2002). Thus, it is important for therapists to explain the biopsychosocial
model (presented in next section) as an alternative, more complete framework
from which women can understand their symptoms and their impact on their
lives (Blake et al., 1998). When presented with a biopsychosocial model, most
women find it an acceptable alternative to traditional medical explanations and
subsequently report more active cognitive and behavioral coping strategies
(Hunter et al., 2002). Specific components of CBT for PMS typically consist
of teaching coping skills, identifying symptom triggers and maintaining factors,
identifying and challenging cognitive distortions, and implementing behavior
change strategies (Blake et al., 1998). Other therapies include cognitive restruc-
turing and assertiveness training (Christensen & Oei, 1995).
Women receiving CBT for PMS have shown greater reductions in symptoms
and irrational thinking compared to placebo control therapy through 9-month
follow up (Kirkby, 1994) and reductions in impairment, depressive symptoms,
and prospective ratings of both physical and mood symptoms when compared to
196 S. T. Sigmon, J. G. Schartel

a wait-list control condition (Blake et al., 1998). However, CBT may be more
effective for managing depressive symptoms than anxiety symptoms, as some
studies have found no reduction in anxiety scores following treatment (Blake
et al., 1998). CBT has demonstrated equal efficacy to medication (i.e., fluoxetine)
in treating PMDD with better maintenance after one year (Hunter et al., 2002 ).
A combination of medication and psychotherapy did not demonstrate any
additional benefits to each alone. It is interesting to note, that women treated
with fluoxetine reported greater improvement in anxiety than those who received
CBT. The authors attributed this to either the known anxiolytic effects of the
medication or the tendency for CBT to focus on depressive symptoms and not
Selective serotonin reuptake inhibitor (SSRI) anti-depressants have been
found to be effective for alleviating symptoms of PMS and PMDD (see
Kornstein & Smith, 2004). Only two controlled trials have demonstrated
efficacy for intermittent dosing of anxiolytic medication for treating PMS
(e.g., Freeman, Rickels, Sondheimer, & Polansky, 1995). However, treatment
with SSRIs can result in decreases in both anxiety and depression symptoms,
with the option to add anxiolytics if residual anxiety symptoms remain
(Rapkin, 2003). Several reviews of the existing literature on antidepressant
treatment of PMS and PMDD are available (e.g., Kornstein & Smith, 2004;
Dimmock, Wyatt, Jones, & O’Brien, 2000). Clearly, more research is needed
on what components of psychological and biological treatments are effective
for anxiety symptoms across the menstrual cycle.

Health Behaviors and the Menstrual Cycle

Health behaviors seem to be significantly impacted by anxiety and menstrual

distress. It is well-established that anxiety problems are associated with negative
health behaviors such as drug and alcohol use (e.g., Strine, Chapman, Kobau,
& Balluz, 2005). Unfortunately, very little research has been done with specific
regard to anxiety across the menstrual cycle and health behavior, thus this brief
section is intended to provide an overview of how general symptoms associated
with menstrual distress influence women’s behavior.
Results of the National Health Interview Survey indicated that women
reporting menstrual problems were more likely to report frequent anxiety,
nervousness, restlessness and were also more likely to engage in health risk
behaviors (e.g., smoking, drinking, overeating) than women reporting no
menstrual problems (Strine, Chapman, & Ahluwalia, 2005). Both physical
and affective menstrual symptoms have been positively associated with
smoking (e.g., Sloss & Frerichs, 1983), alcohol use (e.g., Tobin, Schmidt, &
Rubinow, 1994), and caffeine intake (e.g., MacKay, 1985). Affective symp-
toms may be more strongly associated with negative health behaviors (Woods
et al., 1992).
Anxiety and the Menstrual Cycle 197

There is some evidence to suggest that negative health behaviors may

increase during the premenstrum. Smoking in particular may be likely to
increase during the premenstrual phase both for women without menstrual
disorders (Kritz-Silverstein, Wingard, & Garland, 1999) and for women with
PMDD (Pomerleau, 1996). In addition, women may experience more cravings
or have a more difficult time quitting during the premenstrum (see Carpenter,
Upadhyaya, LaRowe, Saladin, & Brady, 2006 for a review). However, women
may be more likely to relapse during menses (Frye, Ward, Bliss, & Garvey, 1992
as cited in Carpenter, Upadhyaya, LaRowe, Saladin, & Brady, 2006). Some
research has found that women who quit during the luteal phase rated with-
drawal symptoms worse than those who quit during the follicular phase and
shown positive correlations between withdrawal distress and menstrual distress
(O’hara, Portser, & Anderson, 1989).
Alcohol consumption has been shown to increase premenstrually by both
retrospective (e.g., Harvey & Beckman, 1985) and prospective (e.g., Epstein,
Rhines, Cook, Zdep-Mattocks, Jensen, & McCrady, 2006) reports. However
such increases might only be related to those suffering from severe premenstrual
distress (Griffin, Mello, Mendelson, & Lex, 1987), as results have not been
consistent (e.g., Abraham & Mira, 1989). In contrast, Tobin and colleagues
(1994) found that women with PMS prospectively reported more alcohol use
than controls over the course of the menstrual cycle, not just the premenstrum.
Physical activity is a positive health behavior that is supported as a protective
factor both for physical and mental health problems including anxiety (e.g.,
Petruzello, Landers, Hatfield, Kubitz, & Salazar, 1991). Kirkby and Lindner
(1998) found that increases in trait anxiety from mid-cycle to premenstrum were
common in sedentary women with a history of PMS; however there were no
changes in self-reported anxiety among women who exercised. Adolescent girls
who reported severe menstrual symptoms also reported the least amount of
physical activity (Teperi & Rimpela, 1989).
Menstrual distress and health behaviors are thus closely associated; however
it has been difficult for researchers to determine causation. One reason may be
that women are attempting to self-medicate through the use of substances or
through decreasing the amount of physical activity in which they engage.
Conversely, negative health behaviors may make women more vulnerable to a
wide array of health conditions including menstrual distress (Teperi & Rimpela,
1989). In addition, research in this area to needs to distinguish between the
impact of the physical and affective symptoms of menstrual distress.

Summary of Research on Anxiety and the Menstrual Cycle

Inconsistencies abound in the literature on anxiety and the menstrual cycle.

Different definitions and different methodologies used by researchers contribute
to equivocal results. In general, it may be safe to conclude that a significant
198 S. T. Sigmon, J. G. Schartel

number of normal women and those suffering from menstrual or clinical anxiety
disorders may experience increased anxiety symptoms during the premenstrual
phase. Prospective assessment of at least 2 months is necessary to determine if
the anxiety symptoms only increase during the premenstrual phase. Such a
determination is crucial given that different treatment options may be
warranted. If anxiety symptoms persist across all phases of the menstrual
cycle, then treatment may be focused more on anxiety than phase-specific
effects. Thus, assessment and treatment decisions are crucial given the possible
complex relationships between anxiety and the premenstrual phase.

Limitations in Existing Research

In this review, several issues have been delineated that make the literature on
anxiety, its disorders and menstrual cycle phases difficult to summarize.
Retrospective and prospective methods of reporting anxiety symptoms across
the menstrual cycle reveal many inconsistent findings. Researchers are often
unclear about whether they are discussing anxiety as a symptom caused by the
premenstrual phase of the menstrual cycle or the exacerbation of a
pre-existing anxiety disorder. The menstrual literature is replete with differing
definitions of premenstrual complaints, menstrual cycle timing, and
premenstrual exacerbation of existing pathologies. Conceptually, researchers
need to reach consensus on diagnostic criteria for premenstrual difficulties
and premenstrual exacerbation of psychological disorders. In addition,
researchers need to be more consistent in definitions and lengths of menstrual
cycle phases. It may be time for researchers to conduct more translational
research in which both camps agree on these matters.

Future Research

Although anxiety is often associated with premenstrual complaints and

premenstrual exacerbation of anxiety disorders has been documented, more
research needs to be conducted that distinguishes between the symptoms of
anxiety disorders and premenstrual disorders. The complex interactions of
steroidal hormones with neurotransmitters associated with the development
and maintenance of anxiety and its disorders deserves more attention. Following
the tenets of the biopsychosocial approach to understanding premenstrual
difficulties may prove useful in research that attempts to disentangle some of
the complexities of these relationships. In addition, prevention research in
adolescence that targets emotional vulnerability factors, such as anxiety sensi-
tivity may be helpful.
There continues to be a great need for more basic research on furthering our
understanding of the physiological and psychological bases of how menstrual
Anxiety and the Menstrual Cycle 199

cycle phases and anxiety interact. More information is also needed to increase
our understanding of how beliefs, expectations, cultural ‘‘knowledge’’, and
personal experiences influence the experience, reporting, and attribution of
symptoms as well as the perception of control over them. A taxometric
approach may also be helpful in investigating emotional vulnerability factors
in the prediction of premenstrual anxiety. Finally, measurement issues need to
be addressed. The use of prospective monitoring over longer periods of time
coupled with agreement over definitions of premenstrual concepts needs to
standardized. In general, the relationship between menstrual cycle effects and
anxiety symptomatology needs further refinement.


Abraham, S., & Mira, M. (1989). Psychosocial forces in premenstrual syndrome. In L. M.

Demers, J. L. McGuire, A. Phillips, & D. R. Rubinow: Premenstrual, Postpartum, and
Menopausal Mood Disorders. Baltimore: Urban & Schwarzenberg.
American College of Obstetrics and Gynecologists. (2000). Premenstrual Syndrome. ACOG
Practice Bulletin No. 15.
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disor-
ders (4th ed.). Washington, DC: American Psychiatric Association.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disor-
ders (4th ed., Text Revision). Washington, DC: American Psychiatric Association.
Asso, D., & Braier, J. R. (1982). Changes with the menstrual cycle in psychophysiological and
self-report of activation. Biological Psychiatry, 15, 95–107.
Anson, O. (1999). Exploring the bio-psycho-social approach to premenstrual experiences.
Social Science & Medicine, 49, 67–80.
AuBuchon, P. G., & Calhoun, K. S. (1985). Menstrual cycle symptomatology: The role of
social expectancy and experimental demand characteristics. Psychosomatic Medicine,
47(1), 35–45.
Bailey, J. W., & Cohen, L. S. (1999). Prevalence of mood and anxiety disorders in women who
seek treatment for premenstrual syndrome. Journal of Women’s Health, 8, 1181–1184.
Bernstein, A., Zvolensky, M. J., Feldner, M. T., Lewis, S. F., Fauber, A. L., Leen-Feldner, E. W.,
et al. (2005). Anxiety sensitivity taxon and trauma: Discriminant associations for posttrau-
matic stress and panic symptomatology among young adults. Depression and Anxiety, 22,
Blake, F., Salkovskis, P., Gath, D., Day, A., & Garrod, A. (1998). Cognitive therapy for
premenstrual syndrome: a controlled trial. Journal of Psychosomatic Research, 45(4),
Bloch, M., Schmidt, P. J., & Rubinow, D. R. (1997). Premenstrual syndrome: Evidence for
symptom stability across cycles. American Journal of Psychiatry, 154, 1741–1746.
Boyle, G. J., & Grant, A. F. (1992). Prospective versus retrospective assessment of menstrual
cycle symptoms and moods: Role of attitudes and beliefs. Journal of Psychopathology and
Behavioral Assessment, 14(4), 307–321.
Breier, A., Charney, D. S., & Heninger, G. R. (1986). Agoraphobia with panic attacks:
Development, diagnostic stability, and course of illness. Archives of General Psychiatry,
43, 1029–1036.
Brooks-Gunn, J., & Ruble, D. N. (1986). Men’s and women’s attitudes and beliefs about the
menstrual cycle. Sex Roles, 14(5/6), 287–299.
200 S. T. Sigmon, J. G. Schartel

Busch, C. M., Costa, P. T., Whitehead, W. E., & Heller, B. R. (1988). Severe perimenstrual
symptoms: Prevalence and effects on absenteeism and health care seeking in a non-clinical
sample. Women and Health, 14, 59–74.
Cameron, O. G., Kuttesch, D., McPhee, K., & Curtis, G. C. (1988). Menstrual fluctuation in
the symptoms of panic anxiety. Journal of Affective Disorders, 15, 169–174.
Carpenter, M. J., Upadhyaya, H. P., LaRowe, S. D., Saladin, M. E., & Brady, K. T. (2006).
Menstrual cycle phase effects on nicotine withdrawal and cigarette craving: A review.
Nicotine & Tobacco Research, 8, 627–638.
Chrisler, J. C., & Caplan, P. (2002). The strange case of Dr. Jekyll and Ms. Hyde: How PMS
became a cultural phenomenon and a psychiatric disorder. Annual Review of Sex Research,
13, 274–306.
Christensen, A. P., & Oei, T. P. S. (1989). Correlates of confirmed premenstrual dysphoria.
Journal of Psychosomatic Research, 33(3), 307–313.
Christensen, A. P., & Oei, T. (1995). The efficacy of cognitive behaviour therapy in treating
premenstrual dysphoric changes. Journal of Affective Disorders, 33(1), 57–63.
Clare, A. W. (1983). Psychiatric and social aspects of premenstrual complaint. Psychological
Medicine, 4, 1–59.
Clayton, A. H. (2003). Premenstrual dysphoric disorder: diagnosis and interventions. Primary
Psychiatry, 10(8), 33–36.
Cook, B. L., Noyes, R., Garvey, M. J., Beach, V., Sobotka, J., & Chaudhry, D. (1990). Anxiety
and the menstrual cycle in panic disorder. Journal of Affective Disorders, 19, 221–226.
Davydov, D. M., Shapiro, D., & Goldstein, I. B. (2004). Moods in everyday situations:
Effects of menstrual cycle, work, and personality. Journal of Psychosomatic Research, 56,
Dimmock, P. W., Wyatt, K. M., Jones, P. W., & O’Brien, P. M. (2000). Efficacy of selective
serotonin-reuptake inhibitors in premenstrual syndrome: A systematic review. Lancet,
356, 1131–1136.
Dubrovsky, B. (2006). Neurosteroids, neuroactive steroids, and symptoms of affective
disorders. Pharmacology, Biochemistry, and Behavior, 84, 644–655.
Elliott, T. R., & Harkins, S. W. (1992). Emotional distress and the perceived interference of
menstruation. Journal of Psychopathology and Behavioral Assessment, 14(3), 293–306.
Endicott, J., & Halbreich, U. (1982). Retrospective reports of premenstrual depressive changes:
factors affecting confirmation by daily ratings. Psychopharmacological Bulletin, 18, 109–112.
Endicott, J., Nee, J., Cohen, J., & Halbreich, U. (1986). Premenstrual changes: patterns and
correlates of daily ratings. Journal of Affective Disorders, 10, 127–135.
Endicott, J., Nee, J., & Harrison, W. (2006). Daily Record of Severity of Problems (DRSP):
reliability and validity. Archives of Women’s Mental Health, 9, 41–49.
Epstein, E. E., Rhines, K. C., Cook, S., Zdep-Mattocks, B., Jensen, N. K., & McCrady, B. S.
(2006). Changes in alcohol craving and consumption by phase of menstrual cycle in
alcohol dependent women. Journal of Substance Use, 11, 323–332.
Facchinetti, F., Tarabusi, M., & Nappi, G. (1998). Premenstrual syndrome and anxiety
disorders: A psychobiological link. Psychotherapy Psychosomatic, 67, 57–60.
Fava, M., Pedrazzi, F., Guaraldi, G. P., Romano, G., Genazzani, A. R., & Facchinette, F.
(1992). Comorbid anxiety and depression among patients with late luteal phase dysphoric
disorder. Journal of Anxiety Disorders, 6, 325–335.
Fontana, A. M., & Badawy, S. (1997). Perceptual and coping processes across the menstrual
cycle: An investigation in a premenstrual syndrome clinic and a community sample.
Behavioral Medicine, 22, 152–159.
Fontana, A. M., & Palfai, T. G. (1994). Psychosocial factors in premenstrual dysphoria:
Stressors, appraisal, and coping processes. Journal of Psychosomatic Research,
38, 557–567.
Freeman, E. W. (2003). Premenstrual syndrome and premenstrual dysphoric disorder: defini-
tions and diagnosis. Psychoneuroendocrinology, 28, 25–37.
Anxiety and the Menstrual Cycle 201

Freeman, E. W., Rickels, K., Sondheimer, S. J., & Polansky, M. (1995). A double-blind trial
of oral progesterone, alprazolam, and placebo in treatment of severe premenstrual
syndrome. JAMA, 274, 51–57.
Gallant, S. J., Hamilton, J. A., Popiel, D. A., Morokoff, P. J., & Chakraborty, P. K. (1991).
Daily moods and symptoms: Effects of awareness of study focus, gender, menstrual-cycle
phase, and day of the week. Health Psychology, 10(3), 180–189.
Giannini, A. J., Price, W. A., Loiselle, R. H., & Giannini, M. C. (1985). Pseudocholinester-
aseand trait anxiety in premenstrual tension. The Journal of Clinical Psychiatry, 46,
Golub, S. (1976). The magnitude of premenstrual anxiety and depression. Psychosomatic
Medicine, 38, 4–12.
Golub, S., & Harrington, D. M. (1981). Premenstrual and menstrual mood changes in
adolescent women. Journal of Personality and Social Psychology, 41(5), 961–965.
Goodale, I. L., Domar, A. D., & Benson, H. (1990). Alleviation of premenstrual syndrome
symptoms with the relaxation response. Obstetrics and Gynecology, 75(4), 649–655.
Griffin, M. L., Mello, N. K., Mendelson, J. H., & Lex, B. W. (1987). Alcohol use across the
menstrual cycle among marijuana users. Alcohol, 4, 457–462.
Halbreich, U. (2003). The etiology, biology, and evolving pathology of premenstrual
syndromes. Psychoneuroendocrinology, 28, 55–99.
Halbreich, U., Borenstein, J., Pearlstein, T., & Kahn, L. S. (2003). The prevalence, impair-
ment, impact, and burden of premenstrual dysphoric disorder (PMS/PMDD). Psycho-
neuroendocrinology, 28, 1–23.
Halbreich, U., Endicott, J., & Schact, S. (1982). Premenstrual syndromes: A new instrument
for their assessment. Journal of Psychiatric Treatment & Evaluation, 4, 161–164.
Hammarbäck, S., Damber, J. E., & Bäckström, T. (1989). Relationship between symptom
severity and hormone changes in women with premenstrual syndrome. The Journal of
Clinical Endocrinology and Metabolism, 68, 125–130.
Hardie, E. A. (1997). Prevalence and predictors of cyclic and noncyclic affective change.
Psychology of Women Quarterly, 21, 299–314.
Haskett, R. F., Steiner, M., & Carroll, B. J. (1984). A psychoendocrine study of premenstrual
tension syndrome: A model for endogenous depression? Journal of Affective Disorders,
6, 191–199.
Haywood, A., Slade, P., & King, H. (2002). Assessing the assessment measures for menstrual
cycle symptoms. A guide for researchers and clinicians. Journal of Psychosomatic
Research, 52, 223–237.
Hartlage, S. A., & Gehlert, S. (2001). Differentiating premenstrual dysphoric disorder from
premenstrual exacerbations of other disorders: A methods dilemma. Clinical Psychology
of Science Practice, 8, 242–253.
Harvey, S. M., & Beckman, L. J. (1985). Cyclic fluctuation in alcohol consumption among
female social drinkers. Clinical and Experimental Research, 9, 465–467.
Hendrick, V., Altshuler, L. L., & Burt, V. K. (1996). Course of psychiatric disorders across the
menstrual cycle. Harvard Review of Psychiatry, 4, 200–207.
Hightower, M. (1997). Effects of exercise participation on menstrual pain and symptoms.
Women & Health, 26(4), 15–27.
Hillard, P. J., & Deitch, H. R. (2005). Menstrual disorders in the college age female. Pediatric
Clinics of North America. 52, 179–197.
Hsiao, M., Hsiao, C., & Liu, C. (2004). Premenstrual symptoms and premenstrual exacerba-
tion in patients with psychiatric disorders. Psychiatry & Clinical Neurosciences,
58, 186–190.
Hunter, M. S., Ussher, J. M., Cariss, M., Browne, S., Jelley, R., & Katz, M. (2002). Medical
(fluoxetine) and psychological (cognitive-behavioural therapy) treatment for premenstr-
ual dysphoric disorder. A study of treatment processes. Journal of Psychosomatic
Research, 53, 811–817.
202 S. T. Sigmon, J. G. Schartel

Hylan, T. R., Sundell, K., & Judge, R. (1999). The impact of premenstrual symptomatology
on functioning and treatment-seeking behavior: Experience from the United States,
United Kingdom, and France. Journal of Women’s Health and Gender-Based Medicine,
8, 1043–1052.
Johnson, S. R. (2004). The epidemiology of premenstrual syndrome. Primary Psychiatry, 11,
Kaspi, S. P., Otto, M. W., Pollack, M. H., Eppinger, S., & Rosenbaum, J. F. (1994).
Premenstrual exacerbation of symptoms in women with panic disorder. Journal of Anxiety
Disorders, 8(2), 131–138.
Kent, J. M., Papp, L. A., Martinez, J. M., Browne, S. T., Coplan, J. D., Klein, D. F., et al.
(2001). Specificity of panic response to CO2 inhalation in panic disorder: A comparison
with major depression and premenstrual dysphoric disorder. American Journal of
Psychiatry, 158, 58–67.
Kessel, B. (2000). Premenstrual syndrome: Advances in diagnosis and treatment. Obstetrics
and Gynecology Clinics of North America, 27, 625–639.
Kim, D. R., Gyulai, L., Freeman, E. W., Morrison, M. F., Baldassano, C., & Dube, B. (2004).
Premenstrual dysphoric disorder and psychiatric co-morbidity. Archives of Women’s
Health, 7, 37–47.
Kirkby, R. J. (1994). Changes in premenstrual symptoms and irrational thinking following
cognitive-behavioral coping skills training. Journal of Consulting and Clinical Psychology,
62, 1026–1032.
Kirkby, R. J., & Lindner, H. (1998). Exercise is linked to reductions in anxiety but not
premenstrual syndrome in women with prospectively-assessed symptoms. Psychology,
Health, & Medicine, 3, 211–222.
Klein, D. F. (1993). False suffocation alarms, spontaneous panics, and related conditions: An
integrative hypothesis. Archives of General Psychiatry, 50, 306–317.
Koeske, R. D. (1983). Lifting the curse of menstruation: Toward a feminist perspective of the
menstrual cycle. Women and Health, 8, 1–16.
Kornstein, S. G., & Smith, K. C. (2004). Antidepressant treatment of premenstrual syndrome
and premenstrual dysphoric disorder. Primary Psychiatry, 11(12), 53–57.
Kritz-Silverstein, D., Wingard, D. L., & Garland, F. C. (1999). The association of behavior
and lifestyle factors with menstrual symptoms. Journal of Women’s Health and Gender-
Based Medicine, 8, 1185–1193.
Labad, J., Menchon, J. M., Alonso, P., Segalas, C., Jimenex, S., & Vallejo, J. (2005). Female
reproductive cycle and obsessive-compulsive disorder. Journal of Clinical Psychiatry, 66,
Lancer, R. (2005). The effect of aerobic exercise on obsessive compulsive disorder, anxiety,
and depression. Dissertation Abstracts International: Section B: The Sciences and
Engineering, 66(1-B), 599.
Landén, M., & Eriksson, E. (2003). How does premenstrual dysphoric disorder relate to
depression and anxiety disorders? Depression and Anxiety, 17, 122–129.
Layton, C. (1989). Personality, anxiety and general mental health fluctuation, before and
after menstruation in an adolescent group. Personality and Individual Differences, 10,
Le Mellédo, J-M., & Baker, G. B. (2002). Neuroactive steroids and anxiety disorders. Journal
of Psychiatry Neuroscience, 27, 161–165.
Logue, C. M., & Moos, R. H. (1986). Perimenstrual symptoms: Prevalence and risk factors.
Psychosomatic Medicine, 48(6), 388–414.
MacKay, R. A. (1985). Caffeine-containing beverages and premenstrual syndrome in young
women. American Journal of Public Health, 75, 1335–1337.
Magos, A. L., Brincat, M., & Studd, J. W. W. (1986). Trend analysis of the symptoms of 150
women with a history of the premenstrual syndrome. American Journal of Obstetric
Gynecology, 155, 277–282.
Anxiety and the Menstrual Cycle 203

Manger, T. A. (2005). The impact of an exercise program on posttraumatic stress disorder,

anxiety, and depression. International Journal of Emergency Mental Health, 7(1), 49–57.
Marván, M. L., & Cortés-Iniestra, S. (2001). Women’s beliefs about the prevalence of
premenstrual syndrome and biases in recall of premenstrual changes. Health Psychology,
20, 276–280.
McFarland, C., Ross, M., & DeCourville, N. (1989). Women’s theories of menstruation and
biases in recall of menstrual symptoms. Journal of Personality and Social Psychology, 57,
McLeod, D. R., Hoehn-Saric, R., Foster, G. V., & Hipsley, P. A. (1993). The influence of
premenstrual syndrome on ratings of anxiety in women with generalized anxiety disorder.
Acta Psychiatrica Scandinavica, 88, 248–251.
Mira, M., Vizzard, J., & Abraham, S. (1985). Personality characteristics in the menstrual
cycle. Journal of Psychosomatic Obstetrics and Gynecology, 4, 329–334.
Moos, R. H. (1968). The development of a menstrual distress questionnaire. Psychosomatic
Medicine, 30, 853–867.
Morse, G. (1999). Positively reframing perceptions of the menstrual cycle among women with
premenstrual syndrome. Journal of Obstetrics and Gynecological Neonatal Nursing, 28(2),
Mortola, J. F., Girton, L., Beck, L., & Yen, S. S. (1990). Diagnosis of premenstrual syndrome
by a simple, prospective, and reliable instrument: The calendar of premenstrual experi-
ences. Obstetrics and Gynecology, 76, 302–307.
Ninan, P. T., Insel, T. M., Cohen, R. M., Cook, J. M., Skolnick, P., & Paul, S. M. (1982).
Benzodiazepine receptor-mediated experimental "anxiety" in primates. Science, 218,
O’hara, P., Portser, S. A., & Anderson, B. P. (1989). The influence of menstrual cycle changes
on the tobacco withdrawal syndrome in women. Addictive Behavior, 14, 595–600.
Olsen, R. W., & Sapp, D. W. (1995). Neuroactive steroid modulation of GABAa receptors.
Advances in Biochemical Psychopharmacology, 48, 57–74.
Pearlstein, T. B., Frank, E., Rivera-Tover, A., Thoft, J. S. et al. (1990). Prevalence of Axis I
and Axis II disorders in women with late luteal phase dysphoric disorder. Journal of
Affective Disorders, 20, 129–134.
Perkonigg, A., Yonkers, K. A., Pfister, H., Lieb, R., & Wittchen, H. U. (2004). Risk factors
for premenstrual dysphoric disorder in a community sample of young women: the role of
traumatic events and posttraumatic stress disorder. Journal of Clinical Psychiatry, 65,
Petruzello, S. J., Landers, D. M., Hatfield, B. D., Kubitz, K. A., & Salazar, W. (1991). A
meta-analysis on the anxiety-reducing effects of acute and chronic exercise. Outcomes of
mechanisms. Sports Medicine, 11, 143–182.
Pomerleau, C. S. (1996). Smoking and nicotine replacement treatment issues specific to
women. American Journal of Health Behavior, 20, 291–299.
Poromaa, I. S., Smith, S., & Gulinello, M. (2003). GABA receptors, progesterone, and
premenstrual dysphoric disorder. Archives of Women’s Mental Health, 6, 23–41.
Protopopescu, X., Pan, H., Altemus, M., Tuescher, O., Polanecsky, M., McEwen, B., et al.
(2005). Orbitofrontal cortex activity related to emotional processing changes across the
menstrual cycle. Proceedings of the National Academy of Sciences of the United States of
America, 102, 16060–16065.
Rapkin, A. (2003). A review of treatment of premenstrual syndrome & premenstrual dysphoric
disorder. Psychoneuroendocrinology, 28, 39–53.
Rapkin, A. J., Chang, L. C., & Reading, A. E. (1988). Comparison of retrospective and
prospective assessment of premenstrual symptoms. Psychological Reports, 62, 55–60.
Reiss, S., & McNally, R. J. (1985). The expectancy model of fear. In S. Reiss &
R. R. Bootzin (Eds.), Theoretical issues in behavior therapy, 107–121. San Diego:
Academic Press.
204 S. T. Sigmon, J. G. Schartel

Ross, L. E., & Steiner, M. (2003). A biopsychosocial approach to premenstrual dysphoric

disorder. Psychiatric Clinics of North America, 26, 529–546.
Roy-Byrne, P. P., Cowley, D. S., Greenblatt, D. J., Shader, R. I., & Hommer, D. (1990).
Reduced benzodiazepine sensitivity in panic disorder. Archives of General Psychiatry, 47,
Rubinow, D. R. (1985). Premenstrual syndromes: Past and future research strategies. Cana-
dian Journal of Psychiatry, 30, 469–473.
Rubinow, D. R., Hoban, M. C., Grover, G. N., Galloway, D. S., Roy-Byrne, P., Andersen, R.,
et al. (1988). Changes in plasma hormones across the menstrual cycle in patients with
menstrually related mood disorder and in control subjects. American Journal of Obstetrics
and Gynecology, 158, 5–11.
Ruble, D. N. (1977). Premenstrual symptoms: A reinterpretation. Science, 197, 291–292.
Sampson, G. A., & Jenner, F. A. (1977). Studies of daily recordings from the Moos Menstrual
Distress Questionnaire. British Journal of Psychiatry, 130, 265–271.
Schmidt, N. B., Kotov, R., Lerew, D. R., Joiner, T. E., & Ialongo, N. S. (2005). Evaluating
latent discontinuity in cognitive vulnerability to panic: A taxometric investigation.
Cognitive Therapy and Research, 29, 673–690.
Schmidt, N. B., Lerew, D. R., & Trakowski, J. H. (1997). Body vigilance in panic disorder:
Evaluating attention to bodily pertubations. Journal of Consulting and Clinical Psychol-
ogy, 65, 214–220.
Slade, P. (1984). Premenstrual emotional changes in normal women: Fact or fiction? Journal
of Psychosomatic Research, 28(1), 1–7.
Sloss, E. M., & Frerichs, R. R. (1983). Smoking and menstrual disorders. International
Journal of Epidemiology, 12, 107–109.
Strine, T. W., Chapman, D. P., & Ahluwalia, I. B. (2005). Menstrual-related problems and
psychological distress among women in the United States. Journal of Women’s Health, 14,
Strine, T. W., Chapman, D. P., Kobau, R., & Balluz, L. (2005). Associations of self-reported
anxiety symptoms with health-related quality of life and health behaviors. Social Psychiatry
and Psychiatric Epidemiology, 40, 432–438.
Sigmon, S. T., Whitcomb-Smith, S. R., Rohan, K. J., & Kendrew, J. J. (2004). The role of
anxiety level, coping styles, and cycle phase in menstrual distress. Journal of Anxiety
Disorders, 18, 177–191.
Sigmon, S. T., Dorhofer, D. M., Rohan, K. J., & Boulard, N. E. (2000). The impact of
activity sensitivity, bodily expectations, and cultural beliefs on menstrual symptoms
reporting: A test of the menstrual reactivity hypothesis. Journal of Anxiety Disorders, 14,
Sigmon, S. T., Rohan, K. J., Boulard, N. E., Dorhofer, D. M., & Whitcomb, S. R. (2000).
Menstrual reactivity: The role of gender-specificity, anxiety sensitivity, and somatic
concerns in self-reported menstrual distress. Sex Roles, 43, 143–161.
Sigmon, S. T., Fink, C. M., Rohan, K., & Hotovy, L. A. (1996). Anxiety sensitivity and
menstrual cycle reactivity: Psychophysiological and self-report differences. Journal of
Anxiety Disorders, 10, 393–410.
Smith, M. J., Schmidt, P. J., & Rubinow, D. R. (2003). Operationalizing DSM-IV criteria for
PMDD: selecting symptomatic and asymptomatic cycles for research. Journal of Psychiatric
Research, 37, 75–83.
Stein, M. B., Schmidt, P. J., Rubinow, D. R., & Uhde, T. W. (1989). Panic disorder and the
menstrual cycle: panic disorder patients, healthy control subjects, and patients with
premenstrual syndrome. American Journal of Psychiatry, 146, 1299–1303.
Steiner, M. (2000). Premenstrual syndrome and premenstrual dysphoric disorder: Guidelines
for management. Journal of Psychiatry & Neuroscience, 25, 459–468.
Steiner, M., & Born, L. (2000). Diagnosis and treatment of premenstrual dysphoric disorder:
an update. International Clinical Psychopharmacology, 15(s3), S5–S17.
Anxiety and the Menstrual Cycle 205

Steiner, M., Haskett, R. F., & Carroll, B. J. (1980). Premenstrual tension syndrome: The
development of research diagnostic criteria and new rating scales. Acta Psychiatrica
Scandinavica, 62, 177–190.
Stout, A. L., Steege, J. F., Blazer, D. G., & George, L. K. (1986). Comparison of lifetime
psychiatric diagnoses in premenstrual syndrome clinic and community samples. The
Journal of Nervous and Mental Disease, 174, 517–522.
Strine, T. W., Chapman, D. P., Kobau, R., & Balluz, L. (2005). Associations of self-reported
anxiety symptoms with health-related quality of life and health behaviors. Social Psychiatry
and Psychiatric Epidemiology, 40, 432–438.
Symonds C. S., Gallagher, P., Thompson, J. M., & Young, A. H. (2004). Effects of the
menstrual cycle on mood, neurocognitive and neuroendocrine function in healthy pre-
menopausal women. Psychological Medicine, 34, 93–102.
Targum, S. D., Caputo, K. P., & Ball, S. K. (1991). Menstrual cycle phase and psychiatric
admissions. Journal of Affective Disorders, 22, 49–53.
Taylor, J. W. (1979). The timing of menstruation-related symptoms associated by a daily
symptom rating scale. Acta Psychiatrica Scandinavica, 60, 87–105.
Teperi, J. & Rimpela, M. (1989). Menstrual Pain, health and behaviour in girls. Social
Sciences Medicine, 29, 163–169.
Thys-Jacobs, S., Alvir, J. M., & Fratarcangelo, P. (1995). Comparative analysis of three pms
assessment instruments: The identification of premenstrual syndrome core symptoms.
Psychopharmacology Bulletin, 31, 389–396.
Tobin, M. B., Schmidt, P. J., & Rubinow, D. R. (1994). Reported alcohol use in women with
premenstrual syndrome. American Journal of Psychiatry, 151, 1503–1504.
Ussher, J. (2003). The role of premenstrual dysphoric disorder in the subjectification of
women. Journal of Medical Humanities, 24, 131–146.
Veeninga, A. T., & Kraaimaat, F. W. (1995). Causal attributions in premenstrual syndrome.
Psychology and Health, 10, 219–228.
Veeninga, A. T., de Rutter, C., & Kraaimaat, F. W. (1994). The relationship between late luteal
phase dysphoric disorder and anxiety disorders. Journal of Anxiety Disorders, 8, 207–215.
Vickers, K., & McNally, R. J. (2004). Is premenstrual dysphoria a variant of panic disorder? A
review. Clinical Psychology Review, 24, 933–956.
Vulink, N., C., Denys, D., Bus, L., & Westenberg, H. G. (2006). Female hormones affect
symptom severity in obsessive-compulsive disorder. International Clinical Psychopharma-
cology, 21, 171–175.
Watts, S., Dennerstein, L., & Horne, D. J. (1980). The premenstrual syndrome: A psycholo-
gical evaluation. Journal of Affective Disorders, 2, 257–266.
Weems, C. F., Hayward, C., Killen, J., & Taylor, C. B. (2002). A longitudinal investigation of
anxiety sensitivity in adolescence. Journal of Abnormal Psychology, 111, 471–477.
Williams, K. E.., & Koran, L. M. (1997). Obsessive-compulsive disorder in pregnancy, the
puerperium, and the premenstrum. Journal of Clinical Psychiatry, 58(7), 330–334.
Wittchen, H. U., Becker, E., Lieb, R., & Krause, P. (2002). Prevalence, incidence and stability
of premenstrual dysphoric disorder in the community. Psychological Medicine, 32,
Woods, N. F., Mitchell, E. S., & Lentz, M. J. (1995). Social pathways to premenstrual
symptoms. Research in Nursing & Health, 18, 225–237.
Woods, N. F., Taylor, D., Mitchell, E. S., & Lentz, M. J. (1992). Perimenstrual symptoms and
health-seeking behavior. Western Journal of Nursing Research, 14, 418–443.
Yonkers, K. A. (1997). Anxiety symptoms and anxiety disorders: How are they related to
premenstrual disorders? Journal of Clinical Psychiatry, 58(3), 62–69.
Pain and Anxiety Disorders

Gordon J. G. Asmundson, Murray P. Abrams, and Kelsey C. Collimore


There is growing evidence that chronic pain, typically that which is associated
with the musculoskeletal system (e.g., arthritis, low back pain, fibromyalgia),
frequently co-occurs with the anxiety disorders. This co-occurrence is often
overlooked in practice because it is neither standard protocol nor obvious that
clinicians consider pain experiences in the context of screening or assessment of
anxiety disorders. Yet, people with both an anxiety disorder and chronic
musculoskeletal pain typically present with greater distress and functional
impairment compared to those with only one of these conditions. As a result,
both assessment and treatment can be complicated. The goals of this chapter
are several. First, we review the core characteristics of acute and chronic pain,
and present data regarding the extent of its co-occurrence with the anxiety
disorders. Second, we summarize models that have been offered to explain the
co-occurrence of these conditions. Third, we review evidence supporting the
postulates of these models. Fourth, we discuss practical issues that are intended
to improve assessment, treatment, and outcomes for people who present with
an anxiety disorder accompanied by clinically significant pain. Much of the
discussion focuses on the relationship between chronic musculoskeletal pain
and posttraumatic stress disorder (PTSD), as it is the anxiety disorder that has
received the most empirical attention in this context. We conclude with a brief
outline of issues that warrant future research attention.

Gordon J. G. Asmundson
Anxiety and Illness Behaviours Laboratory, University of Regina, Regina, Saskatchewan,
S4S 0A2, Tel: (306) 347-2415, Fax: (306) 585-4854

M. J. Zvolensky, J. A. Smits (eds.), Anxiety in Health Behaviors and Physical Illness. 207
Ó Springer 2008
208 G. J. G. Asmundson et al.

Understanding Pain

Pain was once conceptualized strictly as a sensory experience resulting from

stimulation of specific noxious receptors, such as might occur at the time of
physical injury or from progressive disease. We now understand that pain is
more than sensation. Contemporary models of pain recognize that it is a
complex perceptual experience that is determined by sensory as well as psycho-
logical (i.e., thinking, emotions, behaviors) and social influences (Asmundson
& Wright, 2004). Generally speaking, we experience pain in order to adapt to
and survive in our environment; that is, pain alerts us that potential or actual
tissue damage may be pending, and it motivates us to take action to limit
damage and recover from it (Wall, 1978). Compelling evidence for the adaptive
significance of pain comes from observations of people who have a rare auto-
somal recessive genetic disease called congenital analgesia. These individuals do
not experience pain and, as a consequence, often die in childhood from the
effects of undetected (i.e., painless) life threatening injuries or disease (for
review, see Nagasako, Oaklander, & Dworkin, 2003).
For most people, physical injury or disease is accompanied by pain. This
pain typically abates with recovery. However, for some people pain becomes
chronic (i.e., persists for three months or more; International Association for
the Study of Pain, 1986), losing its adaptive qualities and, instead, causes
considerable emotional distress and impairment of social and occupational
abilities. Many people with chronic pain make frequent physician visits, some-
times undergo inappropriate medical evaluations, and miss work and other
important activities because of their symptoms and associated suffering
(e.g., Gureje, Von Korff, Simon, & Gater, 1998). As a result, chronic pain has
become one of the most common chronic health conditions in North America.
Estimates from the US indicate that approximately 7% of the general popula-
tion has experienced chronic pain in the past 12 months (McWilliams, Cox, &
Enns, 2003) at a cost of about $100 billion annually (Weisberg & Vaillancourt,
1999). While chronic pain is often associated with these negative outcomes, it is
important to note that some people with chronic pain cope effectively with the
pain, and adapt to it in a manner that does not impose limitations on their well-
In recent years it has become increasingly evident that a substantive number
of people who have an anxiety disorder also have chronic pain symptoms.
Likewise, people with chronic pain frequently report significant expressions
of fear and anxiety, often at levels and with impacts that warrant diagnosis of an
anxiety disorder. Fear and anxiety specific to pain, while prevalent in some
people with chronic pain, are not the focus of this chapter; these constructs are
discussed in detail elsewhere (e.g., Asmundson, Vlaeyen, & Crombez, 2004).
Below we review the available data on the epidemiology of co-occurring clini-
cally significant pain and anxiety disorders in both community and treatment-
seeking samples. It is noteworthy that chronic musculoskeletal pain has
Pain and Anxiety 209

received the vast majority of theoretical and empirical attention with regard to
co-occurrence with the anxiety disorders; thus, unless otherwise indicated, it
will be the focus of much of the discussion that follows.


Prevalence of Anxiety Disorders in Pain Samples

Investigators have consistently observed that rates of some anxiety disorders are
elevated in people with chronic musculoskeletal pain. Table 1 and Table 2 show
the 12-month prevalence of various anxiety disorders in people reporting
primarily chronic musculoskeletal or migraine headache pain in community
and treatment seeking samples, respectively. In community samples, the most
prevalent past 12-month anxiety disorders were specific phobia (formerly
simple phobia; ranging from 12.5% to 15.7%), social anxiety disorder (SAD;
ranging from 8.3% to 11.8%), and PTSD (ranging from 7.3% to 10.7%). This
pattern of findings is consistent with, but higher than, the general US popula-
tion 12-month prevalence rates (i.e., specific phobia, 8.7%; SAD, 6.8%; PTSD,
3.5%; Kessler, Chiu, Demler, & Walters, 2005). In a large (N=85,088) survey
of community dwelling adults from 17 countries, Demyttenaere et al. (2007)
most recently reported pooled results indicating that those with back or neck
pain, compared to those without, were approximately two times more likely to
have had past 12-month panic disorder (PD)/agoraphobia and SAD, and
almost three times more likely to have had generalized anxiety disorder
(GAD) or PTSD. Raphael, Janal, Nayak, Schwartz, and Gallagher (2006)
have likewise reported that community-dwelling women with fibromyalgia
were five times more likely than others to have had a lifetime diagnosis of
obsessive-compulsive disorder (OCD), four times more likely to have had
PTSD, and more than four times as likely to have had GAD.
In treatment seeking samples the most prevalent past 12-month anxiety
disorders were phobic disorders (including SAD; ranging from 9% to 13%),
followed generally by GAD (ranging from 0% to 13.4%) and PD (ranging from
1% to 7.2 %). Twelve-month prevalence of any anxiety disorder was 26.5% to
35.1% in community samples with chronic pain, and 8 to 28.8% in treatment
seeking samples with chronic pain, both elevated relative to the general popula-
tion (18.1%; Kessler, Chiu, et al., 2005). Lifetime prevalence of various anxiety
disorders reported by patients with chronic musculoskeletal pain, as illustrated
in Table 2, have been elevated relative to the general population (Kessler,
Berglund, Demler, Jin, & Walters, 2005) in some (Atkinson, Slater, Patterson,
Grant, & Garfin, 1991), but not all (Kinney, Gatchel, Polatin, Fogarty, &
Mayer, 1993; Polatin, Kinney, Gatchel, Lilio, & Mayer, 1993), studies.
It is noteworthy that the 12-month and lifetime prevalence rates for PTSD in
the treatment seeking chronic pain samples described in Table 2 were neither
Table 1 Prevalence of anxiety disorders among persons with pain (community samples)

Study Participants Data set Diagnostic criteria % Meeting criteria for an anxiety disorder
McWilliams et Nationally National DSM-III-R Chronic pain (arthritis; n = 382) General population (n = 5495)
al. (2003) representative Comorbidity Any anxiety disorder 35.1 Any anxiety disorder 18.1
sample of the Survey Part II PD 6.5 PD 1.9
U. S. (NCS) AWHPD 8.4 AWHPD 3.3
population (n SP 11.8 SP 7.8
= 5877)
SiP 15.7 SiP 8.3
PTSD 10.7 PTSD 3.3
GAD 7.3 GAD 2.6
McWilliams et Nationally Midlife DSM-III-R Arthritis (n = 588) No arthritis
al. (2004) representative Development in Panic attacks 11.2 Panic attacks 5.8
sample of the the United States GAD 5.6 GAD 2.7
U. S. Survey (MIDUS) Migraine (n = 340) No migraine
population (n Panic attacks 17.4 Panic attacks 5.5
= 3032)
GAD 9.1 GAD 2.5
Back pain (n = 614) No back pain
Panic attacks 13.0 Panic attacks 5.3
GAD 6.2 GAD 2.5
Von Korff et al. Nationally National DSM-IV Chronic spinal pain (n = 2397) -
(2005) representative Comorbidity Any anxiety disorder 26.5
sample of the Survey PD 4.8
U. S. Replication AWHPD 1.3
population (n (NCS-R) SP 8.3
= 5692)
SiP 12.5
PTSD 7.3
GAD 6.4
Note: PD = Panic Disorder; AWHPD = Agoraphobia Without History of Panic Disorder; SP = Social Phobia (also called social anxiety disorder);
SiP = Simple Phobia; PTSD = Posttraumatic Stress Disorder; GAD = Generalized Anxiety Disorder; Not all studies evaluated all anxiety disorders;
G. J. G. Asmundson et al.

All studies based on 12-month prevalence.

Table 2 Prevalence of anxiety disorders among pain patients (clinical samples)
Study Participants Criteria Lifetime prevalence (%) Current prevalence (%)
Pain and Anxiety

Reich et al. (1983) Mixed chronic pain patients (n = 43) DSM-III - Any anxiety disorder 7.0
(PTSD only)
Katon et al. (1985) Mixed chronic pain patients (n = 37) DSM-III - Any anxiety disorder 16.2
(PD only)
Large (1986) Mixed chronic pain patients (n = 50) DSM-III - Any anxiety disorder 8.0
GAD 4.0
PD 2.0
PTSD 2.0
Fishbain et al. Mixed chronic pain patients (n = 283) DSM-III - Any anxiety disorder 19.4
Atkinson et al. Chronic low back pain patients (n = 97) DSM-III Any anxiety 37.9 Any anxiety disorder 28.8
(1991) disorder
Generalized 22.7 Generalized disorder 13.4
PD 8.2 PD 7.2
OCD 13.4 OCD 8.2
Kinney et al. (1993) Chronic back pain patients (n = 90) DSM-III-R Any anxiety 29.0 Any anxiety disorder 25.0
PD 3.0 PD 3.0
Phobic disorders 17.0 Phobic disorders 13.0
OCD 3.0 OCD 3.0
PTSD 2.0 PTSD 2.0
GAD 4.0 GAD 4.0

Table 2 (continued)
Study Participants Criteria Lifetime prevalence (%) Current prevalence (%)
Polatin et al. (1993) Chronic low back pain patients (n = DSM-III-R Any anxiety 19.0 Any anxiety disorder 17.0
200) disorder
PD 3.0 PD 3.0
Phobic disorders 11.0 Phobic disorders 9.0
OCD 2.0 OCD 2.0
PTSD 1.0 PTSD 1.0
GAD 2.0 GAD 2.0
Asmundson et al. Chronic musculoskeletal pain patients DSM-IV - Any anxiety disorder 17.0
(1996) (n = 200) PD 2.1
SP 11.0
SiP 2.7
OCD 0.0
PTSD 2.1
GAD 0.0
Note: PD = Panic Disorder; OCD = Obsessive-Compulsive Disorder; PTSD = Posttraumatic Stress Disorder; GAD = Generalized Anxiety
Disorder; SP = Social Phobia (also called social anxiety disorder); SiP = Simple Phobia; Not all studies evaluated all anxiety disorders.
G. J. G. Asmundson et al.
Pain and Anxiety 213

remarkable nor elevated relative to the general population. This may be a

product of the nature of structured assessments employed in these studies;
specifically, the PTSD module may have been skipped if the participant did
not respond affirmatively to introductory questions regarding exposure to
traumatic experiences, examples of which do not often include accidental or
painful injury. Contrary to these unremarkable findings, several comprehensive
reviews of the literature indicate that 10% to 50% of patients receiving rehabi-
litation services for chronic musculoskeletal pain and related conditions have
PTSD (Asmundson, Coons, Taylor, & Katz, 2002; Otis, Keane, & Kerns, 2003).
Moreover, the available data indicate that up to 45% of patients with pain
subsequent to burn-related injury exhibit significant posttraumatic stress symp-
toms (Saxe et al., 2001).
Overall, it appears that there is a high prevalence of some anxiety disorders in
people with conditions characterized by chronic musculoskeletal pain. The
most common appear to be PTSD, SAD, GAD, and possibly PD. However,
there are notable limitations in the scope of inquiry to date. Additional inves-
tigation in samples seeking treatment for various chronic pain conditions, using
structured and comprehensive assessment of current and lifetime occurrence
of each anxiety disorder, is warranted.

Prevalence of Pain in Anxiety Disorder Samples

A growing number of studies have assessed the prevalence of clinically signifi-

cant pain conditions in people with anxiety disorders. Two of these studies have
focused on pain reports in patients with PD. Kuch, Cox, Woszczyna, Swinson,
and Shulman (1991) reported that nearly 40% (54 of 141) of their sample of
consecutively referred patients with PD reported chronic pain, most commonly
in the head, shoulders, and lower back. Almost 10% of those in the sample were
using analgesic medications on a daily basis. Similarly, in a study of 71 patients
with PD, Schmidt and Telch (1997) identified a variety of comorbid physical
conditions, including chronic back problems (46%), arthritis (22%), irritable
bowel syndrome (17%), heart conditions (13%), and other conditions such as
migraine, cancer, and diabetes (24%). A substantial number (35%) had seen a
physician in the last month, and most (89%) in the last year; however, only 6%
had been hospitalized in the past six months. These studies provide preliminary
evidence that chronic pain, particularly that of the musculoskeletal system, is
prevalent in patients seeking treatment for PD.
Acute and chronic pain are reported with striking frequency in people with
PTSD (for review, see Asmundson et al., 2002). Upwards of 20% to 30% of
those with PTSD seeking outpatient treatment from community and mental
health clinics report chronic pain (e.g., Hubbard, Realmuto, Northwood, &
Masten, 1995). Prevalence estimates of current chronic pain have, in most cases,
been even higher in military veterans and volunteer firefighters with PTSD,
214 G. J. G. Asmundson et al.

ranging from 50% to 80% (for recent reviews, see Asmundson et al., 2002; Otis
et al., 2003). Recent findings from a sample of female veterans using the
Veterans Administration (VA) Health Care System confirm that these finding
generalize across gender (Asmundson, Wright, & Stein, 2004). Road traffic
collisions, work-related injury, and service in combat and emergency theatres
are the most common events precipitating the development of PTSD accom-
panied by chronic pain (Asmundson, Norton, Allerdings, Norton, & Larsen,
1998; Beckham et al., 1997; Blanchard & Hickling, 2004).
Sareen, Cox, Clara, and Asmundson (2005) recently used data from the US
National Comorbidity Survey Part II to evaluate associations between the
anxiety disorders and diagnoses of general medical conditions, including
those for which pain is often a significant component. After controlling for
socio-demographic variables and other common mental disorders, robust asso-
ciations were found amongst PTSD, panic attacks, and agoraphobia and the
physical disorders. Most strikingly, individuals with PTSD were more than
twice as likely as others to have a past-year neurological disorder (e.g., multiple
sclerosis), roughly twice as likely to have a past-year gastrointestinal disorder
(e.g., ulcer, hernia), more than three times as likely to have a past-year meta-
bolic or immune disorder (e.g., diabetes, lupus), two and one half times as likely
to have a past-year bone or joint conditions (e.g., arthritis, rheumatisms, other
bone/joint disease), and almost twice as likely as others to have one or more
past-year physical disorders.
Overall, it appears that conditions characterized by some degree of persistent
pain are prevalent in people with anxiety disorders, particularly PTSD and
panic-related conditions. This avenue of inquiry is in its infancy; thus, further
investigation geared toward replication of findings in community and treat-
ment-seeking samples, using comprehensive pain assessment batteries, is


Few studies have systematically investigated the extent to which anxiety dis-
orders exist prior to the onset of pain, or vice versa. There is some evidence to
suggest that anxiety disorders precede the onset of pain. In a sample of injured
workers with chronic musculoskeletal pain, Asmundson, Jacobson, Allerdings,
and Norton (1996) found that in all but one case, the anxiety disorder preceded
the pain complaint. Kinney et al. (1993) found that among 90 chronic low back
pain patients, 23% had a preexisting anxiety disorder. In the only prospective
study to date, we recently demonstrated that PTSD symptoms measured prior
to surgery made a unique contribution to the prediction of post-surgical pain
disability in chronic pain patients undergoing general surgery (Martin, Dzyuba,
Halket, Asmundson, Katz, 2007); thus, PTSD symptoms may be important in
the development and/or maintenance of pain disability. There is also evidence
Pain and Anxiety 215

that the probability of onset of an anxiety disorder before versus after pain
onset is comparable. In a study of 97 chronic back pain patients, 30 of whom
had a comorbid anxiety disorder, 46.7% reported onset of anxiety before pain,
and 53.3% reported onset after pain (Atkinson et al., 1991). Additional
research, particularly that which uses prospective methods, is needed to deline-
ate the temporal sequence of co-occurring anxiety disorders and chronic pain.

Theoretical Overview

The substantial degree of co-occurrence of the anxiety disorders and clinically

significant pain experiences suggests that these conditions are related in some
way. Yet, establishing co-occurrence provides neither an understanding of the
nature of the associations between the conditions nor an understanding of
the mechanisms by which they are linked. There are several possible scenarios
that might explain the relationship. For any two variables (or, in this case,
conditions), possible relationship scenarios are as follows: (1) one causes the
other (i.e., the anxiety disorder causes pain or vice versa), (2) they influence one
another in some mutually maintaining way (e.g., pain exacerbates symptoms of
the anxiety disorder and vice versa), or (3) some third factor (e.g., a common
predisposition, a shared environmental event) increases vulnerability to both.
The second and third possibilities are not mutually exclusive.
There are, to the best of our knowledge, no theoretical positions that expli-
cate the first of the possibilities noted above; nor are there data to support the
position that one condition causes the other. Several models, rooted in the
second and third possibilities, have been recently posited to explain the relation-
ship between the anxiety disorders and chronic pain. For the most part, they
have been developed in the context of efforts to understand mechanisms under-
lying co-occurring PTSD and chronic musculoskeletal pain, and are based
on tenets of empirically supported cognitive-behavioral models of each of
these conditions (e.g., Ehlers & Clark, 2000; Foa & Rothbaum, 1998; Norton
& Asmundson, 2003; Vlaeyen & Linton, 2000). Below we review these models
and speculate on their value in explaining the association between the other
anxiety disorders and chronic musculoskeletal pain.

Mutual Maintenance Model

The mutual maintenance model (Sharp & Harvey, 2001) holds that certain
components of PTSD (e.g., physiological arousal, absence of positive emotions,
avoidance of trauma stimuli) maintain or exacerbate symptoms of pain
and, similarly, that certain components of chronic musculoskeletal pain
(e.g., physiological arousal, catastrophizing about pain, avoidance of physical
exertion) maintain or exacerbate symptoms of PTSD. The model predicts, for
216 G. J. G. Asmundson et al.

example, that pain sensations experienced by a person with chronic musculos-

keletal pain will be persistent and arousal-provoking reminders of the trauma
that precipitated the pain. Physiological arousal in response to recollection of
the trauma will, in turn, promote avoidance of pain-related activities and (over
time) physical deconditioning, which makes the experience of pain more likely.
The person thereby becomes trapped in a vicious cycle whereby the symptoms
of PTSD and chronic musculoskeletal pain interact to produce self-perpetuat-
ing distress and functional disability.

Shared Vulnerability (Diathesis)/Stress Models

We (Asmundson et al., 2002; Asmundson, & Taylor, 2006) and others (Otis
et al. 2003; Turk, 2002) have extended the idea of mutual maintenance, suggest-
ing that some maintenance factors may actually denote a shared vulnerability,
or diathesis, for developing both conditions. The shared vulnerability model
(see Fig. 1) holds that there are individual difference factors, possibly geneti-
cally influenced, that predispose people to develop PTSD and chronic muscu-
loskeletal pain when exposed to certain environmental conditions. Specifically,
the model suggests that the interaction of individual difference characteristics –
a psychological vulnerability for feelings of loss of control (and anxiety), and a
lowered physiological threshold for alarm reactions (i.e., activation of physio-
logical processes that prepare one to fight or flee) to stressors – and instigating

(e.g., high injury
sensitivity, high
anxiety sensitivity)

Autonomic Nervous
System and Muscular

Emotional Disabling
Life Event Avoidance
Response Condition
(e.g. traumatic of situations or activities
(e.g. fear, anxiety, (specific or
incident, injury) perceived as negative
worry, agitation) co-occurring)

Hypervigilance and
Cognitive Biases

Low Threshold for

(e.g., sympathetic

Fig. 1 Shared vulnerability model

Pain and Anxiety 217

stressful events (e.g., traumatic incident, injury) explain the development of

PTSD, chronic musculoskeletal pain, and their co-occurrence. To illustrate,
PTSD is likely to occur in trauma-exposed people who are predisposed to react
to traumatic stressors with alarm and feelings that the situation, and their
emotional response to it, are beyond their control. Similarly, chronic muscu-
loskeletal pain is most likely to develop in injured people who believe that the
pain they are experiencing, and their emotional responses (e.g., anxiety, worry,
fear, agitation) to it, are uncontrollable.
The shared vulnerability model further predicts that co-occurring PTSD and
chronic musculoskeletal pain are most likely to develop in cases where vulner-
able people are exposed to an event that is both traumatic and painful, and that
both reminders of the trauma and sensations of pain can serve as triggers for
further alarm reactions. The latter is consistent with postulates of the mutual
maintenance model and further illustrates how predisposing factors can con-
tribute to maintenance of these conditions.
Our understanding of co-occurring PTSD and chronic pain might apply to
other anxiety disorders that frequently co-occur with chronic pain. Symptoms
of physiological arousal and lack of positive emotions – both general character-
istics of the anxiety disorders – may maintain or exacerbate symptoms of pain.
Likewise, one or more aspects of the pain experience (e.g., physiological arou-
sal, pain-related catastrophizing, avoidance of physical exertion) may maintain
or exacerbate clinically significant symptoms of anxiety. For example, avoid-
ance of physical exertion in persons experiencing persistent pain may result
in avoidance of specific social environments (e.g., fitness center, sporting com-
plex) which may, in turn, contribute to the maintenance of symptoms of social
anxiety through rewards (i.e., anxiety reduction) derived from behavioral
avoidance. Likewise, physiological arousal in these social environments
may promote muscle tension, avoidance of pain-related activities, physical
deconditioning, and an associated increase in muscle pain. As with PTSD,
persons with other anxiety disorders may become trapped in a cycle wherein
symptoms of anxiety and pain interact to promote clinically significant distress
or impairment.


It is plausible that postulates of the mutual maintenance and shared vulner-

ability models will prove useful in delineating why some, if not all, of the anxiety
disorders are often accompanied by conditions characterized by chronic pain
(and vice versa). Efforts to understand the mechanisms of co-occurrence are
only just beginning; however, as summarized below, there is a growing body of
support for these postulates emerging from investigations of co-occurring
PTSD and chronic musculoskeletal pain.
218 G. J. G. Asmundson et al.

Overview of Empirical Support

Evidence supporting the postulates of the mutual maintenance and shared

vulnerability models has been culled in the context of pain that accompanies
PTSD (or vice versa). Below we review this literature, with specific focus on
symptom overlap, anxiety sensitivity (AS), selective attention for threat, and
lowered threshold for alarm. These models have not heretofore been applied in
the context of understanding the co-occurrence of chronic musculoskeletal pain
and the other anxiety disorders. While it is likely that the models will prove
useful in this regard, direct empirical evaluation of the model postulates in these
contexts remain a direction for future consideration.

Symptom Overlap

There is considerable symptom overlap between PTSD and chronic musculos-

keletal pain. Both are characterized by somatic hypervigilance and (possibly)
biases in attention toward threatening stimuli, heightened startle, emotional
numbing (e.g., absence of positive emotion), avoidance, and dysregulations
in stress response and pain modulation (for review, see Asmundson et al.,
2002). These findings indicate that PTSD and chronic musculoskeletal pain
share similar response patterns in the cognitive, behavioral, and physiological
domains. There is also evidence to suggest that particular PTSD symptom
clusters are more closely associated with certain aspects of the pain experience;
for example, re-experiencing symptoms are uniquely associated with pain
severity, self-report of physical symptoms, and limitations in functional ability
(Asmundson, Wright, et al., 2004; Beckham et al., 1997; Zoellner, Goodwin, &
Foa, 2000), whereas hyperarousal is associated with detection of pain
(Asmundson, Wright, McCreary, & Pedlar, 2003).

Anxiety Sensitivity

AS (i.e., fear of anxiety based on the belief that it may have harmful conse-
quences) is one of several individual difference factors that increase sense of
danger and fearful responding. AS is elevated in patients with PTSD (Taylor
et al., 2001, 2003) and in some patients with chronic musculoskeletal pain (for
review, see Asmundson, Wright, & Hadjistavropoulos, 2000), is positively
correlated with the severity of PTSD symptoms (Fedoroff, Taylor, Asmundson,
& Koch, 2000), increases the risk of pain-related avoidance and disability
following physical injury (for review, see Keogh & Asmundson, 2004), and
is partly influenced by genetic factors (Stein, Jang, & Livesley, 1999). Conse-
quently, it has been postulated that AS is responsible for the extreme emotional
Pain and Anxiety 219

responses to trauma and pain associated with injury, and that it denotes the
specific vulnerability that predisposes people to develop both PTSD and
chronic musculoskeletal pain (Asmundson et al., 2002; Asmundson & Taylor,
2006; Turk, 2002). It has yet to be established that elevated AS precedes the
development of PTSD and chronic musculoskeletal pain; thus, it remains a
possibility that AS becomes elevated as a consequence of PTSD and chronic
musculoskeletal pain and thereafter serves to maintain symptoms (see Sharp &
Harvey, 2001). Longitudinal studies are needed to assess these possibilities.
Additional study of other potential vulnerability factors (e.g., illness/injury
sensitivity, fear of pain) is also warranted.

Selective Attention to Threat

Cognitive models suggest that people with various forms of psychopathology

or general medical conditions tend to selectively attend to threat-related stimuli
that are representative of the core concerns of their specific disorder; that is,
they direct attention towards objects or situations that they fear. This increases
state anxiety and, according to some, makes one vulnerable for emotional
disorders (e.g., Mathews & MacLeod, 2002). Evidence for syndrome-specific
attentional biases is, with few exceptions, robust across various anxiety disor-
ders (see Williams, Mathews, & MacLeod, 1996). This has been consistently
demonstrated in patients with PTSD using emotional Stroop colour-naming
and fear-potentiated startle tasks (e.g., Paunovic, Lundh, & Öst, 2002) but not
with the dot-probe task (e.g., Elsesser, Sartory, & Tackenberg, 2005). The
findings from chronic musculoskeletal pain patients have been mixed and
there is no clear consensus amongst modified Stroop and dot-probe investiga-
tions as to whether patients with chronic musculoskeletal pain do or do not
selectively attend to pain-related stimuli.
A recent meta-analysis of five modified Stroop investigations suggests that
chronic pain patients, compared to healthy participants, have an attentional bias
to both sensory and affect pain words (Roelofs, Peters, Zeegers, & Vlaeyen,
2002); however, scrutiny of the findings from the individual investigations used in
the meta-analysis as well as findings from controlled investigations (Andersson &
Haldrup, 2003) fail to provide convincing evidence for this conclusion. The
results of dot-probe investigations, with some exceptions, also fail to provide
clear evidence that chronic pain is associated with a pain-specific bias in
attentional processing (Asmundson, Carleton, & Ekong, 2005; Asmundson,
Kuperus, & Norton, 1997, Asmundson, Wright, & Hadjistavropoulos, 2005;
Roelofs, Peters, Fassaert, & Vlaeyen, 2005). Most recently, fear-potentiated
startle has been employed as a means of potentially resolving the mixed findings
yielded using other cognitive tasks. Some of these studies have indicated
increased startle to pain-related stimuli (Flor, Knost, & Birbaumer, 1997) while
others have not (Carleton, Asmundson, Collimore, & Ellwanger, 2006;
220 G. J. G. Asmundson et al.

Kronshage, Kroener-Herwig, & Pfingsten, 2001); however, strong associations

between startle responses to pain-related stimuli and fear-based individual dif-
ference constructs (e.g., AS, fear of pain) were noted in both studies that assessed
the latter (Carleton et al., 2006; Kronshage et al., 2001).
The assumption underlying this area of research is that patients with chronic
musculoskeletal pain are generally fearful of pain and, thus, will selectively
attend to pain-related stimuli; however, it is plausible that investigators have
not identified the specific objects or situations that are feared by these indivi-
duals (for detailed discussion of objects of fear in pain conditions, see Morley &
Eccleston, 2004). That is, pain-related stimuli may not be the object of fear for
the majority of patients with chronic musculoskeletal pain. There is evidence to
suggest that trauma-related stimuli may be the most relevant object of fear for
many patients with chronic musculoskeletal pain. Specifically, we (Asmundson,
Bonin, Frombach, & Norton, 2000) and others (Beck, Gudmundsdottir, &
Shipherd, 2003) have demonstrated that when the heterogeneous nature of
chronic musculoskeletal pain is considered, those patients classified as dysfunc-
tional are far more likely to have co-occurring PTSD (70%) than those
classified as interpersonally distressed (35%) or as adaptive copers (20%).
The empirically-derived system for making these classifications – the Multiaxial
Assessment of Pain (MAP; Turk & Rudy, 1987) – is described in detail in the
Assessment section below. There is also preliminary evidence that patients with
co-occurring PTSD and pain show attentional biases (on a modified Stroop
task) for both pain and accident words (e.g., crash), whereas those with pain
and no PTSD are biased only toward pain words (e.g., throbbing; Beck,
Freeman, Shipherd, Hamblen, & Lackner, 2001). While the latter findings
await replication, they suggest that the object of fear in some chronic pain
patients (i.e., those classified as dysfunctional) may be associated with prior
traumatic and painful injury. Confirmation of these findings may explain the
lack of robustness observed in efforts to identify attentional biases in patients
with chronic musculoskeletal pain and may shed light on cognitive mechanisms
underlying the co-occurrence of these conditions.

Lower Threshold for Alarm

Pain and anxiety are both associated with physiological arousal (e.g., acceler-
ated heart rate, elevated blood pressure, increased respiration, decreased gas-
trointestinal activity, increased muscular tension, increased blood flow to
skeletal muscles; Hoehn-Saric & McLeod, 1993). The bodily changes stemming
from arousal serve a protective function, promoting escape and withdrawal, but
can have detrimental effects if prolonged. Physical injury and traumatic experi-
ences also initiate other complex neural and hormonal processes (e.g., release
of cytokines, b-endorphin, 5-HT-moduline) that, while designed to promote
tissue healing and reinstate homeostasis, can be destructive to muscle, bone,
Pain and Anxiety 221

and neural tissue when prolonged (e.g., Kiecolt-Glaser, McGuire, Robles, &
Glaser, 2002; Melzack & Katz, 2004). In short, prolonged physiological arousal
and activation of neural and hormonal processes, whether initiated by pain or
anxiety, act as a stressor (i.e., they are perceived as threatening and uncontrol-
lable) that can have detrimental effects on various body systems (McEwen,
1998). Illustrating these effects, and as noted in our discussion of epidemiology
of co-occurrence, Sareen et al. (2005) found strong associations between anxiety
disorders, particularly PTSD, and general medical conditions characterized
by pain.
We have been particularly interested in the role that autonomic nervous
system (ANS) dysregulation may play in anxiety disorders and in chronic
musculoskeletal pain. This interest is predicated on the notion that chronic
arousal is, in part, responsible for the symptoms of both conditions. One of the
most robust findings reported in the PTSD literature over the past two decades
is that sympathetic activity is increased and parasympathetic activity decreased,
both in general and in response to trauma-related stimuli. This pattern of
findings has been observed across a wide variety of measures of cardiovascular
reactivity in both traumatized adults (Keane et al., 1998) and children
(Scheeringa, Zeanah, Myers, & Putnam, 2004). Although ANS dysregulation
in chronic musculoskeletal pain has received little empirical scrutiny, available
findings suggest a pattern similar to that observed in PTSD. Rainville, Bao, and
Chretien (2005), for example, used hypnosis to alter mood, perceived pain
unpleasantness, and severity of pain induced in healthy participants, showing
that increases in negative mood and pain unpleasantness were positively asso-
ciated with changes in heart rate variability. This suggests that pain-related
emotion impacts ANS responsivity. We recently completed an investigation of
patients with chronic musculoskeletal pain (n=33), acute musculoskeletal pain
(n=12), and healthy controls (n = 30) using tasks designed to evoke measur-
able, bi-directional ANS responses. Preliminary analyses indicate no between-
group differences with regard to tasks that augment vagal tone; however, the
chronic pain patients exhibited increased sympathetic activity compared to
controls on tasks designed to induce rapid vagal withdrawal followed by
sympathetic discharge (i.e., dysregulated sympathetic discharge). Collectively,
these data indicate both PTSD and chronic musculoskeletal pain are character-
ized by labile sympathetic responsivity.
The literature regarding pain threshold (i.e., the point at which a stimulus is
detected as being painful) and pain tolerance (i.e., the length of time that a pain
stimulus can be endured) in each of PTSD and chronic musculoskeletal pain
may also provide some clues as to the mechanism of association; however, the
findings are mixed and complex. There is, for example, a body of evidence
indicating that hyperalgesia (i.e., reduced pain threshold and tolerance or, in
other words, exaggerated pain perception) is induced by elevations in state and
trait anxiety (e.g., Carter et al., 2002; James & Hardardottir, 2002). Since
elevations in state and trait anxiety are central features of PTSD and chronic
musculoskeletal pain, it is plausible that PTSD and chronic musculoskeletal
222 G. J. G. Asmundson et al.

pain may induce hyperalgesia. On the other hand, there is a body of literature
indicating that conditioned stress-induced hypoalgesia/analgesia (i.e., increased
pain threshold and tolerance or, in other words, attenuated pain perception)
plays an important and potentially causal role in both chronic musculoskeletal
pain (e.g., Flor, Birbaumer, Schulz, Grüsser, & Mucha, 2002) and PTSD (e.g.,
Foa, Zinbarg, & Rothbaum, 1992). This literature suggests that dysregulation
of the endogenous opioid system – perhaps functioning to deactivate fear
structures in the short term through heightened release of endogenous opioids –
may play a role in blunting pain perception (e.g., higher pain threshold and
tolerance), reducing avoidance behavior, and increasing emotional numbing
associated with chronic musculoskeletal pain and PTSD. It is noteworthy that
pain tolerance, but not threshold, is affected by naloxone (a drug that blocks
opioid receptors) only when pain levels are high, suggesting only partial media-
tion by the endogenous opioid system.
These mixed findings are intriguing when placed in the context of evidence
showing that AS does not impact pain tolerance or threshold, but is associated
with pain intensity (for review, see Keogh & Asmundson, 2004). It is possible
that different mechanisms may be operating at different levels of the stimulus-
response range (i.e., from just noticeable sensation through intolerable pain) and
that their operations are partially regulated by individual difference factors that
influence processing of pain sensation as alarming. Given recent observations
that unpredictable and predictable pain are associated with hyperalgeisa and
hypoalgesia/analgesia, respectively (Ploghaus, Becerra, Borras, & Borsook,
2003), it is equally possible that different mechanisms are operative depending
on whether pain evokes anxiety (i.e., response to unpredictable, future threats)
or fear (i.e., response to an immediate threat).
Chronic dysregulation of the ANS and endogenous opioid system appears to
play important, possibly interactive roles in reducing the threshold for alarm in
PTSD and chronic musculoskeletal pain and, in part, may account for their
substantive co-occurrence. This remains to be evaluated in direct comparisons
between those with PTSD, chronic musculoskeletal pain, both PTSD and
chronic musculoskeletal pain, and healthy as well as clinical control partici-
pants. This, combined with evidence that the serotonergic system – an aminer-
gic neurotransmitter system responsible for maintaining homeostasis via
modulation of the release of 5-HT from serotonergic neuron terminals – may
be dysregulated in both PTSD and chronic musculoskeletal pain (e.g., David-
son & Connor, 2001), provides clues as to the peripheral and central physiolo-
gical mechanisms underlying the suggestion of a lower threshold for alarm.


There exists a small and growing body of empirical support for postulates of the
mutual maintenance and shared vulnerability models of co-occurring PTSD
Pain and Anxiety 223

and chronic musculoskeletal pain. However, given that empirical scrutiny of

these models is in its infancy, we remain at a stage where there are more
questions than answers. With regard to the applicability of these models to
understanding, explaining, and guiding treatment for those with other anxiety
disorders that are accompanied by clinically significant pain, even less is known.
Notwithstanding, these models provide a foundation on which to make recom-
mendations for assessment and treatment planning for these potentially com-
plicated cases.

Assessment and Treatment

As noted above, clinically significant pain often goes unnoticed when assessing
and planning treatment for a patient with an anxiety disorder. When over-
looked, pain can make treatment of the anxiety symptoms complicated at best,
and frustrating and ineffective at worst. There is little information available on
assessment and treatment planning for anxiety disorder patients who present
with co-occurring pain symptoms. In order to facilitate clinical and research
efforts, we provide a brief overview of tactics for assessment and treatment in
this context.


Comprehensive assessment of pain requires delineation of pain severity or

intensity, pain location and distribution, attitudes and beliefs about pain and
its effects, ways of coping with pain, pain-specific emotional distress (i.e., fear,
anxiety, mood changes), and pain-related functional abilities and limitations
(Asmundson, 2002; Tait, 1999). Below we highlight specific assessment meth-
ods that may be useful in the context of assessing pain in patients with anxiety
disorders. We assume that (a) a comprehensive assessment of the anxiety
disorders and related Axis I and Axis II psychopathology has been conducted,
(b) individual difference factors pertinent to the anxiety disorders and relevant
in the context of mutual maintenance and shared vulnerability, particularly AS,
have been assessed (see Taylor, 1999), and (c) appropriate steps will or have
been taken to identify and medically address organic pathology or other iden-
tifiable physical factors that might account for the patient’s presenting pain
Assessment methods include reviewing hospital medical records (in order
to chart the course of the patient’s problems), clinician-administered structured
clinical interviews, clinician observation techniques (e.g., pain behaviors, facial
action coding), prospective monitoring forms, and self-report questionnaires.
In the context of patients presenting for treatment of an anxiety disorder, and
for future research efforts, we recommend brief semi-structured interviewing
224 G. J. G. Asmundson et al.

accompanied by select self-report measures as an easy and time efficient means

of getting essential information. Use of semi-structured interviews can provide
a wealth of information regarding pain origins, its impact on current beliefs
and assumptions, its impact on behavior, as well as associated complications
(e.g., marital conflict, depression, substance abuse). Therapist guided inquiries
may facilitate identification of the patient’s conceptualization of his or her
problem. Specific questions (e.g., What makes your pain worse?) can provide
valuable information regarding pain triggers, patterns of avoidance, and the
patient’s treatment goals. Where self-report measures are administered in
advance of interviews, the direction of interviews can be guided to some extent
by information gained from these measures; conversely, information gained
from interviews may guide the choice of instruments.
We recommend that assessment batteries provide maximal information with
minimal overlap. The following measures meet this criterion (also see Mikail,
DeBreuil, & d’Eon, 1993).
Multidimensional Pain Inventory (MPI; Kerns, Turk, & Rudy, 1985). The
MPI assesses physical, psychological, and social factors related to the pain
experience. It is comprised of 52 items and is a reliable, valid self-report
measure. To classify pain patient subgroups, responses on the MPI can be
submitted to the MAP (Turk & Rudy, 1987), an empirically-based and compu-
terized application. Patients can be classified as dysfunctional (i.e., higher than
average pain severity, interference, and distress, and lower levels of self-efficacy
and activity), interpersonally distressed (i.e. lower levels of perceived social
support), or minimizers/adaptive copers (i.e., lower than average pain severity,
interference, and distress, and higher levels of self-efficacy and activity).
Asmundson, Bonin, et al., (2000) found that a substantial proportion of dys-
functional and interpersonally distressed patients were classified as having
PTSD (71.4 and 42.9%, respectively) when compared to minimizers/adaptive
copers (21.3%). These findings suggest that MAP subgroups differ with regard
to their likelihood of having PTSD. Additional research is needed to determine
if they differ with regard to other anxiety disorders.
Short Form McGill Pain Questionnaire (SF-MPQ; Melzack, 1987). The
SF-MPQ is used to measure dimensions of the pain experience. It comprises
15 adjectives describing sensory (e.g. throbbing) and affective (e.g., sickening)
dimensions of pain, a visual analogue scale (VAS) and present pain index [PPI;
based on a scale of 0 (no pain) to 5 (excruciating)]. The SF-MPQ has been widely
used to measure pain experiences in many different types of patients (e.g., low
back pain, fibromyalgia, rheumatoid arthritis), and as a measure of treatment
efficacy (e.g., Frost et al., 2000). A more comprehensive assessment of the pain
experience can be achieved by using the full length MPQ (MPQ; Melzack,
Pain Anxiety Symptoms Scale (PASS; McCracken, Zayfert, & Gross, 1992).
For assessing pain-related anxiety, the 40-item PASS, and a shorter 20-item
version (PASS-20; McCracken & Dhingra, 2002) are commonly used. The
PASS and the PASS-20 both assess four theoretically distinct components of
Pain and Anxiety 225

pain-related anxiety, including cognitive anxiety (e.g., I find it hard to concen-

trate when I hurt), fearful appraisal of pain (e.g., Pain sensations are terrifying),
escape and avoidance behavior (e.g., I try to avoid activities that cause pain),
and physiological anxiety associated with pain (e.g., Pain seems to cause
my heart to pound or race). Individuals rate each item on a 6-point Likert
scale ranging from 0 (never) to 5 (always). The PASS and PASS-20 have been
found to have good psychometric properties. While both versions of the PASS
are widely used, several other measures of pain-related anxiety are available; for
a detailed discussion see Asmundson and Carleton (in press).
Chronic Pain Coping Inventory (CPCI; Jensen, Turner, Romano, & Strom,
1995). The 65-item CPCI, and its abbreviated 42-item short form (Romano,
Jensen, & Turner, 2003), measure cognitive and behavioral strategies that
people might use while experiencing or trying to prevent pain. Eight coping
strategies are assessed, including positive coping self-statements, guarding,
resting, asking others for assistance, seeking social support, relaxing, task
persistence, and exercising. Use of past-week coping strategies is measured on
an 8-point scale expressed in number of days. The CPCI provides a compre-
hensive and psychometrically valid assessment of pain-related coping strategies
(Hadjistavropoulos, MacLeod, & Asmundson, 1999).


Cognitive-behavioural therapy (CBT) is a highly effective treatment for both

anxiety disorders (Butler, Chapman, Forman, & Beck, 2006) and chronic pain
(Morley, Eccleston, & Williams, 1999). Therefore, treatment of clinically sig-
nificant pain in patients with an anxiety disorder may effectively incorporate
elements of CBT for both the anxiety disorder and chronic pain. While the
specific application of CBT for managing pain in patients with anxiety dis-
orders is still in its infancy it is, nonetheless, very promising. Several features of
CBT for managing pain can be used in the context of CBT for the anxiety
disorders. These include psychoeducation, relaxation training, attention diver-
sion strategies, cognitive restructuring, graded activity, and one or more of
several exposure techniques.
Psychoeducation. Patients experiencing pain often desire to be pain-free. This
goal is somewhat unrealistic. Accordingly, education can be used to help the
patient reformulate his or her view of pain as a signal of impending catastrophe
(e.g., permanent disability, disease, reinjury) to one of pain as a common
experience that can be self-managed. A basic understanding of the typical
course of pain (and, if appropriate, healing), combined with appreciation of
the premise that hurt does not always equal harm, can ameliorate reservations
about making pain worse and thereby encourage activity participation. A key
advantage of psychoeducation is that it is simple to administer and can be
delivered in groups.
226 G. J. G. Asmundson et al.

Relaxation Training. We recommend a modified version of the method

described by Taylor and Asmundson (2004). This includes systematic education
about and practice doing release-only relaxation, rapid relaxation, and dia-
phragmatic breathing. Implementation of these strategies may help the patient
to control muscle tension and associated pain, and provide general skills that
can be used in everyday living to become more relaxed. In patients with chronic
musculoskeletal pain or headache we typically omit tense-release relaxation as
the process of repeatedly tensing and releasing certain muscle groups can, rather
than providing relief, increase muscle tension or cramping.
Cognitive Strategies. Cognitive strategies may further enhance pain manage-
ment. Cognitive restructuring strategies involve identification of catastrophic
cognitions about pain (e.g., Hurt equals harm) followed by teaching the patient
strategies to challenge and change their catastrophic thoughts (e.g., My pain is
not harmful, and I can function despite it). Attention diversion strategies may
include focusing attention on external stimuli (e.g., pictures of landscapes),
focusing on neutral (e.g., an instructor attending a lecture) or pleasant (e.g.,
sitting comfortably in a chair) imagery, dramatized reconstruction of the context
in which pain occurs (e.g., imagining pain due to an injury in a sports game),
repetitive or systematized cognitive strategies (e.g., counting backwards from
100 by 3), and pain acknowledging (e.g., reappraisal of pain in terms of objective
sensations). A meta-analysis of 61 studies showed that these strategies enhanced
pain tolerance and reduced pain ratings more than 85% of the time relative to no
treatment (Fernadez & Turk, 1989). Imagery strategies were most effective.
Graded Activity. The purpose of graded activity is to counter the effects of pain-
promoting deconditioning that might occur due to inactivity and pain-related
behavioral avoidance. When applied in the context of patients with an anxiety
disorder and co-occurring chronic pain we recommend a variant of graded
activity that specifically targets exercise and fitness training. The premise is similar
to that of graded activity – exercise and fitness training may serve to promote
activity pacing, recommencement of previous activity levels, and physical recon-
ditioning in patients with a variety of chronic pain conditions (for review, see
Wright & Sluka, 2001). Because exercise can also serve to ameliorate anxiety
symptoms, particularly panic-related symptoms (Smits, Powers, Berry, & Otto, in
press; Stathopoulou, Powers, Berry, Smits, & Otto, 2006), it is an attractive
consideration for the patient with both anxiety and chronic pain symptoms.
Exposure. Many patients with chronic musculoskeletal pain avoid situations
or activities based on the fear that these will provoke further pain or re-injury.
Accordingly, clinicians have used various forms of exposure therapy to reduce
fear of pain and increase involvement in leisure and work activity. Exposure is
conducted within treatment sessions and as homework assignments. Recent
data from studies using replicated single-case experimental (e.g., de Jong et al.,
2005; Vlaeyen, de Jong, Leeuw, & Crombez, 2004) and randomized controlled
(Woods & Asmundson, in press) designs indicate that graded in vivo exposure –
exposure to specific pain-related activities that are feared or avoided – is more
effective than graded activity in reducing fear of pain, avoidance behaviors, and
Pain and Anxiety 227

reports of pain severity. In vivo exposure is explained in detail elsewhere

(Vlaeyen et al., 2004). In vivo exposure may be an effective strategy to incorpo-
rate when a patient with an anxiety disorders reports significant fear and
avoidance of pain-related situations or activities; however, this may require
substantial time (e.g., two weekly session of 45 mins for approximately 8 weeks)
and divergence from standard treatment. Interoceptive exposure (i.e., exposure
to anxiety provoking bodily sensations) is an interesting alternative to in vivo
exposure; it is an empirically-supported intervention often included as part of
CBT for anxiety disorders (e.g., Taylor, 1999) and, more recently, has been
shown effective in reducing both PTSD symptoms (Wald & Taylor, in press)
and fear of pain (Watt, Stewart, Lefaivre, and Uman, 2006).
Other Treatment Considerations. While we have emphasized CBT in our dis-
cussion of treatment options for co-occurring chronic musculoskeletal pain and
anxiety disorders, there are other promising approaches. Pharmacotherapy can be
effective in alleviating pain associated with various conditions (e.g., Portenoy,
2000); thus, combining analgesics with CBT may prove particularly effective in
cases where an anxiety disorder and clinically significant pain co-occur. Combined
pharmacotherapy and CBT is, in fact, recommended in current expert consensus
guidelines for comorbid PTSD and chronic pain (Foa, Davidson, & Frances,
1999). Moreover, recent developments in pharmacotherapy suggest that some
pharmacologic agents may be particularly effective for alleviating co-occurring
PTSD and pain. Specifically, results from two small, randomized trials indicate
that propranolol, a beta-blocker with analgesic effects, reduces PTSD symptoms
(Pitman et al., 2002; Pitman & Delahanty, 2005). Speculations of effectiveness
of combination treatments across the anxiety disorders do, however, await
systematic empirical scrutiny. Acceptance and mindfulness-based interventions
(e.g., Kabat-Zinn, 1990; Hayes, Wilson, & Strosahl, 1999) have also been found to
be effective for a range of medical and psychological problems that include pain
syndromes and anxiety disorders (for reviews, see Baer, 2003; Bishop, 2002;
Melbourne Academic Mindfulness Interest Group, 2006). It is plausible that
these interventions may prove effective, either on their own or as adjuncts to
CBT, for the treatment of anxiety disorders that co-occur with chronic pain. On
the other hand, while standard physical treatments of chronic musculoskeletal
pain (e.g., electrotherapy, muscle manipulation) are potentially therapeutic, there
is a lack of substantive empirical evidence supporting their effectiveness (Wright &
Sluka, 2001); thus, we do not recommend their use in treating a patient with
co-occurring anxiety disorder and chronic musculoskeletal pain.

Outstanding Issues and Conclusion

Generally speaking, relatively little is known about the mechanisms that

underlie the co-occurrence of the anxiety disorders and conditions character-
ized by clinically significant pain. The majority of evidence comes from
228 G. J. G. Asmundson et al.

investigations of the co-occurrence of PTSD and chronic musculoskeletal pain.

Examination in the context of the other anxiety disorders may serve to identify
mechanisms of co-occurrence (e.g., mutual maintenance, shared vulnerability)
that are common to each. Systematic inquiry may also allow further develop-
ment, refinement, and empirical validation of treatments geared specifically
toward those patients who have both an anxiety disorder and chronic pain. As
noted at the outset of this chapter, the co-occurrence of an anxiety disorder and
chronic pain does not bode well for assessment or treatment outcome. It
remains undetermined whether co-occurrence negatively influences health com-
promising behaviors and, if so, how this will affect treatment. However, given
that the anxiety disorders and chronic pain syndromes present with a strikingly
similar cluster of health compromising behaviors (e.g., heavy smoking,
increased alcohol and substance consumption, poor sleep, increased depression
and suicidal ideation, avoidance of leisure activities and exercise), it seems
reasonable to speculate that co-occurrence will, at minimum, be associated
with reduced likelihood that such patients will present with a well developed
arsenal of adaptive health behaviors. This, in turn, may complicate or, at
minimum, prolong treatment.
Below we highlight a number of specific research directions that await
investigation. For those investigators who pursue one or more of these
directions, we urge careful attention to the heterogeneous nature of chronic
pain (e.g., as illustrated by the MAP categories of minimizers/adaptive copers,
interpersonally distressed, dysfunctional; Turk & Rudy, 1987), as failure to do
so may significantly reduce power to detect mechanisms at play. We also urge
that efforts are made to evaluate aspects of both mental and physical health.
First, further elaboration of current and lifetime prevalence of various pain
conditions in each of the anxiety disorders, in community and treatment-
seeking samples, using comprehensive pain assessment batteries is warranted.
Second, and of particular importance, investigation of the temporal sequence of
co-occurring anxiety disorders and chronic pain are required to determine
whether one condition is more likely to precede the other, and whether sequen-
cing of onset is consistent across the anxiety disorders. Third, further empirical
exploration of the mechanisms of co-occurrence, as posited in the mutual
maintenance and shared vulnerability models, for various chronic pain condi-
tions and each of the anxiety disorders will improve our understanding of co-
occurrence. Further empirical scrutiny may also serve to guide treatment for
those with anxiety disorders accompanied by clinically significant pain. The
preliminary research regarding the application of each of propanolol and
interoceptive exposure in this context is promising. Finally, there are a number
of specific studies related to shared vulnerability factors that will advance the
field, including (a) longitudinal studies to examine whether elevated AS
precedes the development of anxiety disorders and chronic musculoskeletal
pain, or whether it becomes elevated as a consequence of these conditions, (b)
additional study of other potential vulnerability factors (e.g., trait negative
affectivity, illness/injury sensitivity, fear of pain) in understanding the
Pain and Anxiety 229

association between anxiety disorders and chronic musculoskeletal pain, (c)

further examination of the object of fear in chronic musculoskeletal pain
patients with and without a co-occurring anxiety disorder (e.g., whether they
are associated with prior traumatic and painful injury), and (d) examination of
the possible interactive roles of ANS dysregulation and the endogenous opioid
system in reducing threshold for alarm in PTSD (and other anxiety disorders)
and chronic musculoskeletal pain.
The aforementioned research directions are neither exhaustive nor presented
in order of importance. There are, instead, ideas that may stimulate research in
this emerging but currently underdeveloped area. Ultimately, the outcomes of
studies targeting theses avenues of inquiry may serve to inform assessment and
treatment that, in turn, will increase the probability that clinicians can posi-
tively impact the mental and physical health of their patients who present with
an anxiety disorder accompanied by clinically significant pain.


Andersson, G., & Haldrup, D. (2003). Personalized pain words and Stroop interference in
chronic pain patients. European Journal of Pain, 7, 431–438.
Asmundson, G. J. G. (2002). Pain assessment: State-of-the-art applications from the cognitive-
behavioural perspective. Behaviour Research and Therapy, 40, 547–550.

Asmundson, G. J. G., Bonin, M., Frombach, I. K., & Norton, G. R. (2000). Evidence of a
disposition toward fearfulness and vulnerability to posttraumatic stress in dysfunctional
pain patients. Behaviour Research and Therapy, 38, 801–812.
Asmundson, G. J. G., & Carleton, R. N. (in press). Fear of pain. In M. M. Antony & M. B. Stein
(Eds.), Handbook of anxiety and the anxiety disorders. (pp. 000–000). New York: Oxford
University Press.
Asmundson, G. J. G., Carleton, R. N., & Ekong, J. (2005). Dot-probe evaluation of selective
attentional processing of pain cues in patients with chronic headaches. Pain, 114, 250–256.
Asmundson, G. J. G., Coons, M. J., Taylor, S., & Katz, J. (2002). PTSD and the experience of
pain: Research and clinical implications of shared vulnerability and mutual maintenance
models. Canadian Journal of Psychiatry, 47, 930–937.
Asmundson, G. J. G., Jacobson, S. J., Allerdings, M D., & Norton, R. G. (1996). Social
phobia in disabled workers with chronic musculoskeletal pain. Behaviour Research and
Therapy, 34, 939–943.
Asmundson G. J. G., Kuperus, J. L., & Norton, G. R. (1997). Do patients with chronic pain
selectively attend to pain-related information? Preliminary evidence for the mediating role
of fear. Pain, 72, 27–32.
Asmundson, G. J. G., Norton, G. R., Allerdings, M. D., Norton, P. J., & Larsen, D. K.
(1998). Posttraumatic stress disorder and work-related injury. Journal of Anxiety Disor-
ders, 12, 57–69.
Asmundson, G. J. G., & Taylor, S. (2006). PTSD and Chronic Pain: Cognitive-Behavioral
Perspectives and Practical Implications. In G. Young, A. W. Kane, & K. Nicholson (Eds.),
Psychological knowledge in court: PTSD, pain, and TBI (pp. 225–241). New York:
Asmundson, G. J. G., Vlaeyen, J. W. S., & Crombez, G. (Eds.). (2004). Understanding and
treating fear of pain. Oxford, UK: Oxford University Press.
230 G. J. G. Asmundson et al.

Asmundson, G. J. G., & Wright, K. D. (2004). Biopsychosocial approaches to pain.

In T. Hadjistavropoulos & K. D. Craig (Eds.), Pain: Psychological perspectives (pp. 13–34),
Mahwah, NJ: Lawrence Erlbaum Associates.
Asmundson, G. J. G., Wright, K. D., & Hadjistavropoulos, H. D. (2000). Invited article (peer-
reviewed). Anxiety sensitivity and disabling chronic health conditions: State of the art and
future directions. Scandinavian Journal of Behaviour Therapy, 29, 100–117.
Asmundson, G. J. G., Wright, K. D., & Hadjistavropoulos, H. D. (2005). Hypervigilance and
attentional fixedness in chronic musculoskeletal pain: Consistency of findings across
modified Stroop and dot-probe tasks. The Journal of Pain, 6, 497–506.
Asmundson, G. J. G., Wright, K. D., McCreary, D. R., & Pedlar, D. (2003). Posttraumatic
stress disorder symptoms in United Nations peacekeepers: An examination of factor struc-
ture in peacekeepers with and without chronic pain. Cognitive Behaviour Therapy, 32, 26–37.
Asmundson, G. J. G., Wright, K., & Stein, M. (2004). Pain and PTSD symptoms in female
veterans. European Journal of Pain, 8, 345–350.
Atkinson, J. H., Slater, M. A., Patterson, T. L., Grant, I., & Garfin, S. R. (1991). Prevalence,
onset, and risk of psychiatric disorders in men with chronic low back pain: a controlled
study. Pain, 45, 111–121.
Baer, R. (2003). Mindfulness training as a clinical intervention: A conceptual and clinical
review. Clinical Psychology: Science and Practice, 10, 125–143.
Beck, J. G., Freeman, J. B., Shipherd, J. C., Hamblen, J. L., & Lackner, J. M. (2001).
Specificity of Stroop interference in patients with pain and PTSD. Journal of Abnormal
Psychology, 110, 536–543.
Beck, J. G., Gudmundsdottir, B., & Shipherd, J. C. (2003). PTSD and emotional distress
symptoms measured after a motor vehicle accident: Relationships with pain coping
profiles. Journal of Psychopathology and Behavioral Assessment, 25, 219–227.
Beckham, J. C., Crawford, A. L., Feldman, M. E., Kirby, A. C., Hertzberg, M. A.,
Davidson, J. R. T., et al. (1997). Chronic posttraumatic stress disorder and chronic pain
in Vietnam combat veterans. Journal of Psychosomatic Research, 43, 379–389.
Bishop, S. R. (2002). What do we really know about Mindfulness-Based Stress Reduction?
Psychosomatic Medicine, 64, 71–84.
Blanchard, E. B., & Hickling, E. J. (2004). After the crash: Psychological assessment and
treatment of survivors of motor vehicle accidents (2nd ed.). Washington: American Psycho-
logical Association.
Butler, A. C., Chapman, J. E., Forman, E. M., & Beck, A. T. (2006). The empirical status of
cognitive-behavioral therapy: A review of meta-analyses. Clinical Psychology Review, 26,
Carleton, R. N., Asmundson, G. J. G., Collimore, K. C., & Ellwanger, J. (2006). Strategic and
automatic threat processing in patients with chronic musculoskeletal pain: A startle probe
investigation. Cognitive Behaviour Therapy, 35, 236–247.
Carter, L. E., McNeil, D. W., Vowles, K. E., Sorrell, J. T., Turk, C. L., Ries, B. J., et al. (2002).
Effects of emotion on pain reports, tolerance and physiology. Pain Research & Manage-
ment, 7, 21–30.
Davidson, J. R. T., & Connor, K. M. (2001). Serotonin and setotonergic drugs in posttrau-
matic stress disorder. In S. M. Montgomery & J. A. den Boer (Eds.), SSRIs in depression
and anxiety. Perspectives in Psychiatry (Vol. 7, pp. 173–188). New York: Wiley.
de Jong, J. R., Vlaeyen, J. W. S., Onghena, P., Cuypers, C., den Hollander, M., & Ruijgrok, J.
(2005). Reduction of pain-related fear in complex regional pain syndrome type I: The
application of graded exposure in vivo. Pain, 116, 264–275.
Demyttenaere, K., Bruffaerts, R., Lee, S., Posada-Villa, J., Kovess, V., Angermeyer, M. C.,
et al. (2007). Mental disorders among persons with chronic back or neck pain: Results
from the world mental health surveys. Pain, 129, 231–232.
Ehlers, A., & Clark, D. (2000). A cognitive model of posttraumatic stress disorder. Behaviour
Research and Therapy, 38, 319–345.
Pain and Anxiety 231

Elsesser, K., Sartory, G., & Tackenberg, A. (2005). Initial symptoms and reactions to trauma-
related stimuli and the development of posttraumatic stress disorder. Depression and
Anxiety, 21, 61–70.
Foa, E. B., Davidson, J. R. T., & Frances, A. (1999). Treatment of PTSD: The NIH expert
consensus guideline series. Journal of Clinical Psychiatry, 60 (Suppl. 16), 4–76.
Fedoroff, I. C., Taylor, S., Asmundson, G. J. G., & Koch, W. J. (2000). Cognitive factors in
traumatic stress reactions: Predicting PTSD symptoms from anxiety sensitivity and beliefs
about harmful events. Behavioural and Cognitive Psychotherapy, 28, 5–15.
Fernadez, E., & Turk, D. C. (1989). The utility of cognitive coping strategies for altering pain
perception: A meta-analysis. Pain, 38, 123 135.
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1996). Structured Clinical
Interview for DSM-IV. New York: New York State Psychiatric Institute, Biometrics
Research Department.
Fishbain, D. A., Goldberg, M., Meager, B. R., Steele, R., & Rosomoff, H. (1986). Male and
female chronic pain patients categorized by DSM-III psychiatric diagnostic criteria. Pain,
26, 181–197.
Flor, H., Birbaumer, N., Schulz, R., Grüsser, S. M., & Mucha, R. F. (2002). Pavlovian
conditioning of opioid and nonopioid pain inhibitory mechanisms in humans. European
Journal of Pain, 6, 395–402.
Flor, H., Knost, B., & Birbaumer, N. (1997). Processing of pain- and body-related verbal
material in chronic pain patients: Central and peripheral correlates. Pain, 73, 413–421.
Foa, E. B., & Rothbaum, B. O. (1998). Treating the trauma of rape: Cognitive-behavioral
therapy for PTSD. Treatment manuals for practitioners. New York: Guilford.
Foa, E. B., Zinbarg, R., & Rothbaum, B. O. (1992). Uncontrollability and unpredictability in
post-traumatic stress disorder: An animal model. Psychological Bulletin, 112, 218–238.
Frost, S., Grossfeld, S., Kirkeley, A., Litchfield, B., Fowler, P., & Amendola, A. (2000). The
efficacy of femoral nerve block in pain reduction for outpatient hamstring anterior
cruciate ligament reconstruction: A double-blind prospective, randomized trial. Arthro-
scopy, 16, 243–248.
Gureje, O., Von Korff, M., Simon, G. E., & Gater, R. (1998). Persistent pain and well being:
A World Health Organization study in primary care. Journal of the American Medical
Association, 280, 147–151.
Hadjistavropoulos, H. D., MacLeod, F. K., & Asmundson, G. J. G. (1999). Validation of the
chronic pain coping inventory in a chronic pain sample. Pain, 80, 471–481.
Hayes, S. C., Wilson, K. G., & Strosahl, K. D. (1999). Acceptance and commitment therapy.
New York: Guilford.
Hoehn-Saric, R., & McLeod, D. R. (Eds.). (1993). Somatic manifestations of normal and
pathological anxiety. In R. Hoehn-Saric & D. R. McLeod, (Eds.), Biology of anxiety
disorders (pp. 177–222). Washington, DC: American Psychiatric Press.
Hubbard, J., Realmuto, G. M., Northwood, A. K., & Masten, A. S. (1995). Comorbidity of
psychiatric diagnoses with posttraumatic stress disorder in survivors of childhood trauma.
Journal of the American Academy of Child & Adolescent Psychiatry, 34, 1167–1173.
International Association for the Study of Pain (1986). Classification of chronic pain:
Descriptions of chronic pain syndromes and definitions of pain terms. Pain, (Suppl. 3),
James, J. E., & Hardardottir, D. (2002). Influence of attention focus and trait anxiety on
tolerance of acute pain. British Journal of Health Psychology, 7, 149–162.
Jensen, M. P., Turner, J. A., Romano, J. M., & Strom, S. E. (1995). The chronic pain coping
inventory: Development and preliminary validation. Pain, 60, 203–216.
Kabat-Zinn, J. (1990). Full catastrophe living: Using the wisdom of your body and mind to face
stress, pain, and illness. New York: Dell Publishing.
Katon, W., Egan, K., & Miller, D. (1985). Chronic Pain: Lifetime Psychiatric Diagnoses and
Family History. American Journal of Psychiatry, 142, 1156–1160.
232 G. J. G. Asmundson et al.

Keane, T. M., Kolb, L. C., Kaloupek, D. G., Orr, S. P., Blanchard, E. B., Thomas, R. G., et al.
(1998). Utility of psychophysiology measurement in the diagnosis of posttraumatic stress
disorder: Results from a department of Veteran’s Affairs cooperative study. Journal of
Consulting and Clinical Psychology, 66, 914–923.
Keogh, E., & Asmundson, G. J. G. (2004). Negative affectivity, catastrophizing, and anxiety
sensitivity. In G. J. G. Asmundson, J. W. S. Vlaeyen, & G. Crombez (Eds.), Understanding
and treating fear of pain (pp. 91–115). Oxford: Oxford University Press.
Kerns, R. D., Turk, D. C., & Rudy, T. E. (1985). The West Haven-Yale Multidimensional
Pain Inventory (WHYMPI). Pain, 23, 345–356.
Kessler, R. C., Berglund, P., Demler, O., Jin, R., & Walters, E. E. (2005). Lifetime prevalence
and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey
Replication. Archives of General Psychiatry, 62, 593–602.
Kessler, R. C., Chiu, W. T., Demler, O., & Walters, E. E. (2005). Prevalence, severity, and
comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Repli-
cation. Archives of General Psychiatry, 62, 617–627.
Kiecolt-Glaser, J. K., McGuire, L., Robles, T. F., & Glaser, R. (2002). Emotions, morbidity,
and mortality: New perspectives from psychoneuroimmunology. Annual Review of
Psychology, 53, 83–107.
Kinney, R. K., Gatchel, R. J., Polatin, P. B., Fogarty, W. T., & Mayer, T. G. (1993).
Prevalence of psychopathology in acute and low back pain patients. Journal of Occupa-
tional Rehabilitation, 3, 95–103.
Kronshage, U., Kroener-Herwig, B., & Pfingsten, M. (2001). Kinesiophobia in chronic low
back pain patients–does the startle paradigm support the hypothesis? International Journal
of Behavioral Medicine, 8, 304–318.
Kuch, K., Cox, B. J., Woszczyna, C. B., Swinson, R. P., & Shulman, I. (1991). Chronic pain in
panic disorder. Journal of Behavior Therapy and Experimental Psychiatry, 22, 255–259.
Large, R. G. (1986). DSM-III Diagnoses in Chronic Pain: Confusion or Clarity? The Journal
of Nervous and Mental Disease, 174, 295–303.
Martin, A., Dzyuba, M., Halket, E., Asmundson, G. J. G., & Katz, J. (2007). Posttraumatic stress
disorder (PTSD) symptoms predict pain-related disability prior to surgery. Abstract accepted
for presentation at the Annual Meeting of the American Pain Society, Washington, May 2–5.
Mathews, A., & MacLeod, C. (2002). Induced processing biases have causal effects on
anxiety. Cognition & Emotion, 16, 331–354.
McCracken, L. M., Zayfert, C., & Gross, R. T. (1992). The Pain Anxiety Symptoms Scale:
Development and validation of a scale to measure fear of pain. Pain, 50, 67–73.
McCracken, L. M., & Dhingra, L. (2002). A short version of the Pain Anxiety Symptoms
Scale (PASS-20): Preliminary development and validity. Pain Research and Management,
7, 45–50.
McEwen, B. S. (1998). Protective and damaging effects of stress mediators. New England
Journal of Medicine, 338, 171–179.
McWilliams, L. A., Cox, B. J., & Enns, M. W. (2003). Mood and anxiety disorders associated
with chronic pain: An examination in a nationally representative sample. Pain, 106, 127–133.
McWilliams, L. A., Goodwin, R. D., & Cox, B. J. (2004). Depression and anxiety associated with
three pain conditions: Results from a nationally representative sample. Pain, 111, 77–83.
Melbourne Academic Mindfulness Interest Group. (2006). Mindfulness-based psychothera-
pies: A review of conceptual foundations, empirical evidence and practical considerations.
Australian and New Zealand Journal of Psychiatry, 40, 285–294.
Melzack, R. (1975). The McGill Pain Questionnaire: Major properties and scoring methods.
Pain, 1, 277–299.
Melzack, R. (1987). The short-form McGill Pain Questionnaire. Pain, 30, 191–197.
Melzack, R., & Katz, J. (2004). The Gate Control Theory: Reaching for the brain.
In T. Hadjistavropoulos & K. D. Craig (Eds.), Pain: Psychological perspectives (pp. 13–34).
Mahwah, NJ, US: Lawrence Erlbaum Associates, Publishers.
Pain and Anxiety 233

Mikail, S. F., DeBreuil, S., & D’Eon, J. L. (1993). A comparative analysis of measures used in
the assessment of chronic pain patients. Psychological Assessment, 5, 117–120.
Morley, S., & Eccleston, C. (2004). The object of pain. In G. J. G. Asmundson,
J. W. S. Vlaeyen, & G. Crombez (Eds.), Understanding and treating fear of pain (pp. 163–188).
Oxford: Oxford University Press.
Morley, S., Eccleston, C., & Williams, A. (1999). Systematic review and meta-analysis of
randomized controlled trials of cognitive behaviour therapy and behaviour therapy for
chronic pain in adults, excluding headache. Pain, 80, 1–13.
Nagasako, E. M., Oaklander, A. L., & Dworkin, R. H. (2003) Congenital insensitivity to pain:
An update. Pain, 101, 213–219.
Norton, P. J., & Asmundson, G. J. G. (2003). Amending the fear-avoidance model of chronic
pain: What is the role of physiological arousal? Behavior Therapy, 34, 17–30.
Otis, J. D., Keane, T. M., & Kerns, R. D. (2003). An examination of the relationship between
chronic pain and post-traumatic stress disorder. Journal of Rehabilitation Research &
Development, 40, 397–406.
Paunovic, N., Lundh, L-G., & Öst, L-G. (2002). Attentional and memory bias for emotional
information in crime victims with acute posttraumatic stress disorder (PTSD). Journal of
Anxiety Disorders, 16, 675–692.
Pitman, R. K., & Delahanty, D. L. (2005). Coceptually driven pharmacologic approaches to
acute trauma. CNS Spectrums, 10, 99–106.
Pitman, R. K., Sanders, K. M., Zusman, R. M., Healy, A. R., Cheema, F., Lasko, N. B., et al.
(2002). Pilot study of secondary prevention of posttraumatic stress disorder with propra-
nolol. Biological Psychiatry, 51, 189–192.
Ploghaus, A., Becerra, L., Borras, C., & Borsook, D. (2003). Neural circuitry underlying
pain modulation: Expectation, hypnosis, placebo. Trends in Cognitive Sciences, 7,
Polatin, P. B., Kinney, R. K, Gatchel, R. J, Lilio, E., & Mayer, T. G. (1993). Psychiatric illness
and chronic low-back pain: The mind and the spine – which goes first? Spine, 18, 66–71.
Portenoy, R. K. (2000). Current pharmacotherapy of chronic pain. Journal of Pain and
Symptom Management, 19, S16–S20.
Raphael, K. G., Janal, M. N., Nayak, S., Schwartz, J. E., & Gallagher, R. M. (2006). Psychiatric
comorbidities in a community sample of women with fibromyalgia. Pain, 124, 117–125.
Rainville, P., Bao, Q. V. H., & Chretien, P. (2005). Pain-related emotions modulate experi-
mental pain perception and autonomic responses. Pain, 118, 306–318.
Reich, J., Tupin, J. P., & Abramowitz, S. J. (1983). Psychiatric diagnosis of chronic pain
patients. American Journal of Psychiatry, 140, 1495–1498.
Roelofs, J., Peters, M. L., Fassaert, T., & Vlaeyen, J. W. S. (2005). The role of fear of
movement and injury in selective attentional processing in patients with chronic low
back pain: A dot-probe evaluation. Journal of Pain, 6, 294–300.
Roelofs, J., Peters, M. L., Zeegers, M. P. A., & Vlaeyen, J. W. S. (2002). The modified Stroop
paradigm as a measure of selective attention towards pain-related stimuli among chronic
pain patients: A meta-analysis. European Journal of Pain, 6, 273–281.
Romano, J. M., Jensen, M. P., & Turner, J. A. (2003). The chronic pain coping inventory-42:
Reliability and validity. Pain, 104, 65–73.
Sareen, J., Cox, B. J., Clara, I., & Asmundson, G. J. G. (2005). The relationship between
anxiety disorders and physical disorders in the U.S. National Comorbidity Survey.
Depression and Anxiety, 21, 193–202.
Saxe, G., Stoddard, F., Courtney, D., Cunningham, K., Chawla, N., Sheridan, R., et al.
(2001). Relationship between acute morphine and the course of PTSD in children with
burns. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 915–921.
Scheeringa, M. S., Zeanah, C. H., Myers, L., & Putnam, F. (2004). Heart period and
variability findings in preschool children with posttraumatic stress symptoms. Biological
Psychiatry, 55, 685–691.
234 G. J. G. Asmundson et al.

Schmidt, N. B., & Telch, M. J. (1997). Nonpsychiatric medical comorbidity, health percep-
tions, and treatment outcomes in patients with panic disorder. Health Psychology, 16,
Sharp, T., & Harvey, A. G. (2001). Chronic pain and posttraumatic stress disorder: Mutual
maintenance? Clinical Psychology Review, 21, 857–877.

Smits, J. A. J., Powers, M. B., Berry, A. C., & Otto, M. W. (in press). Translating empirically-
supported strategies into accessible interventions: The potential utility of exercise for the
treatment of panic disorder. Cognitive and Behavioral Practice.
Stathopoulou, G., Powers, M. B., Berry, A. C., Smits, J. A. J., & Otto, M. W. (2006). Exercise
interventions for mental health: A quantitative and qualitative review. Clinical
Psychology: Science and Practice, 13, 179–193.
Stein, MB., Jang, K. L., & Livesley, J. W. (1999). Heritability of anxiety sensitivity: A twin
study. American Journal of Psychiatry, 156, 246–251.
Tait, R. (1999). Evaluation of treatment effectiveness in patients with intractable pain:
Measures and methods. In R. J. Gatchel, & D. C. Turk, (Eds.), Psychosocial factors in
pain: Critical perspectives (pp. 457–480). New York: Guilford.
Taylor, S. (1999). Anxiety sensitivity: Theory, research, and treatment of the fear of anxiety.
Mahwah, New Jersey: Lawrence Erlbaum Associates.
Taylor, S., & Asmundson, G. J. G. (2004). Treating health anxiety: A cognitive-behavioral
approach. New York: Guilford.
Taylor, S., Fedoroff, I. C., Koch, W. J., Thordarson, D. S., Fecteau, G., & Nicki, R.
(2001). Posttraumatic stress disorder arising after road traffic collisions: Patterns of
response to cognitive-behavior therapy. Journal of Consulting and Clinical Psychology,
69, 541–551.
Taylor, S., Thordarson, D. S., Maxfield, L., Fedoroff, I. C., Lovell, K., & Ogrodniciuk, J.
(2003). Comparative efficacy, speed, and adverse effects of three PTSD treatments:
Exposure therapy, EMDR, and relaxation training. Journal of Consulting and Clinical
Psychology, 71, 330–338.
Turk, D. C. (2002). A diathesis-stress model of chronic pain and disability following traumatic
injury. Pain Research & Management, 7, 9–20.
Turk, D. C., & Rudy, T. E. (1987). Towards an empirically derived taxonomy of chronic pain
patients. Behaviour Research and Therapy, 25, 237–249.
Vlaeyen, J. W. S., de Jong, J., Leeuw, M., & Crombez, G. (2004). Fear reduction in chronic pain:
Graded exposure in vivo with behavioural experiments. In G. J. G. Asmundson,
J. W. S. Vlaeyen, & G. Crombez (Eds.), Understanding and treating fear of pain (pp. 313–346).
Oxford, UK: Oxford University Press.
Vlaeyen, J. W. S., & Linton, S. J. (2000). Fear-avoidance and its consequences in chronic
musculoskeletal pain. Pain, 85, 317–332.
Von Korff, M., Crane, P., Lane, M., Miglioretti, D. C., Simon, G., Saunders, K., et al. (2005).
Chronic spinal pain and physical-mental comorbidity in the United States: Results from
the national comorbidity survey replication. Pain, 113, 331–339.
Wald, J., & Taylor, S. (in press). Efficacy of interoceptive exposure therapy combined with
trauma-related exposure therapy for posttraumatic stress disorder: A pilot study. Journal
of Anxiety Disorders.
Wall, P. W. (1978). On the relation of injury to pain. Pain, 6, 253–264.
Watt, M. C., Stewart, S. H., Lefaivre, M-J., & Uman, L. S. (2006). A brief cognitive
behavioural approach to reducing anxiety sensitivity decreases pain-related anxiety.
Cognitive Behaviour Therapy, 35, 248–256.
Weisberg, J. N., & Vaillancourt, P. D. (1999). Personality factors and disorders in chronic
pain. Seminars in Clinical Neuropsychiatry, 4, 156–166.
Williams, J. M. G., Mathews, A., & MacLeod, C. (1996). The emotional Stroop task and
psychopathology. Psychological Bulletin, 120, 3–24.
Pain and Anxiety 235

Woods, M., & Asmundson, G. J. G. (in press). Randomized, controlled trial of exposure in vivo
versus graded activity and waitlist in treatment of chronic back pain with high fear of pain.
Wright, A., & Sluka, K. A. (2001). Nonpharmacological treatments for musculoskeletal pain.
Clinical Journal of Pain, 17, 33–46.
Zoellner, L. A., Goodwin, M. L., & Foa, E. B. (2000). PTSD severity and health perceptions
in female victims of assault. Journal of Traumatic Stress, 13, 635–649.
Asthma in Anxiety and Its Disorders: Overview
and Synthesis

Lisa S. Elwood and Bunmi O. Olatunji

Asthma is a respiratory disease with symptoms including reversible airway obstruc-

tion, airway inflammation, and hyperactive airways that affects approximately 6%
to 9% of the U.S. population (Carr, 1998; Turkeltaub & Gergen, 1991). Asthma is
one of the most chronic respiratory disorders and the cost of caring for asthma is
estimated to be higher than that of AIDS/HIV and tuberculosis combined (World
Health Organization, 2000). Furthermore, the economic impact of asthma is
considerable, with total US expenditures for 1990 in excess of $6 billion (Weiss,
Gergen, & Hodgson, 1992). Individuals with asthma appear to be at an increased
risk for psychological difficulties, and report high prevalences of anxiety,
depressive, and substance disorders (Scott et al., 2007). Although asthma is a
treatable condition, morbidity and mortality due to asthma have increased
(Afari, Schmaling, Barnhart, & Buchwald, 2001; Sly, 1988; Weiss, Gergen,
Wagener, 1993). The availability of efficacious interventions for the management
of asthma suggests that some proportion of poor outcomes is preventable
(Greineder, Loane, & Parks, 1995). This conclusion has prompted health care
providers to give special attention to risk factors for poor asthma outcomes.
Psychological difficulties may increase the distress associated with asthma symp-
toms and may be linked with poor asthma management (Katon, Richardson,
Lozano, & McCauley, 2004; Lavoie et al., 2005). The current chapter reviews
literature examining the relation between asthma and anxiety symptoms and dis-
orders. The chapter will begin with a brief description of asthma symptomatology,
assessment, treatment, and non-psychological risk factors. The general association
between asthma and anxiety disorders in children and adults will then be reviewed
followed by a more in depth examination of the relation between asthma and panic
disorder. Next, the chapter will review studies that examine the link between anxiety
and asthma severity, control, and quality of life. The relation between psychological
vulnerabilities to anxiety and asthma will then be discussed. The chapter will then

Lisa S. Elwood
National Crime Victims Research and Treatment Center, Medical University of South
Carolina, 165 Cannon Street, P.O. Box 250852, Charleston, SC 29425, Tel: 843-792-2366,
Fax: 843-792-3388

M. J. Zvolensky, J. A. Smits (eds.), Anxiety in Health Behaviors and Physical Illness. 237
Ó Springer 2008
238 L. S. Elwood, B. O. Olatunji

consider the specificity of the relation between asthma and anxiety. Finally,
cognitive behavioral models of the asthma and anxiety relation will be provided
and the efficacy of cognitive behavioral treatments for individuals with asthma will
be reviewed.


During an acute asthma attack, a stimulus initiates an airway response that

includes inflammation, bronchospasm, and increased mucus production (Sims,
2006). Additional asthma symptoms include coughing, wheezing, shortness of
breath, and tightness in the chest. An individual’s asthma symptoms can be
classified as mild intermittent, mild persistent, moderate persistent, or severe
persistent (Banasiak, 2007). Although the diagnosis and classification of
asthma varies, three factors are typically taken into consideration when classi-
fying the severity of asthma: frequency of daytime symptoms, frequency of
nighttime symptoms, and lung functioning. Lung functioning is assessed using
peak flow rates and forced expiratory volume in 1 second (FEV1). The FEV1
is the amount of air the individual is able to forcibly exhale in 1 second. FEV1 is
measured using spirometry, with a lower number indicating increased severity
(Sims, 2006). Peak expiratory flow (PEF) rates compare the individual’s pul-
monary functioning to what is normal for his or her height, age, weight, and sex.
Normal peak flow rates range from 80% to 100%, while severe asthma attacks
yield peak flow rates of less than 60% (Sims, 2006). An individual with mild
intermittent asthma endorses daytime symptoms less than twice a week, night-
time symptoms less than twice a month, FEV1 or PEF greater than 80% of the
predicted amount, and PEF variability of less than 20%. Mild persistent asthma
includes daytime symptoms greater than two times a week, nighttime symptoms
greater than two times a month, and FEV1 or PEF rates greater than 80% with
20–30% variability. Criteria for moderate persistent asthma include daily day-
time symptoms, nighttime symptoms more than once a week, and FEV1 or PEF
greater than 60–80% predicted with PEF variability of greater than 30%.
Finally, severe persistent asthma describes persistent daily symptoms, frequent
nighttime symptoms, and FEV1 or PEF less than 60% of the predicted value
with PEF variability greater than 30% (Banasiak, 2007). Asthma is often treated
using bronchodilators, corticosteroids, and leukotriene antagonists (Sims, 2006).
Research has revealed that there is a considerable amount of variability in
the course of asthma. It has been suggested that asthma should be conceptua-
lized as a syndrome, with many etiologies that may result in different presenta-
tions and outcomes (Reed, 2006). Different asthma etiologies may include
intermittent wheezing with respiratory infection in infants, Immunoglobulin
E (IgE; related to allergic responses) mediated asthma, intrinsic asthma, and
asthma associated with other chronic lung diseases (Reed, 2006). While fre-
quent wheezing in infancy is associated with an increased likelihood of chronic
Asthma and Anxiety 239

asthma symptoms; infants with infrequent wheezing typically do not continue

to display asthma symptoms as they mature. Allergic asthma commonly devel-
ops during the second decade of life when individuals become exposed to
allergens they did not have contact with as children. Allergic asthma is typically
not severe and does not progress over time. Intrinsic asthma appears to develop
frequently in middle-aged and older adults, but can begin at any age. Intrinsic
asthma tends to be persistent, is likely to increase in severity over time, and may
become irreversible.

Non-psychological Risk Factors for Asthma

Research examining risk factors for asthma have identified sex, race, weight,
timing of birth, smoking, environmental factors, and atopy (i.e., allergic
responses) as potential risk factors for the development of asthma. The relation-
ship between sex and asthma appears to fluctuate over time. Studies have found
that asthma appears to be more prevalent in males in childhood (van Merode,
Maas, Twellar, Kester, & van Schayck, 2007), but in females in adulthood
(Apter, 2007; Wenzel & Busse, 2007). Members of ethnic minority groups,
with the exception of Mexican Americans, appear to be at an increased risk
for asthma when compared to Caucasians (Apter, 2007, McCoy et al., 2006).
Specifically, within the US, asthma appears to be most prevalent in Puerto
Rican Americans, followed by Native Americans, African Americans, and then
by Caucasians. In addition, members of minority groups and individuals with
low socioeconomic status (SES) appear to be at an increased risk for not
receiving appropriate medical care for asthma, resulting in higher rates of
medical visits and death related to asthma (Apter, 2007; Joseph, Williams,
Ownby, Saltzgaber, & Johnson, 2006). However, one study found support for
chronic stress as a mediator between low SES and decreased immune processes
and asthma symptoms (Chen, Hanson et al., 2006). A meta-analysis examining
the relation between weight and asthma indicated that both high weight at
birth and at middle childhood appear to be related to increased risk for future
asthma (Flaherman & Rutherford, 2006). The article suggests that diet, gastro-
esophageal reflux, mechanical effects of obesity, atopy, and hormonal influ-
ences may serve as links between body weight and asthma symptoms. The link
between overweight status and asthma has also been supported in adolescents
(OR 1.4, 95% confidence interval (CI) 1.1–1.6; Jones, Merkle, Fulton, Wheeler,
& Mannino, 2006). Research suggests that obesity in adulthood is associated
with the presence of asthma symptoms in females, but not males (McLachlan
et al., 2007). A separate meta-analysis revealed that premature birth may also
serve as a risk factor for asthma (Jaakkola et al., 2006). However, the strength
of the association varied across studies, with cross-sectional design, broad
outcome criteria, small sample size, young sample, and more recent publication
displaying the strongest associations between pre-term birth and asthma
240 L. S. Elwood, B. O. Olatunji

Atopy has consistently been supported as a risk factor for asthma (Shapiro,
2006). Atopy describes an allergic response which frequently affects parts of the
body that are not in contact with the allergen. Atopic responses include eczema,
allergic rhinitis, and allergic conjunctivitis. Both the presence of atopy in the
child and parental atopy or asthma appear to serve as risk factors for the
development or maintenance of asthma symptoms in children (Shapiro,
2006). Studies have consistently shown a high comorbidity between asthma
and rhinitis (i.e., irritation and inflammation of the nose) and have posited that
rhinitis serves as a risk factor for asthma (Bousquet, van Cauwenberge, &
Khaltaev, 2001). It has been proposed that rhinitis and asthma may be two
expressions of a common mucosal susceptibility and may affect and amplify
each other (Jani & Hamilos, 2005). For individuals with both rhinitis and
asthma, there is some evidence that treatment of rhinitis may improve co-
existing asthma symptoms (Bousquet et al., 2001). Other medical conditions
that have been associated with asthma include heartburn and acid regurgitation
(Hancox et al., 2006).
Studies suggest that smoking or being exposed to second hand smoke serves
as a strong risk factor for the development of asthma (McCoy et al., 2006;
Shapiro, 2006). In fact, a recent study found that the risk of developing asthma
was significantly higher among current smokers with an adjusted odds ratio
(OR) of 1.33 (95% CI 1.00–1.77) and among ex-smokers with an adjusted
OR 1.49 (CI 1.12–1.97) compared with never-smokers (Piipari, Jaakkola,
Jaakkola, & Jaakkola, 2004). Similarly, a prospective cohort study among
2,609 children with no lifetime history of asthma or wheezing found that regular
smoking was associated with increased risk of new-onset asthma and the
increased risk from regular smoking was greater in nonallergic than in allergic
children (Gilliand et al., 2006). A third study which assessed individuals over a
eight year period reported that smokers reported a higher frequency of new
onset asthma than non-smokers, OR 2.14 (95% CI 1.30–3.00; Frank, Hazell,
Morris, Linehan, & Frank, 2007). A study examining poor asthma control,
based on patient diaries over a 4 week period, in individuals being treated for
asthma revealed that smoking, along with lung function and ethnicity, was
associated with the presence of asthma episodes (McCoy et al., 2006). Smoking
has been shown to be a unique and independent predictor of the development of
asthma. The smoking-asthma relation may be largely attributable to airway
narrowing and hyperresponsiveness secondary to emphysema and chronic

Asthma and Mental Health

Many studies have outlined how immunological and physiological factors mod-
erate asthma outcomes (e.g., Grupp-Phelan, Lozano, & Fishman, 2001; Watson,
Becker, & Simons, 1993). However, the potential influence of psychological
Asthma and Anxiety 241

factors on asthma has received relatively less attention (Katon et al., 2004). The
available literature does suggest that approximately 40% of individuals with
asthma present with clinically elevated levels of psychological distress (Mullins,
Chaney, Pace, & Hartman, 1997) with 21% to 25% reporting depressive symp-
toms (Badoux & Levy, 1994; Chaney et al., 1999). Furthermore, it has been
shown that asthma is longitudinally associated with a significant increase
in suicidal ideation and suicide attempt, independent of major depression
(Goodwin & Eaton, 2005). These findings highlight the importance of psycho-
logical distress in the conceptualization of the etiology, maintenance, and treat-
ment of asthma.
Recent research has shown that a wide range of mental health problems
are common among patients with asthma (Goodwin, Messineo, Bregante,
Hoven, & Kairam, 2005). However, data from epidemiologic samples suggest
that anxiety disorders, relative to other mental health problems, are especially
prevalent among patients with asthma (e.g., Ortega, McQuaid, Canino,
Ramirez, Fritz, & Klein, 2003).

The Intersection of Anxiety and Asthma

Prevalence and Impact of Anxiety Disorders

Anxiety disorders have the highest overall prevalence rate among psychiatric
disorders, with a 12-month rate of 18.1% and a lifetime rate of 28.8% (Kessler,
Berglund, Demler, Jin, Merikangas, & Walters, 2005; Kessler, Chiu, Demler, &
Walters, 2005). The estimated costs associated with anxiety disorders has been
reported to be over $45 billion, accounting for over 30% of total expenditures in
mental health (DuPont et al., 1996). Anxiety disorders also have a substantial
negative impact on quality of life (Mendlowicz & Stein, 2000; Olatunji, Cisler, &
Tolin, 2007).

Anxiety in Children and Adolescents with Asthma

Asthma is the most common medical disorder among adolescents and is asso-
ciated with increased functional impairment and lost days at school (Weitzman,
Gortmaker, Sobol, & Perrin,1992). There is also evidence that the presence of
asthma increases the risk for the development of an anxiety disorder. As outlined
in Table 1, the presence of asthma in children and adolescents has been linked with
anxious disorders and symptoms in clinical (Butz & Alexander, 1993; Gupta,
Mitchell, Giuffre, & Crawford, 2001; Vila, Nollet-Clemencon, de Blic, Mouren-
Simeoni, & Scheinmann, 2000) and non-clinical samples (Goodwin, Pine, &
Hoven, 2003; Goodwin, Fergusson, & Horwood, 2004; Ortega et al., 2003; Ortega,
Table 1 Findings of studies examining the relation between asthma and anxiety in children and adolescents

Study Sample Asthma Measurement Anxiety Measurement group Main Findings
Butz and Children 7–12, Recruited Self-reported number and State-Trait Inventory for None 65% reported feeling
Alexander from pediatric ER, type of symptoms, Children, reported ‘‘panic’’ at the
(1993) Diagnosis of acute number of hospital and emotion at the beginning of an asthma
asthma ER visits, limitation of beginning of an asthma attack, feeling panic at
activity, medication attack the beginning of an
attack significantly
predicted trait, but not
state, anxiety
Goodwin Data drawn from an Parents report of child DISC; Child global Asthma to no Asthma was associated
et al. (2003) epidemiology study asthma symptoms and assessment scale asthma with an increased
hospitalization for likelihood of panic
asthma attacks; severe asthma
was associated with an
increased likelihood of
panic attacks and
disorder; Number of
panic symptoms was
associated with an
increased likelihood of
asthma and severe
Goodwin, Data collected during the Classified as presence of Items from CIDI; asked None Asthma was related to
Fergusson, course of a longitudinal absence of asthma for the previous two anxiety prior to
& study of an unselected during age 16–18 and years; Assessed GAD, (OR=1.6 (CI
Horwood birth cohort; at ages 18 18–21. Presence was social phobia, specific 1.2–2.2)), but not after
(2004) and 21 participants determined by reported phobia, and controlling for adverse
were asked about diagnosis and agoraphobia; assessed childhood experiences
asthma symptoms during the panic attacks and (e.g., SES, abuse,
L. S. Elwood, B. O. Olatunji
Table 1 (Continued)

Study Sample Asthma Measurement Anxiety Measurement group Main Findings
time period; also symptoms but not parental conflict) OR
obtained self-report of disorder = 1.2 (CI .5–2.9);
number of asthma suggested a third
Asthma and Anxiety

related medical common factor

consultations in past
year, number of
asthma attacks in past
year, frequency of
medication, and
current frequency of
Gupta et al. Children recruited from Diagnosis, needed change Children: Fear Survey Compared to Both groups scored sig
(2001) an asthma clinic, a of 15% in FEV, Schedule-Revised; cardiology higher on medical
cardiology clinic severity determined by RCMAS; Parents: patients fears, fear of injury and
number of CBCL, STAI small animals than
breakthrough attacks norms; both groups
scored sig higher than
norms on the
internalizing T score on
the CBCL and on the
factor of the CBCL;
both groups scored
higher on the
physiological anxiety
subscale of the
RCMAS than norms
Table 1 (Continued)

Study Sample Asthma Measurement Anxiety Measurement group Main Findings
Ortega et al. Community sample, Parent report of asthma DISC, used diagnostic Asthma to no After controlling for
(2003) sample based on symptoms, diagnosis, categories of anxiety, asthma maternal mental
island-wide probability or asthma related affective, and health, income, and
household sample of hospitalization disruptive disorders; education asthma
children diagnoses of separation diagnosis was
anxiety, social phobia, associated with having
major depression, any psychiatric
conduct disorder, disorder, any affective
oppositional defiant disorder, having more
disorder, and ADHD than one disorder, and
ADHD; Asthma
attack associated with
most disorders, but not
social phobia or
conduct disorder;
hospitalization was
associated with
separation anxiety
Ortega et al. Community sample, Parental reports of an DISC; anxiety disorders Children with Children with a history of
(2004) sample based on asthma diagnosis and included: panic and without asthma attacks had
island-wide probability parental reports of the disorder, separation reported higher prevalence of
household sample of child ever having an anxiety disorder, social histories of any anxiety disorder;
children asthma attack phobia, PTSD, and asthma however, separation
GAD anxiety disorder was
the only individual
anxiety disorder that
was significantly
different between
L. S. Elwood, B. O. Olatunji

Table 1 (Continued)

Study Sample Asthma Measurement Anxiety Measurement group Main Findings
Rietveld Adolescents with and Severity of asthma STAI With asthma to No sig differences in state
(2000) without asthma; with classified from one to healthy or trait anxiety
asthma recruited from five; class 1 (if needed controls
Asthma and Anxiety

general and lung bronchodilator) to

physicians class 5 (oral
Rietveld et al. Adolescents with mild to Severity of asthma was STAI Compared Average threshold for
(2000) severe stable asthma classified based on asthmatics breath holding did not
and healthy prescribed medicine: to healthy differ between groups;
participants without severe, moderate- controls state anxiety of
asthma severe, moderate, participants with
moderate-mild, mild asthma was sig higher
than those without; No
sig differences in trait
Vila et al. Outpatients from Moderate to severe Revised Schedule for Asthmatic 35% of the asthmatic
(2000) pediatric allergology persistent asthma Affective Disorders children to children had at least
and pneumology and Schizophrenia for controls one DSM-IV anxiety
department compared School-aged children; disorders (29% GAD,
to healthy community CBCL; Anxiety and 16% separation
participants recruited Fears Behavioral Scale anxiety disorder, 10%
from schools social phobia, 1%
panic); children with
asthma reported
significantly higher
levels of anxiety than

Table 1 (Continued)

Study Sample Asthma Measurement Anxiety Measurement group Main Findings
Wamboldt Children aged 7–19 were Severity was determined RCMAS; CBCL (parent None Neither child-reported
et al. (1998) recruited from asthma by a pediatric report) anxiety nor parent-
related summer camps pulmonologist and reported internalizing
and an inpatient unit; allergist/immunologist symptoms were
both children and a by reviewing medical significantly elevated
parent participated records. Asthma rated
as mild, moderate, or
severe. Parent report of
number of days missed
from school, number of
number of days
hospitalized, and
number of emergency
visits for asthma.

Note. ADHD = Attention Deficit Hyperactivity Disorder, CBCL= Child Behavior Checklist, CIDI = Composite International Diagnostic Inter-
view, DISC = Diagnostic Interview Schedule for Children; GAD = Generalized Anxiety Disorder PTSD = Posttraumatic Stress Disorder, RCMAS
= Revised Children’s Manifest Anxiety Scale, STAI = State-Trait Anxiety Interview, Sig. = p < 0.05.
L. S. Elwood, B. O. Olatunji
Asthma and Anxiety 247

McQuaid, Canino, Goodwin, & Fritz, 2004) samples. The association between
asthma and anxiety in children has been reported by studies using a variety of
methodologies, including cross-sectional (Butz & Alexander, 1993; Gupta et al.,
2001; Vila et al., 2000), epidemiological (Goodwin et al., 2003; Ortega et al., 2003;
Ortega et al., 2004), and longitudinal (Goodwin et al., 2004). However, it should
be noted that some studies have failed to find a relation between asthma and
anxiety in children (e.g., Wamboldt, Fritz, Mansell, McQuaid, & Klein, 1998).
While one study reported that children suffering from asthma report higher levels
of general anxious symptoms than healthy controls (Gupta et al., 2001), two
studies failed to find significant differences between children with asthma and
healthy controls on trait anxiety (Rietveld, 2000; Rietveld, Everaerd, & van Beest,
2000). A detailed review of the literature suggests that the discrepancy between
studies may partially be accounted for by methodological differences. When
relations between asthma and specific anxiety disorders are examined, asthma
has been linked specifically to panic disorder (Goodwin, Pine, et al., 2003),
separation anxiety disorder (Ortega et al., 2003; Ortega et al., 2004; Vila et al.,
2000), and generalized anxiety disorder (Vila et al., 2000) in children.
The inconsistent findings regarding the relationship between asthma and
anxiety in children highlights the importance of considering sampling
approaches as well as the source of data. For example, Wamboldt and collea-
gues (1998) found that child and parent ratings of the child’s psychological
symptoms displayed inconsistent patterns. While the child’s reported level of
anxiety failed to be significantly related to asthma severity, asthma severity was
significantly related to parent’s ratings of the child’s internalizing symptoms. In
addition, parental ratings of the child’s internalizing symptoms were signifi-
cantly related to the parent’s personal level of physical symptoms, suggesting
that parental reports of child symptoms may be biased by their own distress
(Wamboldt et al., 1998).

Anxiety in Adults with Asthma

Although asthma is commonly observed during childhood, the majority of the

research examining the relations between anxiety and asthma has been conducted
using adult samples. As outlined in Table 2, studies examining the relations
between asthma and anxiety disorders in adults have linked asthma with diagnoses
of panic disorder (Afari et al., 2001; Goodwin, Jacobi, & Thefeld, 2003; Brown,
Khan, & Mahadi, 2000; Goodwin & Eaton, 2003; Goodwin, Olfson, et al., 2003;
Goodwin & Pine, 2002; Lavoie et al., 2005; Nascimento et al., 2002; Perna Bertani,
Politi, Colombo, & Bellodi, 1997; Pollack et al., 1996; Shavitt, Gentil, & Mandetta,
1992; Yellowlees, Alpers, Bowden, Bryant, & Ruffin, 1987), social phobia (Afari et
al., 2001; Brown et al., 2000; Goodwin, Jacobi, et al., 2003; Nascimento et al.,
2002; Perna et al., 1997), generalized anxiety disorder (Goodwin, Jacobi et al.,
2003; Goodwin, Olfson et al., 2003; Goodwin & Pine, 2002; Heaney, Conway,

Table 2 Findings of studies examining the relation between asthma and anxiety in adults
Study Sample Asthma Measurement Anxiety Measurement group Findings

Afari et al. Adult patients with Completed a DIS-III-R Compared to Participants endorsed PD,
(2001) diagnosis of mild to methacholine prevalence agoraphobia, and social
moderate asthma inhalation challenge rates phobia at higher rates than
recruited from an test to confirm general population
asthma clinic airway reactivity; prevalence rates
Brown et al. Recruited participants FEV1 SCID-IV None 59% met criteria for at least
(2000) using medication one anxiety disorder; 28%
regularly scheduled to specific phobia, 19%
receive prednisone at an PTSD, 15% PD, 13%
asthma clinic, social phobia
Davis et al. Evidence of asthma, ‘‘objective evidence of ASI, Sheehan Patient None 37% received an anxiety
(2002) recruited from an asthma based on the Rated Anxiety Scale disorder; 11% PD, 6%
asthma clinic Canadian Consensus (SPRAS), ADIS PTSD, 16% GAD, 3%
Guidelines’’ social phobia, 2% specific
Di Marco COPD patients were diagnosis of COPD; Italian version of the COPD a greater percentage of
et al. (2006) recruited from a FEV1 values; Italian STAI patients asthma participants
respiratory unit and versions of the compared endorsed high levels of
controls were recruited modified Medical to healthy anxiety (28%) than the
from annual check-ups Resource Council controls control participants (6%).
dyspnea scale and the
St. George’s
L. S. Elwood, B. O. Olatunji
Table 2 (Continued)

Study Sample Asthma Measurement Anxiety Measurement group Findings

Erhabor & Respiratory unit clinic Diagnosis of asthma, General Health Orthopedic Asthma patients reported a
Mosaku Questionnaire, and healthy significantly higher level of
(2004) STAI 1 and 2 controls state anxiety than
Asthma and Anxiety

orthopedic and healthy

participants; No sig
differences in trait anxiety;
Asthmatic patients scored
significantly higher on the
General Health
Questionnaires (indicating
higher psychiatric
Goodwin & Data was collected as part Self-report of asthma DIS, assessed panic Individuals Individuals with asthma were
Eaton of the Epidemiologic and treatment of attacks with sig more likely to report
(2003) Study asthma asthma PAs at baseline, OR= .75
compared (CI 1.08–2.84), and continued
to those PAs at time 2, but not
without incident PAs at
time 2. Reported treated
asthma was associated with
higher number of
symptoms during PA
Goodwin, Adult participants of an Self-report of current German National Community Current severe asthma was
Jacobi, epidemiological study and past asthma Health Interview individuals associated with any anxiety
et al. (2003) diagnosis, method of and Examination without disorder, specific phobia,
assessment by Survey-Mental asthma PD, and PAs. Lifetime
physician, type of Health Supplement severe asthma was
asthma, treatment associated with any anxiety
Table 2 (Continued)

Study Sample Asthma Measurement Anxiety Measurement group Findings

sought, and severity followed by Munich disorder, PD, PAs, social

of attacks. Rated as CIDI phobia, specific phobia,
current if present in GAD, bipolar, and any
the last 4 weeks. severe mental disorder.
Categorized as severe Current nonsevere asthma
or nonsevere was associated with any
affective disorder, any
severe mental disorder.
Lifetime nonsevere asthma
was associated with any
anxiety disorder, anxiety
NOS, any somatoform
disorder, and any severe
mental disorder.
Goodwin & Adult community sample Self-report. Assessed CIDI short form Community PAs, OR = 2.2 (CI 1.7–3.0)
Pine (2002) both respiratory individuals and GAD, OR = 1.5 (CI
disease (asthma, without .6–2.2) were associated with
chronic bronchitis, respiratory respiratory disease, lung
and emphysema) and & lung disease, and comorbid
lung disease (other disease respiratory and lung
lung disease). disease. When comorbid
mental disorders and
physical problems were
controlled for, only PAs
were associated with
respiratory disease and lung
L. S. Elwood, B. O. Olatunji
Table 2 (Continued)

Study Sample Asthma Measurement Anxiety Measurement group Findings

Goodwin, Adult primary care Asthma diagnoses PRIME-MD; Patient Compared Participants with asthma were
Olfson, et al. patients at an internal obtained from billing Health patients sig more likely to report PA
(2003) medicine clinic encounter data Questionnaire; with and and GAD than those
Asthma and Anxiety

panic attack and without without; PA remained sig

panic disorder were diagnoses after controlling for
collapsed into one of asthma sociodemographic
group characteristics the
association PAs remained
OR = 1.8 (CI 1.2–2.7), but
GAD did not,
OR = 1.4(CI = .9–2.2)
Halser et al. Adult participants of an Self-report of asthma SCL-90-R, Structured Cross-sectional results: 21%
(2005) epidemiological study, symptoms and self- Psychopathological of individuals with asthma
data collected over 20 yrs report of physician Interview and met criteria for PD, 33%
diagnosis Rating of the Social met criteria for any panic
Consequences for symptoms; 20% of
Epidemiology; individuals with PD met
assessed for the criteria for asthma, 12% of
presence of panic individuals with any panic
disorder, any panic, met criteria for asthma;
and childhood asthma was more strongly
anxiety associated with PD than
any panic;
Longitudinal results: asthma
and panic were not
associated at baseline;
asthma was predictive of
Table 2 (Continued)

Study Sample Asthma Measurement Anxiety Measurement group Findings

subsequent PD and any

panic, although when
demographic data was
included the only held up
for women and smokers;
PD, but not any panic, was
a significant predictor of
later asthma; childhood
anxiety also predicted
subsequent asthma
Heaney Participants were recruited Examined difficult to HADS; psychiatric Compared GAD was the most common
et al. (2005) if they had 1) persisting control asthma; interview offered to treatment anxiety disorder (16%),
refractory symptoms Asthma defined on participants (for also found PTSD, acute
prompting referral, the basis of symptom asthma) stress reaction, and specific
2) minimal maintenance with documented responders phobia (no mention of PD,
therapy of long acting b2 reversible airflow to Anxiety did not sig decrease
agonists, and inhaled obstruction (FEV1 of treatment across treatment and was
steroids 3) at least 1 > 12%), skin prick non not sig different between
course of systematic testing to 12 inhalant responders treatment responders and
steroids in preceding 12 allergens, chest x-ray, treatment nonresponders
months and spirometric
testing; Asthma
related quality of life
assessed using the
Juniper scale
Jonas et al. Self-report of doctor General Well Being None For non-smokers, increasing
(1999) diagnosed asthma Schedule, relaxed asthma incidence rates were
L. S. Elwood, B. O. Olatunji
Table 2 (Continued)

Study Sample Asthma Measurement Anxiety Measurement group Findings

Adults 25 to 74 at baseline Spirometry was versus anxious found for increasing

and received a medical measured as  65% scale. anxiety. When split
examination. between presence and
Asthma and Anxiety

FEV1,  65%, or no
best trial. absence of early asthma
Respiratory symptoms, similar pattern
symptoms were for no early symptom non-
coded as present or smokers.
absent: 1) a cough in
morning or winter or,
2) in the summer,
3) any phlegm from
chest in the morning
in the winter, or 4)
the summer, 5) a
period of increased
cough and phlegm