Psychiatr Q (2016) 87:585593

DOI 10.1007/s11126-015-9411-1

ORIGINAL PAPER

Combined Use of Opioids and Antidepressants

in the Treatment of Pain: A Review of Veterans Health
Administration Data for Patients with Pain Both With
and Without Co-morbid Depression

John J. Sellinger1,2 Mehmet Sofuoglu1,2,3 Robert D. Kerns1,2

Robert A. Rosenheck1,2,3

Published online: 8 December 2015

Springer Science+Business Media New York (outside the USA) 2015

Abstract Musculoskeletal pain is prevalent among Veterans treated within the Veterans
Health Administration (VHA). Depression is highly co-prevalent, and antidepressants are
increasingly being used for psychiatric and analgesic benefit. The current study examined
prescribing patterns of antidepressants and opioids in the context of musculoskeletal pain
using a national VHA database. All Veterans diagnosed with musculoskeletal pain who
attended at least one appointment through the VHA during Fiscal Year 2012 were
dichotomized based on the presence or absence of a depression diagnosis. We compared
the proportion in each group that were prescribed antidepressants to the entire sample and
repeated this comparison along a continuum of the number of annual opioid prescriptions
received (ranging in five categories from no opioids up to [20 scripts). Of the 5.1 million
Veterans seen, 19.1 % were diagnosed with musculoskeletal pain, of whom, 27.2 % were
diagnosed with major depressive disorder. Antidepressants were prescribed to 78.41 % of
patients with musculoskeletal pain and depression, compared to 20.23 % of those without
depression. For both groups, antidepressant use increased linearly as annual opioid fills
increased. Across the categories of opioid use, patients with depression showed a 13.98 %
increase in antidepressant use, compared to a 33.97 % increase in the non-depressed group.
Results suggest that antidepressants are frequently prescribed to patients with muscu-
loskeletal pain who are using opioids, consistent with multi-modal pharmacotherapy.
Increasing use of antidepressants in conjunction with escalating opioid prescribing, par-
ticularly in the absence of diagnosed depression, suggests that antidepressants are being
used in both groups to complement opioid therapy.

Keywords Pain management
Opioids
Depression
Antidepressants

& John J. Sellinger

john.sellinger1@va.gov
1
VA Connecticut Healthcare System, Psychology Service (116B), 950 Campbell Avenue,
West Haven, CT 06516, USA
2
Department of Psychiatry, Yale University School of Medicine, New Haven, CT, USA
3
VA New England Mental Illness, Research, Education and Clinical Center, West Haven, CT, USA

123
586 Psychiatr Q (2016) 87:585593

Introduction

Chronic musculoskeletal pain is often complicated by co-occurring psychiatric disorders,

especially depression. In the context of opioid prescribing, the use of antidepressants would
be consistent with the principle of multi-modal therapy, but patterns of such use in real-
world practice have not been subject to empirical study. A literature review by Bair and
colleagues examining the co-prevalence of chronic pain and depression found that rates of
both conditions were higher in the presence of the other condition than when examined in
the absence of either condition [1]. Patients with major depressive disorder often report one
or more physical pain symptoms, the most frequent of which are musculoskeletal pain
complaints [2]. In some cases, the relationship between chronic pain and depression may
be a reciprocal one, with each condition potentially moderating the other to some extent.
For example, some studies have demonstrated that reductions in pain severity are asso-
ciated with an increased probability of remission of major depressive disorder. Conversely,
other studies have shown that changes in the severity of depressive symptoms are pre-
dictive of reductions in pain severity [3, 4]. Thus, the frequent comorbidity of chronic pain
and depression, coupled with evidence suggesting moderating effects of each condition on
the other, suggests that treatments targeting both conditions may result in better overall
outcomes. There is also growing evidence supporting the effectiveness of antidepressant
medications, themselves, in the treatment of a wide array of pain conditions [57], with
rates of use ranging from 12 to 43 % of cases, depending on pain diagnosis [8]. However,
empirical data have not been available on the use of antidepressants in patients who have
pain with or without a diagnosis of depression, and more specifically on the relationship
between the concomitant use of antidepressants and opioid analgesics.
Despite the high co-prevalence of pain and depression, efforts to coordinate treatment of
these conditions are often challenging. One problem is that the centrally involved treatment
disciplines, internal medicine and psychiatry, may tend to take a narrow view of these
problems and focus either on pain or depression, rather than their interrelation. For
example, primary care providers are thought to commonly overlook depressive symptoms
in the context of pain complaints [9]. Likewise, mental health providers are often thought
to focus attention on the emotional elements of a patients presentation without directly
addressing physical pain. The result may be a failure to assess both pain and depression,
which may result in under treatment of one or both conditions [10]. However the extent to
which these two conditions are treated with both opioids and antidepressants in actual
practice has not been subject to pharmaco-epidemiologic study.
A singular challenge for pain management is the tendency to focus on opioids as the
primary management strategy. Evidence-based treatment for chronic pain management
optimally involves a multimodal treatment approach, which ensures that the treatments
utilized to address pain and related comorbidities come from multiple approaches,
including pharmacology, rehabilitation, behavioral, and mental health. Endorsement of this
balanced approach is designed, in part, to counter the significant increase in the use of
opioids as a primary and often mono-therapy approach to pain managementa strategy
that may contribute to a variety of unintended consequences such as psychological and/or
physiological dependence or addiction, hyperalgesia, medication abuse, and increased
mortality [1113]. Though opioids can be effective in the management of acute pain, their
long-term use for chronic pain management has been linked to significant unintended
consequences such as the development of tolerance, hyperalgesia, abuse and dependence,
and reductions in overall function and quality of life. According to a 2007 report, opioid

123
Psychiatr Q (2016) 87:585593 587

medications were the second most commonly abused drugs in the United States [14].
Results from the 2011 National Survey on Drug Use and Health support this finding, and
further indicate that pain medications are the highest category of misused prescription
medications [15]. Thus, efforts to reduce reliance on opioids for pain management may,
paradoxically, help to increase both the efficacy and safety of chronic pain treatment.
In this study we used national administrative data from the Veterans Health Adminis-
tration from Fiscal Year (FY) 2012 to examine the multi-modal use of antidepressants and
opioids in the pharmacotherapy of patients diagnosed with musculoskeletal pain. More
specifically we examined the proportion of patients diagnosed with musculoskeletal pain
both with and without depression who received antidepressant prescriptions along a five-
level continuum of annual opioid scripts received (ranging from no opioid scripts up to 20
or more scripts). The analysis of use of antidepressants among patients receiving various
numbers of opioid prescriptions during the year was thus stratified by depression status.
We sought to evaluate first, among patients diagnosed with musculoskeletal pain and
depression, how frequently antidepressants were prescribed, and second, whether, among
patients both with and without diagnosed depression, antidepressants were complementary
to opioids consistent with the principle of multi-modal treatment (i.e., antidepressant
prescriptions are more likely to be filled the more opioid prescriptions are filled) or are
substitutes for opioids (i.e., antidepressant prescriptions are more likely to be filled when
fewer opioid prescriptions are filled). If the need to treat both pain and depression is
recognized in actual practice, we would expect to see an increased used of antidepressants
in patients with more opioid prescriptions. While we hypothesized that this relationship
would be observed among those diagnosed with depression, we expected the increased rate
of antidepressant use to be most pronounced among patients not diagnosed with depres-
sion, reflecting a recognition that in the presence of more protracted pain even in the
absence of diagnosed depression, co-prescription of antidepressants would increase with
more opioid prescriptions to enhance the treatment of pain itself.

Methods

Data Source

Data for the current analyses come from the VA encounter files of all outpatient encounters
for Veterans who attended at least one outpatient appointment during FY 12 (October 1,
2011 through September 30, 2012) (n = 5.1 million unique individuals). Additional
information available from the encounter file includes demographic and diagnostic
information for each visit. Prescriptions filled for opioids (excluding methadone and
buprenorphine, which are used in opioid substitution programs for addiction) and all
psychotropics were obtained from the VA Drug Benefit Management database. Access to,
and use of these databases for research purposes has been approved by the facilitys
institutional review board.

Sample

Analyses were restricted to Veterans who had an outpatient diagnosis of musculoskeletal

pain (ICD 9 codes 338.xx, 719.4, and 780.96) (n = 974,352, 19.1 % of all veterans treated
by VHA in FY 2012).

123
588 Psychiatr Q (2016) 87:585593

Measures

Patients were classified as having depression if they had one outpatient encounter or
inpatient discharge with any of the following ICD 9 diagnoses during the FY:
296.2296.39, 300.4x, 296.9x, 311.xx, 301.10301.19.
The total number of opioid prescriptions filled at a VA pharmacy during the year were
counted and classified into five groups: no opioid prescriptions, and those who received
12, 310, 1019, and 20 or more prescriptions during the 2012 FY.
The number of antidepressant fills for each patient was summed for the FY and those
receiving at least one antidepressant were identified. Antidepressants included:
amitriptyline, amoxapine, clomipramine, desipramine, doxepin, imipramine, nortripty-
line, protriptyline, trimipramine, isocarboxazid, phenelzine, selegeline, tranyl-
cypromine, bupropion, citalopram, desvenlafaxine, duloxetine, escitalopram, fluoxetine,
fluvoxamine, maprotiline, mirtazapine, nefazodone, paroxetine, sertraline, trazodone,
and venlafaxine.

Data Analysis

As a first step in this descriptive study, we identified Veterans from the larger database who
carried a musculoskeletal pain diagnosis. Those meeting the criteria were subsequently
divided into two groups based on the presence or absence of at least one diagnosis of a
depressive disorder. These groups were then evaluated for having filled any antidepressant
medication during the year. The proportion of patients who received an antidepressant was
calculated for each level of opioid prescription fill and stratified by the presence or absence
of a diagnosis of depression. Because of the very large sample size and the fact that the
sample represented the entire VA population of relevant patients for the FY (i.e., it was not
a statistic sample, but rather the entire population of interest for the FY), tests of statistical
significance were not performed.

100
90
80
70
60
50
40
30
20
10
0
No opioids 1-2 opioids 3-10 opioids 10-19 opioids 20+ opioids

% on andepressants with depression diagnosis

% on andepressants with no depression diagnosis

Fig. 1 Percentage of patients prescribed antidepressants at 5 levels of annual opioid prescribing (number of
scripts filled), stratified by the presence or absence of a diagnosis of major depressive disorder

123
Psychiatr Q (2016) 87:585593 589

Results

During the 2012 FY (October 2011 through September 2012), 5,102,378 unique patients
attended at least one outpatient appointment within the VA healthcare system of whom,
974,352 (19.1 %) carried a diagnosis of musculoskeletal pain, and of those with muscu-
loskeletal pain, 265,471 (27.2 %) also carried a depressive disorder diagnosis. Of those
who carried a diagnosis of musculoskeletal pain and major depressive disorder, 78.41 %
were prescribed an antidepressant, while only one-fourth, (20.23 %) of those diagnosed
with only musculoskeletal pain (no depression) were prescribed an antidepressant.
The proportion of patients who filled a prescription for an antidepressant increased in a
linear fashion as the number of opioid prescriptions issued over the course of the year
increased. This trend was consistent for subjects with and without a diagnosis depressive
disorder (See Fig. 1). For those with a diagnosis of major depressive disorder, rates of
antidepressant use ranged from 73.63 % in the group for whom no opioids were prescribed,
up to a high of 87.61 % in the group for whom 20 or more opioids prescriptions were
issuedan increase of 13.98 %, or by a factor of 1.19 times. In contrast, for those without
a diagnosis of depressive disorder, rates of antidepressant use increased from 15.17 %
when no opioids were prescribed, up to a high of 49.14 % when 20 or more opioid
prescriptions were issueda total increase of 33.97 %, or 3.24 times (see Table 1).

Discussion

This is the first study, to our knowledge, to examine data from a large, national health care
system that demonstrates that the prescribing of antidepressant medications is common
(approaching 80 %) among individuals who suffer from both pain and a depressive dis-
order, and that the utilization rates of antidepressants increase monotonically with the
increasing number of opioid prescriptions over the course of a year. A similar, yet steeper
rate of increase in antidepressant use was found among those not diagnosed with a
depressive disorder, thus confirming our hypothesis and providing evidence of the actual

Table 1 Prevalence of depression, musculoskeletal pain, and antidepressant use at five levels of opioid
prescribing
N (%) Musculoskeletal Musculoskeletal Musculoskeletal Musculoskeletal
pain with no pain with pain with no pain with
depression depression depression depression
N (%) N (%) diagnosis diagnosis
receiving receiving
antidepressant antidepressant
N (%) N (%)

Total sample 5,102,378 708,881 (13.9) 265,471 (5.2) 143,378 (20.23) 208,163 (78.41)
Opioid dose
0 4,127,932 (80) 470,981 (11.4) 136,663 (3.3) 71,468 (15.17) 100,625 (73.63)
12 302,176 (5.9) 68,950 (22.8) 27,022 (8.9) 14,496 (21.02) 21,366 (79.07)
310 419,800 (8.2) 107,164 (25.5) 57,331 (13.7) 32,187 (30.04) 47,903 (83.56)
1019 204,452 (4.0) 49,655 (24.3) 33,030 (16.2) 19,266 (38.80) 28,260 (85.56)
20? 48,018 (0.9) 12,131 (25.3) 11,425 (23.8) 5,961 (49.14) 10,009 (87.61)

123
590 Psychiatr Q (2016) 87:585593

use of multi-modal pharmacotherapy for pain. For both depressed and non-depressed
groups, the overall frequency of antidepressant use increased as the number of annual
prescriptions for opioids increased, with approximately 50 % of the non-depressed group
and 88 % of the depressed group receiving antidepressants when opioid prescriptions were
at the highest level categorized for this study ([20 annual opioid prescriptions). It thus
appears that in the VA system opioids are not the only drug used in the treatment of pain
and that, at least pharmacologically, a more multi-modal approach is being utilized.
In light of evidence demonstrating the potentially exacerbating effect of depression on
pain, including increased pain severity, worsening disability, and poorer health related
quality of life [4, 16], the prevalent use of antidepressants and its association with greater
opioid use in the current study is a promising finding. These findings suggest that opioids
are not being used without attention to potential depressive symptoms, even when a
depression diagnosis is not present in the medical record. In fact, the beneficial effects of
optimized antidepressant therapy have been demonstrated to be one key element of
effective integrated intervention for comorbid chronic pain and depression [17]. As
demonstrated by Kroenke and colleagues, it will be important in future studies on this topic
to use longitudinal pain and depression assessments to evaluate the extent to which
antidepressants are being used effectively in a way that results in clinically meaningful
reductions in both pain and depressive symptoms. Given the fact that the current study
narrowly examined musculoskeletal pain, future studies should also examine patterns of
antidepressant use in the context of other pain conditions for which they have shown
effectiveness, such as neuropathic pain. The prevalence of pain reflected in this study
population (19 %), is far below the estimated 5070 % prevalence rates of chronic pain
found within the Veteran population [18, 19]. A future study using a more inclusive set of
pain diagnoses could further assess antidepressant use in this broader cross section of
Veterans with chronic pain.
Another advantage of the correlated increase in the use of both antidepressants and
opioids is the potential for the use of antidepressants to mitigate the need for higher doses
of opioids, thus potentially serving as a risk reduction approach to pain management.
Dobscha and colleagues found that the presence of major depressive disorder was sig-
nificantly associated with both short-term opioid prescribing and with chronic opioid
therapy, defined as [90 consecutive days [20]. Given this association, as well as the
physical and psychological risks associated with chronic opioid use, such as respiratory
suppression, sleep apnea, hyperalgesia, medication misuse, addiction, and pseudoaddiction
[21], it can be argued that the effective management of depression might serve to reduce
opioid burden for a meaningful number of patients. This reduction in burden may include
lower daily dosing of opioids and/or reductions in chronic opioid prescribing.
In the data evaluated as part of this descriptive study, the use of antidepressants in the
absence of a depression diagnosis may suggest that these medications are being utilized for
their analgesic effects, which may further reduce reliance on more dangerous opioid agents
as monotherapy [22]. Although the current data set does not allow for analysis of the
prescribers intended use of antidepressant medication, clinical trial evidence does support
the use of tricyclic antidepressants (TCAs) and selective serotonin and norepinephrine
reuptake inhibitors (SNRIs) for analgesic effects in the treatment of musculoskeletal pain
[23], and this may account for the greater relative increase in their use in association with
greater use of opioids in the absence of a depression diagnosis.
The large size and national scope of the data used for the present study provides a broad
overview of the care provided to Veterans served across the Veterans Health Adminis-
tration during the 2012 FY. However, the utilization of such a large data set does have

123
Psychiatr Q (2016) 87:585593 591

some limitations that must be acknowledged. First, the absence of a coded diagnosis of
depression in the medical record does not necessarily mean that a patient has not been
diagnosed or is not undergoing treatment for depression [14]. As a result, in the absence of
a depression diagnosis, the current data set does not allow for determination of why
antidepressant agents were being prescribed (treatment of documented or undocumented
depression versus treatment of musculoskeletal pain versus treatment of both conditions).
Another limitation is the absence of dosing data, as current analyses revealed only the
number of filled opioid prescriptions annually per patient. As a result we are unable to
evaluate resultant reductions in depressed mood, direct analgesic effects of antidepressants
or associated reductions of opioid dosing or frequency of use. To fully assess the effects of
antidepressants as an opioid risk mitigation strategy, it would be helpful for future studies
of this nature to examine the relationship between antidepressant use and opioid dosing/
strength.
Another further limitation is that we cannot tell from the current data whether antide-
pressants and opioids are being prescribed by the same clinician as a coordinated strategy,
or in different settings where coordination of treatments for pain and depression is limited.
Results from the current study do suggest that antidepressants are being utilized to com-
plement opioid therapy, as evidenced by the relatively high utilization of antidepressants
among patients without a diagnosis of depression. What is less clear is the extent to which
antidepressants are intentionally being utilized to target depression and pain among those
patients who carry both diagnoses.
Despite these limitations, the present study offers a unique examination of co-pre-
scribing patterns of antidepressants and opioids among a large national sample of veterans
with musculoskeletal pain both with and without a diagnosis of depression. Consistent with
a biopsychosocial approach to pain management, in which multiple complementary ther-
apies are used in an effort to avoid the dangers of opioid monotherapy, results from the
current study reflect a positive approach to the use of antidepressants to treat pain through
direct analgesic effect, and/or indirectly through the reduction of depression.

Acknowledgments This work was supported, in part, by the Veterans Affairs New England Mental
Illness, Research, Education and Clinical Center (MIRECC). Research conducted with the administrative
databases that were utilized in this study was approved by the Institutional Review Board.

Compliance with Ethical Standards

Conflict of interest None of the authors involved in this work have conflicts of interest to disclose.

Disclaimer The views expressed in this article are those of the authors and do not necessarily reflect the
position or policy of the Department of Veterans Affairs or the United States Government.

References
1. Bair MJ, Robinson RL, Katon W, Kroenke K.: Depression and pain comorbidity. Archives of Internal
Medicine 163:24332445, 2003.
2. Ohayon MM.: Specific characteristics of the pain/depression association in the general population.
Journal of Clinical Psychiatry 65:59, 2004.
3. Arnold LM, Meyers AL, Sunderajan P, Montano CB, Kass E, Trivedi M, Wohlreich MM.: The effect of
pain on outcomes in a trial of duloxetine treatment of major depressive disorder. Annals of Clinical
Psychiatry 20:187193, 2008.
4. Kroenke K, Wu J, Bair MJ, Krebs EE, Damush TM, Tu W.: Reciprocal relationship between pain and
depression: A 12-month longitudinal analysis in primary care. Journal of Pain 12:964973, 2011.

123
592 Psychiatr Q (2016) 87:585593

5. Chan HN, Fam J, Ng B.: Use of antidepressants in the treatment of chronic pain. Annals Academy of
Medicine Singapore 38:97479, 2009.
6. Mercier A, Auger-Aubin I, Lebeau J, Schuers M, Boulet P, et al.: Evidence of prescription of antide-
pressants for non-psychiatric conditions in primary care: An analysis of guidelines and systematic
reviews. BMC Family Practice 14: 55, 2013.
7. Murinson BB.: The role of Antidepressants in the Treatment of Chronic Pain. In: Comprehensive
Treatment of Chronic Pain by Medical, Interventional, and Integrative Approaches. New York,
Springer, pp. 4551, 2013.
8. Rasu RS, Sohraby R, Cunningham L, Knell ME.: Assessing chronic pain treatment practices and
evaluating adherence to chronic pain clinical guidelines in outpatient practices in the United States.
Journal of Pain 14:568578, 2013.
9. Panzarino PJ.: The costs of depression: Direct and indirect; treatment versus nontreatment. Journal of
Clinical Psychiatry 59:1114, 1998.
10. Linton SJ, Bergbom S.: Understanding the link between depression and pain. Scandinavian Journal of
Pain 2:4754, 2011.
11. Bohnert AS, Ilgen MA, Trafton JA, Kerns RD, Eisenberg A, Ganoczy D, et al.: Trends and regional
variation in opioid overdose mortality among veterans health administration patients, fiscal year 2001 to
2009. Clinical Journal of Pain 30:605612, 2009.
12. Manchikanti L, Fellows B, Ailinani H, Pampati V.: Therapeutic use, abuse, and nonmedical use of
opioids: A ten-year perspective. Pain Physician 13:401435, 2010.
13. Sites BD, Beach ML, Davis MA.: Increases in the use of prescription opioid analgesics and the lack of
improvement in disability metrics among users. Regional Anesthesia and Pain Medicine 39:612, 2014.
14. Wilson JF.: Strategies to stop abuse of prescribed opioid drugs. Annals of Internal Medicine
146:897900, 2007.
15. U.S. Department of Health and Human Services.: Results from the 2011 national survey on drug use and
health: Summary of national findings (Report No. SMA 12-4713). Retrieved from http://www.samhsa.
gov/data/nsduh/2k11results/nsduhresults2011.htm, 2011.
16. Bair MJ, Wu J, Damush TM, Sutherland JM, Kroenke K.: Association of depression and anxiety alone
and in combination with chronic musculoskeletal pain in primary care patients. Psychosomatic Medi-
cine 70:890897, 2008.
17. Kroenke K, Bair MJ, Damush TM, Wu J, Hoke S, Sutherland J, et al.: Optimized antidepressant therapy
and pain self-management in primary care patients with depression and musculoskeletal pain. JAMA
301:20992110, 2009.
18. Haskell SG, Heapy A, Reid MC, Papas RK, Kerns RD.: The prevalence and age-related characteristics
of pain in a sample of women veterans receiving primary care. Journal of Womens Health 15:862869,
2006.
19. Kerns RD, Otis J, Rosenberg R, Reid MC.: Veterans reports of pain and associations with ratings of
health, health-risk behaviors, affective distress, and use of the healthcare system. The Journal of
Rehabilitation Research and Development 40:371380, 2003.
20. Dobscha SK, Morasco BJ, Duckart JP, Macey T, Deyo RA.: Correlates of prescription opioid initiation
and long-term opioid use in veterans with persistent pain. Clinical Journal of Pain 29:102108, 2013.
21. Gallagher RM, Rosenthal LJ.: Chronic pain and opioids: Balancing pain control and risks in long-term
opioid treatment. Archives of Physical Medicine and Rehabilitation 89:S7782, 2008.
22. Staiger TO, Gaster B, Sullivan MD, Deyo RA.: Systematic review of antidepressants in the treatment of
chronic low back pain. Spine 28:25402545, 2003.
23. Kroenke K, Krebs EE, Bair MJ.: Pharmacotherapy of chronic pain: A synthesis of recommendations
from systematic reviews. General Hospital Psychiatry 31:206219, 2009.

John J. Sellinger, PhD is a Staff Psychologist and Director of the Clinical Health Psychology Service at VA
Connecticut Healthcare System, and Assistant Professor of Psychiatry (Psychology) at the Yale University
School of Medicine.

Mehmet Sofuoglu, MD, PhD is Director of the New England Mental Illness Research, Education and
Clinical Center (MIRECC), and Professor of Psychiatry at the Yale University School of Medicine.

Robert D. Kerns, PhD is Director of the Pain Research, Informatics, Multimorbidities and Education
(PRIME) Center at VA Connecticut Healthcare System, Special Advisor for Pain Research, Veterans health

123
Psychiatr Q (2016) 87:585593 593

Administration, and Professor of Psychiatry, Neurology and Psychology at the Yale University School of
Medicine.

Robert A. Rosenheck, MD is a Senior Investigator in health Services Research at the New England Mental
Illness Research, Education and Clinical Center (MIRECC), and Professor of Psychiatry and of Health
Policy at the Yale University School of Medicine.

123