Testing the Effectiveness of the Exercise Plus Program in Older Women PostHip Fracture

Barbara Resnick, Ph.D., CRNP

University of Maryland School of Nursing

Denise Orwig, Ph.D.

University of Maryland School of Medicine

Janet Yu-Yahiro
Union Memorial Hospital, Baltimore, Maryland

William Hawkes, Ph.D., Michelle Shardell, Ph.D., and J. Richard Hebel, Ph.D.
University of Maryland School of Medicine

Sheryl Zimmerman, Ph.D.

University of North Carolina Chapel Hill

Justine Golden, M.S., Michele Werner, B.S., and Jay Magaziner, Ph.D.
University of Maryland School of Medicine

it was possible to engage these women in a home-based

exercise program and that the plus only, exercise only, and
the exercise plus groups all increased exercise.

ABSTRACT
Background: Exercise is an important strategy with
potential to improve recovery in older adults following a
hip fracture. Purpose: The purpose of this study was to test
the impact of a self-efficacy based intervention, the Exercise
Plus Program, and the different components of the intervention, on self-efficacy, outcome expectations, and exercise
behavior among older women posthip fracture. Methods:
Participants were randomized to one of four groups: exercise
plus, exercise only, plus only (i.e., motivation), or routine
care. Data collection was done at baseline (within 22 days
of fracture), 2, 6, and 12 months posthip fracture. Results:
A total of 209 women were recruited with an average age of
81.0 years (SD 6.9). The majority was White (97.1%),
was widowed (57.2%), and had a high school education
(66.7%). Generalized Estimating Equations were used to
perform repeated measures analyses. No differences in trajectories of recovery were observed for self-efficacy or outcome expectations. A statistically significant difference in
the overall trajectory of time in exercise was seen
(p < .001), with more time spent exercising in all three
treatment groups. Conclusions: The study demonstrated that

(Ann Behav Med

2007, 34(1):6776)

INTRODUCTION
In 2003 there were more than 309,500 hospital admissions for hip fractures, and by the year 2030 over 650,000
hip fractures will occur annually in older adults (1). In
the 1st year of their fracture, approximately 30% of these
individuals will die (1), 25% will have significant functional
decline in activities of daily living such as bathing and
dressing (2), 20% will need help with lower extremity dressing, and 90% will require help climbing the stairs (3). Individuals who have had a hip fracture are noted to be
impaired in their ability to independently rise from an armless chair or to step symmetrically (4), and approximately
38% to 50% need assistance to walk or are unable to walk
at 12 months posthip fracture (5).
An important strategy with potential to improve
recovery in older adults following a hip fracture is exercise
training. Regular exercise, even at low levels of intensity,
can improve mental and physical health in older adults
(6). Despite the well-documented benefits of exercise, only
23% to 35% of men engage in regular exercise, and this is
even lower in women ranging from 17% to 32% (7,8).
Developing successful interventions to help motivate individuals after a hip fracture to adhere to regular exercise
programs has the potential to improve recovery and overall
quality of life.

Support for this project was provided by National Institute

on Aging grants R37 AG09901, R01-AG18668, R01 AG17082,
and the Claude D. Pepper Older Americans Independence Center
P60-AG12583. We thank Thera-Band Academy for their
generous contribution of Thera-Band1 resistive bands used by
study participants, hospitals and personnel participating in the
Baltimore Hip Studies, and research staff who worked with study
patients and their families. We also thank hip fracture patients
and their families for volunteering their time and information
for this work.
Reprint Address: B. Resnick, Ph.D., CRNP, University of Maryland
School of Nursing, 655 West Lombard Street, Baltimore, MD
21201. E-mail: barbresnick@aol.com

Theoretical Basis of the Intervention

Social cognitive theory, based on triadic reciprocity
suggesting that behavior, cognitive, and other personal

# 2007 by The Society of Behavioral Medicine.

67

68

Annals of Behavioral Medicine

Resnick et al.

factors and environmental influences all operate interactively as determinants of each other, incorporates the
factors that most consistently influence exercise behavior
(911). Social cognitive theory considers (a) self-efficacy
expectations, which are the individuals beliefs in their
capabilities to perform a course of action to attain a desired
outcome, and (b) outcome expectations, which are the
beliefs that a certain consequence will be produced by personal action. Exposure to enactive mastery experiences,
such as participation in an exercise class, is the most common intervention used to strengthen efficacy expectations
in older adults (1215). Verbal encouragement from a
trusted, credible source in the form of counseling and education has been used alone, and with enactive mastery, to
strengthen efficacy expectations (16,17). Less frequently,
vicarious experience (15,18), self-modeling (19), or
implementation of interventions to decrease the unpleasant
sensations or augment the pleasant sensations associated
with exercise have been used to strengthen self-efficacy
and increase adherence to exercise (18,20,21).
Behavior can also be altered by the outcome expectations the individual maintains, with expected outcomes
being physical or mental health changes, social effects, or
personal evaluations of behavior. In cross-sectional studies
with older adult samples, outcome expectations related to
exercise have had a significant influence on exercise behavior (2224) and seem to be particularly important to the
prediction of exercise behavior over time (25,26).
Several researchers (15,16,18,27) have utilized a variety
of sources of efficacy information to increase exercise
among older adults. These included verbal encouragement,
education about the benefits of exercise, social support,
manipulation of affective states, self-modeling, and cueing.
McAuley et al. (15,27) noted that the social environment
and affective states of the participants directly and
indirectly influence self-efficacy expectations and exercise
behavior, beyond the effects of enactive mastery. Cumulatively, these studies suggest that in healthy community
dwelling older adults there may be some benefit to adding
sources of efficacy information, in addition to enactive
mastery, when trying to strengthen the individuals beliefs
about their ability to exercise and increase exercise behavior. It is not clear, however, if exposing older individuals
who have sustained an acute medical event (i.e., a hip fracture) to multiple sources of efficacy information will
strengthen self-efficacy and outcome expectations or
increase time spent in exercise over the first 12 months
posthip fracture.
The purpose of this study, therefore, was to test the
Exercise Plus Program and the separate components of
the Exercise Plus Program (exercise only and plus only
components) using a four-group repeated measure design.
Specifically, the following was hypothesized:
1. Participants who were exposed to exercise only (exercise
trainer exercising with the individual) spent more time in

exercise and had stronger self-efficacy and outcome

expectations related to exercise compared to those who
received routine care.
2. Participants who were exposed to plus only (verbal persuasion to exercise through education and goal setting,
self-modeling, and manipulation of affective states)
spent more time in exercise and had stronger selfefficacy and outcome expectations related to exercise
compared to those who received routine care.
3. Participants who were exposed to exercise plus (exercise
trainer exercising with the individual and verbal persuasion, self-modeling, and manipulation of affective
states) spent more time in exercise and had stronger
self-efficacy and outcome expectations related to
exercise compared to those who received routine care.

METHODS
Study Design
This study was a randomized controlled trial using a
repeated measure design with participants randomized to
one of four groups: exercise plus, exercise only, plus only,
or routine care. Baseline testing was conducted within the
first 22 days posthip fracture and random assignment
was done after baseline testing was complete. Follow-up
data collection was done at 2, 6, and 12 months posthip
fracture.
Sample
Participants were recruited from nine hospitals in the
greater Baltimore area (see Figure 1). Eligible participants
were female, were 65 years of age or older, were community
dwelling at the time of fracture, had a nonpathologic fracture within 72 hr of admission, had surgical repair of the
hip fracture, and were free of medical problems that would
potentially put them at particular risk for falls when exercising alone at home alone (e.g., neuromuscular conditions). Participants also had to be walking without
human assistance prior to the fracture and score 20 or
higher on the Folstein Mini Mental State Examination
(28). Institutional Review Board approvals were obtained
from the University of Maryland School of Medicine as
well as the study hospitals, and all enrolled participants
provided their own informed consent. Recruitment was
initiated in August 2000, and data collection on the final
participant was completed in September 2005. A Data
and Safety Monitoring Board met quarterly and reviewed
all adverse events and safety reports.
The Intervention: The Exercise Plus Program
The Exercise Plus Program incorporates the exercise
only and plus only components of the intervention
(described next). The theoretical premise of the Exercise
Plus Program has been described in detail elsewhere (29,30).

Volume 34, Number 1, 2007

Exercise Plus Program

69

FIGURE 1 Flow of participants through the trial: Testing the effectiveness of the Exercise Plus Program.

In all treatment groups, visits from the trainer were initially

twice a week and then decreased to once a month in the
final 4 months of the program, with weekly telephone calls
for those exposed to the plus component of the intervention during the weeks when no visit was scheduled.
All visits lasted 1 hr. The goal was to initiate the intervention as soon as Medicare-covered rehabilitation services
were completed. Assuming that the participant completed
all Medicare-covered rehabilitation services by 1 month
postfracture, the maximum number of anticipated visits
was 38.

Exercise plus intervention. The exercise component

exposes the participant to an exercise session with an
exercise trainer. The exercise sessions incorporated
aerobic exercise using a Stairstep (a specially designed
4-in. stair step with handles on either side for support
and balance) (2931), a comprehensive strengthening
program that covers the main muscle groups relevant to
hip fracture recovery, and stretching exercises (these
were part of the warm up and cool down periods).

Participants were told to perform aerobic activity at least

3 days per week and strength training 2 days per week
for 30 min. Each participant started at her own individual
level with regard to the time spent in aerobic activities
and the amount of repetitions and resistance used in the
strengthening program.
Strength training consisted of a series of 11 exercises
for the upper and lower extremities using Thera-band
products (The Hygenic Corporation, Akron, OH) and=or
ankle and wrist cuff weights. Generally, the duration of
each exercise was increased until the participant could do
three sets of 10 repetitions on the right and left sides. Intensity was then augmented by increasing the resistance of
exercise bands or tubes and=or adding ankle and wrist cuff
weights according to a standard protocol (30). These exercises are available from the first author upon request.
The plus component of the Exercise Plus Program was
implemented by the exercise trainer as part of the same visit
and included education about the benefits of exercise post
hip fracture using an investigator-developed booklet Exercise After Your Hip Fracture, verbal encouragement
through goal setting and positive reinforcement, exposure

70

Resnick et al.

Annals of Behavioral Medicine

to interventions to decrease the unpleasant sensations associated with exercise such as relieving or decreasing pain
with prescribed medications or use of heat=ice treatment,
and cueing with posters describing the exercises, a Goal
Form, and a calendar of daily exercise activities (29). A
detailed description of the plus component is available
from the first author upon request.

exercising. In this study there was evidence of internal

consistency with an alpha coefficient of .91. The
Outcome Expectations for Exercise Scale (34) is a nineitem measure (range 15) that focuses on the perceived
consequences of exercise for older adults and likewise
there was evidence of internal consistency with an alpha
coefficient of .88.

Plus only intervention. Participants in the plus only

group did not exercise with the trainer. Rather the
sessions focused only on the plus components (i.e.,
education about exercise, verbal encouragement, removal
of unpleasant sensations, and cueing).

Exercise behavior. Exercise behavior was measured

based on the Yale Physical Activity Survey (35) and the
Step Activity Monitor (SAM). The Yale Physical Activity
Survey is an interviewer-administered questionnaire
focusing on five major categories of common groups of
activities such as work, exercise, and recreational
activities that older adults engage in. In this study the
Exercise subscale was the only data reported as this was
the focus of the intervention. The SAM provides a count
of the number of steps taken at specified intervals
(3638) and was placed on the participant for 48 hr at the
2-, 6-, and 12-month follow-up period.

Exercise only intervention. Participants in the exercise

only group exercised with the trainer during the session
but were not exposed to the plus components of the
intervention and were not provided with any education
about exercise, verbal feedback, interventions to decrease
unpleasant sensations, or encouragement to exercise
posthip fracture.
Routine care group. Those randomized to routine care
received no intervention-related visits. Routine care for
older adults posthip fracture involved rehabilitation
services based on Medicare guidelines. This generally
included inpatient physical and occupational therapy
based on the functional needs of the individual and in
most cases a single home therapy evaluation for safety.
Randomization
Randomization was performed using a freeware computer program (32). Patient assignment was blocked by
hospital to assure equal probabilities within each hospital
being assigned to each of the four study groups. Patients
were assigned to groups at random with forced balancing
of treatment groups within hospital. The resulting randomization scheme was given to the project coordinator
and patients assigned as they became available at the indicated hospital. The study nurses involved with recruitment
and data collection were blind to randomization. Study
participants were not informed of what specific arm of
the intervention they were randomized to (i.e., exercise
only, plus only, or exercise plus).
Study Outcomes
Study outcomes focused on self-efficacy and outcome expectations related to exercise and time in exercise
activities.
Efficacy expectations. The Self-efficacy for Walking=
Exercise Scale (33) is a nine-item measure (range 010)
that focuses on self-efficacy expectations related to the
ability to continue to exercise in the face of barriers to

Descriptive Variables
Age, marital status, educational level, and living situation prior to hip fracture (whether they live alone or with
others) were obtained. Chart reviews were done to describe
medical history, surgical intervention, and hospital course
(see Table 1).
Statistical Analysis
The intention-to-treat principle was followed in all
analyses used for assessing the effect of treatment on
outcome, that is, all participants who were randomized
were included. Generalized Estimating Equations (39)
(GEEs) were used to perform repeated measures analyses
with the outcome measures described previously as dependent variables. The independent variables included dummy
variables to indicate the exercise plus, exercise only, and
plus only interventions (the control condition serving as
the reference). Additional dummy variables were used to
indicate the 2-, 6-, and 12-month time points (baseline serving as the reference). Interactions between the intervention
and time variables were also included as fixed effects in the
longitudinal model. This model was used to estimate the
mean and standard error of the outcome measure at each
time point for each of the four treatment groups. Robust
standard error estimates were obtained using a technique
described by Huber (40). A global p value for the differences in longitudinal trajectories among the four groups
was obtained from a test of the null hypothesis that all
the Intervention
Time interaction coefficients in the
model are simultaneously zero.
Time-specific between-group contrasts were tested at
the 5% level using Wald statistics derived from the linear
combination of model coefficients used to estimate the dif-

Volume 34, Number 1, 2007

Exercise Plus Program

71

TABLE 1
Descriptive Data for Participants by Treatment Group
Variable
No.
Age
Race
White
African American
Marital status
Never married
Married
Widowed
Divorced
Separated
Education
None
Elementary
High (712)
College
Postgraduate
Fracture site
Intertrochanteric
Subcapital
Subtrochanteric
Other
Anesthesia risk rating
Physical Therapy (PT) sessions
Length of hospital stay
Charlson Co-Morbidity Index
Residence at discharge
Community
Skilled nursing
Rehabilitation
Nursing home: No rehab
Unknown

Exercise Only

Plus Only

Exercise Plus

Routine Care

All Participants

51
82.4  7.9

54
80.6  6.9

52
81.4  5.8

51
79.7  6.7

208
81.0  6.9

51 (100.0%)
0

51 (94.4%)
3 (5.6%)

50 (96.2%)
2 (3.9%)

50 (98.0%)
1 (2.0%)

202 (97.1%)
6 (2.9%)

1 (2.0%)
17 (33.3%)
30 (58.8%)
2 (3.9%)
1 (2.0%)

1 (1.9%)
20 (37.0%)
31 (57.4%)
1 (1.9%)
1 (.9%)

4 (7.7%)
19 (36.5%)
27 (51.9%)
2 (3.9%)
0

1 (2.0%)
15 (29.4%)
31 (60.8%)
3 (5.9%)
1 (2.0%)

7 (3.4%)
71 (34.1%)
119 (57.2%)
8 (3.9%)
3 (1.4%)

0
2 (4.0%)
32 (64.0%)
12 (24.0%)
4 (8.0%)

0
0
37 (68.5%)
13 (24.1%)
4 (7.4%)

0
2 (3.9%)
33 (63.5%)
11 (21.2%)
6 (11.5%)

0
2 (3.9%)
36 (70.6%)
10 (19.6%)
3 (5.9%)

0
6 (2.9%)
138 (66.7%)
46 (22.2%)
17 (8.2%)

23 (45.1%)
22 (43.1%)
5 (9.8%)
1 (2.0%)
2.5  .5
4.3  3.2
3.8  .9
.7  1.3

16 (29.6%)
38 (70.4%)
0
0
2.7  .6
4.9  6.4
4.2  1.1
1.3  1.3

22 (42.3%)
27 (51.9%)
2 (3.9%)
1 (1.9%)
2.6  .6
5.1  4.6
4.3  1.6
1.1  1.1

27 (52.9%)
22 (43.1%)
1 (2.0%)
1 (2.0%)
2.6  .6
6.0  5.9
4.4  1.5
1.6  1.6

88 (42.3%)
109 (52.4%)
8 (3.9%)
3 (1.4%)
2.6  .6
5.1  5.2
4.2  1.3
1.2  1.4

4 (7.8%)
29 (56.9%)
18 (35.3%)
0
0

4 (7.4%)
25 (46.3%)
24 (44.4%)
1 (1.9%)
0

1 (1.9%)
22 (42.3%)
29 (55.8%)
0
0

1 (2.0%)
20 (39.2%)
29 (56.9%)
0
1 (2.0%)

10 (4.8%)
96 (46.2%)
100 (48.1%)
1 (.5%)
1 (.5%)

Note. Patients with angina, myocardial infarction, stroke, transient ischemic attack, deep venous thrombosis, seizures, and gastrointestinal
hemorrhage in the 6 months before admission were not eligible for the study.

ference in means and its standard error. To investigate the

potential for bias due to missing outcome data, sensitivity
analyses were performed using weighted estimating equations (WEEs) (41). Weights were the inverse estimated
probabilities of being observed at each visit derived from
a logistic regression of observed status (yes=no) on time,
group, their interaction, and baseline factors (age, comorbidities, Geriatric Depression Scale (GDS), and prefracture
Activities of Daily Living (ADLs). Results from GEEs are
only unbiased when data are missing completely at random
in the sense of Rubin (42). WEEs using the estimated
weights will produce unbiased results if the data are
missing at random (42).
RESULTS
Table 1 provides descriptive data for participants by
treatment group. Overall, the average patient was 81.0

years old (SD 6.9), all were female, and 97.1% were
White. The majority (57.2%) was widowed and had a
high school education (66.7%). Of the 209 initially
recruited into Testing the Effectiveness of the Exercise
Plus Program, 165 women (79%) were available for 2month assessments, 169 (81%) were available for 6-month
follow up, and 155 (75%) were available for the 12-month
follow up visits. One case was deleted postrandomization
due to being ineligible (no surgery was performed post
hip fracture). Reasons for loss to follow up are shown
in Table 2.
The time from fracture to first intervention visit from
the trainer ranged from 28 to 200 days. Participants generally were not willing to have a visit occur prior to 60 days
postfracture, and only 1 participant had her first visit at 28
days postfracture, 22 (31%) of the participants had their
first visit by 60 days postfracture, 44 (62%) of the participants had their first visit by 90 days postfracture, and

72

Annals of Behavioral Medicine

Resnick et al.
TABLE 2
Reasons for Loss to Follow-up by Treatment Group

Follow-Up Point
Two monthe
Scheduling=unable to contact
Illness
Refusal of study (cumulative)
Refusal of measures
Death (cumulative)
Other
Six monthf
Scheduling=unable to contact
Illness
Refusal of study (cumulative)
Refusal of measures
Death (cumulative)
Other
Twelve monthg
Scheduling=unable to contact
Illness
Refusal of study (cumulative)
Refusal of measures
Death (cumulative)
Other

Usual Carea
2
2
2
3
0
0

(4%)
(4%)
(4%)
(6%)
(0%)
(0%)

4
0
3
0
1
0

(8%)
(0%)
(6%)
(0%)
(2%)
(0%)

1
0
3
2
3
1

(2%)
(0%)
(6%)
(4%)
(6%)
(2%)

Exercise Onlyb

Plus Onlyc

Exercise Plusd

0 (0%)
0 (0%)
4 (10%)
6 (15%)
0 (0%)
0 (0%)
0 (0%)
1 (3%)
10
(26%)
0 (0%)
0 (0%)
1 (3%)

2
0
2
5
1
0

(5%)
(0%)
(5%)
(11%)
(2%)
(0%)

1
3
5
4
0
0

(3%)
(8%)
(13%)
(10%)
(0%)
(0%)

0
0
2
3
1
2

(0%)
(0%)
(4%)
(7%)
(2%)
(4%)

1
0
5
4
1
0

(2%)
(0%)
(12%)
(10%)
(2%)
(0%)

2 (6%)
1 (3%)
10
(29%)
0 (0%)
2 (6%)
1 (3%)

3
0
2
4
1
1

(7%)
(0%)
(5%)
(9%)
(2%)
(2%)

4
0
5
5
2
0

(11%)
(0%)
(14%)
(14%)
(5%)
(0%)

a
n 51. bn 51. cn 54. dn 52. eUsual care, n 42; exercise only, n 40; plus only, n 44; exercise plus, n 39. fUsual care, n 43; exercise only, n 39; plus only, n 46; exercise plus, n 41. gUsual care, n 41; exercise only, n 35; plus only, n 43; exercise plus, n 36.

58 (82%) of the participants had their first visit by 120 days

postfracture.
The mean number of intervention visits was 20.9
(SD 15.3), and the exercise only group participated in
17.3 (SD 15.6) visits, the plus only group participated
in 24.4 (SD 14.5) visits, and the exercise plus group participated in 21.2 (SD 15.3) visits. Based on analysis of
variance these differences were not statistically significant
(F 2.8, p .06). Eighteen of the 51 individuals randomized to exercise only (35%), 15 of the 52 individuals randomized to exercise plus (29%), and 8 of the 54 individuals
randomized to plus only (15%) were not willing to have
any intervention visits.
Results for outcome measures are presented in Table 3.
No differences in trajectories of recovery were statistically
significant at the .05 level for self-efficacy or outcome
expectations. A statistically significant difference in the
overall trajectory of time in exercise (in hours) was seen
(p < .001) with all treatment groups spending more time
in exercise than the control group (see Figure 2).
The SAM also showed a statistically significant
difference (p .03) in the trajectories of the groups (see
Figure 3). At 12 months postfracture, the exercise only
group showed higher activity levels (M 6,459 steps,
SE 968 steps) as did the plus only group (M 6,994
steps, SE 1,012 steps) relative to the control group
(M 4,060 steps, SE 623 steps). The exercise plus group
did not show statistically significantly higher step activity

levels based on the SAM at the .05 level when compared

to the control group.
The estimated trajectories varied slightly when the data
were estimated assuming missing at random using WEE;
however, the directions of changes in the estimates were
not consistent across outcomes (data not shown). The
conclusions derived from the original GEE analyses did
not change and hence are robust to departures from the
missing completely at random assumption.

Treatment Fidelity
Treatment fidelity was considered with regard to study
design, training, delivery of treatment, and receipt of treatment (43). Training of the interventionists was completed
as delineated in the procedure manual, and retraining
was ongoing through monthly meetings of the trainers with
the investigators who developed the exercise only and plus
only interventions. The exercise only group on average
received 45% of the total possible visits, the plus only
group received 63% of the total possible visits, and the
exercise plus group received 55% of the total possible visits. Seventy direct observations (five different interventionists) of the interventionists were completed by two of the
investigators. Overall, there was a 90.8% adherence to
delivery of the intervention across all of the treatment
groups, and in 92% of the observed visits the participants

Volume 34, Number 1, 2007

Exercise Plus Program

73

TABLE 3
Predicted Means (SE) and Sample Sizes for Outcome Measures for Each Treatment Group
Exercise Only
Variable
Self-efficacy expectations
Baseline
2 month
6 month
12 month
Outcome expectations
Baseline
2 month
6 month
12 month
Exercise time in hours
Baseline
2 month
6 month
12 month

Plus Only

Exercise Plus

Routine Care

SE

SE

SE

SE

5.24
6.61
7.35
7.84

0.36
0.45
0.40
0.28

51
40
38
35

5.33
6.30
7.20
7.14

0.34
0.42
0.33
0.45

54
44
46
42

5.54
6.63
7.14
7.30

0.32
0.45
0.45
0.39

52
39
41
36

5.31
6.45
5.89
6.35

0.37
0.36
0.47
0.51

51
42
43
40

.28

3.92
3.91
3.97
4.07

0.10
0.11
0.09
0.09

51
40
39
35

3.99
3.84
3.93
3.89

0.10
0.11
0.09
0.11

54
44
46
42

3.92
4.07
3.86
3.94

0.10
0.09
0.10
0.10

51
39
41
36

3.95
3.88
3.84
3.73

0.08
0.09
0.10
0.10

51
42
43
40

.13

1.21
1.77
2.27
3.34

0.25
0.36
0.29
0.66

51
40
39
35

0.92
1.49
2.91
3.00

0.35
0.28
0.48
0.55

54
44
46
43

0.76
2.08
2.59
3.10

0.25
0.42
0.63
0.65

52
39
41
36

0.66
1.70
1.02
0.92

0.20
0.36
0.25
0.23

51
42
43
40

< .001

Note. n 208.

demonstrated evidence that they received the intervention

as intended.

DISCUSSION
Partially supporting the stated hypotheses, this study
demonstrated that exposure to the full Exercise Plus Program, the exercise only component, or the plus only
component increased reported time in exercise activities
among older women posthip fracture. This subjective
report was further supported by the objective data from
the SAM for the exercise only and motivation only interventions, but not for exercise plus. As has been suggested

FIGURE 2 Longitudinal trajectories of time (hours) spent exercising as measured by the Yale Physical Activity Survey.

in other exercise intervention studies (12,13,44,45), this

study demonstrated that there is benefit to exposing individuals to sources of efficacy information as a way to
increase exercise behavior. It is difficult to interpret the
lack of a statistically significant increase in the number of
steps noted in the exercise plus group. It is possible that
these individuals were spending more time in resistive
exercise activities that were not captured by the SAM.
Conversely, it is possible that combined exercise and plus
components in the same hour-long session resulted in a
less intense exposure to either the exercise activity or the
plus activities thus reducing the impact of the intervention.

FIGURE 3 Longitudinal trajectories of activity (steps) as measured by the Step Activity Monitor (SAM).

74

Resnick et al.

This study further attempted to compare the impact of

different components of the intervention. It should be
recognized, however, that it is impossible to control
exposure to sources of efficacy information in real-world
settings. The plus only participants had mastery experiences if and when they did exercise, although this was
not with the trainer. Likewise the exercise only participants, by virtue of the exercise program they were given,
had goals they could establish for themselves and also
received feedback (albeit not the same as the verbal persuasion provided through the plus component) about successfully or not successfully completing the exercise during
the course of their exercise session. It is therefore impossible to conclude that mastery alone is sufficient or that verbal persuasion, vicarious experience, or altering unpleasant
physiological and affective states is sufficient to motivate
older women posthip fracture to exercise. Future research
is needed to replicate these findings as demonstrating the
beneficial impact of the plus component alone has practical
implications in clinical settings.
On average the treatment group participants reported
that they engaged in 2 to 3 hr of moderate intensity exercise
each week, which approached the recommended 30 min of
moderate intensity physical activity on most if not all days
of the week (46). Conversely, the no-treatment control
group demonstrated a decline in the amount of time they
engaged in exercise activities between 2 and 6 months postfracture, and this remained at a low level of 1 hr weekly at
the 12-month follow-up point. Clinically, the long-term
impact of these differences in time spent exercising cant
be addressed given this study design. The benefits, however, may include a variety of physical and psychological
factors such as improved blood pressure, mood, endurance,
strength and=or bone mineral density.
Although there were trends indicating that individuals
in the treatment groups had a greater increase in their selfefficacy and outcome expectations when compared to controls, these differences were not statistically significant.
This may, in part, be due to measurement issues. Given
that these women volunteered to participate in an exercise
intervention study they generally had strong self-efficacy
and outcome expectations at baseline. It is also possible
changes might have been noted if the participants engaged
in all of the intended visits. In particular it is possible that
intervening earlier (i.e., in the first 60 days postfracture), as
was intended, may have resulted in a greater impact on selfefficacy and outcome expectations. In addition, we focused
only on self-efficacy with regard to challenges associated
with exercise, not confidence in the performance of the
actual exercises prescribed which may have been strengthened given the intervention.
The lack of a statistically significant improvement in
self-efficacy and outcome expectations has, however, been
previously reported (15,45,4749). McAuley et al. (15) suggested that a decline in self-efficacy following exposure to
an exercise intervention can also occur when there is a

Annals of Behavioral Medicine

decrease in exposure to exercise classes (or in this case the
trainer in the home setting), when exposed to a new exercise
program, when there is a change in clinical condition or
ability, or when the exercise program is progressively more
challenging. The study presented here focused on older
women who had just sustained an acute change in clinical
condition due to the hip fracture. We believe these individuals may need time to accurately evaluate their self-efficacy related to exercise following the hip fracture. Their
self-efficacy evaluations may have reflected prefracture
self-efficacy expectations rather than a true reflection of
their confidence related to exercise following the fracture.
The findings from this study are limited in that the
older women in this study were relatively healthy, lived in
the community prior to fracture, and willingly participated
in an exercise intervention study. In addition there was
variability in terms of when the intervention was initiated
(due to participant willingness to allow the trainer to come
out to the home setting) and some group differences
in number of visits to which the participant was exposed.
A major concern in this study was the inability to truly
control the type and amount of efficacy information that
participants were exposed to, limiting our interpretation
of the findings. There were also challenges to measurement
as multiple measures were based on recall. Despite
these limitations, the study demonstrated that it was possible to engage these women in a home-based exercise program and that the plus only, exercise only, or the
combined exercise plus program resulted in the desired outcome of increased time in exercise. For practical purposes,
replication studies are recommended to determine if the
exposure of older women posthip fracture to the plus
component alone is sufficient to increase time spent in exercise as the plus only component only could be easily implemented during follow-up orthopedic visits, daily
encounters with nursing assistants in assisted living and
nursing home settings and by family members and friends
in the home setting.

REFERENCES
(1) National Center for Health Statistics, Centers for Disease
Control and Prevention: Centers for Disease Control and
National Nursing Home Survey (NNHS) Public-Use Data
Files. Retrieved December 2006 from http://www.cdc.gov/
nchs/products/elec prods/subject/nnhs.htm
(2) Cree M, Carriere KC, Soskolne CL, Suarez-Almazor M:
Functional dependence after hip fracture. American Journal
of Physical Medicine and Rehabilitation. 2001, 80:736743.
(3) Magaziner J, Hawkes WG, Hebel JR, et al.: Recovery from
hip fracture in eight areas of function. Journal of Gerontology: Medical Sciences. 2000, 55A:M498M507.
(4) Fox KM, Hawkes WG, Hebel JR, et al.: Mobility after hip
fracture predicts health outcomes. Journal of the American
Geriatrics Society. 1998, 46:169173.
(5) Boonen S, Autier P, Barette M, et al.: Functional outcome
and quality of life following hip fracture in elderly women:

Volume 34, Number 1, 2007

(6)

(7)

(8)

(9)

(10)

(11)
(12)

(13)

(14)

(15)

(16)

(17)

(18)
(19)

(20)

(21)

A prospective controlled study. Osteoporosis International.

2004, 15:8794.
Habris R, OHara P, Harper D: Functional status of geriatric rehabilitation patients: a one year follow-up study.
Journal of the American Geriatric Society. 1995, 36:890896.
United States Department of Health and Human Services,
Public Health System, National Institute of Heart Lung
and Blood Institute: Physical Activity and Health: A Report
of the Surgeon General, NIH Publication No. 013305 M.
Bethesda, MD: National Institute of Heart Lung and Blood
Institute, 2001.
Centers for Disease Control and Prevention, Merck Institute of Aging and Health: The State of Aging and Health
in America. Retrieved March 2, 2007 from http://apps.
nccd.cdc.gov/saha/
King A, Rejeski J, Buchner D: Physical activity interventions targeting older adults: A critical review and recommendations. American Journal of Preventive Medicine.
1998, 15:316333.
Trost SG, Owen N, Bauman AE, Sallis JF, Brown W: Correlates of adults participation in physical activity: Review
and update. Medicine Science Sports and Exercise. 2002,
34:19962001.
Bandura A: Self-Efficacy: The Exercise of Control.
New York: W.H. Freeman, 1997.
Estabrooks PA, Fox EH, Doerksen SE, Bradshaw MH,
King AC: Participatory research to promote physical
activity at congregate-meal sites. Journal of Aging and
Physical Activity. 2005, 13:121144.
Gyurcsik N, Estabrooks P, Frahm-Templar M: Exerciserelated goals and self-efficacy as correlates of aquatic
exercise in individuals with arthritis. Arthritis Rheumatology. 2003, 49:306313.
Li F, Fisher KJ, Harmer P, McAuley E: Falls self-efficacy
as a mediator of fear of falling in an exercise intervention
for older adults. Journal of Gerontology Series B: Psychological Sciences and Social Sciences. 2005, 60:3440.
McAuley E, Konopack JF, Motl RW, et al.: Physical
activity and quality of life in older adults: Influence
of health status and self-efficacy. Annals of Behavioral
Medicine. 2006, 31:99103.
Hiltunen EF, Winder PA, Rait MA, et al.: Implementation
of efficacy enhancement nursing interventions with cardiac
elders. Rehabilitation Nursing. 2005, 30:221229.
Moore SM, Charvat JM, Gordon NH, et al.: Effects of a
CHANGE intervention to increase exercise maintenance
following cardiac events. Annals of Behavioral Medicine.
2006, 31:5362.
Resnick B: Efficacy beliefs in geriatric rehabilitation.
Journal of Gerontological Nursing. 1998, 24:3445.
Resnick B: Testing the impact of the WALC intervention
on exercise adherence in older adults. Journal of Gerontological Nursing. 2002, 28:3240.
Rejeski WJ, Katula J, Rejeski A, Rowley J, Sipe M:
Strength training in older adults: does desire determine confidence? Journal of Gerontology Series B: Psychological
Sciences and Social Science. 2005, 60:P335337.
Resnick B, Luisi D, Vogel A: Reliability and validity of the
SEE and OEE with African American and Hispanic older
adults. Journal of Nursing Measurement. 2005, 12:235247.

Exercise Plus Program

75

(22) Jette A, Rooks, D, Lachman, M, et al.: Home-based resistance training: Predictors of participation and adherence.
The Geronotologist. 1998, 38:412421.
(23) Resnick B, Palmer MH, Jenkins LS, Spellbring AM:
Path analysis of efficacy expectations and exercise behavior
in older adults. Journal of Advanced Nursing. 2000, 31:
13091315.
(24) Resnick B, Luisi D,Vogel A: Reliability and validity of the
SEE and OEE with African American and Hispanic older
adults. Journal of Nursing Measurement. 2005, 12:235247.
(25) Neff K, King A: Exercise program adherence in older
adults: The importance of achieving ones expected benefits.
Medicine and Exercise in Nutrition and Health. 1995,
4:355362.
(26) Wilcox S, Castro C, King A: Outcome expectations and
physical activity participation in two samples of older
women. Journal of Health Psychology. 2006, 11:6577.
(27) McAuley E, Jerome GJ, Elavsky S, Marquez DX, Ramsey
SN: Predicting long-term maintenance of physical activity
in older adults. Preventive Medicine. 2003, 37:110118.
(28) Folstein M, Folstein S, McHugh P: Mini-Mental state
A practical method for grading the cognitive state of
patients for the clinician. Journal of Psychiatric Research.
1975, 12:189198.
(29) Resnick B, Magaziner J, Orwig D, Zimmerman S: Evaluating
the components of the Exercise Plus Program: Rationale,
theory and implementation. Health Education Research.
2002, 17:648659.
(30) Resnick B, Orwig D, Magaziner J, et al.: Innovative Strategies for Research with Frail Older Adults: The Baltimore
HIP Studies. Society of Behavioral Medicine Conference
Seminar. Baltimore: 2004.
(31) Yu-Yahiro J, Resnick B, Orwig D, Hicks G, Magaziner J:
Design and implementation of a home-based exercise
program post-hip fracture: The Baltimore hip studies
experience. Society of Behavioral Medicine Conference
Seminar. Baltimore: 2004.
(32) Dallal GE: PC-PLAN, Randomization Plans, Version 1.21.
Malden, MA: 1985.
(33) Resnick B, Jenkins L: Testing the reliability and validity of
the Self-Efficacy for Exercise scale. Nursing Research. 2000,
49:154159.
(34) Resnick B, Zimmerman S, Orwig D, Furstenberg A,
Magaziner J: Outcome expectations for exercise scale:
Utility and psychometrics. Journal of Gerontology Social
Sciences. 2000, 55B:S352S356.
(35) Dipietro L, Caspersen CJ, Ostfeld AM, Nadel ER: A survey
for assessing physical activity among older adults. Medicine
Science Sports and Exercise. 1993, 25:628642.
(36) Coleman KL, Smith DG, Boone DA, Joseph AW, del
Aguila MA: Step activity monitor: long-term, continuous
recording of ambulatory function. Journal of Rehabilitation
Research and Development. 1999, 36:818.
(37) Resnick B, Nigg C, Riebe D, Castro C: Measurement
of activity in older adults: A triangulation approach. Gerontological Society of America Conference Symposium.
Washington, DC: 2000.
(38) Shaughnessy M, Michael K, Sorkin JD, Macko RF: Steps
after stroke: Capturing ambulatory recovery. Stroke.
2005, 36:13051307.

76

Resnick et al.

(39) Liang KY, Zeger SL: Longitudinal data analysis using

generalized linear models. Biometrika. 1986, 73:1322.
(40) Huber PJ: The behavior of maximum likelihood estimates
under non-standard conditions. Proceedings of the Fifth
Berkeley Symposium on Mathematical Statistics and
Probability. Berkeley, CA: 1967:221223.
(41) Robins JM, Rotnitzky A, Zhao LP: Analysis of semiparametric regression models for repeated outcomes in the presence of missing data. Journal of the American Statistical
Association. 1995, 90:106121.
(42) Rubin DB: Inference and missing data. Biometrika. 1976,
63(3):581592.
(43) Resnick B, Inguito P, Hawkes W, et al.: Treatment fidelity
in behavior change research: A case example. Nursing
Research. 2005, 54:139143.
(44) Cox KL, Gorely TJ, Puddey IB, Burke V, Beilin LJ:
Exercise behaviour change in 40 to 65-year-old women:
The SWEAT Study (Sedentary Women Exercise Adherence
Trial). British Journal of Health Psychology. 2003, 8:
477495.

Annals of Behavioral Medicine

(45) Norweg AM, Whiteson J, Malgady R, Mola A, Rey M: The
effectiveness of different combinations of pulmonary
rehabilitation program components: A randomized controlled trial. Chest. 2005, 128:663672.
(46) Agency for Healthcare Research and Quality and the
Centers for Disease Control: Physical Activity and Older
Americans: Benefits and Strategies. Retrieved December
2006 from http:// www.ahrq.gov/ppip/activity.htm
(47) Jones F, Harris P, Waller H, Coggins A: Adherence to an
exercise prescription scheme: The role of expectations, self
efficacy stage of change and psychological well-being.
British Journal of Health Psychology. 2005, 10:359378.
(48) Vancouver JB, Thompson CM, Williams AA: The changing
signs in the relationships among self efficacy, personal goals
and performance. Journal of Applied Psychology. 2001,
86:605620.
(49) Vancouver JB, Thompson CM, Tischner CE, Putka DJ:
Two studies examining the negative effect of self-efficacy
on performance. Journal of Applied Psychology. 2002,
87:506516.