Você está na página 1de 7

Arthritis & Rheumatism (Arthritis Care & Research)

Vol. 53, No. 5, October 15, 2005, pp 659 665


DOI 10.1002/art.21466
2005, American College of Rheumatology
ORIGINAL ARTICLE

Exercise, Self-Efcacy, and Mobility Performance


in Overweight and Obese Older Adults With
Knee Osteoarthritis
BRIAN C. FOCHT,1 W. JACK REJESKI,2 WALTER T. AMBROSIUS,2 JEFFREY A. KATULA,2 AND
STEPHEN P. MESSIER2

Objective. To examine changes in mobility-related self efcacy following exercise and dietary weight loss interventions
in overweight and obese older adults with knee osteoarthritis (OA), and to determine if self efcacy and pain mediate the
effects of the interventions on mobility task performance.
Methods. The Arthritis, Diet, and Activity Promotion Trial was an 18-month, single-blind, randomized, controlled trial
comparing the effects of exercise alone, dietary weight loss alone, a combination of exercise plus dietary weight loss, and
a healthy lifestyle control intervention in the treatment of 316 overweight or obese older adults with symptomatic knee
OA. Participants completed measures of stair-climb time and 6-minute walk distance, self efcacy for completing each
mobility task, and self-reported pain at baseline, 6 months, and 18 months during the trial.
Results. Mixed model analyses of covariance of baseline adjusted change in the outcomes demonstrated that the
exercise dietary weight loss intervention produced greater improvements in mobility-related self efcacy (P 0.0035),
stair climb (P 0.0249) and 6-minute walk performance (P 0.00031), and pain (P 0.09) when compared with the
healthy lifestyle control intervention. Mediation analyses revealed that self efcacy and pain served as partial mediators
of the benecial effect of exercise dietary weight loss on stair-climb time.
Conclusion. Exercise dietary weight loss results in improved mobility-related self efcacy; changes in these task-
specic control beliefs and self-reported pain serve as independent partial mediators of the benecial effect of exercise
dietary weight loss on stair-climb performance.

KEY WORDS. Arthritis; Efcacy beliefs; Physical activity; Physical function.

INTRODUCTION factor for the development and progression of knee OA (3).


Epidemiologic evidence suggests that weight loss may pre-
Knee osteoarthritis (OA) is a chronic degenerative disease vent the incidence of knee OA and alleviate adverse symp-
that affects approximately one-third of all older adults in toms accompanying the onset of the disease (4). Conse-
the United States. The joint damage and pain accompany- quently, weight loss interventions are now advocated in
ing knee OA are primary causes of activity restriction and the treatment of overweight or obese patients with OA of
physical disability (1) and have a profound impact on the knee (57).
quality of life in the elderly (2). Obesity is a modiable risk With the growing recognition of the importance of
weight loss interventions for arthritis patients, the recently
completed Arthritis, Diet, and Activity Promotion Trial
Supported by the National Institute of Aging (grants (ADAPT) examined the effects of exercise and dietary
AG14131 and 5P60 AG10484) and the General Clinical Re-
search Center (grant M01-RR07122).
weight loss interventions, both separately and in combi-
1
Brian C. Focht, PhD: Ohio State University, Columbus; nation, in the treatment of overweight or obese older
2
W. Jack Rejeski, PhD, Walter T. Ambrosius, PhD, Jeffrey A. adults with knee OA. Findings from ADAPT demonstrated
Katula, PhD, Stephen P. Messier, PhD: Wake Forest Univer- that combining exercise and dietary weight loss resulted in
sity, Winston-Salem, North Carolina.
Address correspondence to Brian C. Focht, PhD, Division signicant improvements in self-reported measures of
of Health Behavior and Health Promotion, School of Public physical function and pain symptoms and performance
Health, Ohio State University, 320 West 10th Avenue, Co- measures of mobility (8). The current investigation ad-
lumbus, OH 43210. E-mail: bfocht@sph.osu.edu. dresses the inuence of the interventions in ADAPT on
Submitted for publication May 19, 2005; accepted in re-
vised form June 2, 2005. changes in mobility-related self efcacy and examines
whether changes in self efcacy and self-reported pain

659
660 Focht et al

mediate and/or are independent predictors of improve- determined by a single observer on the basis of weight-
ment in performance-based measures of mobility. bearing anteroposterior radiographs; and willingness to
Perceptions of personal capabilities are important cog- undergo testing and intervention procedures.
nitive mechanisms that inuence the health status and Exclusion criteria included a serious medical condition
physical functioning of patients with arthritis. For exam- that precluded safe participation in an exercise program
ple, self-efcacy beliefs and knee pain are independent such as coronary artery disease, severe hypertension, pe-
predictors of activity restriction among older adults with ripheral vascular disease, stroke, congestive heart failure,
knee OA (9). Performance-related self-efcacy beliefs have chronic obstructive pulmonary disease, insulin-dependent
also been shown to be prospectively related to functional diabetes, psychiatric disease, renal disease, liver disease,
decline among older adults with knee pain (10). Recent active cancer other than skin cancer, and anemia; a Mini-
research with physical activity also suggests that changes Mental score 24; inability to complete the 18-month
in performance-related control beliefs possess particularly study or unlikely to be compliant; inability to walk with-
important implications for the effectiveness of exercise
out a cane or other assistive device; participation in an-
interventions in the treatment of knee OA. Notably,
other research study; excessive alcohol consumption of
Rejeski and colleagues (11) demonstrated that exercise
14 drinks per week; or inability to complete the trial
therapy resulted in signicant improvements in self ef-
protocol, in the opinion of the clinical staff, because of
cacy for the performance of functional tasks. Moreover,
frailty, illness, or other reasons.
changes in these mobility-related control beliefs and knee
A total of 2,209 older adults were prescreened via tele-
pain were found to mediate the improvements in perfor-
mance measures of mobility achieved with exercise ther- phone interviews. Of this population, 1,596 individuals
apy. did not meet 1 or more of the eligibility criteria and an-
Taken collectively, these ndings reinforce the position other 297 refused to be contacted any further. A total of
that changes in mobility-related self-efcacy beliefs and 316 participants were randomized into the study with the
perceptions of relevant physical symptoms, such as pain, following group assignments: 82 in dietary weight loss
are determinants of the functional benets accompanying alone, 80 in exercise alone, 76 in the combination exercise
exercise participation. Therefore, in the present investiga- and dietary weight loss, and 78 in the healthy lifestyle
tion, we examined the effects of exercise and dietary control.
weight loss interventions, both separately and in combi-
nation, on mobility-related self efcacy in the ADAPT Measures. Measures of mobility. Participants com-
trial. A secondary objective of this study was to examine pleted 2 performance-based mobility tasks: a 6-minute
whether self efcacy and pain mediated the benecial walk task and timed stair-climb task. The 6-minute walk
effects of the exercise and dietary weight loss interven- task was conducted in a gymnasium measuring 70 feet by
tions on performance measures of mobility. 88 feet in area. Each individual was instructed to walk as
far as possible in 6 minutes. Participants began walking at
the command go and continued walking until they re-
MATERIALS AND METHODS
ceived the command stop. Participants were not allowed
to carry a watch and were not provided any feedback
Design. Complete details of the ADAPT design and during the test. Performance was measured as the total
methodology have been reported elsewhere (8,12,13). In distance covered in feet. The stair-climb task involved
brief, ADAPT was a single-blind, 18-month, randomized,
ascending a set of 5 stairs, turning around on the top of the
controlled trial examining the effects of 4 interventions on
platform, and then descending. Performance was mea-
various measures of physical function: exercise alone, di-
sured as the total time (in seconds) necessary to complete
etary weight loss alone, exercise in combination with di-
the task. These performance-based tasks have been shown
etary weight loss, or healthy lifestyle control. ADAPT was
to be valid and reliable tests of physical function in older
conducted at the Claude D. Pepper Older Americans In-
dependence Center of Wake Forest University, Winston- adults with knee OA in previous research (9).
Salem, NC. All participants provided written informed Walking self efcacy. Prior to performing the 6-minute
consent that had received approval of the university insti- walk task, participants were asked to rate their condence
tutional review board prior to participation. in their ability to walk around the gymnasium 2 times
without stopping. This measurement was subsequently
Participants. The eligibility criteria for participation in repeated for 5 additional levels of difculty for the antic-
the study were age 60 years; calculated body mass index ipated distances of completing 4 laps, 6 laps, 8 laps, 10
28 kg/m2; self-reported knee pain on most days of the laps, and 12 laps without stopping. For each level of dif-
month; sedentary activity pattern with 20 minutes of culty, participants were presented with a condence lad-
formal exercise per week during the past 6 months; self- der with 10 steps ranging from 0 (completely uncertain) to
reported difculty with at least 1 of the following activities 10 (completely certain). Walking self-efcacy scores were
due to knee pain: walking 0.25 miles (3 4 city blocks), calculated by summing the participants condence rat-
climbing stairs, bending, stooping, kneeling, shopping, ings across the 6 levels of difculty and multiplying this
house cleaning, getting in or out of bed, standing up from result by 2 to produce a score ranging from 0 to 100. This
a chair, lifting and carrying groceries, or getting in or out of hierarchical measurement protocol is consistent with the
a bathtub; radiographic evidence of tibiofemoral OA as protocol developed by Bandura (14), and the walking self-
Exercise and Self-Efcacy in Knee OA 661

efcacy scale has demonstrated adequate psychometric participants who were having difculty complying with
properties in several previous investigations (9,11,15). the home-based exercise intervention. Attendance logs for
Stair-climbing self efcacy. Stair-climbing self efcacy center-based exercise and self-reported exercise logs for
was assessed using the same hierarchical measurement home-based sessions were used to gather adherence data
approach used for walking self efcacy. Specically, prior and monitor progress. Exercise adherence was dened as
to completing the stair-climb task, participants were asked the number of exercise sessions completed divided by the
to rate their level of certainty on the 0 10 condence total number of prescribed sessions.
ladder that they could complete the task 2 times, 4 times, Dietary weight loss. The dietary intervention was de-
6 times, 8 times, and 10 times without stopping. Stair- signed to produce and maintain an average weight loss of
climb self-efcacy scores were also calculated by summing 5% across the 18-month intervention. The intervention
across the 5 levels of difculty and multiplying this result was based on principles from the group dynamics litera-
by 2, yielding a score ranging from 0 to 100. ture (17) and social-cognitive theory (14). The major em-
Pain. Pain was assessed with the pain subscale of the phasis of the intensive phase was to heighten awareness of
Western Ontario and McMaster Universities Osteoarthritis the importance of and need to change eating habits to
Index (WOMAC) (16). Participants were asked to indicate lower caloric intake. Behavior change was facilitated
the pain severity they had experienced during the past 48 through the use of self-regulatory skills including self
hours due to knee OA on a scale ranging from 0 (none) to monitoring, goal setting, cognitive restructuring, problem
4 (extreme). The WOMAC pain subscale consists of 5 items solving, and environmental management. One introduc-
and total scores range from 0 to 20, with higher scores tory individual session was followed by 16 weekly ses-
indicating greater pain. sions consisting of 3 group sessions and 1 individual ses-
sion each month. The transition phase included a session
Procedures. Following recruitment, participants com- every other week for a total of 8 weeks. The goals for this
pleted a series of clinic visits to verify their eligibility for phase included assisting participants who had not reached
the study. The baseline assessments were obtained during their weight loss goals in establishing new goals, and
the screening visits and eligible participants were as- maintaining and preventing relapse in those participants
signed, using a variable-block randomization procedure, who had reached their weight loss goals. The maintenance
into 1 of the 4 treatment arms: exercise alone, dietary phase included monthly meetings and phone contacts,
weight loss alone, exercise and dietary weight loss, or alternated every 2 weeks.
healthy lifestyle control. Assessments of mobility-related Exercise dietary weight loss. The exercise dietary
self efcacy and physical function were obtained at data weight loss intervention involved completing the proce-
collection visits conducted at baseline and at 6 and 18 dures previously described for both the exercise alone and
months after randomization. the dietary weight loss alone programs. Delivery of the
combined intervention was provided to participants con-
Interventions. Exercise. Each participant assigned to secutively on the same day and at the same location.
the exercise alone or exercise and dietary weight loss Healthy lifestyle control. The healthy lifestyle control
treatment arms participated in three 60-minute exercise intervention served as a usual care comparison group with
sessions per week. Specically, each session comprised an the 3 treatment arms and was designed to provide atten-
aerobic phase (15 minutes), a resistance training phase (15 tion, social interaction, and health education. The group
minutes), a second aerobic phase (15 minutes), and a cool met monthly for 1 hour during the rst 3 months. A health
down (15 minutes). The rst 4 months of the 18-month educator, who scheduled videotaped presentations and
intervention were facility based. At any time after the physician discussions on topics concerning OA, obesity,
initial 4 months, participants who wished to exercise at and exercise, organized the healthy lifestyle program.
home underwent a 2-month transition phase in which the Monthly phone contacts were maintained during months
participants alternated between exercising at the facility 4 6 and bimonthly phone contacts during months 718.
and exercising at their home. Hence, some participants Compliance was dened as the number of sessions at-
remained in the facility-based program, others chose the tended divided by the total number of sessions offered.
home-based program, and some participants opted to en-
gage in a combined facility/home-based program. At the Statistical analysis. The effects of the diet and/or exer-
beginning of a home-based phase, exercise leaders visited cise interventions on changes in self efcacy for walking
homes to work with participants in tailoring their individ- and stair climbing at the 6-month and 18-month assess-
ualized exercise regimens that were consistent with the ments were analyzed using mixed-model analyses of co-
study protocol. variance (ANCOVA). Time (2 levels) and group (4 levels)
The aerobic exercise involved walking within a heart were included as factors. We used the logarithm of stair-
rate range of 50 75% of heart rate reserve. The resistance climb time and the Arcsin of the square root of self efca-
training consisted of 2 sets of 12 repetitions of the follow- cy/100 as variance stabilizing transformations (18). All
ing exercises: leg extension, leg curl, heel raise, and step analyses were conducted using PROC Mixed in the SAS
up. Cuff weights and weighted vests were used to provide software version 8 (SAS Institute, Cary, NC), a procedure
the resistance and a 11.5-minute rest interval was main- that analyzes all available followup information by pro-
tained between each exercise. Additional home visits or viding maximum likelihood estimates of missing data. In
facility-based booster sessions were scheduled to assist each of these models, change from baseline in self efcacy
662 Focht et al

Table 1. Mean SD self-efcacy scores by task and treatment condition

Adjusted score
Treatment group Baseline Followup Change at followup

Self-efcacy scores for stair climb


(range 0100)
Health education 70.24 28.75 72.28 27.11 2.04 83.20
Diet only 68.29 27.94 72.01 27.12 3.72 81.64
Exercise only 70.19 26.22 72.90 26.02 2.71 83.07
Diet plus exercise 63.84 29.25 77.25 24.28 13.41 89.70
Self-efcacy scores for 6-minute walk
(range 0100)
Health education 67.88 31.95 72.90 31.89 5.01 83.85
Diet only 68.91 32.53 74.03 30.16 5.11 83.79
Exercise only 67.49 34.21 82.03 24.71 14.54 92.02
Diet plus exercise 66.44 33.26 84.95 24.59 18.51 95.24

was used as the outcome and age, sex, and the baseline (F[3,206] 2.65, P 0.05). Inspection of the group means
self-efcacy value were included as covariates. The Tukey- provided in Table 1 demonstrates that participants in the
Kramer adjustment for multiple comparisons was used exercise dietary weight loss intervention signicantly
when examining pair-wise differences. We examined in- improved their stair-climb self efcacy compared with the
tervention by time interactions and found that none were healthy lifestyle control group. Analysis of change in
signicant (all P 0.35); therefore, they were not included walking self efcacy also revealed a signicant main effect
in the analyses. We calculated the adjusted means for for treatment group (F[3,210] 6.04, P 0.0006). As
treatment averaging over time. Participants who com- illustrated in Table 1, participants in the exercise di-
pleted baseline and at least one followup assessment were etary weight loss and exercise alone interventions signi-
included in the analyses. Additionally, all analyses were cantly improved their walking self efcacy compared with
conducted using the intent-to-treat principle. Following the healthy lifestyle control group.
the primary analyses that tested the effects of the interven- Although changes in mobility-related control beliefs are
tions on change in mobility-related self efcacy, secondary relevant outcomes of lifestyle interventions among older
analyses were conducted to determine if self efcacy and patients with knee OA, the minimal clinically signicant
pain mediated the effects of interventions on 6-minute difference (MCSD) associated with changes in mobility-
walk and stair-climb performance. Specically, a series of related self efcacy has yet to be established. Accordingly,
separate mixed-model ANCOVA models were tested to we estimated the MCSD for the mobility-related self-ef-
examine the effects of the interventions on change in self- cacy measures. To estimate the MCSD, we calculated the
reported pain and 6-minute walk and stair-climb perfor- mean and SD of the transformed data for each self-efcacy
mance; the effects of the potential mediators (self efcacy outcome at baseline. We subsequently calculated small,
for walking and stair climbing and self-reported pain) on medium, and large effect sizes. The following MCSD val-
6-minute walk and stair-climb performance; and a com- ues represent the back transformed differences obtained by
posite model testing the effects of the treatments on using half of the small, medium, and large effect sizes
6-minute walk and stair-climb performance while control- above and below the observed baseline mean: for walking
ling for self efcacy for mobility and self-reported pain. self efcacy the effect sizes were 8.08, 20.04, and 31.60,
whereas for stair-climb self efcacy the effect sizes were
6.70, 16.65, and 26.38, respectively.
RESULTS
Effects of the interventions on performance-related mo-
Participant attrition and adherence. Of the 316 partic- bility and knee pain. As reported in previous ndings
ipants randomized into the trial, 252 (80%) completed the from the ADAPT trial (8), signicant group main effects
trial. As reported previously (8), retention of participants were observed for pain (F[2,232] 3.37, P 0.0193),
did not differ among the 4 intervention groups. Adherence 6-minute walk distance (F[3,196] 12.75, P 0.0001),
was also not signicantly different among the groups, with and stair-climb time (F[3,212] 3.18, P 0.0249). Specif-
rates of 75% in the healthy lifestyle control, 72% in the ically, the exercise dietary weight loss intervention re-
dietary weight loss, 60% in the exercise only, and 64% in sulted in signicantly greater improvements in pain
the combined intervention. (t[232] 2.35, adjusted P 0.09) relative to the healthy
lifestyle control group. In regard to the performance mea-
Effects of the interventions on self efcacy for mobility. sures of mobility, the exercise dietary weight loss inter-
The main analyses in this study were conducted to exam- vention yielded a more favorable improvement in stair-
ine the effect of the exercise and dietary weight loss inter- climb time (t[212] 2.85, adjusted P 0.0249) relative to
ventions on change in the measures of self efcacy for the healthy lifestyle control group, and both the exercise
mobility. Analysis of change in stair-climbing self efcacy dietary weight loss intervention (t[196] 4.97, adjusted
yielded a signicant main effect for treatment group P 0.0001) and the exercise alone intervention (t[196]
Exercise and Self-Efcacy in Knee OA 663

Table 2. Mean SD for mobility performance by task and treatment condition

Adjusted score
Treatment group Baseline Followup Change at followup

Stair-climb time
Health education 9.44 4.91 9.86 5.56 0.41 8.60
Diet only 9.77 5.70 9.86 8.78 0.09 8.24
Exercise only 9.85 4.56 9.15 4.70 0.70 7.80
Diet plus exercise 10.38 7.26 8.85 5.35 1.53 7.54
6-minute walk distance
Health education 1,422 269 1,411 261 11 1,417
Diet only 1,406 254 1,433 260 27 1,447
Exercise only 1,417 251 1,551 297 134 1,559
Diet plus exercise 1,360 280 1,524 316 163 1,575

4.55, adjusted P 0.0001) resulted in signicantly better was evidence that change in self efcacy and pain had
changes in walking distance when compared with the independent effects on the outcomes of interest. First,
healthy lifestyle control group (Table 2). mixed-model ANCOVA analyses were conducted to test
the effect of the proposed mediators on the measures of
The independent and mediational role of self efcacy mobility task performance. Results of the model examin-
and pain on performance. To evaluate the independent ing change in 6-minute walk performance demonstrated
effects of self efcacy and pain on change in stair-climb that baseline walking self efcacy (F[1,178] 25.75, P
time, composite mixed-model ANCOVAs were con- 0.0001), change in walking self efcacy (F[1,178] 35.81,
structed using change in stair-climb time and 6-minute P 0.0001), and change in pain (F[1,178] 6.43, P
walk distance as outcome variables. In these models, co- 0.0121) were signicant predictors of change in baseline-
variates included baseline scores for each outcome (visit, adjusted 6-minute walk distance. Additionally, results of
age, sex, and treatment effect) whereas baseline status of the model examining change in stair-climb performance
pain or self efcacy, as well as change in pain or self demonstrated that baseline stair-climb self efcacy
efcacy, were considered for any independent effect that (F[1,192] 7.96, P 0.0053), change in stair-climb self
they had on change in each outcome. Results of these efcacy (F[1,192] 38.63, P 0.0001), baseline pain
analyses revealed that baseline pain (F[1,191] 7.60, P (F[1,192] 7.45, P 0.0069), and change in pain (F[1,192]
0.0064) and baseline self efcacy (F[1,191] 7.03, P 9.90, P 0.0019) were signicant predictors of change
0.0087), together with change in pain (F[1,191] 10.44, P in baseline-adjusted stair-climb time. Thus, the ndings
0.0015) and change in self efcacy (F[1,191] 35.93, P from these models demonstrate that self efcacy and pain
0.0001), were all independent predictors of stair-climb are signicant independent predictors of change in walk-
time beyond the treatment effect and other covariates men- ing and stair-climb performance.
tioned above. In regard to the baseline predictors, these Having established the required univariate relationships
ndings suggest that participants reporting less pain and between the exercise and dietary interventions, self ef-
higher self efcacy at baseline demonstrated superior stair- cacy and pain, and the performance measures of mobility,
climb performance at followup. Additionally, the observa- we then tested a composite model examining the effect of
tion that change in each outcome was an independent the interventions on change in the measures of mobility
predictor of performance suggests that participants dem- performance after controlling for self efcacy and pain in
onstrating the greatest increase in self efcacy and de- the model. Results of the mixed-model ANCOVA analyses
crease in pain exhibited the most favorable change in of stair-climb performance revealed that controlling for
stair-climb performance. Similar results were found for stair-climb self efcacy and pain in the original model
6-minute walk distance except that baseline pain was not attenuated, but did not eliminate, the signicance of the
statistically signicant (F[1,178] 22.27 for baseline self treatment group main effect (F[3,192] 2.78, P 0.0425).
efcacy, P 0.0001; F[1,178] 9.05 for change in pain, P Results of the ANCOVA analyses of walking performance
0.0030; F[1,178] 28.47 for change in self efcacy, P revealed that controlling for walking self efcacy and pain
0.0001). in the original model did not reduce the signicance of the
Collectively, the primary results of this study and the treatment group main effect (F[3,178] 11.46, P
ndings reported previously by Messier et al (8) demon- 0.0001). However, in each analysis, the changes in self
strate that the ADAPT interventions had a signicant effect efcacy and pain remained independent predictors of mo-
on both the proposed mediators (self efcacy for mobility bility-related performance after including terms for the
and knee pain) and the outcomes of interest (performance group effects.
measures of mobility). Thus, we conducted secondary ana-
lyses in the present investigation to examine whether
DISCUSSION
change in self efcacy and pain mediated the effects of the
exercise and dietary weight loss interventions on change In the ADAPT trial, the combination of exercise and di-
in the performance measures of mobility or whether there etary weight loss resulted in signicant improvements in
664 Focht et al

self-reported measures of physical functioning, pain, and driven by the combined exercise and dietary weight loss
performance measures of mobility. The present study ex- group.
amined the inuence of the ADAPT intervention on In conclusion, the present study demonstrated that a
changes in mobility-related self-efcacy beliefs. Moreover, combined dietary weight loss and physical activity inter-
we were interested in whether baseline levels or change in vention had unique effects on changes in self efcacy for a
self-efcacy beliefs and pain were independent predictors weight-dependent stair-climb task as compared with exer-
of change in performance-based measures of function cise alone. Additionally, both baseline values and changes
and/or whether change in self efcacy and/or pain medi- in self efcacy and pain were signicant predictors of
ated the effect that the interventions had on these indices improvement in mobility disability above and beyond the
of mobility disability. effects of the interventions. These ndings add to the
Analyses revealed that both the exercise alone and the growing body of evidence demonstrating that control be-
diet exercise groups experienced signicant increases in liefs and physical symptoms are crucial to understanding
walking self efcacy. However, only the diet exercise and intervening the process of physical disablement (18).
group reported signicant improvements in self efcacy Furthermore, when treating obese, older adults who have
for stair climbing. Based on the estimated MCSD for the compromised function due to knee OA, it is highly pref-
self-efcacy measures, the improvements in mobility-re- erable to use a combined dietary weight loss and physical
lated self efcacy following the exercise alone and diet activity intervention as compared with a physical activity
exercise interventions represented small to moderate ef- intervention alone.
fect size differences. The differential effects of the exercise
alone and combination interventions on change in self
efcacy for 6-minute walk and stair-climb performance REFERENCES
underscore the importance of specicity in the measure- 1. Leveille SG, Fried LP, McMullen W, Guralnik JM. Advancing
ment of mobility-related self efcacy (14). Thus, the the taxonomy of disability in older adults. J Gerontol A Biol
present ndings suggest that although exercise alone in- Sci Med Sci 2004;59:86 93.
2. Rejeski WJ, Shumaker S. Knee osteoarthritis and health-re-
creases self efcacy for walking, exercise must be com- lated quality of life. Med Sci Sports Exerc 1994;26:14415.
bined with weight loss to inuence efcacy beliefs involv- 3. Ettinger WH, Davis MA, Neuhaus JM, Mallon KP. Long-term
ing the performance of more challenging mobility tasks physical functioning in persons with knee osteoarthritis from
such as stair climbing. Because the exercise only treatment NHANES I: effects of comorbid medical conditions. J Clin
Epidemiol 1994;47:809 15.
did not differ from the healthy education control group on
4. Felson DT, Naimark A, Anderson J, Kazis L, Castelli W,
changes in stair-climb time in the ADAPT study (8), it Meenan RF. The prevalence of knee osteoarthritis in the
appears that physical activity and weight loss should in- elderly: the Framingham Osteoarthritis Study. Arthritis
deed be the intervention of choice in older, overweight, Rheum 1987;30:914 8.
5. American College of Rheumatology Subcommittee on Osteo-
and obese adults with knee OA.
arthritis Guidelines. Recommendations for the medical man-
The results of this study underscore the importance of agement of osteoarthritis of the hip and knee: 2000 update.
self-efcacy beliefs and pain in the treatment of knee OA, Arthritis Rheum 2000;43:190515.
even within the context of lifestyle interventions. That is, 6. Minor M, Stenstrom CH, Klepper SE, Hurley M, Ettinger WH.
both baseline levels and changes in self efcacy and knee Work Group Recommendations: 2002 exercise and physical
activity conference, St. Louis, Missouri. Arthritis Rheum
pain were signicant independent predictors of improve- 2003;49:453 4.
ments in mobility performance. The only exception was 7. Pendleton A, Arden N, Dougados M, Doherty M, Bannwarth
that baseline levels of pain did not predict change in B, Bijlsma JW, et al. EULAR recommendations for the man-
6-minute walk times. These data are consistent with grow- agement of knee osteoarthritis: report of a task force of the
Standing Committee for International Clinical Studies Includ-
ing empirical evidence indicating that control beliefs and ing Therapeutic Trials (ESCISIT). Ann Rheum Dis 2000;59:
pain play key roles in the disablement process (10,11,19). 936 44.
Therefore, lifestyle interventions aimed at improving 8. Messier SP, Loeser RF, Miller GD, Morgan TM, Rejeski WJ,
physical functioning in patients with knee OA should Sevick MA, et al. Exercise and dietary weight loss in over-
weight and obese older adults with knee osteoarthritis: the
include components that target self-efcacy beliefs and Arthritis, Diet, and Activity Promotion Trial. Arthritis Rheum
pain management (19). In addition to maximizing effects 2004;50:150110.
on physical functioning, these types of intervention strat- 9. Rejeski WJ, Craven T, Ettinger WH Jr, McFarlane M, Shu-
egies have the added potential of enhancing adherence maker S. Self-efcacy and pain in disability with osteoarthri-
tis of the knee. J Gerontol B Psychol Sci Soc Sci 1996;51:
and the long-term maintenance of physical activity P24 9.
(15,20). 10. Rejeski WJ, Miller ME, Foy C, Messier S, Rapp S. Self-efcacy
In the present study, self efcacy and knee pain partially and the progression of functional limitations and self-re-
mediated the inuence of the interventions on change in ported disability in older adults with knee pain. J Gerontol B
Psychol Sci Soc Sci 2001;56:S2615.
stair-climb performance. A previous study involving older 11. Rejeski WJ, Ettinger WH Jr, Martin K, Morgan T. Treating
adults with knee OA, the Fitness and Arthritis in Seniors disability in knee osteoarthritis with exercise therapy: a cen-
Trial, found that self efcacy and pain completely medi- tral role for self-efcacy and pain. Arthritis Care Res 1998;11:
ated the effect of a physical activity intervention on stair- 94 101.
12. Miller GD, Rejeski WJ, Williamson JD, Morgan T, Sevick MA,
climb time (10). In the current study, the exercise only Loeser RF, et al. The Arthritis, Diet, and Activity Promotion
group did not differ from the health education control Trial (ADAPT): design, rationale, and baseline results. Con-
group on stair-climb time; thus, the effect in this study was trol Clin Trials 2003;24:462 80.
Exercise and Self-Efcacy in Knee OA 665

13. Rejeski WJ, Focht BC, Messier SP, Morgan T, Pahor M, Pen- for measuring clinically important patient relevant outcomes
ninx B. Obese, older adults with knee osteoarthritis: weight to antirheumatic drug therapy in patients with osteoarthritis
loss, exercise, and quality of life. Health Psychol 2002;21: of the hip or knee. J Rheumatol 1988;15:1833 40.
419 26. 17. Cartwright DC, Zander A. Group dynamics: research and the-
14. Bandura A. Theoretical perspectives: self-efcacy: the exer- ory. New York: Harper and Row; 1953.
cise on control. New York: W.H. Freeman and Company; 18. Box GE, Hunter WG, Hunter JS. Statistics for experimenters.
1997. New York: John Wiley & Sons; 1978.
15. Rejeski WJ, Brawley LR, Ambrosius WT, Brubaker PH, Focht 19. Rejeski WJ, Focht BC. Aging and physical disability: on inte-
BC, Foy CG, et al. Older adults with chronic disease: benets grating group and individual counseling with the promotion
of group-mediated counseling in the promotion of physically of physical activity. Exerc Sport Sci Rev 2002;30:166 71.
active lifestyles. Health Psychol 2003;22:414 23. 20. McAuley E, Jerome GJ, Elavsky S, Marquez DX, Ramsey SN.
16. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt Predicting long-term maintenance of physical activity in
LW. Validation study of WOMAC: a health status instrument older adults. Prev Med 2003;37:110 8.