The Journal of Rheumatology Volume 35, no.

Influence of patient education on exercise compliance in rheumatoid arthritis:

a prospective 12-month randomized controlled trial.
Anne Mayoux-Benhamou, Janine-Sophie Giraudet-Le Quintrec, Philippe Ravaud, Karine
Champion, Emmanuelle Dernis, Djamila Zerkak, Carine Roy, Andre Kahan, Michel Revel
and Maxime Dougados

J Rheumatol 2008;35;216-223
http://www.jrheum.org/content/35/2/216

1. Sign up for our monthly e-table of contents

http://www.jrheum.org/cgi/alerts/etoc

2. Information on Subscriptions
http://jrheum.com/subscribe.html

3. Have us contact your library about access options

Refer_your_library@jrheum.com

4. Information on permissions/orders of reprints

http://jrheum.com/reprints.html

The Journal of Rheumatology is a monthly international serial edited by Earl D.

Silverman featuring research articles on clinical subjects from scientists working
in rheumatology and related fields.

Downloaded from www.jrheum.org on September 12, 2016 - Published by The

Journal of Rheumatology
Influence of Patient Education on Exercise Compliance
in Rheumatoid Arthritis: A Prospective 12-month
Randomized Controlled Trial
ANNE MAYOUX-BENHAMOU, JANINE-SOPHIE GIRAUDET-LE QUINTREC, PHILIPPE RAVAUD, KARINE
CHAMPION, EMMANUELLE DERNIS, DJAMILA ZERKAK, CARINE ROY, ANDRE KAHAN, MICHEL REVEL,
and MAXIME DOUGADOS

ABSTRACT. Objective. To determine the effect of education on the exercise habits of patients with rheumatoid arthri-
tis (RA) after 6 and 12 months.
Methods. We studied 208 outpatients recruited between June 2001 and December 2002. This was a
prospective controlled randomized trial. The active group received a multidisciplinary education pro-
gram, including training in home-based exercises and guidelines for leisure physical activity (PA). The
control group received a booklet added to usual medical care. Compliance with home-based exercises
was defined as a practice rate 30% of the prescribed training. Compliance with leisure PA was defined
as 20% increase in Baecke questionnaire score. Additional assessments involved possible predictors
of compliance and changes with regard to the compliance.
Results. At 6-month followup, home-based exercise and leisure PA compliance were significantly high-
er [13.5% vs 1%, respectively (p = 0.001); and 28.2% vs 13.8% (p = 0.02)], but were not at 12 months.
Predictors of leisure PA compliance at 6 months included participating in the active group (odds ratio
2.74, 95% CI 1.17 to 6.38) and previous low leisure PA (OR 6.01, 95% CI 2.47 to 14.61), with
decreased fatigue (FACIT-F mean 2.94 8.04 vs 0.1 7.25 for noncompliant subjects; p = 0.04) and
improved psychological status (Arthritis Impact Measurement Scale mean 1.25 3.12 vs 0.11 3.39;
p = 0.03).
Conclusion. Education of patients with RA may increase compliance especially with leisure PA, par-
ticularly when it is poor at baseline, but these effects are limited and short-term. (First Release Dec 15
2007; J Rheumatol 2007;35:21623)

Key Indexing Terms:

RHEUMATOID ARTHRITIS EDUCATION PHYSICAL ACTIVITY
EXERCISE COMPLIANCE RANDOMIZED TRIAL

Regular exercise is encouraged among healthy people to pre- toid arthritis (RA) who have increased susceptibility for these
vent death and disease from cardiovascular disease, osteo- comorbid conditions2-4 may benefit from regular exercise5
porosis, anxiety, and depression1. Thus, people with rheuma- and may be encouraged to exercise. However, the reduced
physical capacity found frequently among patients with RA
From the Institute of Rheumatology, Assistance PubliqueHpitaux de may be attributable in part to inadequate levels of physical
Paris (AP-HP), Cochin Hospital, Paris Descartes University, Paris, activity (PA). Indeed, traditionally, exercise restriction has
France. been recommended because of concerns about aggravating
Supported in part by grants from the direction de la Recherche Clinique, joint inflammation and accelerating joint damage in these
AP-HP, France (PHRC AOM00027).
A. Mayoux-Benhamou, MD, PhD, Department of Rehabilitation, Cochin
patients6.
Hospital; J-S. Giraudet-Le Quintrec, MD, Department of Rheumatology Recent systematic reviews7 suggest that patients with RA
A, Cochin Hospital; P. Ravaud, MD, PhD, Professor, Department of could benefit from regular physical exercise, including
Epidemiology and Biostatistics, Bichat Hospital, Paris 7 University;
K. Champion, MD, Department of Rheumatology A, Cochin Hospital;
dynamic and weight-bearing exercises, and could improve
E. Dernis, MD; D. Zerkak, MD, Department of Rheumatology B, Cochin aerobic capacity, muscle strength, functional ability, and psy-
Hospital; C. Roy, Department of Epidemiology and Biostatistics, Paris 7 chological well-being with moderate-intensity exercise with-
University; A. Kahan, MD, PhD, Professor, Department of Rheumatology
A, Cochin Hospital; M. Revel, MD, Professor, Department of
out detrimental effects on disease. The American College of
Rehabilitation, Cochin Hospital; M. Dougados, MD, Professor, Rheumatology (ACR) recommended regular participation in
Department of Rheumatology B, Cochin Hospital. dynamic exercise programs in its recent update of treatment
Address reprint requests to Dr. A. Mayoux-Benhamou, Department of guidelines for RA8.
Rehabilitation, Cochin Hospital, 27 Rue du Faubourg Saint Jacques,
75679 Paris Cedex 14, France.
Therefore, development of educational programs and
E-mail: anne.mayoux-benhamou@cch.aphp.fr strategies is needed to promote PA for patients with RA.
Accepted for publication September 18, 2007. Supervised training, such as that in the Rheumatoid Arthritis

Personal non-commercial use only. The Journal of Rheumatology Copyright 2008. All rights reserved.

216 The Journal of Rheumatology 2008; 35:2

Downloaded from www.jrheum.org on September 12, 2016 - Published by The

Journal of Rheumatology
Patients In Training (RAPIT) study, is not cost-effective as patients were assigned randomly to the active or control groups. The alloca-
compared with usual care9. Exercise programs with less ther- tion sequence was generated by random placement of thoroughly shuffled
marked cards into sequentially numbered, sealed, opaque envelopes by a sta-
apist supervision and more self-care may reduce costs10,11. tistics assistant not involved in the trial.
Education interventions for RA patients have shown benefi- The active group participated in a multidimensional educational program
cial effects on health behavior and physical and psychosocial added to usual medical care. The control group received usual care. All
health status11,12. Higher self-efficacy, which is amenable patients received 2 information booklets31,32.
through an education program11, is associated with better Educational program (active group). The educational program was delivered
health status13 and lower arthritis-related costs. But data on within the month following the randomization and included 8 weekly, 5-hour
changes in leisure PA after such interventions are sparse. sessions for outpatients. Participants were organized into classes of 8 to 10 for
the sessions. Four sessions consisted of comprehensive information about RA
Disease activity, disability, or medication have not been iden- and its medical management. Four sessions were devoted to physical program
tified as predictors of loss of performed leisure activities that and were conducted by health professionals. Each session was initiated by a
occurs in early RA14,15, and other factors, such as psycho- physicians lecture focused specifically on guidelines for practicing adequate
social status, may be important for predicting changes in PA and a discussion that aimed to enhance positive attitudes and beliefs related
leisure activities. to exercise (1 hour). Tailored advice and individual approaches were provided
to offset physical and psychological barriers to exercise, instructing patients on
These concerns prompted us to conduct a post-hoc analy- how to incorporate moderate PA into their usual day, find enjoyable and attain-
sis of a trial aimed at evaluating the effect of an educational able activities, and modify the program according to their current health
program on functional disability in patients with RA16. This because of the variable course of RA. Then participants were split into sub-
randomized controlled trial aimed to assess compliance with a groups to participate in workshops. The occupational therapists intervention (1
home-based exercise program and recommendations for hour) included education on joint protection positioning, proper footwear, and
use of splints and adaptive aids that participants could test. The physical thera-
leisure PA delivered through a multidisciplinary educational pists intervention (1 hour) included the practice of the home-based exercise and
intervention added to usual medical care. As well, we sought aerobic activities such as cycling. After a break, the participants attended class-
to investigate baseline physical and psychological determi- es devoted to aquatic (1 hour) or relaxation training (1 hour).
nants of compliance with exercise in RA patients, and to The home-based exercise program included 10 exercises: 3 hand and
assess the health changes in participants who comply with wrist range-of-motion exercises; 5 isometric strengthening exercises to rein-
force limb muscles with elastic bands; a foot-roll exercise involving a tennis
exercise and leisure PA recommendations. ball; and 1 towel-grabbing exercise to recruit plantar muscles. For each exer-
cise, a set of 10 repetitions was prescribed. From our previous experience33,
MATERIALS AND METHODS participants were encouraged to practice at least and alternatively 3 exercises
Study design. Our randomized controlled prospective study was a single-cen- each day according to their joint involvement and their current physical activ-
ter trial of 12 months duration. The trial was approved by the local ethics ities. The program could be performed in one daily session or split into short
committee, and we obtained informed consent from patients for participation sessions throughout the day.
before randomization. The home-based exercise program and leisure PA recommendations were
Patients. The medical records of RA patients at our institution were screened described in the booklets given to both active and control groups.
between June 2001 and December 2002, and patients were contacted by mail After a 6-month followup assessment, patients attended a meeting (at 6
or asked by their physicians to participate in the trial. months after the first educational session) to reinforce the program.
Eligible patients had RA according to the 1987 revised ACR criteria17, Usual medical care (control group). Usual medical care was given by the
were between 18 and 80 years of age, had received stable dosages of disease- rheumatologist in charge of the patient and could include individual physical
modifying drugs in the previous 3 months, and had a Steinbrocker functional therapy only if considered necessary by the attending physician.
status of class I, II, or III18. Patients were excluded if their previous general Followup visits. Followup visits were performed by 3 physicians blinded to
health conditions disallowed their participation in an exercise program, they group assignment and independent physicians (KC, ED, DZ) 6 and 12 months
were pregnant, or they misunderstood the French language. Eligible patients after the randomization. The physicians collected the exercise sheets from
who refused to participate were asked to explain their decision by selecting each patient but did not check whether patients could actually perform the
one of the following items: lack of motivation, remote from home, time-con- exercises.
suming intervention, or other personal reasons. Eligible patients who agreed Exercise compliance assessment at a given visit. The compliance rate for
to participate attended a baseline visit with one of 2 investigators (JSGLQ or home-based exercise was measured as described33. The mean weekly practice
AMB). was calculated as the proportion of self-reported mean weekly number of
Baseline evaluation. At the baseline medical visit, data were collected on exercises to total number of exercises included in the home-based program.
medical history, disease management, and leisure behaviors. Disease activity To be compliant, each participant had to have a compliance rate 30%,
was assessed by the Disease Activity Score (DAS28)19. The 50-foot walk test meaning at least a daily mean practice of a set of 3 different exercises what-
was performed20. ever the exercises performed and have disrupted training less than 1 month
Additionally, this multidimensional evaluation involved patients answer- before the 6-month followup visit and less than 2 months before the 12-month
ing several questionnaires in their French version: the functional Health followup visit.
Assessment Questionnaire (HAQ)21,22 validated in French, the Hospital Leisure PA compliance was measured by comparing the baseline and fol-
Anxiety and Depression Scale (HADS)23, the Arthritis Helplessness Index lowup (6- or 12-month) level of leisure PA as assessed by the Baecke ques-
(AHI) for coping measurement24, the short-form of the Arthritis Impact tionnaire. Because identification of a minimal clinically important difference
Measurement Scale (AIMS2)25,26 validated in French, the Functional is lacking for the Baecke score, we decided that compliant participants had to
Assessment of Chronic Illness TherapyFatigue (FACIT-F) question- have increased their score by at least 20% over that at baseline. This thresh-
naire27,28, and the Baecke questionnaire29,30 validated in French, which old was chosen because of its clinical relevance and out of respect to the 5-
assesses usual leisure-time PA. point scale of the Baecke questionnaire.
Study procedure. Randomization. In the week following the baseline visit, Sample size calculation. The sample size calculation was based on the mean

Personal non-commercial use only. The Journal of Rheumatology Copyright 2008. All rights reserved.

Mayoux-Benhamou, et al: Patient education and RA 217

Downloaded from www.jrheum.org on September 12, 2016 - Published by The

Journal of Rheumatology
change in HAQ score (score at baseline minus score at 12 months) because attend the educational classes. Their baseline characteristics
the HAQ was the primary outcome of the main study16. Assuming a mean did not differ from those of the participants. Among the 92
change from baseline of 0.05 for the control group and 0.20 for the inter-
vention group (and equal variance, 0.41), a sample size of 118 patients in each
patients who began the education procedure, the attendance
group would provide 80% power to detect a difference in means of 0.15, rate was excellent (97.8%). The physical training was well tol-
assuming a common standard deviation of 0.41, using a 2-group t-test with a erated, and no participant experienced a clinical detrimental
0.05 2-sided significance level. effect related to prescribed exercises. Thus, we gathered data
Statistical analysis. Statistical analysis was conducted in 4 consecutive steps. on 6-month home-based exercise compliance for 180 patients
The first compared the baseline characteristics of patients per treatment and 6-month leisure PA compliance for 172 patients.
group. Quantitative variables were compared by use of Student t-test and cat-
egorical variables by the chi-square test. The second step evaluated the effect Compliance rates (Table 2). 6-month compliance.
of the educational intervention on the 2 main outcome variables: proportion Home-based exercise compliance: 12 patients (13.5%) in the
of patients compliant with home-based exercise and leisure PA by treatment active group and 1 (1.1%) in the control group complied with
group (see description of the outcome variable). For this purpose, we used the the home-based exercise program as defined (p = 0.002), the
chi-square test. The first- and second-step analyses tested data obtained at 6
and 12 months.
compliance rate being 15.8% 24.9% and 4.8% 18.2%,
The third step aimed to explore predisposing factors for leisure PA com- respectively (p < 0.0001).
pliance at 6 months. We conducted both uni- and multivariate analysis (logis- Leisure PA compliance: 24 patients (28.2%) in the active
tic regression), with the independent variable, leisure PA compliance, defined group and 12 (13.8%) in the control group were compliant at
as a dichotomous variable (yes/no). The potential predisposing factors (inde-
pendent variables) were defined as dichotomous, with the median used as a
6 months (i.e., they increased their leisure PA by at least 20%;
cutoff. The factors were patient educational intervention (1 = active group, 0 p = 0.02). The level of leisure PA was significantly increased
= control group); living alone at home (1 = yes, 0 = no); DAS28 (1 4.2, 0 at 6-month followup only in the active group (+15.2%
= no); Steinbrocker functional class (1 = class 1, 0 = class 2 or 3); HAQ (1 34.9% vs 0.20% 24.8%; p = 0.0001).
1.2, 0 = no); 50-foot walk time (1 = time < 10 s, 0 = no); baseline level of
usual leisure PA score as assessed by the Baecke questionnaire (1 16 score, 12-month compliance.
2 16); HADS anxiety score (1 6, 0 = no); HADS depression score (1 Home-based exercise compliance: Only 7 patients (7.9%) in
10, 0 = no); AHI coping score (1 18, 0 = no); FACIT-F score (1 18, 0 = the active group and 3 (3.4%) in the control group were com-
no); AIMS physical subscore (1 22, 0 = no); and AIMS psychological sub- pliant with home-based exercise at 12-month followup (p =
score (1 8, 0 = no). The variables with p < 0.20 after univariate analysis
0.19). The 12-month compliance rate was 11.8% 25.5% ver-
were entered into the multivariate analysis. For the predisposing factors
revealed by the multivariate model, the median cutoff was selected before cal- sus 4% 16%, respectively (p < 0.0001).
culation of an odds ratio. Leisure PA compliance: 8 patients (9.2%) in the active group
The fourth step involved comparing changes in outcome variables during and 7 (7.1%) in the control group were leisure PA compliant
the first 6 months in terms of compliance. We classified patients as compliant at 12-month followup (p = 0.61), and the level of leisure PA
or not using the same definition as above. The outcome variables were 50-
was 10% 31.7% in the active group versus 16.8%
foot walk time and scores on the DAS, HAQ, AIMS physical domain, AIMS
psychological domain, AIMS social domain, HADS anxiety and depression, 26.2% in the control group (p = 0.10).
AHI coping, and FACIT-F. Predictors of 6-month leisure PA compliance. Of the 13 vari-
All analyses used SAS version 8 (SAS Institute Inc., Cary, NC, USA) and ables assessed in the univariate analysis (Table 3), 3 were
related to leisure PA compliance: participating in the active
p < 0.05 was considered significant.

group (p = 0.02), having a low baseline level of PA (p <

RESULTS
0.0001), and having good psychological status (p = 0.04).
Patients and study course. The process used to enroll the tar-
Five variables (patient educational intervention, Steinbrocker
get population and the patient characteristics have been
functional class, baseline level of usual leisure PA score,
described16. Among 1242 patients invited to participate, 932
AIMS physical subscore, and AIMS psychological subscore)
(90.1%) refused (316 lived too far from the institute or trans-
portation was a problem, 398 found the study too time-con- were entered in the multivariate model. The multivariate
suming, 218 lacked motivation) and 102 (9.9%) were not eli- analysis identified only 2 variables as predisposing factors:
gible after screening (Figure 1). Thus, 208 patients were participating in the active (educated) group (odds ratio 2.74,
included in the study. Among them, 165 were recruited via 95% confidence interval 1.17 to 6.38) and having a low base-
their physician. The mean age was 54.7 13.1 years; disease line level of leisure PA [i.e., a Baecke score < 16 (OR 6.01,
duration at enrolment was 12.7 9.8 years. Most patients had 95% CI 2.47 to 14.61)].
a Steinbrocker functional class II score (72.6%) and no cur- Effect of home-based exercise and leisure PA compliance on
rent occupation (65.4%). After randomization, the active (edu- clinical outcomes (Table 4). Compliance with home-based
cated) and control groups each included 104 participants. exercise and selected outcome change were not associated.
Characteristics of the 2 groups were similar (Table 1). However, we found an association between compliance with
Eight patients from the active group and 11 from the con- leisure PA and 2 outcomes: AIMS psychological domain
trol group were lost to followup at 12-month evaluation (n = (mean 1.25 3.12 vs 0.11 3.39 for noncompliant patients,
9, 4%). p = 0.03, a negative score indicating better psychological out-
Twelve patients allocated to the active group refused to look) and fatigue on the FACIT vitality scale (mean 2.94

Personal non-commercial use only. The Journal of Rheumatology Copyright 2008. All rights reserved.

218 The Journal of Rheumatology 2008; 35:2

Downloaded from www.jrheum.org on September 12, 2016 - Published by The

Journal of Rheumatology
 Figure 1. Progress of participants included in the trial.

Table 1. Baseline characteristics of patients with RA assigned to an active group (participated in an education-
al program about exercise) and control groups to determine compliance with home-based exercises and leisure
physical activity. Differences were examined by chi-square or Mann-Whitney U test when appropriate.

Feature Active Group, Control Group,

n = 104 n = 104

Age, yrs, mean SD 55.32 11.80 54.31 14.37

Female, % 89.86 88.85
Weight, kg, mean SD 65.35 13.98 66.25 16.59
Positive for rheumatoid factor, % 74.74 79.38
Duration of RA, yrs, mean SD 11.85 9.44 14.25 10.27
Functional class, %
1 16.83 20.79
2 70.3 60.4
3 12.87 18.81
DAS28, mean SD 4.32 1.31 4.13 1.57
HAQ, mean SD 1.22 0.67 1.12 0.69
50-foot walk time, s, mean SD 11.19 3.77 11.14 3.37
Leisure physical activity (Baecke score) 16.37 4.62 17.31 7.96

Personal non-commercial use only. The Journal of Rheumatology Copyright 2008. All rights reserved.

Mayoux-Benhamou, et al: Patient education and RA 219

Downloaded from www.jrheum.org on September 12, 2016 - Published by The

Journal of Rheumatology
 Table 2. Compliance rates for home-based exercise and leisure physical activities for RA patients in the active
group (participated in an educational program about exercise) and control group at 6- and 12-month followup.

Compliance Rate
Group N Active N Control p

Home-based exercise
6 months, n (%) 89 12 (13.5) 91 1 (1.1) 0.002
12 months, n (%) 89 7 (7.9) 89 3 (3.4) 0.190
Leisure physical activity
6 months, n (%) 85 24 (28.2) 87 12 (13.8) 0.020
12 months, n (%) 87 8 (9.2) 85 7 (7.1) 0.610

N: number of assessed participants.

Table 3. Baseline clinical status, functional status, psychological status, leisure behaviors and quality of life of
patients compliant or not with leisure physical activity program. For most determinants, a dichotomic definition
was used, with median as cutoff.

Feature Compliant Group, Noncompliant Group, p

n = 36 (%) n = 136 (%)

Active group 24 (66.7) 61 (44.8) 0.02

Living alone 11 (30.6) 57 (42.2) 0.20
DAS28 < 4.2 17 (53.1) 61 (51.7) 0.85
Functional class = 1 4 (11.1) 31 (23.7) 0.10
HAQ < 1.2 20 (55.6) 69 (50.7) 0.61
50-foot walk time < 10 s 12 (34.3) 48 (35.8) 0.87
Leisure physical activity < 16
(Baecke questionnaire) 28 (77.8) 49 (36.3) < 0.0001
Anxiety (HADS) < 6 15 (41.7) 64 (47.1) 0.56
Depression (HADS) < 10 16 (44.4) 72 (52.9) 0.36
Coping (AHI) < 18 14 (38.9) 67 (49.3) 0.27
Fatigue (FACIT-F) < 18 14 (38.9) 69 (50.7) 0.21
AIMS physical < 22 13 (36.1) 71 (52.2) 0.09
AIMS psychological < 8 12 (34.3) 73 (53.7) 0.04

Differences were examined by use of the chi-square or Mann-Whitney U test when appropriate.

Table 4. Effect of compliance with leisure physical activity program on clinical outcomes of patients with RA
at 6-month followup.

Measure Compliant Group, n = 36 Noncompliant Group, n = 136

Mean SD Mean SD p

DAS28 0.38 1.41 0.48 1.44 0.75

HAQ 0.14 0.45 0.08 0.43 0.47
50-foot walk time, s 1.06 4.17 0.24 4.45 0.34
Anxiety (HAD) 0.42 1.90 0.10 2.43 0.24
Depression (HAD) 1.19 4.01 0.10 2.71 0.07
Coping (AHI) 1.25 4.21 0.10 4.09 0.08
Fatigue (FACIT) 2.94 8.04 0.10 7.25 0.04
AIMS physical 0.89 5.13 0.04 5.15 0.38
AIMS psychological 1.25 3.12 0.11 3.39 0.03
AIMS social 0.47 2.14 0.35 2.32 0.77

8.04 vs 0.1 7.25 for noncompliant patients, p = 0.04, a neg- particularly those whose baseline leisure PA was poor.
ative score indicating reduction of fatigue). Moreover, patients who were leisure PA-compliant showed
improved psychological status and decreased fatigue. But these
DISCUSSION effects were limited and short-term, which is in agreement with
Patients with RA who were assigned to receive an educational results of other studies showing that, in general, patient educa-
program showed increased compliance with the home-based tion has a limited effect on physical and psychosocial health sta-
exercise program and especially with the leisure PA program, tus and health behavior in patients with RA12,34.

Personal non-commercial use only. The Journal of Rheumatology Copyright 2008. All rights reserved.

220 The Journal of Rheumatology 2008; 35:2

Downloaded from www.jrheum.org on September 12, 2016 - Published by The

Journal of Rheumatology
 Since our study was a randomized controlled trial, we care responsibilities and physical exercises and by enhancing
ensured the internal validity of the study. However, the trial a positive attitude toward the leisure physical activities and
enrolment process before randomization might have resulted self-efficacy, which are determinants of adherence to short-
in a highly selected population who may not be representative term exercise regimens in patients with RA41.
of all people with RA, particularly in terms of age, duration of The multivariate analysis revealed that a baseline leisure
disease, and level of information about RA16. The effect of PA as assessed by the Baecke questionnaire predicted compli-
therapeutic education may diminish in older and well ance with the leisure PA program. Previous exercise habits
informed patients with long-lasting RA. Therefore this high have yet to be identified as a determinant of home-based exer-
selection could limit the generalizability of the study results35. cise compliance in RA41, but in contrast to Stenstrms
This multidisciplinary educational program was based on results, the initially less physically active patients were more
previous programs in other countries10,11,36 and did not focus compliant. This result could be related to a ceiling effect. A
exclusively on physical exercise. However, the exercise lack of sensitivity to change in physical function in response
behaviors after such interventions are usually documented to an exercise program, especially in high-functioning sub-
briefly, and our study provided a more precise analysis of jects, has been suggested for the results of other self-reporting
them. questionnaires42. This result also suggested that our educa-
The results concerning home-based exercise indicated a tional intervention allowed inactive people with arthritis to get
relatively low rate of compliance. Poor compliance with treat- moving and to approach an adequate level of activity.
ment recommendations is found across various health states Several authors have reported lower PA levels among
and treatments37. As expected, compliance with home-based patients with RA than among the general population40. This
exercise was poorer than that found with supervised programs leisure PA program, formulated with the expertise of a rehabil-
for RA35. But unexpectedly, it was also poorer than that with itation unit staff, aimed to keep RA patients physically active
an exercise program of minimal supervision used in and to counteract the negative consequences, primarily car-
Hakkinens trial of early RA38; however, lifestyle habits might diorespiratory effects, of decreased physical fitness. Significant
be more entrenched in patients with long-lasting RA. The health benefits can be achieved by moving people from inactive
most common deterrent to exercise is usually lack of motiva- to active status43, and the ultimate goal of this intervention was
tion33,39, so the effects of exercise disappear soon after to keep patients as physically active as possible for them to
patients finish a supervised exercise program. develop and maintain cardiorespiratory fitness44.
Leisure PA compliance was better than that with home- The univariate analysis demonstrated a significant relation
based exercise. The level of leisure activity was significantly between leisure PA compliance and AIMS psychological sub-
increased at 6-month followup in the active group, but was scores, so better psychological status was associated with bet-
considered stable in the control group. These results confirm ter leisure PA compliance. But this variable was not identified
that patients tended to adopt appropriate leisure physical as a predictor in the multivariate analysis. However, besides
behaviors more easily than they would regularly practice physical benefits, leisure PA influenced psychological status
home-based exercises33. Home-based training is less time- and decreased fatigue, identified as one of the most problem-
consuming and can be fit to any lifestyle, but it is certainly atic aspects of RA45. Not exercising enough could trap
less attractive than self-selected leisure activities, and particu- patients into a downward spiral, whereby inactivity stimulates
larly group activities34, which allow for social interactions pain levels and pain results in more inactivity. Thus, empow-
that are thought to play an important role in compliance. At ering RA patients through self-management exercise could
12-month followup, the Baecke score decreased in both disrupt this spiral.
groups, although a recent study found no difference in total Our intervention, as with other educational interventions12,
energy expenditure for RA patients over 12 months40. resulted in modest, short-term exercise behavior changes.
Data regarding determinants of adherence to longterm However, the low rate of inclusion among the eligible patients
exercise programs in RA are sparse and inconsistent15,35,40. and the excellent attendance rate in the active group suggest-
We did not perform multivariate analysis to identify predis- ed that the participants were highly motivated. That most par-
posing factors for compliance with home-based exercise ticipants had long-lasting RA in our institution, where they
because of the low number of participants who were compli- would have previously received a great deal of education
ant. However, univariate analysis demonstrated a strong rela- regarding their condition, could explain why few changes
tion between compliance with home-based exercise and par- occurred over the followup for both groups. Whatever the
ticipating in the active group. However, many more patients behavior change, a modest benefit for a single person trans-
allocated to the active group were leisure PA-compliant at 6- lates to large benefits for society in the long term. Maintaining
month followup. Usual medical care, including the delivery of individuals motivation to exercise over prolonged periods is
information booklets, did not change entrenched habits in not easy, but longterm benefits could be achieved with regular
most patients. Our educational program was an integral part of incentives. Reinforcement sessions are not effective in
patient management by helping patients to assimilate the self- encouraging RA patients36. Exercise classes are attractive and

Personal non-commercial use only. The Journal of Rheumatology Copyright 2008. All rights reserved.

Mayoux-Benhamou, et al: Patient education and RA 221

Downloaded from www.jrheum.org on September 12, 2016 - Published by The

Journal of Rheumatology
beneficial34, but not cost-effective9. Thus, adoption of recom- 5. Manson JE, Greenland P, LaCroix AZ, et al. Walking compared
mended leisure behaviors in RA is a longterm, cost-effective with vigorous exercise for the prevention of cardiovascular events
in women. N Engl J Med 2002;347:71625.
challenge facing all healthcare workers. Exercise consulta- 6. Swezey RL. Essentials of physical management and rehabilitation
tions face to face or by telephone, performed in exercise trials, in arthritis. Semin Arthritis Rheum 1974;3:349-68.
can help patients maintain high PA levels46, but in usual med- 7. de Jong Z, Vliet Vlieland TPM. Safety of exercise in patients with
ical practice, most physicians aware of the health benefits of rheumatoid arthritis. Curr Opin Rheumatol 2005;17:17782.
exercise admit that they rarely address the subject with their 8. Guidelines for the management of rheumatoid arthritis: 2002
update. Arthritis Rheum 2002;46:32846.
patients47,48. Moreover, most rheumatologists and physical 9. van den Hout WB, de Jong Z, Munneke M, Hazes JMW, Breedveld
therapists still recommend conventional RA exercise pro- FC, Vliet Vlieland PM. Cost-utility and cost-effectiveness analyses
grams with exercise restrictions49, and half the patients who of a long-term, high-intensity exercise program compared with
receive an exercise prescription are referred to physical thera- conventional physical therapy in patients with rheumatoid arthritis.
py50, with its questionable cost-effectiveness51. Arthritis Rheum 2005;53:39-47.
10. Lorig K, Lubeck D, Kraines RG, Seleznick M, Holman HR.
Our study showed that a multidisciplinary educational Outcomes of self-help education for patients with arthritis. Arthritis
intervention, added to usual medical practice, could convince Rheum 1985;28:680-5.
only a minority of patients with RA to be exercise-compliant 11. Lorig K, Ritter PL, Plant K. A disease-specific self-help program
in the long term, and that patients tended to adopt appropriate compared with a generalized chronic disease self-help program for
leisure physical behaviors more easily than they would regu-
arthritis patients. Arthritis Rheum 2005;53:950-7.
12. Riemsma RP, Kirwan JR, Taal E, Rasker JJ. Systematic review of
larly practice home-based exercises. Short bouts of self- rheumatoid arthritis patient education. Arthritis Rheum
selected moderate-intensity activity, such as walking, incor- 2004;51:1045-59.
porated into the daily routine can provide physical and mental 13. Cross MJ, March LM, Lapsley HM, Byrne E, Brooks PM. Patient
health benefits and decrease the susceptibility for cardiovas- self-efficacy and health locus of control: relationships with health
status and arthritis-related expenditure. Rheumatology Oxford
cular and other comorbid conditions among patients with RA. 2006;45:92-6.
Identification of psychological profiles as well as physical 14. Wikstrom I, Book C, Jacobsson LT. Difficulties in performing
health factors could be investigated in patients with RA to leisure activities among persons with newly diagnosed rheumatoid
help them overcome barriers to exercise. In our study, a pre- arthritis: a prospective, controlled study. Rheumatology Oxford
vious low level of leisure PA was identified as a determinant 2006;45:1162-6.
15. Wikstrom I, Jacobsson LT. Change in and predictors of leisure
of leisure PA compliance. This study suggested that educa- activities among patients with rheumatoid arthritis: a prospective
tional interventions fostering appropriate physical activity study. Scand J Rheumatol 2005;34:367-71.
could improve lifestyle habits particularly in RA patients with 16. Giraudet-Le Quintrec JS, Mayoux-Benhamou A, Ravaud P, et al.
a low level of PA, as a consequence of reluctance to exercise Impact of a collective educational program for rheumatoid arthritis:
and inadequate physical activity. a prospective 12-month randomized controlled trial. J Rheumatol,
in press.
Longterm exercise compliance rate was low and motiva- 17. Arnett FC, Edworthy SM, Bloch DA, et al. The American
tional strategies are needed to convince participants to under- Rheumatism Association 1987 revised criteria for the classification
go longterm physical training. Tailored advice and regular of rheumatoid arthritis. Arthritis Rheum 1988;31:315-24.
incentives from all healthcare professionals could be a first 18. Steinbrocker O, Traeger CH, Batterman RC. Therapeutic criteria in
step toward improving longterm RA patients awareness of rheumatoid arthritis. JAMA 1949;140:659-62.
19. van Gestel AM, Prevoo ML, van t Hof MA, van Rijswijk MH, van
exercise benefits besides pharmacological therapy. This de Putte LB, van Riel PL. Development and validation of the
requires convincing both healthcare professionals and patients European League Against Rheumatism response criteria for
to become engaged in exercise therapy. rheumatoid arthritis. Comparison with the preliminary American
College of Rheumatology and the World Health
ACKNOWLEDGMENT Organization/International League Against Rheumatism criteria.
Arthritis Rheum 1996;39:34-40.
We acknowledge the health professionals of the rehabilitation department
20. Harkcom TM, Lampman RM, Banwell BF, Castor CW. Therapeutic
who were involved in the education of patients.
value of graded aerobic exercise training in rheumatoid arthritis.
Arthritis Rheum 1985;28:329.
REFERENCES 21. Fries JF, Spitz P, Kraines RG, Holman HR. Measurement of patient
1. Macera CA, Hootman JM, Sniezek JE. Major public health benefits outcome in arthritis. Arthritis Rheum 1980;23:137-45.
of physical activity. Arthritis Rheum 2003;49:122-8. 22. Guillemin F, Briancon S, Pourel J. Mesure de la capacit
2. Wong JB, Ramey DR, Singh G. Long-term morbidity, mortality, fonctionnelle dans la polyarthrite rhumatode; Adaptation franaise
and economics of rheumatoid arthritis. Arthritis Rheum du Health Assessment Questionnaire (HAQ). Rev Rhum Mal
2001;44:2746-9. Osteoartic 1991;58:459-65.
3. Haugeberg G, Orstavik RE, Uhlig T, Falch JA, Halse JI, Kvien TK. 23. Bjelland I, Dahl AA, Haug TT, Neckelmann D. The validity of the
Bone loss in patients with rheumatoid arthritis: results from a Hospital Anxiety and Depression Scale. An updated literature
population-based cohort of 366 patients followed up for two years. review. J Psychosom Res 2002;52:69-77.
Arthritis Rheum 2002;46:17208. 24. Nicassio PM, Wallston KA, Callahan LF, Herbert M, Pincus T. The
4. Turesson C, Jarenros A, Jacobson L. Increased incidence of measurement of helplessness in rheumatoid arthritis. The
cardiovascular disease in patients with rheumatoid arthritis: results development of the Arthritis Helplessness Index. J Rheumatol
from a community based study. Ann Rheum Dis 2004;63:952-5. 1985;12:462-7.

Personal non-commercial use only. The Journal of Rheumatology Copyright 2008. All rights reserved.

222 The Journal of Rheumatology 2008; 35:2

Downloaded from www.jrheum.org on September 12, 2016 - Published by The

Journal of Rheumatology
25. Meenan RF, Mason JH, Anderson JJ, Guccione AA, Kazis LE. 37. Dunbar-Jacob J, Mortimer-Stephens MK. Treatment adherence in
AIMS2: the content and properties of a revised and expanded chronic disease. J Clin Epidemiol 2001;54:57-60.
Arthritis Impact Measurement Scales Health Status Questionnaire. 38. Hakkinen A, Malkia E, Hkkinen K, Jappinen I, Laitinen L,
Arthritis Rheum 1992;35:1-10. Hannonen P. Effects of detraining subsequent to strength training
26. Pouchot J, Guillemin F, Coste J, Bregeon C, Sany J. Validation of on neuromuscular function in patients with inflammatory arthritis.
the French version of the Arthritis Impact Measurement Scales 2 Br J Rheumatol 1997;36:1075-81.
and comparison with the French version of the Nottingham Health 39. Johnson CA, Corrigan SA, Dubbert PM, Gramling SE. Perceived
Profile. Quality of Life in Rheumatology Task Force. Rev Rhum barriers to exercise and weight control practice in community
Engl Ed 1996;63:389-404. women. J Women Health 1990;16:177-91.
27. Webster K, Cella D, Yost K. The Functional Assessment of Chronic 40. Mancuso CA, Rincon M, Sayles W, Paget SA. Comparison of
Illness Therapy (FACIT) Measurement System: properties, energy expenditure from lifestyle physical activities between
applications, and interpretation. Health Qual Life Outcomes patients with rheumatoid arthritis and healthy controls. Arthritis
2003;12:1-79. Rheum 2007;57:672-8.
28. Cella D, Yount S, Sorensen M, Chartash E, Sengupta N, Grober J. 41. Stenstrom CH, Arge B, Sundbom A. Home exercise and
Validation of the Functional Assessment of Chronic Illness Therapy compliance in inflammatory rheumatic diseases: a prospective
Fatigue Scale relative to other instrumentation in patients with clinical trial. J Rheumatol 1997;24:4706.
rheumatoid arthritis. J Rheumatol 2005;32:811-9. 42. Bilek LD, Venema M, Camp KL, Lyden ER, Meza JL. Evaluation
29. Baecke JAH, Burema J, Frijters JER. A short questionnaire for the of the human activity profile for use with persons with arthritis.
measurement of habitual physical activity epidemiological studies. Arthritis Rheum 2005;53:756-63.
Am J Clin Nutr 1982;36:932-42. 43. Blair SN, Wei M. Sedentary habits, health, and function in older
30. Bigard AX, Duforez F, Portero P, Gezennec CY. Dtermination de women and men. Am J Health Promot 2000;15:1-8.
lactivit physique par questionnaire: validation du questionnaire de 44. Pate RR, Pratt M, Blair SN, et al. Physical activity and public
Baecke. Sci Sport 1992;7:215-21. health: a recommendation from the Centers for Disease Control and
31. Dougados M, Giraudet-Le Quintrec JS, Mayoux-Benhamou MA, et Prevention and the American College of Sports Medicine. JAMA
lquipe de lInstitut de Rhumatologie de lhpital Cochin. La 1995;273:402-7.
polyarthrite rhumatode en 100 questions. Paris: Assistance 45. Tack BB. Self reported fatigue in rheumatoid arthritis. A pilot study.
Publiquehpitaux de Paris, dition NHA; 1997. Arthritis Care Res 1990;3:154-7.
32. Mayoux-Benhamou MA, Giraudet-Le Quintrec JS, et lquipe du 46. Kirk A, Mutrie N, MacIntyre P, Fisher M. Increasing physical
service de rducation de lInstitut de Rhumatologie de lhpital activity in people with type 2 diabetes. Diabetes Care
Cochin. Guide pratique de la polyarthrite rhumatode. Paris: 2003;26:1186-92.
Assistance Publiquehpitaux de Paris, dition NHA; 1999. 47. Pinto BM, Goldstein MG, Marcus BH. Activity counseling by
33. Mayoux-Benhamou MA, Roux C, Perraud A, Fermanian J, primary care physicians. Prev Med 1998;27:506-13.
Rahali-Kachlouf H, Revel M. Predictors of compliance with a 48. Iversen MD, Fossel AH, Daltroy LH. Rheumatologist-patient
home-based exercise programme added to usual medical care in communication about exercise and physical therapy in the
preventing postmenopausal osteoporosis: an 18-month prospective management of rheumatoid arthritis. Arthritis Care Res
study. Osteoporosis Int 2005;16:325-31. 1999;12:18092.
34. Taal E, Rasker JJ, Wiegman O. Group education for rheumatoid 49. Munneke M, de Jong Z, Zwinderman AH, et al. High intensity
arthritis patients. Semin Arthritis Rheum 1997;26:805-16. exercise or conventional exercise for patients with rheumatoid
35. Munneke M, de Jong Z, Zwinderman AH, et al. Adherence and arthritis? Outcome expectations of patients, rheumatologists, and
satisfaction of rheumatoid arthritis patients with a long-term physiotherapists. Ann Rheum Dis 2004;63:804-8.
intensive dynamic exercise program (RAPIT Program). Arthritis 50. Li LC, Bombardier C. Utilization of physiotherapy and
Rheum 2003;49:66572. occupational therapy by Ontario rheumatologists in managing
36. Riemsma RP, Taal E, Rasker JJ. Group education for patients with rheumatoid arthritis - a survey. Physiother Can 2003;55:2330.
rheumatoid arthritis and their partners. Arthritis Rheum 51. Li LC, Iversen MD. Outcomes of patients with rheumatoid arthritis
2003;49:556-66. receiving rehabilitation. Curr Opin Rheumatol 2005;17:172-6.

Personal non-commercial use only. The Journal of Rheumatology Copyright 2008. All rights reserved.

Mayoux-Benhamou, et al: Patient education and RA 223

Downloaded from www.jrheum.org on September 12, 2016 - Published by The

Journal of Rheumatology