1. Top of page
  2. Abstract
  8. Acknowledgments


To evaluate the effectiveness of a brief supervised education, self-management, and global upper extremity exercise training program, supplementing a home exercise regimen, for people with rheumatoid arthritis (RA; the Education, Self-Management, and Upper Extremity Exercise Training in People with Rheumatoid Arthritis [EXTRA] program).


Adults with RA of ≤5 years' duration were randomized to receive either usual care or the EXTRA program comprising 4 (1-hour) group education, self-management, and global upper extremity exercise training sessions supplementing the first 2 weeks of a 12-week individualized, functional home exercise regimen in addition to usual care. Outcome measures were assessed at baseline, 12 weeks (primary end point), and 36 weeks and included the Disabilities of the Arm, Shoulder, and Hand questionnaire (primary outcome measure), the Grip Ability Test, handgrip strength (N), the Arthritis Self-Efficacy Scale (pain, function, and symptoms subscales), and the 28-joint Disease Activity Score.


One hundred eight participants (26 men, mean ± SD age 55 ± 15 years, mean ± SD disease duration 20 ± 19 months) were randomized to receive either usual care (n = 56) or the EXTRA program (n = 52). At 12 weeks, there was a significant between-group difference in the mean change in disability (−6.8 [95% confidence interval (95% CI) −12.6, −1.0]; P = 0.022), function (−3.0 [95% CI −5.0, −0.5]; P = 0.011), nondominant handgrip strength (31.3N [95% CI 9.8, 52.8]; P = 0.009), self-efficacy (10.5 [95% CI 1.6, 19.5]; P = 0.021 for pain and 9.3 [95% CI 0.5, 18.2]; P = 0.039 for symptoms), and disease activity (−0.7 [95% CI −1.4, 0.0]; P = 0.047), all favoring the EXTRA program.


The EXTRA program improves upper extremity disability, function, handgrip strength, and self-efficacy in people with RA, with no adverse effects on disease activity.


  1. Top of page
  2. Abstract
  8. Acknowledgments

Rheumatoid arthritis (RA) is a systemic, inflammatory disease causing pain, joint destruction, and disability. Hand articular damage occurs early in RA, often within 5 years of diagnosis ([1]), and upper extremity function deteriorates as the disease progresses, impacting independence ([2, 3]).

Exercise is a key component in the management of people with RA ([4]), and general exercise programs improve sensorimotor control and aerobic capacity, which are essential for function ([5-7]). Effective upper extremity function requires good proximal motor control to stabilize the upper extremity, allowing hand placement and manual dexterity. Upper extremity exercises may be incorporated into general exercise programs, but studies evaluating exercise programs frequently utilize global health outcomes that do not measure upper extremity function in isolation, making it difficult to assess the effectiveness of exercise on upper extremity function ([8]). To date, studies evaluating the effectiveness of upper extremity exercise in people with RA have targeted single joint complexes, i.e., the shoulder or hand ([9-11]), and have not addressed potential global motor deficits that are associated with upper extremity dysfunction ([12]). Exercise regimens are frequently supervised, which is resource intensive and expensive to implement, and once exercise ceases, many of the benefits diminish ([13, 14]). Individually tailored home exercise regimens are required to address global upper extremity dysfunction that, if completed in the longer term, could encourage self-management and be cost effective to implement.

However, the long-term effectiveness of any exercise program relies on adherence to the regimen, which is poor in people with rheumatic diseases ([15]). Psychosocial variables can influence participation in habitual exercise and disease self-management ([16, 17]), and specific behavior change techniques, based on psychological models of health-related behavior, have been developed to facilitate behavior change ([18]). Although no one theoretical perspective is effective, techniques underpinned by the social cognition theory ([19]) have been successfully integrated into exercise prescription and delivered by health care professionals in people with chronic musculoskeletal conditions with long-term benefits ([20, 21]).

The social cognition theory incorporates 5 core constructs: knowledge of the risks and benefits of a health behavior, outcome expectations, self-regulation methods (e.g., goal setting and action planning), perceived barriers to and facilitators of a health behavior, and self-efficacy (confidence in one's ability to perform an action or behavior) ([22]). Self-efficacy, a key factor influencing health behavior ([23, 24]), may be developed through performance accomplishments such as successfully experiencing an exercise regimen, verbal persuasion, e.g., receiving encouragement from group members and therapists, vicarious experience, e.g., observing others' progress, and physiologic cues such as experiencing improvement following exercise ([22]). Therefore, experiencing an exercise regimen within a group that can then be performed conveniently at home, enhanced by information and individualized behavioral change techniques, may facilitate uptake and long-term adherence to exercise and promote self-management.

This study evaluated a brief, supervised education, self-management, and global upper extremity exercise training (EXTRA) program, supplementing a functional home exercise regimen, aimed at improving global upper extremity disability in people with RA.

Box 1. Significance & Innovations

  • This is the first study to evaluate the Education, Self-Management, and Upper Extremity Exercise (EXTRA) program in people with rheumatoid arthritis.
  • The EXTRA program improves upper extremity disability, function, and strength in people with rheumatoid arthritis for at least 12 weeks, and improves arthritis self-efficacy and pain for at least 36 weeks.
  • The EXTRA program is safe and does not adversely affect disease activity, even in participants with high disease activity.
  • The EXTRA program requires minimal therapist training and may be easily implemented into clinical practice.


  1. Top of page
  2. Abstract
  8. Acknowledgments

Study design

This assessor-blind, randomized controlled trial received ethical and research governance approval from King's, Guy's, St. Thomas', and Lewisham Hospital NHS Foundation Trusts and the London (Dulwich) Research Ethics Committee (08/H0808/118).

Study participants

People with RA (by American College of Rheumatology 1987 revised criteria [25]) of ≤5 years' duration who were age >18 years and had no contraindications to upper extremity exercise were recruited from the rheumatology clinics and physiotherapy departments of 4 public hospitals between February 2009 and September 2010, either by the researchers (VLM, DLS, LMB) or by direct referrals from the multidisciplinary team. People who had received intramuscular or upper extremity intraarticular steroid injections in the previous 4 weeks, those who had received upper extremity surgery or physiotherapy during the previous 6 months, and those who were unable to provide written informed consent were excluded. Initially, people with changes in disease-modifying antirheumatic drugs or therapy with a biologic agent within the previous 3 months were excluded from the study. However, to facilitate recruitment and ensure that people with a range of disease durations and disease activities were enrolled into the study, this exclusion criterion was removed for the final 6 months of the recruitment period, and only people commencing therapy with a biologic agent within the previous 3 months were excluded from the study.

Sample size

Based on a 12-week minimum clinically important difference (MCID) of 10 points in the Disabilities of the Arm, Shoulder, and Hand (DASH) questionnaire (primary outcome measure) ([26]), a 5% level of confidence (P < 0.05), 90% power, and assuming an SD of 21 ([27]), 100 participants were required (50 per study arm). To allow for an expected 20% attrition rate, we aimed to recruit 120 participants.

Outcome measures

At baseline, sociodemographic and disease characteristics (including age [years], height [cm], weight [kg], body mass index [kg/m2], smoking status, RA disease duration [months], number of comorbidities, and employment status) were recorded. All other outcome measures were assessed at baseline and 12 and 36 weeks from baseline except erythrocyte sedimentation rate (ESR), which was tested during routine clinical care and obtained from participant records. All outcomes were collected by a researcher (VLM) who was masked to treatment allocation. Participants who were unable to attend reassessments at 12 or 36 weeks completed the questionnaires by post.

Primary outcome measure

Disability was assessed with the 30-item DASH questionnaire ([28]). This brief, self-administered measure of global upper extremity symptoms and function is valid and reliable in people with early RA ([27, 29]). Items are rated on a Likert scale ranging from 1–5, where 1 = no difficulty and 5 = unable to do; a higher total DASH score (range 0–100) indicates greater disability.

Secondary outcome measures

Hand functional ability was assessed with the Grip Ability Test (GAT), a valid and reliable measure of handgrip function in people with RA comprising the timed assessment (weighted seconds) of 3 activities involving 4 common hand grips ([30]). Global upper extremity functional ability was assessed by timed dressing and eating (seconds) ([12]). Lower scores represent better function in both tests.

Handgrip strength was measured with a hydraulic handgrip dynamometer (Jamar model J00105, Lafayette Instrument) 3 times, using standardized protocols. The peak force (N) produced by each hand was used for analysis ([31]).

Quality of life was estimated with the 30-item Rheumatoid Arthritis Quality of Life (RAQoL) questionnaire ([32]). This brief, self-administered, disease-specific questionnaire requires participants to rate the applicability of statements to themselves “at the moment” (answers “yes” or “no”). A higher total RAQoL score indicates poorer quality of life (range 0–30).

The confidence people had in their ability to influence their disease symptoms and daily activities was measured with the self-administered, valid, and reliable Arthritis Self-Efficacy Scale (ASES), comprising subscales for pain (5 items), function (9 items), and other symptoms (6 items) ([33, 34]). Items are scored on a visual analog scale (VAS; 10–100 mm, where 10 = very uncertain and 100 = very certain). A higher mean subscale score indicates greater self-efficacy.

Disease activity was estimated with the valid and reliable 28-joint Disease Activity Score (DAS28), incorporating 28 swollen and tender joint counts, patient's assessment of disease activity (0–100-mm VAS, where 0 = not active at all and 100 = extremely active), and ESR (mm/hour) ([35]).

Additionally, participant-reported pain, fatigue (0–100-mm VAS, where 0 = no pain/fatigue and 100 = pain/fatigue as bad as it could be), morning stiffness in the last week (mean minutes), and assessor-rated disease activity (5-point Likert scale ranging from 1–5, where 1 = asymptomatic and 5 = very severe) were recorded.

Study protocol


Randomization was conducted via computer-generated random number generation and held by a third party unconnected to the study. Following the baseline assessment, participants were randomly assigned to receive either usual medical care or the EXTRA program in addition to usual care. A researcher not involved with the assessment (LMB) contacted the randomization administrator and then informed the participant of their treatment allocation. Participants were reminded not to discuss their treatment allocation with the assessing researcher (VLM) prior to each assessment. Success of assessor blinding was evaluated by asking the assessor to identify each participant's treatment allocation following their final assessment.

Usual care control group

Participants randomized to receive usual care continued to be managed by their medical team. Any pharmacologic, physical, or other therapy interventions required during the study were documented.

Education, Self-Management, and Upper Extremity Exercise Training for People with Rheumatoid Arthritis (EXTRA) program

Participants randomized to receive the EXTRA program continued to receive usual care but, in addition, received the EXTRA program facilitated by a senior clinical physiotherapist. Any supplementary therapies required during the study period were documented.

Development of the EXTRA program

Underpinned by the social cognition theory ([19]), the EXTRA program was developed from the successful Enabling Self-Management and Coping with Arthritis Knee Pain through Exercise program ([20]) and refined in collaboration with clinicians (physicians and allied health professionals) experienced in the management of people with RA, researchers, and patient representatives. Following a pilot study that assessed the patients' and therapists' experiences and acceptability of the intervention and feasibility of the study protocol (data not shown), the EXTRA program was refined to comprise 4 supervised group (4–6 participants per group) education, self-management, and global upper extremity exercise training sessions (delivered twice weekly for the first 2 weeks of the intervention) supplementing a functional daily home exercise regimen.

Supervised group sessions

The supervised sessions commenced with a 15-minute interactive discussion/seminar designed to increase participants' knowledge of RA and exercise, self-efficacy, and disease self-management, and facilitate uptake and longer-term exercise participation. Behavior change strategies were integrated into the seminars and reviewed, where necessary, in subsequent sessions (Table 1).

Table 1. Behavior change techniques applied during the EXTRA program as defined by Michie et al ([18])*
SessionTopics/activitiesBehavior change techniquesMethod of implementation
  1. EXTRA = Education, Self-Management, and Upper Extremity Exercise Training in People with Rheumatoid Arthritis; RA = rheumatoid arthritis.

  2. a

    Techniques applied in all sessions.

1EXTRA program aims and objectivesRA and exerciseExercise adviceSupervised exerciseProvide information on consequences of behavior in generalProvide normative information about others' behavioraProvide feedback on performanceaProvide instruction on how to perform behavioraDemonstrate behavioraPrompt practiceaFacilitate social comparisonaGroup discussion on the outcomes of exercise and inactivity in RA, facilitated by a physiotherapistInformation about others' behavior through group discussionExercise instruction and feedback from a physiotherapist and supported by the handbookExercise demonstration by a physiotherapistEncouraged to exercise daily and record exercise in the handbook diarySocial comparison facilitated by group discussion and exercise
2Managing flaresCoping with pain and tirednessSupervised exerciseProvide information on consequences of behavior in generalBarrier identification/problem solvingRelapse prevention/coping planningGroup discussion on the outcomes of exercise and inactivity in RAGroup discussion on exercise barriers and coping strategies and relapse preventionProblem solving section and contact details of clinical and research staff in the handbook
3Monitoring exercise intensityPersonal objectives and goal settingSupervised exercisePrompt self-monitoringGoal setting and action planning (“when,” “where,” “how”)Set graded tasksGroup discussion and advice from a physiotherapist on monitoring exercise intensity, goal setting, and action planningEncouraged to set and record short-term (weekly) and long-term goals and action plans
4Maintaining motivationExercise modificationSupervised exercisePrompt self-monitoringBarrier identification/problem solvingRelapse prevention/coping planningPrompt review of goalsPrompt generalization of target behaviorGroup discussion on exercise modification supported by the handbookGroup discussion to identify barriers to maintaining exercise and means of overcomingGroup discussion on relapse preventionEncouraged to review goals and action plansEncouraged to participate in physical activity

The interactive seminars were followed with a standardized exercise warm-up (including cardiovascular exercise and upper extremity and trunk range of movement/stretching exercises), a personalized exercise circuit of 6 upper extremity exercises selected from a core set of 16 functional/strengthening exercises (e.g., shoulder flexion against elastic resistance bands, handgrip against therapeutic putty resistance), and an exercise cool down (program available upon request from the corresponding author). The sessions were facilitated by a senior clinical physiotherapist who amended the participants' exercises during the sessions, if required (e.g., changed to other exercises from the core set or altered exercise intensity). Attendance at the supervised sessions was recorded.

Therapist training and fidelity to the EXTRA program

The physiotherapists conducting the EXTRA program sessions received 2 hours of training by the research team (VLM, LMB) on the aims and content of the program and techniques to facilitate the interactive discussions/seminars. A therapist manual (available upon request from the corresponding author) was developed to support accurate program delivery. A member of the research team (LMB) regularly attended the supervised sessions to monitor the accuracy of program delivery and fidelity using a checklist (available upon request from the corresponding author).

Home exercise program

Participants were asked to complete the exercise warm-up, their individually prescribed upper extremity exercises, and the exercise cool down on a daily basis at home (except the days when the participants attended a supervised session) for 12 weeks. Equipment (i.e., therapeutic putty and elastic resistance bands) was provided and the participants were advised to monitor and progress their exercise intensity using the Borg Rating of Perceived Exertion (RPE) scale (range 6–20 [36]). Participants were encouraged to maintain an RPE of 13–17 (equivalent to 50–80% of maximal exertion).

Participants were provided with an EXTRA program handbook containing information to support the interactive discussions/seminars, pictorial and written descriptions of all upper extremity exercises, advice on exercise modification, a 12-week exercise diary to record participation and intensity (including the Borg RPE scale [36]), a troubleshooting section, and contact details of the clinical and research staff. The exercise diary was returned to a researcher not masked to treatment allocation (LMB) after 12 weeks.

Statistical analysis

Analysis was by intent to treat. Values are shown as the mean (95% confidence interval [95% CI]) or as the median (interquartile range [IQR]) of raw and/or change scores (baseline − followup).

Data distribution was assessed by calculating Z scores for skewness and kurtosis (Z scores >2.58 were taken to represent significant deviations from normality) and data not normally distributed were transformed for analyses using reciprocal transformation (GAT timed dressing and eating). To aid interpretation of the transformed data, these values were transformed back to their original scales ([37]).

Data missingness was evaluated with Little's Missing Completely at Random test ([38]). Because there was not enough evidence to suggest data were missing at random (χ2 = 8146.3, 13,199 df, P = 1.000), complete cases were used for all descriptive and inferential statistics for the primary analyses of treatment effects.

Estimates of treatment effect were obtained using full factorial mixed-method analysis of variance with treatment, time, and treatment × time interaction as fixed effects. The assumption of sphericity was examined using Mauchly's test; where sphericity was violated, Greenhouse-Geisser correction was applied. Simple, first-order contrast effects were used to identify between-group differences from baseline to 12 and 36 weeks and confirmed with independent t-tests. Post hoc analyses using dependent t-tests with Bonferroni adjustment for multiple comparisons were conducted to determine within-group changes from baseline to 12 and 36 weeks.

Standardized effect sizes of between-group differences were calculated using Cohen's d (95% CI) and interpreted as small (d = 0.2), medium (d = 0.5), and large (d = 0.8) ([39, 40]). Number needed to treat (NNT [95% CI]) for the primary outcome measure was calculated ([40]) using a difference of 10 DASH points ([26]). Associations between variables were determined using Pearson's correlation coefficient.

Sensitivity analyses were completed to explore the potential effect of participant withdrawal from the study and the effect of alteration in medication 12 weeks prior to study enrollment on the primary outcome measure. To explore the effect of participant withdrawal, imputed data sets were generated using multiple imputation via a Bayesian fully conditional specification algorithm (Markov chain Monte Carlo) linear regression model using all variables as predictor variables ([38]), and estimation of the treatment effect of the complete case and imputed data sets was conducted (as above). To investigate the potential effect of changes in medication 12 weeks prior to study enrollment, analysis of treatment effect in participants with stable medication prior to study entry versus those who had medication changes within 12 weeks of baseline assessment were completed. All analyses were completed with SPSS for Windows, version 17, and significance was accepted at a 5% level of confidence (P values of 0.05 or less).


  1. Top of page
  2. Abstract
  8. Acknowledgments



Three hundred sixteen potential participants were identified from February 2009 to September 2010. One hundred twenty-two people agreed to participate; however, 14 failed to attend their baseline assessment. Therefore, 108 people (26 men) were assessed at baseline and randomized to receive the EXTRA program (n = 52) or to continue with usual care (n = 56) (Figure 1).


Figure 1. Trial profile and participant progression following the Education, Self-Management, and Upper Extremity Exercise Training in People with Rheumatoid Arthritis (EXTRA) program or usual care. Participants provided more than one reason for declining participation. RA = rheumatoid arthritis.

Download figure to PowerPoint


The overall mean ± SD age and disease duration of the study participants were 55 ± 15 years and 20 ± 19 months, respectively. There were no substantial between-group differences in the sociodemographic or disease characteristics of the participants, except that there were more men in the usual care group (Table 2). There were also no substantial differences in the participants who were lost to followup (n = 19) compared to those who completed the study (n = 89) (data not shown).

Table 2. Baseline characteristics of participants randomized to either the EXTRA program or usual care*
 EXTRA program (n = 52)Usual care (n = 56)
  1. EXTRA = Education, Self-Management, and Upper Extremity Exercise Training in People with Rheumatoid Arthritis; BMI = body mass index; IQR = interquartile range.

  2. a

    N = 50 per study arm.

Sex, no.  
Age, mean ± SD years53 ± 1657 ± 15
Weight, mean ± SD kga77 ± 1980 ± 19
Height, mean ± SD cma162 ± 8165 ± 10
BMI, mean ± SD kg/m229 ± 730 ± 7
Disease duration, mean ± SD months20 ± 1820 ± 19
Smokers, no.611
Comorbidities, median (IQR)2 (2)2 (2)
Employment status, no.  
Full time1010
Part time1110
Sick leave910
Medication prior to study enrollment, no.  
Medication change >3 months3635
Medication change ≤3 months1621

Primary outcome measure

The between-group difference in mean change in DASH scores was significant at 12 weeks, even following sensitivity analyses, favoring those in the EXTRA program. This equated to an NNT of 9 (95% CI 3, 16) people. This difference was no longer significant at 36 weeks (Figure 2 and Table 3). There were no substantial differences in response to treatment between those with stable mediation prior to study enrollment and those who had medication changes within 12 weeks of commencing the study (Figure 1 and Table 3).


Figure 2. Mean (95% confidence interval) change in the Disabilities of the Arm, Shoulder, and Hand (DASH) questionnaire following the Education, Self-Management, and Upper Extremity Exercise Training in People with Rheumatoid Arthritis (EXTRA) program or usual care.

Download figure to PowerPoint

Table 3. Baseline values and mean changes (95% confidence intervals) in the DASH questionnaire (primary outcome measure) at 12 and 36 weeks following the EXTRA program or usual care*
DASHWithin-group comparisonBetween-group comparison
EXTRA programUsual careBetween-group differenceF (df, error df)PaEffect size (d)
  1. DASH = Disabilities of the Arm, Shoulder, and Hand; EXTRA = Education, Self-Management, and Upper Extremity Exercise Training in People with Rheumatoid Arthritis; medication change >3 months = stable medication for at least 3 months prior to study enrollment; medication change ≤3 months = medication change within 3 months of study enrollment.

  2. a

    Derived from contrast effect from mixed-model two-way analysis of variance to investigate between-group difference in change from baseline to followup (12 or 36 weeks).

  3. b

    P < 0.05.

Complete case analysis      
Baseline44.6 (37.2, 52.0)40.8 (33.6, 48.0)3.8 (−6.6, 14.1)   
Change after 12 weeks−5.3 (−10.4, −0.2)b1.5 (−3.5, 6.5)−6.8 (−12.6, −1.0)b5.44 (1, 84)0.022b0.50 (0.07, 0.93)
Change after 36 weeks−2.7 (−9.5, 4.2)−1.4 (−8.0, 5.3)−1.3 (−9.1, 6.5)0.12 (1, 84)0.7360.07 (−0.35, 0.49)
Sensitivity analyses      
Multiple imputation analysis      
Change after 12 weeks−5.0 (−10.0, 0.0)b1.4 (−3.4, 6.2)−6.4 (−12.1, −0.8)b5.06 (1, 106)0.027b0.43 (0.05, 0.81)
Change after 36 weeks−3.4 (−10.0, 3.2)−1.2 (−7.5, 5.2)−2.2 (−9.7, 5.2)0.35 (1, 106)0.5560.13 (−0.25, 0.51)
Medication change >3 months      
Change after 12 weeks−4.6 (−10.9, 1.8)2.4 (−4.0, 8.9)−7.0 (−14.3, 0.3)3.65 (1, 55)0.0610.51 (−0.03, 1.03)
Change after 36 weeks−4.1 (−12.3, 4.2)−1.7 (−10.2, 6.7)−2.3 (−11.9, 7.3)0.24 (1, 55)0.6280.13 (−0.39, 0.65)
Medication change ≤3 months      
Change after 12 weeks−6.9 (−16.2, 2.4)−0.1 (−8.5, 8.3)−6.8 (−16.9, 3.3)1.91 (1, 27)0.1790.51 (−0.24, 1.24)
Change after 36 weeks0.4 (−13.0, 13.8)−0.7 (−12.7, 11.4)1.1 (−13.4, 15.5)0.02 (1, 27)0.879−0.06 (−0.79, 0.68)

At 12 weeks, upper extremity disability was associated with function (r = 0.61, P < 0.001) and disease activity (r = 0.63, P < 0.001) and negatively related to nondominant grip strength (r = −0.55, P < 0.001) and self-efficacy (pain: r = −0.61, function: r = −0.69, symptoms: r = −0.65; P < 0.001 for all).

Secondary outcome measures

Between-group differences in secondary outcome measures were significant at 12 weeks for function (GAT), nondominant handgrip strength, self-efficacy (ASES pain and symptoms subscales), DAS28, and participant-reported pain, and at 36 weeks for self-efficacy (pain) and participant-reported pain, in favor of the EXTRA program (Table 4).

Table 4. Baseline values and mean changes (95% confidence intervals) in secondary outcome measures at 12 and 36 weeks following the EXTRA program or usual care*
 Within-group comparisonBetween-group comparison
EXTRA programUsual careBetween-group differenceF (df, error df)PaEffect size (d)
  1. EXTRA = Education, Self-Management, and Upper Extremity Exercise Training in People with Rheumatoid Arthritis; GAT = Grip Ability Test; DAS28 = 28-joint Disease Activity Score; RAQoL = Rheumatoid Arthritis Quality of Life.

  2. a

    Derived from contrast effect from mixed-model two-way analysis of variance to investigate between-group difference in change from baseline to followup (12 or 36 weeks).

  3. b

    P < 0.01.

  4. c

    P < 0.05.

GAT, weighted seconds      
Baseline20.9 (23.3, 18.9)20.0 (22.2, 18.2)1.0 (−2.5, 4.5)   
Change after 12 weeks−2.5 (−4.4, −0.5)b0.5 (−1.5, 2.6)−3.0 (−5.0, −0.5)c6.85 (1, 77)0.011c0.59 (0.13, 1.04)
Change after 36 weeks−3.1 (−5.6, −0.5)b−0.9 (−2.8, 1.4)−2.0 (−4.4, 0.5)2.315 (1, 77)0.1320.34 (−0.11, 0.78)
Timed dressing, seconds      
Baseline21.7 (24.4, 20.0)20.8 (23.3, 18.9)1.2 (−2.5, 5.6)   
Change after 12 weeks−0.7 (−3.5, 2.1)0.5 (−1.1, 2.7)−1.9 (−4.3, 1.2)1.47 (1, 78)0.2290.27 (−0.17, 0.71)
Change after 36 weeks−0.7 (−3.3, 2.0)1.2 (−1.2, 3.0)−1.9 (−4.4, 1.3)1.61 (1, 78)0.2090.28 (−0.16, 0.72)
Timed eating, seconds      
Baseline7.9 (8.9, 7.2)7.4 (8.2, 6.8)0.6 (−0.6, 1.7)   
Change after 12 weeks−0.6 (−1.3, 0.1)−0.2 (−0.9, 0.6)−0.4 (−1.2, 0.4)1.14 (1, 77)0.2890.24 (−0.20, 0.68)
Change after 36 weeks−0.8 (−1.6, 0.1)−0.3 (−1.0, 0.5)−0.5 (−1.4, 0.5)0.93 (1, 77)0.3390.22 (−0.23, 0.66)
Baseline5.3 (4.7, 5.9)4.9 (4.4, 5.5)0.4 (−0.4, 1.2)   
Change after 12 weeks−0.8 (−1.4, −0.2)b−0.1 (−0.7, 0.4)−0.7 (−1.4, 0.0)c4.14 (1, 54)0.047c0.54 (0.00, 1.07)
Change after 36 weeks−0.8 (−1.4, −0.1)c−0.2 (−0.8, 0.4)−0.5 (−1.2, 0.1)2.51 (1, 54)0.1190.42 (−0.11, 0.95)
Pain, mm      
Baseline50.6 (42.6, 58.5)40.2 (32.3, 48.0)10.4 (−0.8, 21.6)   
Change after 12 weeks−13.0 (−23.0, −2.9)b1.7 (−8.2, 11.6)−14.7 (−26.2, −3.2)c6.48 (1, 79)0.0130.57 (0.12, 1.00)
Change after 36 weeks−8.0 (−18.0, 2.0)3.5 (−6.4, 13.4)−11.5 (−23.0, −0.1)c4.01 (1, 79)0.0490.45 (0.00, 0.88)
Fatigue, mm      
Baseline49.6 (40.7, 58.5)45.0 (36.2, 53.8)4.6 (−7.9, 17.0)   
Change after 12 weeks−7.9 (−18.3, 2.6)1.2 (−9.2, 11.5)−9.0 (−21.0, 2.9)2.27 (1, 79)0.1360.33 (−0.11, 0.77)
Change after 36 weeks−8.6 (−19.6, 2.5)−4.5 (−15.4, 6.4)−4.0 (−16.7, 8.6)0.41 (1, 79)0.5260.14 (−0.30, 0.58)
Morning stiffness, minutes      
Baseline155.9 (55.0, 256.7)92.9 (−8.0, 193.8)62.9 (−80.2, 206.1)   
Change after 12 weeks−115.9 (−249.7, 18.0)4.1 (−129.7, 137.9)−120.0 (−274.0, 34.0)2.41 (1, 78)0.1250.34 (−0.10, 0.78)
Change after 36 weeks−76.1 (−243.0, 90.8)56.9 (−110.0, 223.8)−133.0 (−325.0, 59.1)1.90 (1, 78)0.1720.31 (−0.14, 0.75)
Assessor-rated disease activity      
Baseline2.7 (2.4, 3.0)2.5 (2.2, 2.9)0.2 (−0.3, 0.6)   
Change after 12 weeks−0.2 (−0.5, 0.2)−0.2 (−0.5, 0.1)0.0 (−0.4, 0.4)0.01 (1, 79)0.915−0.02 (−0.46, 0.41)
Change after 36 weeks−0.3 (−0.6, 0.1)−0.1 (−0.4, 0.3)−0.2 (−0.6, −0.2)1.02 (1, 79)0.3150.22 (−0.22, 0.66)
Pain self-efficacy      
Baseline57.5 (50.7, 64.2)59.2 (52.9, 65.6)−1.7 (−11.0, 7.5)   
Change after 12 weeks4.8 (−3.1, 12.8)−5.7 (−13.2, 1.8)10.5 (1.6, 19.5)c5.54 (1, 79)0.021c0.52 (0.08, 0.96)
Change after 36 weeks6.6 (−0.8, 14.0)−1.8 (−8.8, 5.2)8.4 (0.1, 16.7)c4.06 (1, 79)0.0470.45 (0.00, 0.89)
Function self-efficacy      
Baseline62.8 (54.6, 71.0)63.6 (55.9, 71.3)−0.8 (−12.0, 10.4)   
Change after 12 weeks2.6 (−3.9, 9.1)−4.7 (−10.8, 1.5)7.2 (0.0, 14.5)3.92 (1, 79)0.0510.44 (0.00, 0.88)
Change after 36 weeks4.9 (−2.0, 11.7)−0.9 (−7.4, 5.5)5.8 (−1.8, 13.5)2.29 (1, 79)0.1340.33 (−0.11, 0.77)
Symptoms self-efficacy      
Baseline60.9 (53.6, 68.2)62.4 (55.2, 69.7)−1.5 (−11.8, 8.8)   
Change after 12 weeks4.6 (−3.1, 12.3)−4.7 (−12.4, 3.0)9.3 (0.5, 18.2)c4.43 (1, 74)0.039c0.48 (0.02, 0.93)
Change after 36 weeks4.5 (−3.6, 12.7)−3.2 (−11.3, 5.0)7.7 (−1.7, 17.0)2.69 (1, 74)0.1050.38 (−0.08, 0.83)
RAQoL questionnaire      
Baseline14.1 (11.2, 17.0)14.1 (11.4, 16.7)0.0 (−3.9, 3.9)   
Change after 12 weeks−1.4 (−3.3, 0.5)−0.8 (−2.5, 1.0)−0.6 (−2.8, 1.5)0.37 (1, 79)0.5450.13 (−0.30, 0.57)
Change after 36 weeks−0.7 (−2.7, 1.3)−1.3 (−3.1, 0.5)0.6 (−1.5, 2.8)0.31 (1, 79)0.581−0.12 (−0.56, 0.32)
Dominant handgrip, N      
Baseline184.8 (144.1, 225.5)223.8 (184.1, 263.5)−39.0 (−95.9, 17.8)   
Change after 12 weeks16.8 (−8.2, 41.7)3.7 (−20.6, 28.0)13.1 (−15.3, 41.4)0.84 (1, 76)0.3610.21 (−0.24, 0.65)
Change after 36 weeks16.5 (−16.7, 49.7)−6.1 (−38.5, 26.2)22.6 (−15.1, 60.3)1.43 (1, 76)0.2350.28 (−0.18, 0.71)
Nondominant handgrip, N      
Baseline160.5 (122.6, 198.3)227.0 (189.6, 264.4)−66.5 (−119.7, −13.3)   
Change after 12 weeks22.4 (3.7, 41.2)−8.9 (−27.4, 9.7)31.3 (9.8, 52.8)b8.41 (1, 79)0.005b0.64 (0.19, 1.08)
Change after 36 weeks12.4 (−19.6, 44.3)−17.2 (−48.8, 14.3)29.6 (−6.9, 66.1)2.60 (1, 79)0.1110.36 (−0.08, 0.79)

Changes in medication and the prescription of supplementary therapy during the study

During the study, there were fewer changes in medication at 12 weeks (n = 20 versus n = 25) and 36 weeks (n = 35 versus n = 50) and fewer referrals for supplementary therapies at 12 weeks (n = 2 versus n = 3) and 36 weeks (n = 4 versus n = 5) in favor of those participating in the EXTRA program (Figure 1).

Effectiveness of assessor masking

At 36 weeks, the assessing researcher (VLM) successfully identified participant group allocation in 42% of cases (n = 45 of 108; n = 32 in the EXTRA program and n = 13 in usual care).

Attendance and adherence to the EXTRA program

Seventy-one percent of those randomized to the EXTRA program (n = 37 of 52) attended ≥3 supervised sessions (median 3, IQR 2 sessions). The exercise diary was returned by 70% (n = 30 of 43) of the EXTRA program participants; 73% (n = 22 of 30) of those who returned the diary completed a median of 6 (IQR 0) prescribed exercises on ≥6 days/week (median 6, IQR 3 days/week) at a median RPE of 13 (IQR 2).


  1. Top of page
  2. Abstract
  8. Acknowledgments

In people with RA, a brief education, self-management, and upper extremity exercise training program (EXTRA) supplementing an individualized functional home exercise regimen improved upper extremity disability, function, nondominant handgrip strength, and arthritis self-efficacy for at least 12 weeks, with no adverse effects on disease activity or pain.

This is one of the first trials to evaluate a clinically applicable education, self-management, and global upper extremity exercise training program where treatment fidelity was monitored during delivery. The trial recruited an inner-city population of people with RA with a range of disease and sociodemographic characteristics, and the supervised sessions were delivered in a typical outpatient department by physiotherapists who required minimal additional training. This brief intervention supplemented an individualized functional home exercise regimen that utilized inexpensive equipment, and therefore may be more readily integrated into clinical practice than resource-intensive inpatient programs ([41]) or lengthy outpatient supervised regimens ([42]).

To achieve good upper extremity function, all joints must be anatomically adequate and have good sensorimotor feedback, effective motor control, and sufficient range of movement to stabilize the shoulder and accurately position the arm and hand for manipulative activities. Because upper extremity disability is associated with global sensorimotor deficits in people with RA ([12]), it is noteworthy that improvements in disability (d = 0.50; 95% CI 0.07, 0.93) following the EXTRA program appear greater than in studies that target single upper extremity joint complexes, although differing study outcomes hinder direct comparison ([9, 10]). Although there may be difficulties interpreting the clinical relevance of our findings, since MCIDs for many outcomes have not been established in an RA population, the change in DASH score following the EXTRA program is similar to the MCIDs reported in other chronic musculoskeletal conditions (4–15 DASH points) ([29]), which were associated with increased work capacity ([43]) and the NNT comparable to rheumatology drug trials ([44]).

This study enrolled participants with RA who had moderate to high disease activity, and one-third had recent medication changes prior to the baseline assessment. Although recent medication changes may confound the effects of the EXTRA program, there were no substantial differences between those with stable medication and those with changes in medication 3 months prior to study enrollment throughout the study, suggesting that the EXTRA program had no detrimental effects on disease activity. Because medication changes are frequent in people with relatively recently diagnosed RA, the effects of the EXTRA program may be suitable for people with RA even if they have moderate to high disease activity or recent changes in medication.

Sustaining exercise participation is demanding and frequently poor ([15, 45]). Attendance at our supervised sessions and initial adherence to the home exercise regimen were good, concurring with other studies ([20, 46]). This may be because the EXTRA program contained theoretically underpinned behavior change techniques that increased participants' self-efficacy, a key factor in both exercise uptake ([23]) and maintenance ([24]). Adherence to exercise was monitored by daily exercise diaries, and the majority of participants who returned their diary reported completing their prescribed exercises for 12 weeks, consistent with objective improvements in strength and function. Although there is a risk that participants may overestimate their exercise participation ([47]) and that diaries may facilitate adherence to exercise, this effect may be small ([48]).

Apart from improvements in self-efficacy and pain, the majority of the benefits following the EXTRA program were not sustained. This may be because the participants did not continue exercising, contrary to advice, and despite increased self-efficacy for self-management. However, exercise participation was only monitored for 12 weeks in this study (the primary study end point); therefore, longer-term exercise participation is not known. Although the EXTRA program incorporated a number of behavior change techniques, based on the social cognition theory, it may be that supplementary or alternative behavior change strategies targeting motivation for exercise or readiness to change exercise behavior (transtheoretical model) may be needed to encourage longer-term adherence.

There are methodologic limitations to our study; the trial did not reach the target recruitment, but attrition was low and significant between-group differences were found at the primary end point. Despite strategies to conceal allocation, the assessor frequently identified treatment allocation, potentially introducing bias, but consistent with other studies ([20]). Although the success of assessor masking was evaluated following each participant's final assessment, a more detailed account of when the assessor became aware of a participant's treatment allocation was not recorded, and this may have influenced the assessment of some outcomes. Masking interventions that require active participation is challenging, but several outcomes (DASH, ASES, and RAQoL) were self-administered, so these may be less influenced by assessor unmasking. However, self-reported measures are at risk of threats to their validity, such as demand characteristics, although the improvements in DASH score (self-reported disability) are consistent with changes in the GAT (objectively measured function), lending validity to the self-administered primary outcome.

The EXTRA program is a complex health care intervention; therefore, it is impossible to distinguish the independent effects of each component of the EXTRA program or the effect of therapist attention in this study, which confounds the interpretation of our findings. Future studies should utilize active comparison groups to explore this further.

As the societal costs of RA remain high ([49]) despite contemporary pharmacologic management, the EXTRA program provides an effective, low-cost, exercise-based intervention that requires little additional therapist training and no specialized equipment. The EXTRA program could be provided to large numbers of people with RA and is concordant with clinical guidelines for the management of RA ([4]) and public health physical activity guidelines ([50]), and promotes disease self-management.


  1. Top of page
  2. Abstract
  8. Acknowledgments

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Bearne had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Manning, Hurley, Scott, Bearne.

Acquisition of data. Manning, Scott, Bearne.

Analysis and interpretation of data. Manning, Hurley, Scott, Coker, Choy, Bearne.


  1. Top of page
  2. Abstract
  8. Acknowledgments

The authors would like to thank all of the participants who gave their time and effort during this study; the clinical teams at King's, Guy's, St. Thomas', and Lewisham NHS Foundation Trusts; and the physiotherapists who facilitated the EXTRA program supervised sessions.


  1. Top of page
  2. Abstract
  8. Acknowledgments
  • 1
    Johnsson PM, Eberhardt K.Hand deformities are important signs of disease severity in patients with early rheumatoid arthritis. Rheumatology (Oxford)2009;48:1398401.
  • 2
    Adams J, Burridge J, Mullee M, Hammond A, Cooper C.Correlation between upper limb functional ability and structural hand impairment in an early rheumatoid population. Clin Rehabil2004;18:40513.
  • 3
    Horsten NC, Ursum J, Roorda LD, van Schaardenburg D, Dekker J, Hoeksma AF.Prevalence of hand symptoms, impairments and activity limitations in rheumatoid arthritis in relation to disease duration. J Rehabil Med2010;42:91621.
  • 4
    National Collaborating Centre for Chronic Conditions.Rheumatoid arthritis: national clinical guideline for management and treatment in adults. London: National Institute for Health and Clinical Excellence (NICE); 2009.
  • 5
    Sainburg RL, Poizner H, Ghez C.Loss of proprioception produces deficits in interjoint coordination. J Neurophysiol1993;70:213647.
  • 6
    Bearne LM, Scott DL, Hurley MV.Exercise can reverse quadriceps sensorimotor dysfunction caused by rheumatoid arthritis without exacerbating disease activity. Rheumatology (Oxford)2002;41:15766.
  • 7
    Minor MA, Hewett JE, Webel RR.Efficacy of physical conditioning exercise in patients with rheumatoid arthritis and osteoarthritis. J Rheumatol1989;15:90511.
  • 8
    O'Connor D, Kortman B, Smith A, Ahern M, Smith M, Krishnan J.Correlation between objective and subjective measures of hand function in patients with rheumatoid arthritis. J Hand Ther1999;12:3239.
  • 9
    Brorsson S, Hilliges M, Sollerman C, Nilsdotter A.A six-week hand exercise programme improves strength and hand function in patients with rheumatoid arthritis. J Rehabil Med2009;41:33842.
  • 10
    O'Brien AV, Jones P, Mullis R, Mulherin D, Dziedzic K.Conservative hand therapy treatments in rheumatoid arthritis: a randomized controlled trial. Rheumatology (Oxford)2006;45:57783.
  • 11
    Bostrom C, Harms-Ringdahl K, Karreskog H, Nordemar R.Effects of static and dynamic shoulder rotator exercises in women with rheumatoid arthritis: a randomised comparison of impairment, disability, handicap, and health. Scand J Rheumatol1998;27:28190.
  • 12
    Bearne LM, Coomer AF, Hurley MV.Upper limb sensorimotor function and functional performance in patients with rheumatoid arthritis. Disabil Rehabil2007;29:10359.
  • 13
    Lemmey AB, Marcora SM, Chester K, Wilson S, Casanova F, Maddison PJ.Effects of high-intensity resistance training in patients with rheumatoid arthritis: a randomized controlled trial. Arthritis Rheum2009;61:172634.
  • 14
    Hakkinen A, Malkia E, Hakkinen K, Jappinen I, Laitinen L, Hannonen P.Effects of detraining subsequent to strength training on neuromuscular function in patients with inflammatory arthritis. Br J Rheumatol1997;36:107581.
  • 15
    Manning VL, Hurley MV, Scott DL, Bearne LM.Are patients meeting the updated physical activity guidelines? Physical activity participation, recommendation, and preferences among inner-city adults with rheumatic diseases. J Clin Rheumatol2012;18:399404.
  • 16
    Lorig K, Gonzalez VM, Laurent DD, Morgan L, Laris BA.Arthritis self-management program variations: three studies. Arthritis Care Res1998;11:44854.
  • 17
    Gecht MR, Connell KJ, Sinacore JM.A survey of exercise beliefs and exercise habits among people with arthritis. Arthritis Care Res1996;9:828.
  • 18
    Michie S, Ashford S, Sniehotta FF, Dombrowski SU, Bishop A, French DP.A refined taxonomy of behaviour change techniques to help people change their physical activity and healthy eating behaviours: the CALO-RE taxonomy. Psychol Health2011;26:147998.
  • 19
    Bandura A.Social foundations of thought and action: a social cognitive theory. Englewood Cliffs (NJ): Prentice Hall; 1986.
  • 20
    Hurley MV, Walsh NE, Mitchell HL, Pimm TJ, Patel A, Williamson E, et al.Clinical effectiveness of a rehabilitation program integrating exercise, self-management, and active coping strategies for chronic knee pain: a cluster randomized trial. Arthritis Rheum2007;57:12119.
  • 21
    Hurley MV, Walsh NE, Mitchell H, Nicholas J, Patel A.Long-term outcomes and costs of an integrated rehabilitation program for chronic knee pain: a pragmatic, cluster randomized, controlled trial. Arthritis Care Res (Hoboken)2012;64:23847.
  • 22
    Bandura A.Self-efficacy: toward a unifying theory of behavioral change. Psychol Rev1977;84:191215.
  • 23
    Dawson KA, Brawley LR.Examining the relationship between exercise goals, self-efficacy, and overt behavior with beginning exercisers. J Appl Soc Psychol2000;30:31529.
  • 24
    Luszczynska A, Sutton S.Physical activity after cardiac rehabilitation: evidence that different types of self-efficacy are important in maintainers and relapsers. Rehabil Psychol2006;51:31421.
  • 25
    Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al.The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum1988;31:31524.
  • 26
    Gummesson C, Atroshi I, Ekdahl C.The Disabilities of the Arm, Shoulder and Hand (DASH) outcome questionnaire: longitudinal construct validity and measuring self-rated health change after surgery. BMC Musculoskelet Disord2003;4:11.
  • 27
    Adams J, Burridge J, Mullee M, Hammond A, Cooper C.Self-reported hand functional ability measured by the DASH in individuals with early rheumatoid arthritis. Hand Ther2005;10:214.
  • 28
    Hudak PL, Amadio PC, Bombardier C.Development of an upper extremity outcome measure: the DASH (Disabilities of the Arm, Shoulder, and Hand) [published erratum appears in Am J Ind Med 1996;30:372]. Am J Ind Med1996;29:6028.
  • 29
    Beaton DE, Katz JN, Fossel AH, Wright JG, Tarasuk V, Bombardier C.Measuring the whole or the parts? Validity, reliability, and responsiveness of the Disabilities of the Arm, Shoulder and Hand outcome measure in different regions of the upper extremity. J Hand Ther2001;14:12846.
  • 30
    Dellhag B, Bjelle A.A grip ability test for use in rheumatology practice. J Rheumatol1995;22:155965.
  • 31
    Grindulis KA, Calverley M.Grip strength: peak or sustained pressure in rheumatoid arthritis. J Chronic Dis1983;36:8558.
  • 32
    De Jong Z, van der Heijde D, McKenna SP, Whalley D.The reliability and construct validity of the RAQoL: a rheumatoid arthritis-specific quality of life instrument. Br J Rheumatol1997;36:87883.
  • 33
    Barlow JH, Williams B, Wright C.The Generalized Self-Efficacy Scale in people with arthritis. Arthritis Care Res1996;9:18996.
  • 34
    Lorig K, Chastain RL, Ung E, Shoor S, Holman HR.Development and evaluation of a scale to measure perceived self-efficacy in people with arthritis. Arthritis Rheum1989;32:3744.
  • 35
    Prevoo ML, van Riel P, van ‘t Hof MA, van Rijswijk MH, van Leeuwen MA, Kuper HH, et al.Validity and reliability of joint indices: a longitudinal study in patients with recent onset rheumatoid arthritis. Br J Rheumatol1993;32:58994.
  • 36
    Borg G.Perceived exertion as an indicator of somatic stress. Scand J Rehabil Med1970;2:928.
  • 37
    Jorgensen E, Pedersen AR.How to obtain those nasty standard errors from transformed data: and why they should not be used. Pennsylvania: Pennsylvania State University, College of Information Sciences and Technology; 1998.
  • 38
    Little R.Statistical analysis with missing data. New York: John Wiley and Sons; 1981.
  • 39
    Cohen J.Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale (NJ): Lawrence Erlbaum; 1988.
  • 40
    Durlak JA.How to select, calculate, and interpret effect sizes. J Pediatr Psychol2009;34:91728.
  • 41
    Ronningen A, Kjeyen I.Effect of an intensive hand exercise programme in patients with rheumatoid arthritis. Scand J Occup Ther2008;15:17383.
  • 42
    Lemmey AB, Williams SL, Marcora SM, Jones J, Maddison PJ.Are the benefits of a high-intensity progressive resistance training program sustained in rheumatoid arthritis patients? A 3-year followup study. Arthritis Care Res (Hoboken)2012;64:715.
  • 43
    Schakenraad CH, Vendrig L, Sluiter JK, Veenstra W, Frings-Dresen MH.Evaluation of a multidisciplinary treatment for patients with chronic non-specific upper-limb musculoskeletal disorders: a pilot study. Occup Med (Lond)2004;54:5768.
  • 44
    Osiri M, Suarez-Almazor ME, Wells GA, Robinson V, Tugwell P.Number needed to treat (NNT): implication in rheumatology clinical practice. Ann Rheum Dis2003;62:31621.
  • 45
    Dunbar-Jacob J, Erlen JA, Schlenk EA, Ryan CM, Sereika SM, Doswell WM.Adherence in chronic disease. Annu Rev Nurs Res2000;18:4890.
  • 46
    De Jong Z, Munneke M, Zwinderman AH, Kroon HM, Jansen A, Ronday KH, et al.Is a long-term high-intensity exercise program effective and safe in patients with rheumatoid arthritis? Results of a randomized controlled trial. Arthritis Rheum2003;48:241524.
  • 47
    Prince SA, Adamo KB, Hamel ME, Hardt J, Connor Gorber S, Tremblay M.A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review. Int J Behav Nutr Phys Act2008;5:56.
  • 48
    Moseley GL.Do training diaries affect and reflect adherence to home programs? Arthritis Rheum2006;55:6624.
  • 49
    Neovius M, Simard JF, Askling J.How large are the productivity losses in contemporary patients with RA, and how soon in relation to diagnosis do they develop? Ann Rheum Dis2011;70:10105.
  • 50
    Department of Health, Physical Activity, Health Improvement and Protection.Start active, stay active: a report on physical activity for health from the four home countries’ chief medical officers. London: Department of Health, Physical Activity, Health Improvement and Protection; 2011.