Arthritis & Rheumatism (Arthritis Care & Research) Vol. 51, No. 6, December 15, 2004, pp. 1045–1059 DOI 10.1002/art.20823 © 2004, American College of Rheumatology REVIEW Systematic Review of Rheumatoid Arthritis Patient Education ROBERT P. RIEMSMA,1 ERIK TAAL,2 JOHN R. KIRWAN,3 AND JOHANNES J. RASKER4 Introduction Rheumatoid arthritis (RA) is a common chronic condition characterized by uncertain disease progression and an unpredictable course of exacerbations and remissions. Approximately 1–3% of the population in Western countries is affected by RA (1). Various interventions may alleviate its course, and patients come into contact with a large number and variety of health professionals. For many patients, pain, disability, deformity, and reduced quality of life persist in spite of treatment. There is room for new approaches to enhance current treatment effectiveness. Patient education is one such approach that is thought to be beneﬁcial in helping patients cope with their disease and cooperate with its complex management (2,3). As with other chronic diseases, there is no cure for most types of arthritis, including RA. Furthermore, the disease This article is a summary of a Cochrane systematic review: Riemsma R, Kirwan J, Taal E, Rasker J. Patient education for adults with rheumatoid arthritis. Cochrane Database Syst Rev 2002;(3):CD003688. An editorial discussing the main results of the review has been published: Riemsma RP, Taal E, Kirwan JR, Rasker JJ. Patient education programmes for adults with rheumatoid arthritis [editorial]. BMJ 2002;325:558 –9. The results of a Cochrane review can be interpreted differently, depending on people’s perspectives and circumstances. Please consider the conclusions presented carefully. They are the opinions of review authors, and are not necessarily shared by the Cochrane Collaboration. The views expressed in this report are those of the authors and not necessarily those of the Dutch Arthritis Association. Any errors are the responsibility of the authors. Supported by a grant from the Dutch Arthritis Association. 1 Robert P. Riemsma, PhD: Centre for Reviews and Dissemination, University of York, York, United Kingdom; 2Erik Taal, PhD: University of Twente, Enschede, The Netherlands; 3John R. Kirwan, BSc, MD: Bristol Royal Inﬁrmary, Bristol, United Kingdom; 4Johannes J. Rasker, MD, PhD: University of Twente and Medisch Spectrum Twente, Enschede, The Netherlands. Address correspondence to Robert P. Riemsma, PhD, Centre for Reviews and Dissemination, University of York, York, YO10 5DD, UK. E-mail: email@example.com. Submitted for publication July 4, 2003; accepted in revised form November 30, 2003. course is often unpredictable and the symptoms that patients experience can vary from day to day or even from hour to hour. Because of the nature of pain and disability, the partial and inconsistent effects of treatment, and the unpredictability people with arthritis face on a daily basis, patient education programs have become a complement to traditional medical treatment (4). Thus, patient education programs have given people with arthritis the strategies and the tools necessary to make daily decisions to cope with the disease (5,6). Patient education has been deﬁned as “any set of planned educational activities designed to improve patients health behaviors and/or health status” (7). Lorig has further stated “the purpose of patient education is to maintain or improve health, or, in some cases, to slow deterioration” (7). The focus of arthritis patient education programs is to teach patients to adjust their daily activities as dictated by disease symptoms. In other words, in addition to teaching patients what they should do, patients are also instructed on how to approach situations and to make adjustments that are appropriate for each individual and his or her own needs. There has been increasing research in the ﬁeld of patient education, and major reviews of published studies have been conducted on the value of education in general (8) and, more recently, on education in arthritis (5,6,9,10). Two reviews on arthritis patient education reported combined effect estimates on main outcome measures, such as pain, functional disability, and psychological outcomes (9,10). Hawley reviewed 34 rheumatic disease patient education clinical trials performed between 1985 and 1995 (9). She reported average effect sizes for treatment groups compared with nonintervention controls for RA patient pain, functional ability, and depression at postintervention of 0.13, – 0.16, and 0.01, respectivley; effect sizes for osteoarthritis (OA) patient pain, functional ability, and depression at postintervention were 0.44, 0.28, and 0.56, respectively (conﬁdence intervals were not reported). These data show a trend to greater improvement for OA compared with RA patients (9). Superio-Cabuslay et al compared the effects of 19 patient education trials and 28 nonsteroidal antiinﬂammatory drug trials between 1966 and 1993 among patients with OA and RA (10). They found an average effect size for treatment compared with nonintervention controls for RA patient pain and func1045 1046 tional disability at postintervention of 0.16 (95% conﬁdence interval [95% CI] – 0.18, 0.50) and 0.18 (95% CI – 0.27, 0.63); the comparable effect sizes for OA patients were 0.16 (95% CI – 0.69, 1.02) and 0.0 (95% CI – 0.61, 0.61), suggesting that RA patients show greater improvements compared with OA patients after completion of an educational intervention (10). Although the data are ambiguous, available data suggest that there may be a difference in efﬁcacy by diagnosis. From the available literature, it is not clear what type of educational interventions are most effective in improving health status for patients with chronic diseases (11). Education strategies can vary from the provision of information only to the use of cognitive-behavioral strategies. Our objective was to determine the effectiveness of patient education in patients with RA based on a systematic review of the evidence from randomized controlled trials (RCTs). We focused on the effects of patient education on pain, functional disability, joint counts, patient and physician global assessment, affect scores, and measures of acute phase reactants. These outcome measures are based on the core set of outcome measures to be used in clinical trials in RA as identiﬁed and agreed upon by Outcome Measures in Rheumatology Clinical Trials (OMERACT) (12). This set of outcome measures has been acknowledged as the gold standard for outcome measures in RA by the World Health Organization (WHO) and the International League of Associations for Rheumatology (ILAR) (13). To examine the differential effects of various types of patient education, we have made a distinction between 3 types of patient education in our analyses: information only, counseling, and behavioral treatment. Information only includes all interventions aimed primarily at the exchange of information by means of persuasive communication or informational brochures; these interventions do not include a behavioral component and are not aimed at generating support. Counseling includes interventions mainly aimed at social support and giving patients the opportunity to discuss their problems. Behavioral treatment refers to interventions that include techniques aimed at behavioral change, such as behavioral instruction, skills training, and biofeedback. Literature search A systematic search using the recommendations of the Cochrane Collaboration was conducted using 3 databases (Medline, EMBASE, and PsycInfo) using entries from 1966 forward to September 2002. A similar search was performed in the Cochrane database, and a selection of review articles were examined to identify additional relevant publications (5,6,9,10,14 –23). Search terms included arthritis, health promotion, patient education, behavior therapy, occupational therapy, self care, psychological adaptation, counseling, and exercise therapy. No restrictions were made concerning the language of publication (24,25). RCTs in which patients with a conﬁrmed diagnosis of RA were included. Studies with mixed populations were included, but only data regarding RA patients were used in Riemsma et al the analyses. Furthermore, RCTs had to include a nonintervention control group and pretest and posttest results had to be available separately for RA, either in the publication or from the studies’ authors. Studies that did not include data on the predeﬁned set of outcome measures were reported, but excluded from the meta-analysis. If data necessary for the calculation of weighted mean differences (WMDs) or standardized mean differences (SMDs) were unavailable, either in the publication or from the studies’ authors, the study was also excluded from the analysis. Studies that did include data on the predeﬁned set of outcome measures, but only for speciﬁc parts of the body, e.g., pain in the hand, were also excluded. Outcomes included in the review are scores on pain, functional disability, joint counts, patient and physician global assessment, psychological status (anxiety and depression), and disease activity (acute phase reactants). All patient education interventions were eligible for inclusion in this review. We deﬁned a patient education intervention as one that includes formal structured instruction on arthritis and ways to manage arthritis symptoms. Studies that used modern psychobehavioral methods to promote changes in health behaviors were also included. As a complement to an instructional component, interventions could include exercise, biofeedback, or psychosocial support. The title and abstract of each citation were examined by 2 reviewers (RPR and ET) and the trials were retrieved that, according to at least 1 of the reviewers, cited RCTs. If it was unclear from the title and abstract whether allocation of the intervention had been conducted in a randomized manner or whether the intervention included an educational component or whether RA patients were involved, the full report was retrieved. Both reviewers then examined the full reports. Disagreements regarding inclusion status were resolved by discussion. Types of outcome measures A core set of outcome measures to be used in clinical trials in RA have been identiﬁed and agreed upon by OMERACT (12). This set of outcome measures has been acknowledged as the gold standard for outcome measures in RA by the WHO and ILAR (13). For RA, the preliminary core set of outcomes identiﬁed by OMERACT including validated measures of pain, functional disability, joint pain/tenderness, joint swelling, patient and physician global assessment, and acute phase reactants (erythrocyte sedimentation rate and C-reactive protein) were selected as outcome measures to be included in this review. Because psychological status is an important aspect of health status, we also included affect scores (psychological status, anxiety, and depression). Assessment of study quality Methodologic quality of the included trials was assessed independently by 2 assessors (RPR and JRK) using an adapted version of the instrument developed by Jadad et al (26). This was done by evaluating the methods and results of the reports without knowledge of the authors. Disagreement among the reviewers regarding the quality of the RA Patient Education: Review 1047 Table 1. Criteria for the assessment of methodologic quality of randomized controlled trials* Description of criteria Selection bias Randomization reported but not speciﬁed On-site computer, random number tables Centralized or in prenumbered/coded/ identical boxes or containers Attrition (losses to followup) Followup ⬍80% overall or not reported Followup ⱖ80% Intention-to-treat, explicit and clear Detection bias (blinding of outcome assessment) Not reported Reported but not fully blinded Outcome assessment fully blinded Performance bias (cointerventions) Cointerventions allowed but not reported Cointerventions allowed, reported Cointerventions allowed, reported, analyzed, or not allowed Scoring 0 1 2 0 1 2 0 1 2 0 1 2 * Each criterion was scored from 0 to 2; therefore, a maximum score of 8 and a minimum score of 0 could be achieved for each trial. articles was readily resolved by discussion and consensus. Our quality scale comprises the 3 criteria proposed by Jadad et al (26), which cover 3 of 4 criteria outlined in the Cochrane Collaboration handbook (27): selection bias, attrition bias, and detection bias. We added 1 item concerning cointerventions to cover the fourth criterion: performance bias (see Table 1). Performance bias refers to systematic differences in care provided to comparison groups other than the intervention of interest. Data extraction and presentation The details of the included reports were scrutinized by 1 reviewer (RPR) and a standardized form was used for data abstraction. For continuous variables, WMDs, SMDs, and their 95% CIs were calculated using RevMan 4.2.2 software (28). Effect sizes were calculated for each outcome as the value of the mean change from baseline to postintervention (or ﬁnal followup) in the experimental group minus the mean change from baseline to postintervention (or ﬁnal followup) in the control group, divided by the pooled standard deviation of the change scores in both groups. In RevMan, the weight given to each study is the inverse of its variance, i.e., more precise estimates (from larger studies with more events) are given more weight (29). The calculation of effect sizes (WMDs and SMDs) was such that negative values favor the patient education intervention. If absolute values were reported, we calculated mean differences in scores from baseline to postintervention. The results for each trial were tested for heterogeneity using the chi-square statistic. Effect estimates were analyzed using ﬁxed effects models unless heterogeneity, due to differences in the outcome measures, was signiﬁ- cant (at P ⬍ 0.05); in which case a random effects model was used. Results are reported at ﬁrst followup and at ﬁnal followup. First followup refers to the assessment immediately after the intervention. Depending on the duration of the intervention, this ranged from 3 weeks to 9 months after baseline assessment. Final followup refers to the last assessment in the study, this ranged from 3 to 14 months after the intervention. For studies with a waiting-list control group, assessments that were performed after the crossover are not included in this review. Results are reported separately for 3 types of interventions: information only, counseling, and behavioral treatment. Potential bias in meta-analytic research is publication bias, which occurs when trials showing no effect are selectively not published (30). One method used to detect publication bias is to draw a funnel plot. Funnel plots show the distribution of effect sizes according to sample size: it is to be expected that the points will ﬁll a funnel shape, with there being more variability in reported effect sizes for smaller studies. Large gaps in the funnel indicate a group of possibly “missing” publications. These omissions are usually small studies with point estimates suggesting a different effect from those available and are unlikely to be missing at random (31). We investigated whether publication bias existed among these studies by plotting sample sizes versus effect sizes for the outcomes that were most often reported: pain and functional disability. Sensitivity analyses Other sources of bias in the meta-analysis were dealt with by several sensitivity analyses. The results are shown with and without use of quality scores to examine the effect of quality scores and we have run the analysis with only the larger studies to help determine the extent to which publication bias affected the conclusions. The main analyses focused on outcomes reported at ﬁrst followup, i.e., immediately after the intervention irrespective of the duration of the intervention. However, as some studies assessed ﬁrst followup after 3 weeks and others after 9 months, depending on the duration of the intervention, we did a separate analysis for studies that reported outcomes 2– 4 months after baseline assessment. A way to reduce heterogeneity is using only one, the most commonly used, instrument to measure each outcome; therefore a separate sensitivity analysis was performed using the most commonly used instrument to measure each outcome. Some studies included 2 or 3 experimental conditions. Because we included comparisons of each experimental condition versus the control condition, the control conditions for these studies were included 2 or 3 times, thus overestimating the results of the control condition. To see whether this overestimation seriously inﬂuenced results, we have done separate analysis including only 1 (the most extreme, see Table 2) educational intervention. Studies selected The search strategies identiﬁed 1,423 references, which were ﬁrst examined on the basis of titles and abstracts (see 0 NR 30 15 15 23 6 3 3 15 37 12 35 13 4 3 7 46 9 1 6 0 n 18 108 602 150 65 60 35 105 79 68 100 105 34 63 100 60 219 98 33 84 141 59 Author, year (ref.)* Appelbaum, 1988 (67) Barlow, 1997 (40,41,pers com) Barlow, 2000 (48) Bell, 1998 (49) Brus, 1998 (50,92,pers com) Geissner, 1994 (51,pers com) Multimodal pain management** Visualization techniques Relaxation training Hammond, 1999 (52) Helewa, 1991 (68,pers com) Helliwell, 1999 (53) Hewlett, 1998 (pers com) Hill, 2001 (54) Huiskes, 1991 (44,45,pers com) Combination therapy** Cognitive behavioral therapy Occupational therapy Kaplan, 1981 (69) Leibing, 1999 (55) Lindroth, 1997 (56,pers com) Maisiak 1996 (70,pers com) Maisiak 1996 (58, pers com) Treatment counseling** Symptom monitoring Neuberger, 1993 (57) Nurse patient contracts** Practice time and demonstrations Self instruction O’Leary, 1988 (71) Parker, 1988 (59,pers com) Cognitive behavioral group** Attention placebo Parker, 1995 (60,pers com) Stress management course** Patient education course Radojevic, 1992 (61) Behavioral therapy with family support** Behavioral therapy without family support Education family support Dropout %† 54 60 49 61 53 49 53 55 54 60 55 54 53 57 63 57 62 59 58 56 59 48 Age, mean, years 76 43 100 4 66 100 75 83 100 92 83 87 66 69 73 68 11 81 84 80 80 78 Female % 4 4 4 10 ⫹ 5 10 ⫹ 5 4 4 4 10 ⫹ 65 10 ⫹ 65 1 ⫹ 52 1 ⫹ 52 13 ⫹ 26 13 ⫹ 26 5⫹6 5⫹6 1 wk clinic ⫹ 6 1 wk clinic ⫹ 6 10 10 10 12 12 8 26 10 10 10 12 12 8 5–8 16 16 16 5 4–6 4–6 4–6 6 6 4 5 24 6 6 6 4(⫹1) 6 4 5 7 4 4 4 5 6 3 6 6 39 10 Mailed leaﬂets 6 4 6 No. of sessions‡ Program duration, weeks BT BT Inf BT Inf BT Inf BT BT Inf BT C Inf BT BT BT C BT BT C BT BT BT BT BT Inf BT Inf BT Inf BT BT BT Intervention type§ Table 2. Arthritis patient education trials included in the meta-analysis G, Pa, A G, A G, V, Pa F, CP F, CP F ⫹ G, W F ⫹ G, W, V F, W F F, W G, W T T G, W G, W G, W G G G, W T G G G G, W, V, Pa F G, W, Pa G F, W F W G, W F, W, G, W, Pa Education mode¶ Ps Ps Ps Ps, CP Ps, CP In In N N N FE TC US (continued) R, Ps, OT R, Ps, BT R, OT PC, Ps Ps R, N, Ps, OT, SW, Di PC Ps Ps Ps OT OT HP N, OT, PT, Ps RN SP NA LL PT R Program facilitator# 1048 Riemsma et al 21 19 NR 8 0 15 23 5 24 48 238 50 183 68 53 105 42 75 n 55 50 48 55 56 52 52 45 56 Age, mean, years 70 74 79 70 75 80 75 97 62 Female % 20 5 ⫹ 39 5 ⫹ 39 13 13 13 2 8 10 10 12 5 20 5⫹3 5⫹3 ⱖ1 10 10 9 8 10 10 36 5 No. of sessions‡ Program duration, weeks BT C BT BT BT C BT BT BT BT Inf BT Intervention type§ G G HE, W, A G, Pa, W, A G, W G, W G, Pa, W F, W G, Pa, W, A G, W, A F G, A Education mode¶ Ps Ps NA HP BT, N/SW PC R, O, PT, Ps, SW Ps HP HP R PC, Ps Program facilitator# * pers com ⫽ personal communication. † NR ⫽ not reported. ‡ Number of sessions: 4(⫹1) ⫽ 4 consecutive sessions plus 1 additional session if required; 5 ⫹ 6 ⫽ 5 sessions during the ﬁrst period and 6 during the second period; 1 wk clinic⫹6 ⫽ 1 week clinic stay and 6 sessions during the following 12 months; 10 ⫹ 5 ⫽ 10 sessions during the ﬁrst period and 5 during the second period; 5 ⫹ 3 ⫽ 5 sessions during the ﬁrst period and 3 during the second period (see program duration for length of each period). § Intervention type: BT ⫽ behavioral treatment; Inf ⫽ information only; C ⫽ counselling. ¶ Education mode: A ⫽ audiocassettes; CP ⫽ computer program; F ⫽ individual face-to-face contact; G ⫽ group sessions; HE ⫽ home exercises; Pa ⫽ partner participation; T ⫽ telephone contact; V ⫽ video; W ⫽ written materials. # Programme facilitator: BT ⫽ behavioral therapist; CP ⫽ computer program; Di ⫽ dietician; FE ⫽ female experimenters; HP ⫽ health professional; In ⫽ instructors; LL ⫽ lay leader; N ⫽ nurse; O ⫽ orthopedist; OT ⫽ occupational therapist; PC ⫽ patient counselor; Ps ⫽ psychologist/psychiatrist; PT ⫽ physiotherapist; R ⫽ rheumatologist; RN ⫽ rheumatology nurse; SP ⫽ student in psychology; SW ⫽ social worker; TC ⫽ trained and experienced arthritis patient counselor; US ⫽ untrained student; NA ⫽ not applicable. ** Interventions included in the analysis using only one intervention for each trial. Rhodes, 1988 (72) Riemsma, 1999 (pers com) Group education with partner** Group education without partner Rodriguez-Lozano, 1996 (62) Savelkoul, 2000 (47, pers com) Coping intervention group** Mutual support Scholten, 1999 (63) Sharpe, 2001 (94) Shearn, 1985 (73) Self management** Mutual support Stenstrom, 1994 (65,pers com) Taal, 1993 (66,93,pers com) Author, year (ref.)* Dropout %† Table 2. Arthritis patient education trials included in the meta-analysis (Continued) RA Patient Education: Review 1049 1050 Riemsma et al (Hewlett S: personal communication; Riemsma R: personal communication; Savelkoul M: personal communication). Subsequently, 2 of these have been published (46,47). In total, 50 studies are suitable for inclusion in this review. For these 50 studies, we found complete data on 24 studies (refs. 40, 44, 47– 66, and Hewlett S: personal communication; Riemsma R: personal communication) and 7 other studies gave incomplete data (67–73), the authors of these studies did not reply to our requests for more information. Therefore, 31 studies were included in this review and these trials are summarized in Table 2. Of the remaining 19 studies that were suitable for inclusion, we have no data yet for 2 studies (74,75), but the authors replied that the information requested will be sent as soon as possible. On 8 studies, we found no data relating to the outcomes under investigation in the report (76 – 83), and the authors have not replied to our requests for more information. Finally, the relevant data are not available for 9 studies according to the authors (43, 84 –91). We received additional data through personal communication for 18 of the 31 included studies (refs. 92 and 93, and Barlow J: personal communication; Bradley L: personal communication; Brus H: personal communication; McEvoy-DeVellis B: personal communication; Fries J: personal communication; Stenstrom C: personal communication; Geissner E: personal communication; Goeppinger J: personal communication; Helewa A: personal communication; Kraaimaat F: personal communication; Lindroth Y: personal communication; Lorig K: personal communication; Maisiak R: personal communication; Oermann M: personal communication; Smarr K: personal communication; Hewett J: personal communication; Taal E: personal communication). Figure 1. Flow diagram of search strategy. RCT ⫽ randomized controlled trial; RA ⫽ rheumatoid arthritis. Figure 1). For 229 references, the full report was retrieved. Eighty-six publications were not RCTs; in 32 publications, the subjects involved were not RA patients; in 29 publications, the intervention did not include an educational component; 11 publications involved secondary analysis; 8 publications did not include a nonintervention control group; 2 publications presented only preliminary results; in 1, the intervention was education for health professionals; 2 turned out to be conference abstracts (we have not been able to ﬁnd more information about these 2 studies); and 1 publication could not be retrieved (32). One publication is awaiting assessment because we need more information from the authors (33). In 6 studies, the outcome variables did not include any of the selected outcome measures (34 –39). The remaining 50 publications are suitable for inclusion in this review. Among the 50 references, we found 3 studies with double publications (40 – 45); therefore, 47 trials were suitable for inclusion in the review. We also searched for unpublished studies, and were able to retrieve data from 3 additional trials that have recently been completed Publication bias We have drawn funnel plots showing sample sizes versus effect sizes for the 2 outcomes that were assessed most often: pain and functional disability (see Figures 2 and 3). The distribution appears symmetric and small studies with comparatively negative outcomes are as frequently reported as those with comparatively more positive outcomes. Methodologic quality of the studies The quality of the 31 included studies was assessed (Table 3). For the studies on which we had 2 publications or more, we used all available information from all publications to assess the quality of each study. If it was possible to retrieve additional information from the authors concerning the quality of the study, this was incorporated in the score as well. If it was not possible to retrieve additional information, the quality score reported reﬂects the quality of the study as it is reported in the article. This may not reﬂect the true quality of the study. The quality of studies on average was not very high. The mean score from all 31 studies was 3.26 (of a possible 8). Of all 31 randomized controlled trials, only 8 received the full 2 points for description of the randomization proce- RA Patient Education: Review 1051 Figure 2. Funnel plot showing sample sizes versus effect sizes for scores on pain. SMD ⫽ standardized mean difference. Figure 3. Funnel plot showing sample sizes versus effect sizes for scores on functional disability. SMD ⫽ standardized mean difference. 1052 Riemsma et al Table 3. Methodologic quality of included studies Study* Selection Performance Attrition Blinding Total score Appelbaum, 1988 (67) Barlow, 1998 (46, 41, pers com) Barlow, 2000 (48) Bell, 1998 (49) Brus, 1998 (50, 92, pers com) Geissner, 1994 (51, pers com) Hammond, 1999 (52) Helewa, 1991 (68, pers com) Helliwell, 1999 (53) Hewlett, 1998 (pers com) Hill, 2001 (54) Huiskes, 1991 (44, 45, pers com) Kaplan, 1981 (69) Leibing, 1999 (55) Lindroth, 1997 (56, pers com) Maisiak, 1996 (70, pers com) Maisiak, 1996 (58, pers com) Neuberger, 1993 (57) O’Leary, 1988 (71) Parker, 1988 (59, pers com) Parker, 1995 (60, pers com) Radejovic, 1992 (61) Rhodes, 1988 (72) Riemsma, 1999 (pers com) Rodriguez-Lozano, 1996 (62) Savelkoul, 2000 (47, pers com) Scholten, 1999 (63) Sharpe, 2001 (94) Shearn, 1985 (73) Stenstrom, 1994 (65, pers com) Taal, 1993 (66, 93, pers com) Total of all 31 studies 0 0 2 2 0 0 2 1 2 2 2 0 0 0 0 1 0 0 0 1 0 0 0 0 0 2 1 2 0 0 1 21 0 0 0 2 1 2 1 0 2 1 2 0 0 2 1 0 1 0 0 0 1 1 0 0 1 1 1 0 0 0 0 20 0 0 2 2 1 0 1 2 2 2 0 1 1 1 2 2 1 0 1 1 1 1 0 2 1 2 1 2 0 1 0 33 0 0 1 1 2 0 1 1 1 1 2 1 1 1 0 1 2 0 1 0 1 0 1 1 0 2 2 1 0 1 1 27 0 0 5 7 4 2 5 4 7 6 6 2 2 4 3 4 4 0 2 2 3 2 1 3 2 7 5 5 0 2 2 101 * pers com ⫽ personal communication. dure, these were all published in the last 5 years. Only 5 other studies received 1 point for randomization, making randomization together with cointerventions the 2 least well-reported elements of the 4 quality items, with a mean of 0.68 and 0.65 (of a possible score of 2) for randomization and cointerventions, respectively. Most studies scored higher on attrition; with a mean of 1.06 (of 2), this item showed the highest scores of the quality items. A separate analysis was undertaken including only the 17 studies with a quality score of 3 or higher and on which we have data (refs. 47–50, 52–56, 58, 60, 63, 68, 70, and 94, and Hewlett S: personal communication; Riemsma R: personal communication) to check whether the quality of studies seriously inﬂuences the results. Effectiveness of patient education We found signiﬁcant effects of patient education at ﬁrst followup, i.e., immediately after the intervention (depending on the duration of the intervention, this ranged from 3 weeks to 9 months after baseline assessment), for scores on functional disability (SMD ⫽ – 0.17; 95% CI ⫺0.25, ⫺0.09; Z ⫽ 3.97; P ⫽ 0.00007; N ⫽ 2,275) involving 25 trials and 37 comparisons; joint counts (SMD ⫽ ⫺0.13; 95% CI ⫺0.24, ⫺0.01; Z ⫽ 2.14; P ⫽ 0.03; N ⫽ 1,158) involving 15 trials and 23 comparisons; patient global assessment (SMD ⫽ ⫺0.28; 95% CI ⫺0.49, ⫺0.07; Z ⫽ 2.65; P ⫽ 0.008; N ⫽ 358) involving 4 trials and 6 comparisons; psychological status (SMD ⫽ ⫺0.16; 95% CI ⫺0.28, ⫺0.04; Z ⫽ 2.66; P ⫽ 0.008; N ⫽ 1,138) involving 8 trials and 18 comparisons; and depression (SMD ⫽ – 0.14; 95% CI ⫺0.23, ⫺0.05; Z ⫽ 2.90; P ⫽ 0.004; N ⫽ 1,770) involving 18 trials and 29 comparisons. The number of comparisons is more than the number of trials for each outcome because a number of trials included ⬎1 intervention arm. Physician global assessment was not assessed in any of the included studies. Of anxiety, pain, and disease activity, none showed signiﬁcant effects. A trend was found in favor of patient education for pain; however (SMD ⫽ – 0.08; 95% CI – 0.16, 0.00; Z ⫽ 1.86; P ⫽ 0.06; N ⫽ 2,219), this involved 24 trials and 37 comparisons (see Figure 4). Heterogeneity was not signiﬁcant for all measures; therefore, in all cases the ﬁxed effect model was used. RA Patient Education: Review 1053 Figure 4. Forest plots of main results (pain, functional disability, joint count, patient global assessment, psychological status, anxiety, depression, and disease activity) at ﬁrst followup. Pooled effect sizes (center points) and 95% conﬁdence intervals (horizontal bars) for A, all patient education interventions, B, information-only interventions, C, counseling interventions, and D, behavioral treatment interventions. Exp ⫽ experimental; SMD ⫽ standardized mean difference; 95% CI ⫽ 95% conﬁdence interval. *The random effects model was used because heterogeneity was signiﬁcant (P ⬍ 0.05). 1054 At ﬁnal followup, no signiﬁcant effects of patient education were found (see Figure 5). A trend was seen in favor of patient education, however, for scores on functional disability (SMD ⫽ – 0.09; 95% CI – 0.20, 0.02; Z ⫽ 1.66; P ⫽ 0.10; N ⫽ 1,308) involving 15 trials and 23 comparisons. Analyses by type of intervention We found 9 studies that included 9 comparisons (and 687 patients) of an intervention aimed at information only versus control. The pooled analyses showed no signiﬁcant effects at ﬁrst followup for scores on pain, functional disability, joint counts, patient global assessment, anxiety, depression, and disease activity. However, scores on pain and psychological status showed a trend in favor of the information-only group: pain scores involved 8 trials and 8 comparisons (SMD ⫽ – 0.15; 95% CI – 0.32, 0.02; Z ⫽ 1.71; P ⫽ 0.09; N ⫽ 524) and psychological status scores involved 3 trials and 3 comparisons (SMD ⫽ – 0.24; 95% CI – 0.48, 0.01; Z ⫽ 1.88; P ⫽ 0.06; N ⫽ 257) (Figure 4). At ﬁnal followup, no signiﬁcant effects of information only were found (see Figure 5). We found 5 studies that included 5 comparisons (and 430 patients) of an intervention aimed at counseling versus control. The pooled analyses showed no signiﬁcant effects for scores on pain, functional disability, joint counts, patient global assessment, anxiety, depression, or disease activity (Figure 4). However, a trend was found in favor of the intervention for scores on psychological status (SMD ⫽ – 0.25; 95% CI – 0.52, 0.03; Z ⫽ 1.74; P ⫽ 0.08; N ⫽ 203) involving only 2 trials and 2 comparisons. At ﬁnal followup, no signiﬁcant effects of counseling were found (see Figure 5). We found 24 studies that included 31 comparisons (and 2,493 patients) of an intervention aimed at behavioral treatment versus control. The pooled analyses showed signiﬁcant effects for scores on functional disability (SMD ⫽ – 0.23; 95% CI – 0.36, – 0.10; Z ⫽ 3.52; P ⫽ 0.0004; N ⫽ 1,532) involving 21 trials and 27 comparisons; patient global assessment (SMD ⫽ – 0.30; 95% CI – 0.55, – 0.04; Z ⫽ 2.25; P ⫽ 0.02; N ⫽ 236) involving 4 trials and 4 comparisons; and depression (SMD ⫽ – 0.14; 95% CI – 0.25, – 0.04; Z ⫽ 2.63; P ⫽ 0.009; N ⫽ 1,350) involving 16 trials and 21 comparisons (Figure 4). Heterogeneity was found to be signiﬁcant for scores on functional disability, therefore the random effects model was used. A trend in favor of the intervention was found for scores on pain (SMD ⫽ – 0.09; 95% CI – 0.19, 0.02; Z ⫽ 1.67; P ⫽ 0.10; N ⫽ 1,453) involving 19 trials and 26 comparisons. At ﬁnal followup, no signiﬁcant effects of behavioral treatment were found (see Figure 5). However, trends in favor of behavioral treatment were found for scores on functional disability (SMD ⫽ ⫺0.10; 95% CI – 0.23, 0.02; Z ⫽ 1.64; P ⫽ 0.10; N ⫽ 1,003) involving 14 trials and 18 comparisons and depression (SMD ⫽ – 0.12; 95% CI – 0.25, 0.01; Z ⫽ 1.80; P ⫽ 0.07; N ⫽ 911) involving 12 trials and 16 comparisons. Sensitivity analyses results Looking only at higher-quality studies and large studies, we found signiﬁcant effects favoring patient education for Riemsma et al scores on functional disability, patient global assessment, psychological status, and depression. These results conﬁrm the results from the main analyses and show that the effects found are quite robust. Three other sensitivity analyses were performed: 1) using only 1 instrument to measure each outcome to reduce heterogeneity in outcome measures; 2) including only 1 educational intervention, thus not overestimating the results of the control condition in studies with ⬎1 experimental condition; and 3) comparing trials at a ﬁxed time point (after 2– 4 months) to reduce heterogeneity in outcome assessment. The results of these 3 sensitivity analyses showed similar effects for scores on functional disability, patient global assessment, and depression, which again show that the results are quite robust. However, standardized effect sizes ranged from – 0.11 to – 0.32, indicating that the effect is very small. The results of the sensitivity analyses are presented in Table 4. Discussion In this review we examined the effectiveness of patient education interventions on health status (disease activity, patient global assessment, joint counts, pain, functional disability, and psychological wellbeing) in patients with RA. We found a small beneﬁcial effect of patient education at ﬁrst followup for functional disability (SMD ⫽ – 0.17; 95% CI – 0.25, – 0.09), joint counts (SMD ⫽ – 0.13; 95% CI – 0.24, – 0.01), patient global assessment (SMD ⫽ – 0.28; 95% CI – 0.49, – 0.07), psychological status (SMD ⫽ – 0.16; 95% CI – 0.28, – 0.04), and depression (SMD ⫽ – 0.14; 95% CI – 0.23, – 0.05). At ﬁnal followup (3–14 months), there were no signiﬁcant beneﬁts. These results suggest signiﬁcant effects of patient education for scores on functional disability, and moreover these effects were quite robust, as most sensitivity analyses showed signiﬁcant effects. However, standardized effect sizes ranged from – 0.11 to – 0.21, indicating that the effect is very small. An SMD of 0.17 in favor of patient education can be translated into an improvement on the Stanford Health Assessment Questionnaire (HAQ; range 0 –3) of 0.10, assuming the mean score in the control group remains the same and a standard deviation of 0.60 in both groups. Assuming a start level of 1.00 on the HAQ, an SMD of – 0.17 translates to a 10% (95% CI 5, 16) improvement on the HAQ. Similarly, compared with no intervention, patient education produced a 4% (95% CI 0, 9) decrease in pain visual analog scale; 9% (95% CI 1, 16) improvement on the Ritchie Articular Index; 12% (95% CI 3, 22) improvement on the Arthritis Impact Measurement Scales (AIMS) arthritis impact subscale; 5% (95% CI 1, 9) improvement on the AIMS2 affect subscale; and 12% (95% CI 4, 19) improvement on depression scores. In summary, the statistically signiﬁcant beneﬁts of patient education at ﬁrst followup are modest (5–12%). The most important beneﬁt was for functional disability, with an effect size of – 0.17. This compares with the following effect sizes for disease-modifying drug treatments: – 0.09 (95% CI – 0.45, 0.27) for antimalarials (95); – 0.19 (95% CI – 0.39, 0.02) for auranoﬁn (96); – 0.29 (95% CI – 0.77, 0.19) RA Patient Education: Review 1055 Figure 5. Forest plots of main results (pain, functional disability, joint count, patient global assessment, psychological status, anxiety, depression, and disease activity) at ﬁnal followup. Pooled effect sizes (center points) and 95% conﬁdence intervals (horizontal bars) for A, all patient education interventions, B, information-only interventions, C, counseling interventions, and D, behavioral treatment interventions. For deﬁnitions of abbreviations, see Figure 4. 1056 Riemsma et al Table 4. Summary of statistically signiﬁcant results (P < 0.05) or trends (P < 0.10) for SMD and WMD for main study and various sensitivity analyses* Functional Joint Patient global Psychological Disease disability counts assessment status Anxiety Depression activity Measure Pain SMD WMD ⫺0.08† ⫺0.38‡ ⫺0.17‡ ⫺0.19† ⫺0.13‡ ⫺1.79‡ Sensitivity analysis: one experimental condition Sensitivity analysis: high quality studies Sensitivity analysis: large studies Sensitivity analysis: 2–4 months Sensitivity analysis: comparable interventions: information Sensitivity analysis: comparable interventions: counseling Sensitivity analysis: comparable interventions: behavioral Final followup Main analysis Sensitivity analysis: one instrument SMD VAS ⫺0.10† HAQ ⫺0.23‡ RAI ⫺0.15‡ Sensitivity analysis: one experimental condition Sensitivity analysis: high quality studies Sensitivity analysis: large studies Sensitivity analysis: 2–4 months Sensitivity analysis: comparable interventions: information Sensitivity analysis: comparable interventions: counseling Sensitivity analysis: comparable interventions: behavioral SMD First followup Main analysis Sensitivity analysis: one instrument ⫺0.20‡ SMD SMD SMD SMD ⫺0.10† ⫺0.15† ⫺0.15‡ ⫺0.14‡ ⫺0.17† ⫺0.28‡ ⫺0.16‡ ⫺0.45† ⫺0.14‡ ⫺0.62‡ ⫺0.30‡ AIMS ⫺0.16† HAD ⫺0.18‡ ⫺0.32‡ ⫺0.18‡ ⫺0.21‡ ⫺0.31‡ ⫺0.22† ⫺0.13† ⫺0.13‡ ⫺0.11‡ ⫺0.25† SMD SMD ⫺0.24† ⫺0.09† ⫺0.23 ⫺0.09† ⫺0.11‡ SMD WMD ⫺0.13† HAQ ⫺0.12† ⫺0.30‡ ⫺0.14‡ ⫺1.55‡ RAI ⫺0.14† SMD SMD SMD SMD SMD SMD ⫺0.12† ⫺0.12† * SMD ⫽ standardized mean difference; WMD ⫽ weighted mean difference; VAS ⫽ visual analog scale; HAQ ⫽ Health Assessment Questionnaire; RAI ⫽ Ritchie Articular Index; AIMS ⫽ Arthritis Impact Measurement Scales; HAD ⫽ Hospital Anxiety and Depression Scale. † P ⬍ 0.10. ‡ P ⬍ 0.05 for penicillamine (97); – 0.31 (95% CI –1.06, 0.44) for azathioprine (98); – 0.78 (95% CI –1.10, – 0.47) for cyclosporine (99); and –1.48 (95% CI –1.82, –1.14) for methotrexate (100). Glucocorticoids, when given in addition to diseasemodifying drugs, have an additional effect size of – 0.57 (95% CI – 0.92, – 0.22) (101). Our results suggest signiﬁcant effects of patient education for scores on functional disability. It is important to keep in mind that this outcome refers to the functional ability to perform certain tasks. Patient education interventions can enable patients to perform these tasks more easily and with less pain through education and training. Unfortunately, the effects are short lived. Therefore, strategies to preserve the effects over time need to be explored. In evaluating clinical effects of patient education, it must be taken into account that patient education was provided in addition to standard medical care so the effects of patient education are always supplementary to the beneﬁts of standard medical care. Also, it needs to be taken into account that in all these studies, patients were invited to take part in an experimental procedure and randomly allocated to intervention or control groups. This contrasts with routine clinical practice in which patients may be more likely to select themselves for education sessions. As reported, the quality of studies was not very high. Most studies (18 of 31) failed to score any points for the way randomization and concealment of allocation were reported. Therefore, the ﬁndings are subject to selection bias. Sixteen of 31 studies did not score any points for performance bias. Studies have shown that contamination (provision of the intervention to the control group) and cointervention (provision of unintended additional care to either comparison group) can affect study results (102,103). Although the quality scores reported in this review reﬂect the quality of the study as it is reported in the article and may not reﬂect the true quality of the studies, these limitations have to be taken into account when interpreting the ﬁndings of this review. RA Patient Education: Review Patient education has been advocated in arthritis for information provision itself, and for its therapeutic potential (4). In practice, many patient education programs have not been disease speciﬁc and there has been the assumption that all beneﬁts would be generic. This analysis raises doubts over the achievement of meaningful beneﬁts in patients with RA speciﬁcally, who are recruited via invitation to participate usually through a hospital outpatient department. Future research should be disease speciﬁc and should seek to identify which patient characteristics are relevant to beneﬁcial outcomes from educational intervention, and which components of patient education programs are effective (104). In this review, we performed separate analyses for 3 types of interventions, with behavioral programs showing the best results compared with no intervention. However, behavioral programs themselves show great variation in content (e.g., physical exercise, relaxation, pain coping, biofeedback), methods (e.g., group versus individual programs or brochures, homework assignments, contracting), and organization (professional or lay leaders, numbers and lengths of sessions, etc.). Further investigation to assess what components are most effective is warranted. A review of educational beneﬁts in other speciﬁc forms of arthritis, particularly OA, would be worthwhile as well. In this review, we assessed the effects of patient education on core outcomes as deﬁned by the OMERACT group for clinical trials of medical therapies. It should be considered whether these outcomes are appropriate for patient education interventions and other health care interventions, such as occupational or physical therapy. Other measures that might be more appropriate include fatigue and social participation in life situations as deﬁned in the new international classiﬁcation of functioning, functional disability and health of the WHO (105). Many patient education programs are aimed at teaching patients how to cope with stress and pain or how to manage consequences of their disease in daily life. Measuring change in social participation, e.g., fulﬁllment of social or work roles, is often omitted from education studies. Trials of education should include as outcome measures the WHO-ILAR core set as deﬁned by the OMERACT group (12), together with measures of psychological status, such as the Hospital Anxiety and Depression Scale, AIMS2 affect subscale, or Center for Epidemiologic Studies Depression Scale, and measures of social participation. ACKNOWLEDGMENTS The authors would like to thank the Cochrane Musculoskeletal Group editorial team and the Dutch Cochrane Centre for their support in the preparation of this systematic review. 1057 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. REFERENCES 1. Akil M, Amos R. ABC of rheumatology: rheumatoid arthrits—I. Clinical features and diagnosis. BMJ 1995;310:587– 90. 2. Kirwan J. Patient education in rheumatoid arthritis. Curr Opin Rheumatol 1990;2:336 –9. 3. Taal E, Rasker J, Wiegman O. Patient education and self- 27. 28. 29. management in the rheumatic diseases: a self-efﬁcacy approach. Arthritis Care Res 1996;9:229 –38. Tucker M, Kirwan J. Does patient education in rheumatoid arthritis have therapeutic potential? Ann Rheum Dis 1991; 50:422– 8. Hirano P, Laurent D, Lorig K. Arthritis patient-education studies, 1987–1991: a review of the literature. Patient Educ Couns 1994;24:9 –54. Taal E, Rasker J, Wiegman O. Group education for rheumatoid arthritis patients. Semin Arthritis Rheum 1997;26:805– 16. Lorig K. Common sense patient education. Ivanhoe, Victoria, Australia: Fraser Publications; 1992. Mazucca S. Does patient education in chronic disease have therapeutic value? J Chron Dis 1982;35:521–9. Hawley D. Psycho-educational interventions in the treatment of arthritis. Bailliere’s Clin Rheumatol 1995;9:803–23. Superio-Cabuslay E, Ward M, Lorig K. Patient education interventions in osteoarthritis and rheumatoid arthritis: a meta-analytic comparison with non-steroidal antiinﬂammatory drug treatment. Arthritis Care Res 1996;9:292–301. Cooper H, Booth K, Fear S, Gill G. Chronic disease patient education: lessons from meta-analyses. Patient Educ Couns 2001;44:107–17. Tugwell P, Boers M, for the OMERACT Committee. Developing consensus on preliminary core efﬁcacy endpoints for rheumatoid arthritis clinical trials. J Rheumatol 1993;20: 555– 6. Brooks P, Hochberg M. Outcome measures and classiﬁcation criteria for the rheumatic diseases: a compilation of data from OMERACT (Outcome Measures for Arthritis Clinical Trials), ILAR (International League of Associations for Rheumatology), regional leagues and other groups. Rheumatology 2001;40:896 –906. Bradley L. Behavioral interventions for managing chronic pain. Bull Rheum Dis 1994;43:2–5. DeVellis R, Blalock S. Psychological and educational interventions to reduce arthritis disability. Baillieres Clin Rheumatol 1993;7:397– 416. Glazier R. Managing early presentation of rheumatoid arthritis: systematic overview. Can Fam Phys 1996;42:913– 22. Goeppinger J, Lorig K. Interventions to reduce the impact of chronic disease: community-based arthritis patient education. Annu Rev Nurs Res 1997;15:101–22. Hill J. Patient education in rheumatic disease. Nurs Stand 1995;9:25– 8. Keefe F, Caldwell D. Cognitive behavioral control of arthritis pain. Med Clin North Am 1997;81:277–90. Langer H, Mattussek S. Patient education in rheumatology. Wien Med Wochenschr 1990;140:349 –351. Lorig K, Konkol L, Gonzalez V. Arthritis patient education: a review of the literature. Patient Educ Couns 1987;10:207–52. Mullen P, Laville E, Biddle A, Lorig K. Efﬁcacy of psychoeducational interventions on pain, depression, and disability in people with arthritis: a meta-analysis. J Rheumatol 1987;14:33–9. Ytterberg S, Mahowald M, Krug H. Exercise for arthritis. Baillieres Clin Rheumatol 1994;8:161– 89. Gregoire G, Derderian F, LeLorier J. Selecting the language of the publication included in a meta-analysis: is there a tower of Babel bias? J Clin Epidemiol 1995;48:158 – 63. Egger M, Davey Smith G. Meta-analysis: bias in location and selection of studies. BMJ 1998;316:61– 6. Jadad A, Moore A, Carroll D, Jenkinson C, Reynolds D, Gavaghan D, et al. Assessing the quality of reports of randomized clinical trials: Is blinding necessary? Control Clin Trials 1996;17:1–12. Clarke M, Oxman A. Cochrane reviewers handbook 4.0 [updated July 1999]. Oxford: Update Software; 2000. Review Manager (RevMan) [computer program]. Version 4.2 for Windows. Oxford: The Cochrane Collaboration; 2003. Laird N, Mosteller F. Some statistical methods for combining 1058 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. Riemsma et al experimental results. Int J Tech Assess Health Care 1990;6: 5–30. Felson D. Bias in meta-analytic research. J Clin Epidemiol 1992;45:885–92. Khan K, Riet GT, Glanville J, Sowden A, Kleijnen J. Undertaking systematic reviews of research on effectiveness: CRD’s guidance for those carrying out or commissioning reviews. 2nd ed. York, UK: Centre for Reviews and Dissemination, University of York; 2001. Sebro B, Dubravica M, Jajicz. Spontaneous use of active and passive coping strategies for pain in patients with rheumatoid arthritis. Reumatizam 1993;40:1– 4. Newman A. Self-help care in older African Americans with arthritis. Geriatr Nurs 2001;22:135– 8. Darmawan J, Muirden K, Wigley R, Valkenburg H. Arthritis community education by leather puppet (wayang kulit) shadow play in rural Indonesia (Java). Rheumatol Int 1992; 12:97–101. Feinberg J. Effect of the arthritis health professional on compliance with use of resting hand splints by patients with rheumatoid arthritis. Arthritis Care Res 1992;5:17–23. van Deusen J, Harlowe D. The efﬁcacy of the ROM dance program for adults with rheumatoid arthritis. Am J Occup Ther 1987;41:90 –5. Young L, Bradley L, Turner R. Decreases in health care resource utilization in patients with rheumatoid arthritis following a cognitive behavioral intervention. Biofeedback Self Regul 1995;20:259 – 68. Pope J, Stevens A, Rooks M. A randomized double blind trial of verbal NSAID education compared to verbal and written education. J Rheumatol 1998;25:771–5. Linne A, Lennart H. The effects on knowledge of the systematic education of patients with joint diseases treated with NSAIDs and diuretics. Patient Educ Couns 2001;42:165–74. Barlow J, Pennington D, Bishop P. Patient education leaﬂets for people with rheumatoid arthritis: a controlled study. Psychol Health Med 1997;2:221–35. Barlow J, Wright C. Knowledge in patients with rheumatoid arthritis: a longer term follow-up of a randomized controlled study of patient education leaﬂets. Br J Rheumatol 1998;37: 373– 6. Bradley L, Young L, Anderson K, Turner R, Agudelo C, McDaniel L, et al. Effects of cognitive-behavioral therapy on rheumatoid arthritis pain behavior: one-year follow-up. In: Dubner R, Gebhart G, Bond M, editors. Proceedings of the Vth world congress on pain. New York: Elsevier; 1988. p. 310 –14. Bradley L, Young L, Anderson K, Turner R, Agudelo C, McDaniel L, et al. Effects of psychological therapy on pain behavior of rheumatoid arthritis patients: treatment outcome and six-month follow-up. Arthritis Rheum 1987;30:1105–14. Huiskes C, Kraaimaat F, Brons M, Bijlsma J. The effect of cognitive behavior therapy and occupational therapy in patients with rheumatoid arthritis. Gedragstherapie 1991;24: 253– 68. Kraaimaat F, Brons M, Geenen R, Bijlsma J. The effect of cognitive behavior therapy in patients with rheumatoid arthritis. Behav Res Ther 1995;33:487–95. Riemsma R, Taal E, Rasker J. Group education for patients with rheumatoid arthritis and their partners. Arthritis Rheum 2003;49:556 – 66. Savelkoul M, de Witte L, Candel M, van der Tempel H, van der Borne B. Effects of a coping intervention on patients with rheumatic diseases: results of a randomized controlled trial. Arthritis Rheum 2001;45:69 –76. Barlow JH, Turner AP, Wright CC. A randomized controlled study of the Arthritis Self-Management Programme in the UK. Health Educ Res 2000;15:665– 80. Bell M, Lineker S, Wilkins A, Goldsmith C, Badley E. A randomized controlled trial to evaluate the efﬁcacy of community based physical therapy in the treatment of people with rheumatoid arthritis. J Rheumatol 1998;25:231–7. Brus H, van de Laar M, Taal E, Rasker J, Wiegman O. Effects of patient education on compliance with basic treatment 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. regimens and health in recent onset active rheumatoid arthritis. Ann Rheum Dis 1998;57:146 –51. Geissner E, Jungnitsch G, Schmitz J. Psychological treatment approaches in pain: a comparative study of therapies in patients with chronic polyarthritis. Z Klin Psychol Psychopathol Psychother 1994;42:319 –38. Hammond A, Lincoln N, Sutcliffe L. A crossover trial evaluating an educational-behavioural joint protection programme for people with rheumatoid arthritis. Patient Educ Couns 1999;37:19 –32. Helliwell PS, O’Hara M, Holdsworth J, Hesselden A, King T, Evans P. A 12-month randomized controlled trial of patient education on radiographic changes and quality of life in early rheumatoid arthritis. Rheumatology (Oxford) 1999;38: 303– 8. Hill J, Bird H, Johnson S. Effect of patient education on adherence to drug treatment for rheumatoid arthritis: a randomized controlled trial. Ann Rheum Dis 2001;60:869 –75. Leibing E, Pﬁngsten M, Bartmann U, Rueger U, Schuessler G. Cognitive-behavioral treatment in unselected rheumatoid arthritis outpatients. Clin J Pain 1999;15:58 – 66. Lindroth Y, Brattstrom M, Bellman I, Ekestaf G, Olofsson Y, Strombeck B, et al. A problem-based education program for patients with rheumatoid arthritis: evaluation after three and twelve months. Arthritis Care Res 1997;10:325–32. Neuberger G, Smith K, Black S, Hassasein R. Promoting self-care in clients with arthritis. Arthritis Care Res 1993;6: 141– 8. Maisiak R, Austin J, Heck L. Health outcomes of two telephone interventions for patients with rheumatoid arthritis or osteoarthritis. Arthritis Rheum 1996;39:1391–9. Parker J, Frank R, Beck N, Smarr K, Buescher K, Phillips L, et al. Pain management in rheumatoid arthritis patients: a cognitive-behavioral approach. Arthritis Rheum 1988;31:593– 601. Parker J, Smarr K, Buckelew S, Stucky-Ropp R, Hewett J, Johnson J, et al. Effects of stress management on clinical outcomes in rheumatoid arthritis. Arthritis Rheum 1995;38: 1807–18. Radojevic V, Nicassio P, Weisman M. Behavioral intervention with and without family support for rheumatoid arthritis. Behav Ther 1992;23:13–30. Rodriguez-Lozano C, Bilbao A, Naranjo A, Ojeda S, Francisco F. Patient education in rheumatoid arthritis: its inﬂuence in the disease outcome. Rev Esp Reumatol 1996;23: 40 – 8. Scholten C, Brodowicz T, Graninger W, Gardavsky I, Pils K, Pesau B, et al. Persistent functional and social beneﬁt 5 years after a multidisciplinary arthritis training program. Arch Phys Med Rehabil 1999;80:1282–7. Sharpe L, Sensky T, Timberlake N, Ryan B, Allard S. Longterm efﬁcacy of a cognitive behavioural treatment from a randomized controlled trial for patients recently diagnosed with rheumatoid arthritis. Rheumatology (Oxford) 2003;42: 435– 41. Stenstrom C. Home exercise in rheumatoid arthritis functional class II: goal setting versus pain attention. J Rheumatol 1994;21:627–34. Taal E, Riemsma R, Brus H, Seydel E, Rasker J, Wiegman O. Group education for patients with rheumatoid arthritis. Patient Educ Couns 1993;20:177– 87. Appelbaum K, Blanchard E, Hickling E, Alfonso M. Cognitive behavioral treatment of a veteran population with moderate to severe rheumatoid arthritis. Behav Ther 1988;19: 489 –502. Helewa A, Goldsmith C, Lee P, Bombardier C, Hanes B, Smythe H, et al. Effects of occupational therapy home service on patients with rheumatoid arthritis. Lancet 1991;337: 1453– 6. Kaplan S, Kozin F. A controlled study of group counseling in rheumatoid arthritis. J Rheumatol 1981;8:91–9. Maisiak R, Austin J, West S, Heck L. The effect of personcentered counseling on the psychological status of persons with systemic lupus erythematosus or rheumatoid arthritis: RA Patient Education: Review 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. a randomized, controlled trial. Arthritis Care Res 1996;9: 60 – 6. O’Leary A, Shoor S, Lorig K, Holman H. A cognitive-behavioral treatment for rheumatoid arthritis. Health Psychol 1988;7:527– 44. Rhodes JT, Foard T, Dickstein L. Professional peer counseling in the management of rheumatoid arthritis: a clinical trial. In: Ahmed PI, editors. Coping with arthritis. Springﬁeld (IL): Charles C. Thomas; 1988. p. 73–106. Shearn M, Fireman B. Stress management and mutual support groups in rheumatoid arthritis. Am J Med 1985;78: 771–5. Cohen J, van Houten-Sauter S, DeVellis R, McEvoy-DeVellis B. Evaluation of arthritis self-management courses led by laypersons and by professionals. Arthritis Rheum 1986;29: 388 –93. Daltroy LH, Morlino CI, Eaton HM, Poss R, Liang MH. Preoperative education for total hip and knee replacement patients. Arthritis Care Res 1998;11:469 –78. Balmer D. The CARE project: the evaluation of group counseling as a therapeutic intervention for patients with rheumatoid arthritis. Br J Guidance Couns 1989;17:304 –16. Branch V, Lipsky K, Nieman T, Lipsky P. Positive impact of an intervention by arthritis patient educators on knowledge and satisfaction of patients in a rheumatology practice. Arthritis Care Res 1999;12:370 –5. Cziske R, Jaeckel W, Jacobi E. Effects of a short pain-coping training programme for RA-patients during rehabilitation. Z Klin Psychol 1987;16:115–23. Lorig KR, Sobel DS, Stewart AL, Brown BW Jr, Bandura A, Ritter P, et al. Evidence suggesting that a chronic disease self-management program can improve health status while reducing hospitalization. Med Care 1999;37:5–14. Lorig K, Gonzalez VM, Ritter P. Community-based Spanish language arthritis education program: a randomized trial. Med Care 1999;37:957– 63. Maggs F, Jubb R, Kemm J. Single-blind randomized controlled trial of an educational booklet for patients with chronic arthritis. Br J Rheumatol 1996;35:775–7. Strauss G, Spiegel J, Daniels M, Spiegel T, Landsverk J, Roy-Byrne P, et al. Group therapies for rheumatoid arthritis: a controlled study of two approaches. Arthritis Rheum 1986; 29:1203–9. Wetstone S, Sheenan T, Votaw R, Peterson M, Rothﬁeld N. Evaluation of a computer based education lesson for patients with rheumatoid arthritis. J Rheumatol 1985;12:907–12. Fries J, Carey C, McShane D. Patient education in arthritis: randomized controlled trial of a mail-delivered program. J Rheumatol 1997;24:1378 – 83. Goeppinger J, Arthur M, Baglioni A, Brunk S, Brunner C. A reexamination of the effectiveness of self-care education for persons with arthritis. Arthritis Rheum 1989;32:706 –16. Lorig K, Lubeck D, Kraines R, Seleznick M, Holman H. Outcomes of self-help education for patients with arthritis. Arthritis Rheum 1985;28:680 –5. Lorig K, Feigenbaum P, Regan C, Ung E, Chastain R, Holman H. A comparison of lay-taught and professional-taught arthritis self-management courses. J Rheumatol 1986;13: 763–7. 1059 88. Lorig K, Seleznick M, Lubeck D, Ung E, Chastain R, Holman H. The beneﬁcial outcomes of the arthritis self-management course are not adequately explained by behavior change. Arthritis Rheum 1989;32:91–5. 89. McEvoy-DeVellis B, Blalock S, Hahn P, DeVellis R, Hochbaum G. Evaluation of a problem-solving intervention for patients with arthritis. Patient Educ Couns 1988;11:29 – 42. 90. Oermann M, Doyle T, Clark L, Rivers C, Rose V. Effectiveness of self-instruction for arthritis patient education. Patient Educ Couns 1986;8:245–54. 91. Parker J, Singsen B, Hewett J, Walker S, Hazelwood S, Hall P, et al. Educating patients with rheumatoid arthritis: a prospective analysis. Arch Phys Med Rehabil 1984;65:771– 4. 92. Brus H. Compliance and patient education in rheumatoid arthritis [thesis]. Enschede, the Netherlands: Univ. of Twente; 1997. 93. Taal E, Seydel E, Riemsma R, Brus H, Rasker J, Wiegman O. Omgaan met Reumatoide Arthritis: Ontwikkeling en evaluatie van een groepsprogramma voor patienten met reumatoide artritis. Enschede, The Netherlands: Universiteit Twente (Aspekt 35); 1992. 94. Sharpe L, Sensky T, Timberlake N, Ryan B, Brewin CR, Allard S. A blind, randomized, controlled trial of cognitivebehavioural intervention for patients with recent onset rheumatoid arthritis: preventing psychological and physical morbidity. Pain 2001;89:275– 83. 95. Suarez-Almazor M, Belseck E, Shea B, Homik J, Wells G, Tugwell P. Antimalarials for treating rheumatoid arthritis [Cochrane review]. Oxford: Update Software; 2001. 96. Suarez-Almazor M, Spooner C, Belseck E, Shea B. Auranoﬁn versus placebo in rheumatoid arthritis [Cochrane review]. Oxford: Update Software; 2001. 97. Suarez-Almazor M, Spooner C, Belseck E. Penicillamine for treating rheumatoid arthritis [Cochrane review]. Oxford: Update Software; 2001. 98. Suarez-Almazor M, Spooner C, Belseck E. Azathioprine for treating rheumatoid arthritis [Cochrane review]. Oxford: Update Software; 2001. 99. Wells G, Haguenauer D, Shea B, Suarez-Almazor M, Welch V, Tugwell P. Cyclosporine for treating rheumatoid arthritis [Cochrane review]. Oxford: Update Software; 2001. 100. Suarez-Almazor M, Belseck E, Shea B, Wells G, Tugwell P. Methotrexate for treating rheumatoid arthritis [Cochrane review]. Oxford: Update Software; 2001. 101. Criswell L, Saag K, Sems K, Welch V, Shea B, Wells G, et al. Moderate-term, low-dose corticosteroids for rheumatoid arthritis [Cochrane review]. Oxford: Update Software; 2001. 102. Sackett D. Bias in analytic research. J Chronic Dis 1979;32: 51– 63. 103. CCSG, The Canadian Cooperative Study Group. The Canadian trial of aspirin and sulﬁnpyrazone in threatened stroke. N Engl J Med 1978;299:53–9. 104. Newman S, Mulligan K, Steed L. What is meant by selfmanagement and how can its efﬁcacy be established? [editorial]. Rheumatology (Oxford) 2001;40:1– 6. 105. WHO. International classiﬁcation of functioning, disability and health (ICF). Geneva: World Health Organization; 2001.