Effects of a Weight and Pain Management Programin Patients With Rheumatoid Arthritis With Obesity

A Randomized Controlled Pilot Investigation

Tamara J. Somers, PhD; James A. Blumenthal, PhD; Caroline S. Dorfman, PhD; Kim M. Huffman, MD, PhD; Sara N. Edmond, PhD; Shannon N. Miller, BPH; Anava A. Wren, PhD; David Caldwell, MD; Francis J. Keefe, PhD


J Clin Rheumatol. 2022;28(1):7-13. 

In This Article



The Figure presents the study CONSORT diagram. Three hundred eighty-seven patients were approached for this study. Of these, 104 (26.9%) agreed to be assessed for eligibility. Forty-nine patients screened were ineligible or excluded. Fifty-five of 57 patients (96.5%) screened met full eligibility criteria. Two hundred eighty-three (73.1%) declined study participation prior to eligibility screening. Reasons for declining participation included (1) lack of interest (26.9%), (2) distance (23.7%), (3) lack of time (11.0%), (4) transportation (4.6%), and (5) reason unknown (33.9%).


CONSORT diagram. Color online-figure is available at http://www.jclinrheum.com.

Fifty patients (90.9%) who enrolled in the study at baseline were randomized: 21 to the standard care arm and 29 to the intervention arm. Five (9.1%) were not randomized because they did not complete the exercise treadmill test (n = 2) or dropped out prior to randomization (n = 3). Of those randomized to the intervention arm, 79.3% (n = 23) initiated treatment. Five of the 6 participants who did not initiate treatment completed study assessments at baseline and follow-up. Twenty-six individuals randomized to the intervention arm (89.7%) had data available at both pretreatment and posttreatment. Of those randomized to the standard care condition, 14 (66.7%) completed both the pretreatment and posttreatment assessments.

Participants initiating treatment (n = 23) attended 74.3% of group skills sessions and 64.2% of exercise sessions. More than 80% (n = 19, 82.6%) of intervention arm participants attended at least half of the treatment sessions and half of the exercise sessions.

Sample Characteristics

Sociodemographic and baseline characteristics for participants are presented in Supplemental Table 2 (http://links.lww.com/RHU/A364). There were no differences between participants randomized to the intervention or standard care arms on sociodemographic, medical, or outcome variables. Participants were primarily female (92.0%), White (66.0%), and married/cohabiting (70.0%). Participants were 56.1 (SD, 11.0) years old on average and had received diagnosis of RA an average of 12.6 (SD, 9.7) years prior to accrual. More than a third of participants (34.0%) were on medical disability. The physician assessment revealed the average number of swollen and tender joints to be 4.0 (SD, 4.2) and 15.1 (SD, 8.7), respectively. On average, participants' BMIs fell into the obese range (mean, 35.1 [SD, 5.2] kg/m2). At baseline, participants reported a total Brief Pain Inventory score of 4.72 (SD, 1.97) over the past week, with their worst pain at mean 7.02 (SD, 2.56) and least pain at mean 2.67 (SD, 1.98). Average arthritis self-efficacy scores were a mean of 56.36 (SD, 19.19).

Study Outcomes

Within-group comparisons from baseline to posttreatment are presented in Supplemental Tables 3 and 4 (http://links.lww.com/RHU/A365; http://links.lww.com/RHU/A366) for the intervention and standard care groups, respectively. Data from completed assessments are presented.

Primary Outcomes

Weight. While individuals in both arms evidenced improvements in weight from pretreatment to posttreatment, much larger improvements were seen among the intervention arm (mean difference, 2.28 [SD, 4.38] kg; d = 0.52) participants when compared with standard care arm (0.77 [SD, 2.61] kg; d = 0.30) participants.

Pain Intervention arm participants experienced improvements in pain severity from pretreatment to posttreatment. When examining participants' usual, worst, least, and average pain over the last week, intervention arm participants experienced improvement across all domains, with the greatest improvements seen for worst pain (pretreatment: mean, 7.52 [SD, 2.50] to posttreatment: mean, 5.84 [SD, 3.02]; d = 0.67). Individuals in the standard care arm saw an increase in the severity of their pain over the course of the study (pretreatment: mean, 4.35 [SD, 1.61] to posttreatment: mean, 5.13 [SD, 1.04]; d = 0.56). When examining change in specific pain ratings, standard care participants experienced increased pain across all domains, with the magnitude of the effect sizes ranging from small (worst pain: d = 0.20) to large (least pain: d = 0.87).

Secondary Outcomes

Waist Circumference. Improvements were seen among intervention arm participants for waist circumference (mean difference, 4.76 [SD, 6.18] cm; d = 0.77). Slight improvements in waist circumference were also found among standard care participants (mean difference, 1.23 [SD, 4.70] cm; d = 0.26).

Physical Functioning. Individuals in the intervention arm saw improvements in physician-reported disease severity as assessed by the study rheumatologist from pretreatment to posttreatment. On average, scores for intervention arm participants decreased by close to 4.5 points (mean difference, 4.43 [SD, 11.38] points; d = 0.39). Disease severity scores for the standard care arm remained relatively stable (mean difference, 0.62 [SD, 8.02]; d = 0.08).

Participants in the intervention arm also experienced improvements on the physical functioning subscale of the AIMS-2 as well as the 6MWT, our objective measure of physical disability (physical functioning: mean difference, 0.46 [SD, 0.91]; d = 0.51; 6MWT: mean difference, −50.19 [SD, 63.44] m; d = 0.79). The standard care arm reported slight declines in physical functioning over time (mean difference, −0.28 [SD, 0.84]; d = 0.25), and scores on the 6MWT remained relatively stable (mean difference, −14.25 [SD, 44.48] m; d = 0.32).

Eating Behavior. Intervention arm participants evidenced improvements in binge-eating behaviors (mean difference, 5.63 [SD, 6.88]; d = 0.82) from baseline to posttreatment. Standard care arm participants also reported slight improvements in eating behaviors (mean difference, 2.14 [SD, 4.86]; d = 0.44).

Self-efficacy for Weight Control. Intervention arm participants evidenced improvements in self-efficacy for weight control of slightly over 1 point (mean difference, −1.09 [SD, 1.57]; d = 0.69). Standard care arm participants experienced slight improvements in self-efficacy as well, with an average change from baseline to follow-up of approximately half a point (mean difference, −0.52 [SD, 1.32]; d = 0.39).

Self-efficacy for Arthritis. Improvements were found for both groups on the ASES, suggesting that participants' confidence in their abilities to control arthritis pain and minimize disability improved over time (intervention: mean difference, −4.14 [SD, 13.73]; d = 0.20; standard care: mean difference, −6.41 [SD, 17.91]; d = 0.35).