Arthritis & Rheumatism, Volume 60,
October 2009 Abstract Supplement

The 2009 ACR/ARHP Annual Scientific Meeting
Philadelphia October 16-21, 2009.

Adherence to An 8-Week Yoga Program for RA and OA: Who Drops out and Why?

Haaz1,  Steffany, Bingham III2,  Clifton O., Bartlett3,  Susan J.

Johns Hopkins School of Medicine, Baltimore, MD,
Johns Hopkins University, Baltimore, MD,
Johns Hopkins, Baltimore, MD


Physical activity is integral to comprehensive arthritis management, yet long-term adherence averages 50–60% over several months. We have previously shown yoga is a safe, effective and enjoyable physical activity for arthritis patients, though long term-adherence in unknown.


To identify factors associated with adherence to yoga tailored for people with rheumatoid (RA) and osteoarthritis (OA).


In this RCT, 75 persons ages 18–65 with OA or RA were recruited through arthritis clinics and area rheumatologists. Participants were 96% female, 55% Caucasian, aged 52.3±11.9 years. 36 (48%) had RA and 39 (52%) had OA and were randomized to immediate yoga (IY) or waitlist control (WC) followed by participation in the yoga program. Yoga classes, held biweekly for 8 weeks, included poses (asanas), breathing (pranayama), chanting, relaxation and meditation. Poses were designed to accommodate joint limitations and individual abilities. Instruction was provided for weekly home practice.

Adherence was assessed as: 1) attendance at >=13 classes (protocol) and 2) completion of study requirements (study). Baseline predictors included sociodemographic and disease characteristics, depressive symptoms (CES-D), self-efficacy (ASES) and selected markers of physical fitness (4) and health status (4). Logistic regression was used to identify baseline predictors of protocol and study adherence.


Of 75 patients screened, 40 were assigned to IY and 35 were assigned to WC. There were no differences between treatment groups in sex, race, disease diagnosis or duration. The WC group was slightly older than the IY group (55.9±1.5 vs. 49.2±2.1, p=.01).

There was a trend for those randomly assigned to IY to attend any classes (83% vs. 63% WC respectively p=.059), which did not change when controlling for differences in age between groups.

Overall, 69% were classified as protocol adherent (73% IY vs. 64% WC, p=.48). Caucasian race was the only baseline characteristic associated with starting or completing the yoga intervention (odds ratio [OR]=.33, 95% CI=.11,.97 and OR=.25 95% CI=.09,.94 respectively). Study adherence was 71% (63% IY vs. 80% WC, p=.10). Attending college and Caucasian race were the only predictors associated with study adherence (OR=2.4 95% CI=.10,.75 and OR=.26 95% CI=1.1, 4.9 respectively). However, after controlling for age and race, education was no longer predictive. While race was repeatedly associated with adherence, the contribution to predicting any form of adherence was low. Trends toward significance were found for hypothesized markers of physical function, physical fitness and self-efficacy. Common reasons for attrition were change in eligibility, life events and class logistics (time, location).


Arthritis patients who begin yoga may be likely to persist with it over time. Risk of drop-out appears highest in minority persons and before starting classes. Better understanding of minority barriers to participation and other factors affecting adherence is needed. Clinicians may wish to consider recommending yoga to arthritis patients, particularly when adherence to exercise has been problematic.

To cite this abstract, please use the following information:
Haaz, Steffany, Bingham III, Clifton O., Bartlett, Susan J.; Adherence to An 8-Week Yoga Program for RA and OA: Who Drops out and Why? [abstract]. Arthritis Rheum 2009;60 Suppl 10 :1978
DOI: 10.1002/art.27051

Abstract Supplement

Meeting Menu