Background/Purpose: Measures of chronic pain typically do not account for individuals’ physical activity (PA) levels.  Although PA is essential for managing conditions like osteoarthritis (OA), some people may reduce activity to manage their pain.  Recent research showed that a PA-adjusted pain measure was more strongly associated with radiographic OA severity than an unadjusted pain measure.  We extend this area of research by examining whether PA-adjusted pain is also more closely associated with key function and quality of life outcomes.

Methods:  In a subset of 140 Veterans (M age=61.8 years, 87.1% male) enrolled in a clinical trial of group vs. individual physical therapy, we used the Western Ontario and McMaster Universities Osteoarthritis (WOMAC) Pain Scale and calculated four composite WOMAC pain and Physical Activity (WOPA) scores using accelerometer-derived data.  Specifically, WOMAC pain scores were adjusted for: 1) step counts, 2) time in sedentary activity, 3) time in moderate-intensity activity, and 4) energy expenditure (kilocalories).  All data were from baseline assessments.  We examined associations of WOMAC pain score and each of the four WOPA scores with six OA-related outcomes: 6-minute walk test, 8-foot walk test, chair stand test, self-reported satisfaction with physical function, fatigue (brief fatigue inventory), and anxiety / depressive symptoms (single item).  Analyses were partial correlations, controlling for age, gender and body mass index.

Results: Significant (p<0.05) associations were found between WOMAC / WOPA scores and OA-related outcomes in the majority (22/30) of models (Table 1).  In all cases greater pain was associated with poorer outcomes.  For the four OA-related outcomes that measure aspects of physical function (six-minute walk, chair stands, 8-foot walk, satisfaction with function), the step-count adjusted and energy expenditure adjusted WOMAC pain scores had stronger associations (partial r’s=0.24-0.45) than the unadjusted WOMAC pain scores (partial r’s=0.15-0.25).  For fatigue, unadjusted WOMAC pain and energy expenditure adjusted WOMAC pain scores had similar associations (partial r’s=0.27 and 0.28), and for anxiety and depressive symptoms, the unadjusted WOMAC pain score had the strongest association (partial r=0.31). 

Conclusion:   Results suggest PA-adjusted pain measures may add increased value in predicting some OA-related outcomes and should be explored further, particularly in longitudinal studies.  Step-count adjusted pain (which was associated with radiographic OA severity in previous research) and energy expenditure adjusted pain may be particularly useful in predicting functional outcomes.