DISTRIBUTION AND PREDICTORS OF DEPRESSIVE SYMPTOMS IN OSTEOARTHRITIS

Objective. To determine the prevalence and correlates of depressive sy mptoms in a sample of elderly persons with osteoarthritis (OA). Method s. Secondary analysis of cross sectional interview data from 108 commu nity living persons age 50 or over who met American College of Rheumat ology clinical criteria for OA of the hip and/or knee. Results. The pr evalence of substantial depressive symptomatology appeared to be no gr eater than would be expected in the general population for this age gr oup. However, persons currently under a physician's care for OA were s ignificantly more depressed than those not under care, and, with the e xception of those who had posthigh school education, depression was a better predictor of being under care than OA symptoms. The direct and interactive effects of 3 variables - education, age and self-perceived impact of the OA problem - were found to explain 40% of variance in d epression scores in the total sample and 50% among those currently rec eiving care. Younger and less educated subjects had relatively more de pressive symptoms and, furthermore, in this subgroup the correlation b etween OA impact and depression was high. Conclusions. The importance of assessing depression in persons who seek care for OA is supported. Furthermore, our results suggest that the observed relationship betwee n education and arthritis outcomes could be mediated, at least in part , by depression.