Impact of joint impairment on longitudinal disability in elderly persons.

Recent longitudinal data indicating that arthritis is a major contributor to disability in elderly persons are based on self-reported diagnostic information. This longitudinal study included baseline physical examinations of joints of 541 persons over age 60. Previous results from a cross-sectional multivariate model of disability in this sample found that joint impairment (and, its absence, arthritis pain) explained a significant proportion of variance in overall disability. We have retested this model using generalized estimation equations (GEE) analysis to estimate the effect of joint impairment and arthritis pain on baseline and Year 2 disability. Findings indicate that baseline joint impairment contributes substantially to longitudinal disability. If direct measures of baseline joint impairment are unavailable, concurrent self-reported arthritis pain also predicts longitudinal disability well. These findings indicate that longitudinal studies should monitor arthritis pain and that symptomatic arthritis is a risk factor for future disability.