Although there is a large literature examining the effects of distress on the demand for medical care, the data on which this literature is based are equivocal. Nonetheless, this literature is cited frequently by those who advocate a national mental health policy designed to produce a cost-effective "medical offset effect." In this study, longitudinal data on illness episodes, physician visits, and depressive symptoms were collected from 940 Medicare recipients enrolled in a health maintenance organization (HMO) under a Tax Equity and Fiscal Responsibility Act (TEFRA) contract. Seven waves of interviews were conducted over a period of 1 year. This article presents two sets of analyses. In the first, controlling for chronic conditions and demographics reported at baseline, the relationships between depressive symptoms reported at baseline, and all illness episodes and physician visits that occurred over the subsequent year are examined. In the second, controlling for depressive symptoms and demographics reported at baseline, the relationships between illness episodes and physician visits over the study year, and depressive symptoms recorded at the final interview are examined. The data indicate that, whereas depressive symptoms at baseline are virtually unrelated to subsequent illness episodes and physician visits, illness episodes and physician visits are related to subsequent depressive symptoms. These data indicate, therefore, that policies aimed at diverting the distressed from seeking medical care may result in further inequities in the receipt of needed care.