OBJECTIVES Arthritis is one of the most common chronic diseases among elderly. Approximately 50% of all elderly in the United States suffer from the disease. Drug therapy plays an important role in decreasing morbidity and mortality associated with the disease. The objectives of this study are to evaluate the effect of prescription drug insurance coverage on prescription drug use, and health among the elderly patients with arthritis. METHODS Estimates are obtained using multivariate regression, and a fixed-effect (within person) research design that controls for the unmeasured person-specific effects. Analyses were based on the Medicare Current Beneficiary Survey for years 1992 - 2004. RESULTS Estimates show that prescription drug coverage is associated with a 2% - 15% increase in utilization of prescription drugs depending on the type and generosity of the coverage. In addition, the effect of drug coverage on drug use differed depending on the type of drugs, and other co-morbid conditions. For example, among arthritic patients, drug coverage increased use of diabetic drugs by 20%, whereas it had relatively low impact on use of cold medications (2%). We found no evidence that drug coverage improved hospitalization and general health status. CONCLUSIONS Drug coverage plays a crucial role on the use of essential medications among elderly patients with arthritis. The results of the analysis also suggest the importance of controlling for selection bias. Our estimates on drug coverage were reduced markedly when we accounted for selection into plans.