Current literature suggests that nearly 18% of lifetime costs for medical care (>$40,000 per person) is estimated to be incurred in the last year of life and nearly 28% of U.S. Medicare payments for those over age sixty-five are for persons in their last year of life [Scitovsky, 1994]. Little has been reported, however, on total medical costs at the end-of-life for prostate cancer (PCa) patients. OBJECTIVE: The aim of this research was to compare the last year of medical costs in PCa patients dying at home or in an health care institution. METHODS: Cost data was extracted on 293 men from CaPSURE, a longitudinal disease registry of patients with PCa. Standard charges (1997 dollars) were applied to all health care services in the last 4 quarters of life. Not all patients contributed data to each quarter. Descriptive and multivariate analyses were used to examine cost differences between groups. RESULTS: About 63% of patients died in institutions, while 31% died at home. PCa was the reported cause of death in 32.8% of cases. After controlling for time period, location and cause of death, total medical costs were lower in the last 3 months of life for PCa patients who died at home ($1975 vs $3179, p<.05). No statistical differences were found in the preceding three quarters. CONCLUSIONS: Total medical costs in the last 3 months of life were lower for PCa patients who die at home when compared with those dying in health care institutions. Inpatient care costs were nearly 3 times higher for patients dying in health care institutions. These cost savings may inform the way we meet the challenge to improve the quality of life, but also the quality of death among PCa patients at the end-of-life.