This study was a quasi-experimental time series design to test the hypothesis that an expanded program of home care for the terminally ill would reduce hospital utilization and costs of care during the last month of life. The data base was derived by comparing cancer deaths in persons less than 65 years of age over a 7-year period from a regional tumor registry against Blue Cross enrollment and claims files, to form claims histories for the last 6 months of life for 1,874 decedents who had made at least one claim during this period. Forty-six percent (46%) of the decedents had used home-care services, and a subgroup of high-intensity users was identified from daily home care charges during the last month of life. As the home care program expanded to provide more intensive home-hospice services, a trend was observed of greater cost savings among home-care users than among nonusers. These savings among users were achieved by a reduction in hospital days, and by a reduction in the mean daily cost of hospitalization. Home-care users also showed significantly less variability in costs than nonusers.