BACKGROUND: Beginning October 1, 1997, Medicare implemented a series of major changes to the Home Health (HH) reimbursement system. Reimbursements were first significantly reduced under the Interim Payment System (IPS) and then relaxed slightly until implementation of the HH Prospective Payment System (PPS) on October 1, 2000. OBJECTIVE: The objective of this study was to examine the impact of reimbursement policy on HH care utilization. RESEARCH DESIGN: We postulated that in response to the initial changes, there would be reductions in both the probability of any HH use and the number of HH visits per HH user. Under PPS, we postulated there would be further reduction in number of HH visits. We tested whether the policy response differed by HH agency structure and whether subgroups of patients were differentially affected. An interrupted time-series analysis was conducted to examine month-to-month probability of HH selection and the number of HH visits among users. SUBJECTS: A 100% sample of all Medicare recipients undergoing either elective joint replacement (1.6 million hospital discharges) or surgical management of hip fracture (1.2 million hospital discharges) between January 1996 and December 2001 was selected. RESULTS: Under the IPS, the probability of any HH use and number of visits per episode of HH care fell until the IPS was refined in October 1998. With implementation of the PPS, HH visits fell commensurately. Differentially larger reductions in care were noted at for-profit HH agencies, for the elderly, women, patients receiving state assistance, and patients first discharged to skilled nursing facility or rehabilitation hospitals. CONCLUSIONS: Changes in month-to-month utilization of HH services were sharp and well correlated with policy implementation dates, strengthening the evidence for a causal association between policy and patient care in the midst of a sea of concurrent policy changes. Greater reductions in HH visits were noted for vulnerable groups.