Miller SC, Gozalo P, Mor V; Association for Health Services Research. Meeting.
Abstr Book Assoc Health Serv Res Meet. 1999; 16: 439.
Center for Gerontology and Health Care Research, Brown University, Providence, RI, USA.
RESEARCH OBJECTIVE: There has been concern regarding the cost to Medicare of the provision of Medicare hospice care to nursing facility (NF) patients. However, studies examining Medicare utilization and expenditures for these patients have not been performed. Previous research on non-NF populations has shown hospice enrollment to result in large Medicare Part A cost savings in the last month of life, due primarily to reductions in hospitalization. In this study our objective was to examine whether hospitalization of NF patients in the last 30 days of life was less for those patients enrolled in Medicare hospice. STUDY DESIGN: All NF patients in Kansas, Maine, Mississippi, New York and South Dakota having at least one NF minimum data set (MDS) record between 1992 and 1996 and dying while on the Medicare hospice benefit were included. Three non-hospice decedents were matched to each hospice decedent by state of NF residence, one of three diagnostic groups, and by the time interval from last MDS to death. MDS data from the 5 HCFA Case-Mix Reimbursement and Quality Demonstration states were merged with Medicare claims, On-line Survey Certification Automated Survey (OSCAR) data and with hospice provider of service files. Multivariate analyses were performed; the dependent variable being the incidence of hospitalization and the independent variables including both patient and provider level variables. PRINCIPAL FINDINGS: NF decedents not enrolled in Medicare hospice had an over 5 times greater incidence of hospitalization than those decedents enrolled in hospice, even when controlling for patient and provider characteristics. For non-hospice decedents, older age was independently associated with a reduced risk of hospitalization while the presence of pain and dyspnea and residence in a for-profit NF were independently associated with an increased risk. For hospice decedents, older age and greater functional impairment were independently associated with a reduced risk of hospitalization. There was also variation in hospitalization across states, with hospitalization less likely in states with higher Medicaid per diems. CONCLUSIONS: These preliminary findings suggest that the provision of Medicare hospice care can reduce the costs to Medicare of caring for terminal NF residents, at least in the last month of life. Other analyses comparing Medicare expenditures across the entire hospice episode (and a comparable episode for non-hospice decedents) can provide a more definitive answer as to the comparative costs of providing Medicare hospice are in NFs. IMPLICATIONS FOR POLICY, DELIVERY OR PRACTICE: Information provided by this study suggests that hospice care in NFs presents an opportunity for Medicare savings. Given this, and considering the disruptive effects a terminal hospitalization are thought to have on patients and their families, it appears that a focus on increased access to hospice or hospice-like care in NFs is needed.
Publication Types:
Keywords:
- Breast Feeding
- Case-Control Studies
- Cost Savings
- Health Expenditures
- Hospice Care
- Hospices
- Hospitalization
- Humans
- Kansas
- Maine
- Medicaid
- Medicare
- Mississippi
- New York
- Pain
- South Dakota
- economics
- nursing
- hsrmtgs
Other ID:
UI: 102193791
From Meeting Abstracts