INFLUENCE OF PATIENT PREFERENCES AND LOCAL HEALTH SYSTEM CHARACTERISTICS ON THE PLACE OF DEATH

OBJECTIVE: To examine the degree to which variation in place of death is explained by differences in the characteristics of patients, includ ing preferences for dying at home, and by differences in the character istics of local health systems. DESIGN: We drew on a clinically rich d atabase to carry out a prospective study using data from the observati onal phase of the Study to Understand Prognoses and Preferences for Ou tcomes and Risks of Treatments (SUPPORT component). We used administra tive databases for the Medicare program to carry out a national cross- sectional analysis of Medicare enrollees place of death (Medicare comp onent). SETTING: Five teaching hospitals (SUPPORT); All U.S. Hospital Referral Regions (Medicare). STUDY POPULATIONS: Patients dying after t he enrollment hospitalization in the observational phase of SUPPORT fo r whom place of death and preferences were known. Medicare beneficiari es who died in 1992 or 1993. MAIN OUTCOME MEASURES: Place of death (ho spital vs non-hospital). RESULTS: In SUPPORT, most patients expressed a preference for dying at home, yet most died in the hospital. The per cent of SUPPORT patients dying in-hospital varied by greater than 2-fo ld across the five SUPPORT sites (29 to 66%). For Medicare beneficiari es, the percent dying in hospital varied from 23 to 54% across U.S. Ho spital Referral Regions (HRRs). In SUPPORT, variations in place of dea th across site were not explained by sociodemographic or clinical char acteristics or patient preferences. Patient level (SUPPORT) and nation al cross-sectional (Medicare) multivariate models gave consistent resu lts. The risk of in-hospital death was increased for residents of regi ons with greater hospital bed availability and use; the risk of in-hos pital death was decreased in regions with greater nursing home and hos pice availability and use. Measures of hospital bed availability and u se were the most powerful predictors of place of death across HRRs. CO NCLUSIONS: Whether people die in the hospital or not is powerfully inf luenced by characteristics of the local health system but not by patie nt preferences or other patient characteristics. These findings may ex plain the failure of the SUPPORT intervention to alter care patterns f or seriously ill and dying patients. Reforming the care of dying patie nts may require modification of local resource availability and provid er routines.