Where do older women and men die, and what factors are associated with their healthcare utilization at the end-of-life and site of death? In their recent report, the Alliance for Aging Research (1998:7) calls for the use of the 1993 National Mortality Followback Survey (NMFS) """"""""to ask important unanswered questions about end-of-life experiences for older women and men and to determine unmet needs of dying people."""""""" The proposed study uses the 1993 NMFS to accomplish the following aims: (1) to describe gender differences in end-of-life care including length of time in a hospital in the last year of life, length of time residing in a nursing home in the last year of life, and site of death; (2) to assess factors associated with the type and amount of health service use for men and women in the last year of life, use of hospice care, and of site of death; (3) to determine whether gender differences in health service utilization and site of death vary by socioeconomic status, race/ethnicity, marital status, characteristics of other family ties, and the availability of a family member as a caregiver. The strengths of the proposed study are its cost effectiveness, the ability to simultaneously assess the determinants of site of death (e.g., home, nursing home, or hospital) and to test for interaction effects associated with gender and other demographic characteristics. Data from the (NMFS) provide an unprecedented opportunity to study factors associated with patterns of care and site of death for a random sample of older men and women. Additional benefits of this large nationally-representative sample are its oversample of Blacks and the wealth of data on the decedent's medical history including diseases and disability and whether the decedent had a living will. This retrospective cohort study will set the stage for a prospective study of men's and women's end-of-life care. This research agenda promises to raise new possibilities for social interventions designed to improve end- of-life care.