Utilization patterns of end-of-life services by injection drug users who have died with HIV/AIDS

Rationale: The changing demographics of HV/AIDS have implications for the future use of health care resources. Objective: To determine the location of death (hospital, community hospice, home) of HIV-infected injection drug users (IDU's) and type of care (intensive, acute, palliative) at time of death; and to examine demographic, behavioral, and other variables in order to identify factors associated with receiving palliative care. Methods: IDU's who died with HIV/AIDS in Vancouver between 1991 and 1996 were identified through seven sources, including hospital data bases and direct services workers. Data collection consisted of a retrospective review of client records from Vancouver hospitals, community hospices, community organizations, and the Drug Treatment Program of BC Centre for Excellence in HIV/AIDS. Results: There were 121 deaths documented that indicated that the individual was HTV positive and an IDU (92 were male, 24 were female, and 5 were transgendered). Median age at death was 35 years. Fifty-three percent were Aboriginal, and 61 percent lived in unstable housing. Sixty-one percent died in an acute care hospital setting, 27 percent died in a palliative care unit or in a community hospice, 11 percent died at home, and one percent on the street. Eleven of the thirteen home deaths were attributed to drug overdoses. AIDS was the most common cause of death (66%), followed by drug overdose (9%). Excluding deaths due to drug overdose, suicide or unknown causes, 71% of the IDU's died while receiving some form of palliative treatment. However, the median length of palliative treatment was only two days. Having an AIDS diagnosis was positively associated with receiving palliative care for more than two days, and with dying in a hospice or palliative care unit. Other variables, such as stable housing, being male and a period of longer than 36 months between HIV diagnosis and death were also positively associated with one of the three sets of criteria used for palliative care in this study. Multivariate analysis revealed that an AIDS diagnosis was independently associated with receipt of palliative care. Conclusions: Analysis suggests that IDU's are receiving palliative treatment, but for the majority, it is only for a very short period of time, and often when already bedridden or unconscious. Results should assist in planning future end-of-life services in light of changing demographics of HIV/AIDS in Vancouver, and in light of the moral and ethical imperative to provide compassionate care at the end of life.Medicine, Faculty ofPopulation and Public Health (SPPH), School ofGraduat