Health Care Utilization in Homeless People: Translating Research into Policy and Practice

Homelessness is a major problem across North America: Approximately 3 million Americans have experienced homelessness within the last year, and parents accompanied by children account for about one-fifth of the U.S homeless population. Homeless people suffer from high rates of illness and frequently encounter barriers to accessing care. Comprehensive information on health care utilization for homeless people will help to guide the development of health policy, however, these data have not been attainable through administrative databases in the U.S., because only one-third of the homeless have Medicaid and more than half are uninsured. The Agency for Healthcare Research and Quality (AHRQ) has funded investigatorinitiated research projects for promoting developments related to Translating Research into Policy and Practice. This research study describes and summarizes the background, objectives, methods, research findings and implications for policy and practice from an AHRQ grant entitled “Health Care Utilization in Homeless People” (1 R01 HS04129-01), Stephen W. Hwang, M.D., M.P.H., Principal Investigator, St. Michael‟s Hospital, University of Toronto. The aims and primary objectives of this study were as follows: 1) To enroll a representative sample of homeless men, women, and families to obtain baseline data on these individuals, including demographic characteristics, history of homelessness, physical and mental health status, health conditions, usual source of health care, mental health problems, alcohol and drug problems, health beliefs, social supports, self-reported health care utilization, unmet needs for health care, and self-reported barriers to accessing health care; 2) To precisely quantify these homeless individuals‟ total and annualized health care utilization (office-based visits, emergency department visits and hospitalizations) and health care costs over a 4-5 year (2004­ 2009) follow-up period within a system of universal health insurance; 3) To compare the intensity of health care utilization among homeless individuals in the general population of Toronto; 4) To identify individual characteristics that are predictive of high levels of health care utilization among the population; 4) To determine whether the use of office-based services by homeless people is associated with a likelihood of emergency department visits and/or hospitalization, after adjustment for baseline characteristics. The study was conducted in Toronto, Ontario, where there is both a large homeless population and comprehensive administrative databases that capture the delivery of all health services for every resident. The study enrolled a random sample of 1,190 homeless individuals in Toronto, Canada, in 2004-2005. Sampling was stratified by sex and family status, resulting in the enrollment of 603 single men, 303 single women, and 284 heads of families (plus 421 children in these 284 families). Homeless participants were matched 1:1 to low income controls from the general population based on year of birth and sex. Data for the control group were obtained from the Registered Persons Database (RPDB) which includes all persons in Toronto, Ontario. The research team constructed a survey instrument that focused on factors likely to be associated with future health care utilization. These factors included demographic characteristics, history of