Effects of health insurance on the use of health services and antiretroviral therapy among homeless and marginally housed HIV-positive adults

OBJECTIVE: To determine the impact of health insurance continuity and duration on the use of health services and antiretroviral therapy (ART) among HIV-positive homeless and marginally housed individuals in San Francisco, CA. METHODS: Between 1996 and 2006, quarterly interviews were conducted. Marginal structural models were used to estimate the causal effect of insurance coverage on subsequent ambulatory care visits, inpatient hospitalizations, emergency department visits and use of ART. RESULTS: Among 330 HIV positive respondents, 19% were female, 42% were African American, and the median age was 41 years. On average, 77% of study participants used ambulatory care during each 3-month quarter, 10% received inpatient care, 17% were seen in an emergency department and 54% of those eligible took ART medication. After adjusting for effects of significant confounding variables, ambulatory care was predicted by both intermittent and continuous insurance, as well as increasing durations of insurance coverage. ART was most consistently predicted by continuous insurance coverage, with only slight increases for longer durations of observed coverage. Effects of health insurance on inpatient care and emergency department use neither resulted in consistent patterns nor reached levels of statistical significance. CONCLUSIONS: Among homeless and marginally housed HIV-positive adults, the probability of using ambulatory care and ART increase as the continuity and duration of insurance coverage increase. Ambulatory care and ART are the factors with the most potential impact on the course of HIV, indicating that securing continuous insurance coverage would have a large impact on the morbidity and mortality of this vulnerable population.