Association Between Community Health Center Usage and Emergency Department Utilization among California’s HIV-Infected Medicaid Beneficiaries, 2009

Importance: Community Health Centers (CHC) are important sites of care for people living with HIV (PLWH) and play an increasing role in their care under the Affordable Care Act. Little is known about the relationship between CHC usage and emergency department (ED) utilization in this population. Objective: To determine the association between CHC usage and ED utilizationDesign: Retrospective, cross-sectional study of diagnosed PLWH enrolled in California’s Medicaid program in 2008 and 2009. Zero-inflated Poisson models were used to estimate the odds of being an ED user and the number of ED visits in 2009. We controlled for demographics (age, gender, race, urban residence, income, education), service characteristics (managed care enrollment, provider HIV experience), and medical characteristics (mental health, substance abuse, tobacco, medical comorbidity, antiretroviral therapy). Setting: Emergency department Participants: We included 6284 adult, full-term 2008-2009 beneficiaries with strong evidence of HIV diagnosis and excluded pregnant and dual-eligible beneficiaries. Exposures: CHC users were patients who had ≥1 CHC outpatient claim in 2008. Non-CHC users had outpatient claims only at non-CHCs. Those with no outpatient usage had no 2008 outpatient claims. Main Outcomes and Measures: Number of ED claims on separate days per beneficiary in 2009Results: CHC users averaged significantly greater numbers of ED visits than non-CHC users and those with no outpatient usage (1.91, 1.58, and 1.70, respectively; P=0.022). CHC users had higher odds of being ED users (OR=1.16; 95%CI 1.04-1.30). Controlling for demographic and service characteristics did not alter this result (OR=1.16; 95%CI 1.03-1.31). The difference was mitigated once medical characteristics were included (OR=1.09; 95%CI 0.96-1.25). The association between CHC status and number of ED visits, conditional on using the ED at all, was not significant in the bivariate (rate ratio (RR)= 1.12; 95%CI 0.97-1.28) or multivariate models (RR=1.01; 95%CI 0.87-1.17). The overall differences in mean ED visits observed between CHC and non-CHC groups were reduced to insignificance (1.77; 95% CI 1.60-1.93 vs 1.68; 95%CI 1.53-1.84) after adjusting for demographic, service, and medical characteristics. Conclusions and Relevance: CHC users had higher ED utilization than non-CHC users, but the disparity was largely driven by differences in medical characteristics