A multisite study of the prevalence of HIV using rapid testing in inpatient and outpatient mental health settings

Context: Rates of HIV are much higher among persons with mental illness than in the general population. Despite CDC recommendations that annual HIV testing be conducted routinely in all clinical settings, such testing is rarely conducted in mental health settings. Objective: Estimate HIV prevalence and risk factors among persons receiving treatment in mental health settings in Philadelphia and Baltimore between January 2009 and August 2011. Design, Setting, and Participants: We tested 1062 individuals for HIV who were receiving services in three settings: University-based inpatient psychiatric units (N=288), community mental health centers (CMHCs) (N=501), and intensive case management (ICM) programs (N=273). We used univariate Poisson models to examine the association between HIV status and risk behaviors, HIV testing history, psychiatric symptoms and psychiatric diagnoses. We used multivariable Poisson models to estimate adjusted prevalence ratios. Results: Overall, 51 (4.8%) individuals were HIV-infected. The proportions with confirmed positive HIV test results were 5.9% for inpatient units, 4.0% for CMHCs and 5.1% for ICM programs. HIV infection was associated with black race (adjusted prevalence ratio [aPR]: 5.71, 95% confidence interval [CI]: 1.58-20.67), homosexual or bisexual identity (aPR: 2.27, CI: 1.03-5.00), and overall psychiatric symptom severity (aPR: 1.69, CI: 1.01-2.83). Conclusions: HIV seroprevalence for individuals receiving mental health services in this study was approximately 15 times higher than that estimated for the general population. As a result, we recommend that HIV testing and linkage to infectious disease care care be integrated into routine clinical care in all mental health services delivery settings.