Serious mental illness and risk of new HIV/AIDS diagnoses: an analysis of Medicaid beneficiaries in eight states.

Objective: A longitudinal analysis was used to explore the relationship between diagnosis of serious mental illness and subsequent new diagnoses of HIV.

Methods: Logistic regression was used to predict HIV/AIDS diagnoses in 2002–2004 among Medicaid beneficiaries in eight states (N=6,417,676) who were without HIV in 2001. Results for beneficiaries with and without serious mental illness, a substance use disorder, and psychiatric comorbidities in 2001 were compared.

Results: After controlling for substance abuse or dependence and other factors, the analyses indicated that the odds of new HIV/AIDS diagnoses among beneficiaries with or without serious mental illness did not differ significantly. Compared with beneficiaries without a substance use disorder or serious mental illness, individuals with a substance use disorder but without serious mental illness were 3.1 times (OR=3.13, p<.001) more likely, and those with both substance abuse or dependence and serious mental illness were 2.1 times (OR=2.09, p<.001) more likely, to receive a new HIV diagnosis in 2002–2004. However, people with serious mental illness but without a substance use disorder in 2001 were 23% less likely (OR=.77, p<.001) than people without serious mental illness or a substance use disorder in 2001 to receive a new HIV diagnosis.

Conclusions: After substance abuse or dependence was controlled for longitudinally, little independent association between serious mental illness and the risk of new HIV diagnoses was found. HIV-prevention services for low-income individuals should be delivered to all persons with serious mental illness, but especially those with comorbid substance use disorders.