Objective: To investigate the incidence and consistency of antiretroviral (ARV) treatment in the period before the introduction of protease inhibitors among Medicaid beneficiaries in New Jersey who had both the human immunodeficiency virus (HIV) and schizophrenia. Method: HIV-infected Medicaid beneficiaries were identified using the HIV and acquired immunodeficiency syndrome (AIDS) registries for New Jersey; claims histories were used to identify patients diagnosed with ICD-9-CM schizophrenia and affective psychoses and to examine use of ARV drugs. Results: Bivariate and multivariate analysis found no difference in the likelihood of receiving ARV drugs between patients with HIV and schizophrenia and HIV-infected patients without schizophrenia. However, once the therapy was initiated, patients with schizophrenia were more consistent users of ARV drugs. Conclusion: Results do not indicate that HIV-seropositive (HIV+) patients with schizophrenia are less adherent to HIV therapies than HIV+ patients without schizophrenia. In our study population, consistency of use was actually higher among HIV+ patients with schizophrenia, perhaps because their multiple diagnoses place them under closer medical scrutiny.