OBJECTIVE To ascertain whether initiation of protease inhibitors was associated with a change in depressive symptoms among persons infected with HIV. METHODS Study subjects included men and women who were enrolled in the HIV/AIDS Drug Treatment Program and who had completed an annual participant survey before and after initiating triple combination therapy with a protease inhibitor. Depressive symptoms were assessed using the Centre for Epidemiologic Studies-Depression scale (CES-D). Statistical analyses to determine the change in CES-D total and subscale scores before and after protease inhibitor use were conducted using parametric and multivariate methods. RESULTS Our analysis was restricted to 453 participants. Of these 234 (52%) were depressed at baseline (CES-D score > or = 16). Compared with nondepressed participants, depressed participants were slightly younger (p = .048), less likely to be employed (p < .001) and more likely to have an annual income less than $10,000 per annum (p < .001). After adjusting for CD4 count, employment status, income, age, and CES-D total or subscale score at baseline, we found a significant improvement in total scale score (p = .001) and depressive mood (p = .002), positive affects (p = .005), and somatic symptoms (p = .011) subscale scores at follow-up. There was no significant change in the interpersonal relations score over the study period. CONCLUSIONS Our findings indicate that in addition to conferring impressive clinical benefits, protease inhibitor use is associated with a significant improvement in HIV-positive individuals' mental health.