Background: The prevalence of depression is elevated among HIV-infected individuals and there is evidence that depression exerts a negative impact on HIV medication adherence. Methods: Merged HIV/AIDS surveillance data and Medicaid claims data from January 1996 to December 1998 were used to identify AIDS-infected patients with diagnosed depression, and filled prescriptions were used to identify their antidepressant use, and highly active antiretroviral therapy (HAART). Chi-square tests and robust logistic regressions were used to examine antidepressant use after HAART initiation, and a person-month approach was used to estimate the association between antidepressant treatment and adherence to HAART after its initiation. Results: Of the 406 AIDS-infected patients diagnosed with depression who initiated HAART during this period, 81% (N = 329) were treated with an antidepressant. The HAART adherence rate was low overall. After HAART initiation; only 63% of the person-months had a prescription for it. However, use of an antidepressant in the prior month was significantly associated with HAART in the current month. After controlling for other factors, the odds of current-month HAART adherence were increased by almost 30% for those with antidepressant use in the prior month (Adjusted OR = 1.28, 95% CI [1.16, 1.41]). Conclusions: While the HAART adherence rate was low among patients with AIDS diagnosed with depression, prior month's antidepressant use increases odds of adherence. Unmeasured factors may influence the reported association between antidepressant use and HAART adherence, but our findings point to the need to investigate directly the impact of antidepressant therapy on HAART adherence found among patients with AIDS and depression.