Efficacy of antidepressant medication among HIV-positive individuals with depression: a systematic review and meta-analysis

Abstract

Depression is highly prevalent among HIV-positive individuals yet some, but not all, have not found antidepressant medication to be efficacious in this population. We performed a systematic review and meta-analysis of double-blinded, randomized controlled trials to examine efficacy of antidepressant treatment among HIV-positive depressed individuals and evaluate whether the results are generalizable to women and minorities. We used PubMed, the Cochrane Database, a search of bibliographies, and consultation with experts to find double-blinded, controlled clinical trials with random assignment to antidepressants or control condition for which HIV-positive patients met standard diagnostic criteria for depression. The principal measure of effect size was the standard difference between means on the Hamilton Depression Score (HDS). We identified 7 studies that included 494 subjects. Three of the 7 studies reported significant antidepressant effects. The pooled effect size from the random effects model was 0.57 (95% confidence interval [CI]: [0.28-0.85]). Heterogeneity across studies was significant (Q, 13.22; p = 0.07; I(2) = 47.1%). When stratified by placebo response, the pooled effect size for placebo response greater than 33% was 0.20 (-0.11-0.52) and not significant while the pooled effect size for placebo response greater than 33% was 0.80 (0.52-1.08) and was significant. Placebo response explained nearly 62% of the variance in effect sizes across studies. Women were nearly absent from and minorities were underrepresented in the studies investigated. Antidepressant medication is efficacious in treating depression among depressed, HIV-positive individuals. However, the underrepresentation of women and minorities limits the generalizability of these findings and suggests that future studies be directed to address this disparity.