Objective Second-generation antidepressants are commonly used to treat major depression in late-life. This systematic review and meta-analysis was undertaken to assess the evidence for efficacy of second-generation antidepressants in late-life major depression. Methods The Cochrane Library (2006 [3]), MEDLINE (1966 to August 2006), and meeting presentations were searched for trials of second-generation antidepressants (nontricyclics) marketed in the United States. Published and unpublished placebo-controlled randomized clinical trials in outpatients 60 years and older, with nonpsychotic, unipolar major depression were selected. Clinical characteristics and outcomes were extracted. Outcomes were expressed as odds ratios (OR), risk differences, and weighted mean differences. Results Ten unique trials (four unpublished) with 13 contrasts met selection criteria. Trials were 6–12 weeks duration, and included 2,377 patients who received active drug and 1,788 received placebo. The ORs by meta-analysis for response and remission were 1.40 (95% confidence interval [CI] 1.24–1.57, z = 5.45, N = 13, p <0.001) and 1.27 (CI 1.12–1.44, z = 3.67, N = 13, p <0.001), respectively, with significant heterogeneity for response and remission among the trials. Mean pooled response rates for antidepressant and placebo were 44.4% and 34.7%, respectively. The OR for response was significantly higher in the 10–12 week trials (OR = 1.73, CI 1.42–2.09, z = 5.51, N = 5, p <0.001) than the 6–8 week trials (OR = 1.22, CI 1.05–1.42, z = 2.60, N = 8, p = 0.01). ORs for discontinuation for any reason and for adverse events were significantly higher with drugs than with placebo. Conclusions Antidepressants are more effective than placebo in elderly depressed subjects although effects are modest and vary. Identification of the characteristics of responders and nonresponders will be crucial to improving treatment outcomes. Second-generation antidepressants are commonly used to treat major depression in late-life. This systematic review and meta-analysis was undertaken to assess the evidence for efficacy of second-generation antidepressants in late-life major depression. The Cochrane Library (2006 [3]), MEDLINE (1966 to August 2006), and meeting presentations were searched for trials of second-generation antidepressants (nontricyclics) marketed in the United States. Published and unpublished placebo-controlled randomized clinical trials in outpatients 60 years and older, with nonpsychotic, unipolar major depression were selected. Clinical characteristics and outcomes were extracted. Outcomes were expressed as odds ratios (OR), risk differences, and weighted mean differences. Ten unique trials (four unpublished) with 13 contrasts met selection criteria. Trials were 6–12 weeks duration, and included 2,377 patients who received active drug and 1,788 received placebo. The ORs by meta-analysis for response and remission were 1.40 (95% confidence interval [CI] 1.24–1.57, z = 5.45, N = 13, p <0.001) and 1.27 (CI 1.12–1.44, z = 3.67, N = 13, p <0.001), respectively, with significant heterogeneity for response and remission among the trials. Mean pooled response rates for antidepressant and placebo were 44.4% and 34.7%, respectively. The OR for response was significantly higher in the 10–12 week trials (OR = 1.73, CI 1.42–2.09, z = 5.51, N = 5, p <0.001) than the 6–8 week trials (OR = 1.22, CI 1.05–1.42, z = 2.60, N = 8, p = 0.01). ORs for discontinuation for any reason and for adverse events were significantly higher with drugs than with placebo. Antidepressants are more effective than placebo in elderly depressed subjects although effects are modest and vary. Identification of the characteristics of responders and nonresponders will be crucial to improving treatment outcomes.