BACKGROUND: Separate reports from the maintenance treatment for late-life depression (MTLD) trials have shown that low socioeconomic status (SES) and anxiety symptoms at the time of treatment initiation predict lower levels of response to antidepressant treatment and higher levels of suicidal ideation in older adults. AIM: To determine whether SES and anxiety independently contribute to worse treatment outcomes, as indicated by persistence of depressive symptoms during treatment and the persistence of suicidal ideation. Consistent with prior evidence that sociodemographic factors and clinical history are both prognostic of depression treatment efficacy, we hypothesized that SES and pre-existing anxiety symptoms will both predict lower levels of response to treatment and higher levels of suicidal ideation. METHOD: Secondary analyses of data from the MTLD trials. RESULTS: Regression analyses which controlled for comorbid anxiety indicated that residents of middle- and high-income census tracts were more likely to respond to treatment (HR, 1.63; 95%CI, 1.08-2.46) and less likely to report suicidal ideation during treatment (OR, 0.51; 95%CI, 0.28-0.90) than residents of low income census tracts. The same regression models indicated that pre-existing anxiety symptoms were independently related to lower treatment response (HR, 0.73; 95%CI, 0.60-0.89) and higher risk of suicidal ideation (OR, 1.45; 95%CI, 0.98-2.14). CONCLUSION: These findings demonstrate the importance of treating anxiety symptoms during the course of treatment for late-life depression and, at the same time, addressing barriers to treatment response related to low SES.