The purpose of this study was to assess ethnic differences in use of psychotherapy (having met at least once with a psychotherapist) for late-life depression in primary care. Participants were identified through a two-stage, age-stratified (60–74, ≥75) depression screening of randomly sampled patients from 20 practices in New York City, Philadelphia, and Pittsburgh in a practice-randomized trial. Practices were randomly assigned to usual care or to an intervention with a depression care manager who worked with primary care physicians to provide algorithm-based care. Depression status based on clinical interview and any use of psychotherapy within the 2-year follow-up interval were the primary dependent variables under study. The focus was on 582 persons with complete data. Participants were sorted into major depression (n=385, 112 African American and 273 white) and clinically significant minor depression (n=197, 51 African American and 146 white) based on clinical diagnostic assessment. Persons who self-identified as African American were less likely than whites to use interpersonal therapy (IPT) if they had minor depression, even after adjusting for potentially influential variables including age, cognitive functioning, and whether the dose of antidepressant was adequate (adjusted odds ratio (AOR)=0.22, 95% confidence interval (CI)=0.06–0.80). Ethnicity was not significantly associated with IPT use in persons with major depression (AOR=0.71, 95% CI=0.37–1.37). Older African Americans with minor depression were less likely than whites to use psychotherapy. Targeted strategies are needed to mitigate the disparity in use of psychotherapy.