OBJECTIVE: Low-income white older adults and those of color are at greater risk for depression but less likely to receive care. We evaluated the impact of a one-to-one peer support intervention compared to active control delivered by nonpeers for this population. DESIGN: Mixed methods, single-blind randomized controlled trial. SETTING: Community-based social service and aging organizations and geriatric primary care. PARTICIPANTS: Low-income white older adults and those of color 50+ with depression. INTERVENTIONS: Peer Enhanced Depression Care and nonpeer, social interaction control. MEASUREMENTS: Primary outcome was depression (PHQ-9). Data were collected at baseline, postinterventions, 3, 6, 9, and 12 months. Poststudy interviews were conducted with both groups. RESULTS: Among 149 randomized participants, the mean age was 70, 84% were women, 52% Black and 41% White. Both groups experienced an average decrease of 3.7 (SE: 0.55, 95% CI: [-4.77, -2.63]) points in depression from baseline to postintervention and 2.56 (SE: 0.71, 95% CI: [-3.95, -1.17]) points from baseline to 12 months. Effect sizes at postintervention (Cohen's d = 0.81) and at 12-months (Cohen's d = 0.52) indicate large and medium effects, respectively. Both groups experienced decreases in loneliness and increases in adaptive coping and self-efficacy. Qualitative findings suggest the intervention group learned coping skills and experienced behavior change, whereas control group described a general positive experience. CONCLUSIONS: Peer support intervention was not superior to social interactions delivered by nonpeers. Divergent quantitative and qualitative results suggest the need for additional effectiveness trials of peer support interventions outside of pandemic conditions. Trial Registration ClinicalTrials.gov Identifier: NCT04319094.