Background: Existing research on refugee mental health is heavily skewed towards refugees in high-income countries, even though most refugees (83%) are hosted in low-income and middle-income countries. This problem is further compounded by the unrepresentativeness of samples, small sample sizes and low response rates. Objective: To present representative findings on the prevalence and correlates of depression among different refugee subgroups in East Africa. Methods: We conducted a multicountry representative survey of refugee and host populations in urban and camp contexts in Kenya, Uganda and Ethiopia (n=15 915). We compared the prevalence of depression between refugee and host populations and relied on regression analysis to explore the association between violence, depression and socioeconomic outcomes. Findings: We found a high prevalence of elevated depressive symptoms (31%, 95% CI 28% to 35%) and functional impairment (62%, 95% CI 58% to 66%) among the refugee population, which was significantly higher than that found in the host population (10% for depressive symptoms, 95% CI 8% to 13% and 25% for functional impairment, 95% CI 22% to 28%) (p<0·001). Further, we observed a dose-response relationship between exposure to violence and mental illness. Lastly, high depressive symptoms and functional impairment were associated with worse socioeconomic outcomes. Conclusion: Our results highlight that refugees in East-Africa-particularly those exposed to violence and extended exile periods-are disproportionately affected by depression, which may also hinder their socioeconomic integration. Clinical implications: Given the high prevalence of depression among refugees in East Africa, our results underline the need for scalable interventions that can promote refugees' well-being.