INTRODUCTION: Most studies on the epidemiology of personality disorders (PDs) have been conducted in high-income countries and may not represent what happens in most part of the world. In the last decades, population growth has been concentrated in low- and middle-income countries, with rapid urbanization, increasing inequalities and escalation of violence. Our aim is to estimate the prevalence of PDs in the Sao Paulo Metropolitan Area, one of the largest megacities of the world. We examined sociodemographic correlates, the influence of urban stressors, the comorbidity with other mental disorders, functional impairment and treatment. METHODS: A representative household sample of 2,942 adults was interviewed using the WHO-Composite International Diagnostic Interview and the International Personality Disorder Examination-Screening Questionnaire. Diagnoses were multiply imputed, and analyses used multivariable regression. RESULTS AND DISCUSSION: Prevalence estimates were 4.3% (Cluster A), 2.7% (Cluster B), 4.6% (Cluster C) and 6.8% (any PD). Cumulative exposure to violence was associated with all PDs except Cluster A, although urbanicity, migration and neighborhood social deprivation were not significant predictors. Comorbidity was the rule, and all clusters were associated with other mental disorders. Lack of treatment is a reality in Greater Sao Paulo, and this is especially true for PDs. With the exception of Cluster C, non-comorbid PDs remained largely untreated in spite of functional impairment independent of other mental disorders. CONCLUSION: Personality disorders are prevalent, clinically significant and undertreated, and public health strategies must address the unmet needs of these subjects. Our results may reflect what happens in other developing world megacities, and future studies are expected in other low- and middle-income countries.