Objectives: Assessing the prevalence of sexual violence in Brazil and its association with individual and contextual factors.
Methods: A multilevel analysis performed with cross-sectional data from 2010. The adjusted prevalence ratio was estimated through Poisson multilevel modelling. Cross-level interactions were evaluated by the inclusion of interaction terms between socio-economic variables from the two levels.
Results: Sexual violence is more prevalent in federal units that presented lower years of schooling expectancy at 18 years old (PR 1.27; CI95% 1.09-1.48), lower per capita income (PR 1.23; CI95% 1.05-1.43), lower HDI (PR 1.17; CI95% 1.01-1.37), higher proportion of poverty vulnerability (PR 1.18; CI95% 1.02-1.38), higher proportion of unemployment (PR 1.68; CI95% 1.45-1.96) and higher proportion of people who neither work nor study (PR 1.26; CI95% 1.08-1.46). Higher prevalence of sexual violence was associated with lower Gini index and greater coverage of primary health services. In analysing the tendency demonstrated a strong contextual effect between the Brazilian federative units in relation to sexual violence CONCLUSIONS: The strategy to reduce sexual violence in Brazil must be integrated with other social policies, considering both individual risk factors and macro-social determinants.
Keywords: Multilevel; Public health policy; Sexual violence; Social determinants.