Intimate Partner Violence and Perpetrator Characteristics: Insights from a Regional Analysis of Women's Perspectives

Background: Intimate Partner Violence (IPV) is a widespread issue that transcends demographic boundaries, affecting individuals regardless of age, gender, ethnicity, environment, or sexual orientation. Globally, women are particularly vulnerable, especially in certain regions, and predominantly in low- and middle-income countries. Prior research has often concentrated on specific regions or individual countries, limiting the scope of analysis. This study aims to provide a broader, cross-regional analysis of IPV prevalence while exploring perpetrator characteristics across different regions—an area that remains under-researched, partly due to the inherent challenges in obtaining reliable data. Methods: This study utilizes IPV data from Demographic and Health Surveys (DHS) conducted between 2014 and 2022 across five regions and 19 countries (n = 41,392 women). The analysis investigates how various factors, including age, education, employment, residence, and marital status, influence IPV prevalence. Additionally, sub-analysis focuses on decision-making characteristics, while Random Forest modelling is employed to determine the importance of specific perpetrator traits. Results: Regionally, the highest IPV prevalence was observed in Latin America and the Caribbean (15.64%; 15.00-16.27), followed by Central Asia (14.39%; 13.72–15.05). The lowest prevalence was identified in Sub-Saharan Africa (1.3%; 1.23–1.36). Analysis of victims' perceptions of perpetrator characteristics revealed regional differences. In Sub-Saharan Africa, income disparity and educational attainment were ranked as the most significant factors contributing to IPV, whereas in other regions, accusations of infidelity and jealousy were deemed more important. Alcohol consumption by the perpetrator was found to be a less significant factor across all regions. Conclusions: The findings reveal substantial regional disparities in IPV prevalence, underscoring the need for region-specific interventions, particularly in areas with high IPV rates. The profiling of perpetrator behaviours suggests that valuable insights can be obtained directly from the experiences of survivors. Given the limited data available on perpetrators, we recommend that future surveys, such as the DHS, incorporate more detailed questions regarding perpetrator characteristics, while maintaining survey-based safeguards to protect the well-being of respondents.