Prevalence, disparities, and trends in intimate partner violence against women living in urban slums in 34 low-income and middle-income countries: a multi-country cross-sectional study

Background: Intimate partner violence (IPV) is a significant public health issue, closely tied to social and neighborhood environments. The impoverished, overcrowded, and stressful conditions in urban slums may heighten IPV risk, yet evidence remains limited. This study aims to assess the prevalence, disparities, and trends of IPV in urban slums across low- and middle-income countries (LMICs). Methods: This cross-sectional study used nationally representative Demographic and Health Surveys data from 2006 to 2023, focusing on countries with available domestic violence data for women aged 15-49. The outcomes measured include past-year exposure to any IPV (primary outcome) and physical IPV, sexual IPV, and psychological IPV (secondary outcomes). We examined both absolute and relative disparities between urban slums, non-slum urban, and rural areas using differences and ratios. Additionally, we used Fairlie decomposition analysis based on a multivariable logistic regression model to examine the contributions of IPV risk factors (i.e., poor partner relationships, gender inequality, and poverty) to the disparities. For countries with multiple surveys, we conducted trend analysis by assessing annual changes in IPV prevalence in urban slums and the disparities. Findings: Among 283,658 women from 34 LMICs, 14,111 (5.0%) lived in urban slums. IPV prevalence in urban slums was notably high, with 18 of the studied countries above 30% for any IPV. Women in urban slums experienced higher IPV rates than those in non-slum urban and rural areas. For example, the prevalence of any IPV in urban slums was 31.4% (95% confidence interval [CI]: 30.7-32.2), which was 5.9 percentage points (95% CI: 5.1-6.7, P < 0.0001) higher than that in non-slum urban areas and 1.2 percentage points (95% CI: 0.4-2.0, P = 0.0022) higher than that in rural areas. Controlling behavior by husbands explained the largest proportion of disparities in all IPV types between urban slums and other areas. For example, 27.2% (95% CI: 25.1-29.3) of the any IPV disparities between urban slums and non-slum urban areas could be explained by this factor. In ten countries with multiple surveys, trend analysis showed rising any IPV prevalence in urban slums of four countries-Sierra Leone, Tanzania, Mali, and Nigeria-with Sierra Leone having the most significant increase (4.6 percentage points, 95% CI: 2.5-6.6, P < 0.0001). Interpretation: Our findings suggest that IPV is more prevalent in urban slums than other areas, underscoring the need for targeted public health strategies, such as addressing harmful partner's behaviors. Funding: This research was supported by China National Natural Science Foundation and the Research Fund, Vanke School of Public Health, Tsinghua University.