Correlates of intimate partner violence among urban women in sub-Saharan Africa.

Introduction: The dynamics of intimate partner violence (IPV)-one of the world's leading public health problems-in urban Africa remain poorly understood. Yet, urban areas are key to the future of women's health in Africa.

Study Objectives: We explored survivor-, partner-, and household-level correlates of prevalence rates for types of IPV in urban SSA women.

Method: The study uses DHS data from 42,143 urban women aged 15-49 in 27 SSA countries. Associations at the bivariate level were examined using the Pearson Chi-square test. The modified Poisson regression test estimated the relative prevalence of IPV subtypes in the study population at the multivariate level.

Results: Approximately 36% of women in urban SSA experienced at least one form of IPV; 12.8% experienced two types; and 4.6% experienced all three types. SSA urban women who had only primary-level education, had 3 or more living children, were informally employed, were in polygynous unions, or who approved of wife-beating similarly displayed higher adjusted prevalence rates for all three forms of IPV compared respectively to their counterparts without formal education, without a living child, were unemployed, in monogamous unions, or who do not approve of wife-beating. On the other hand, the region's urban women who began cohabiting between ages 25 and 35 years or who lived in higher wealth households showed consistently lower adjusted prevalence rates for all three forms of IPV relative to their counterparts who began cohabiting before 18 years or who lived in lower wealth households. Compared to their counterparts without formal education, without a living child, or whose partners did not have formal education, women with secondary and higher education, with 1-2 living children, or whose partners had only primary level schooling displayed higher adjusted prevalence rates for both IPEV and IPPV, but not for IPSV. However, relative to their counterparts whose partners were aged 25 years or below, living with a partner aged 40 years and above was associated with statistically significant reduced prevalence rates for IPPV and IPSV, but not for IPEV. Only for IPPV did women with partners educated at secondary and above levels display statistically significant higher adjusted prevalence rates relative to their counterparts with uneducated partners. Also, solely for IPPV did women who began cohabiting between ages 18 and 24 years or whose partners were employed (whether formally or informally) show decreased adjusted prevalence rates relative to their counterparts who started cohabiting before 18 years or whose partners were unemployed. In addition, only for IPSV did women aged 40 years and above or living in middle wealth households show statistically significant reduced adjusted prevalence rates relative to their counterparts aged less than 25 years or living in lower wealth households.

Discussion And Conclusion: By 2030, the majority of SSA women will be urban dwellers. Complexities surround IPV in urban SSA, highlighting the unique dynamics of the problem in this setting. While affirming the link between IPV and marital power inequities and dynamics, findings suggest that the specific correlates of prevalence rates for different IPV sub-types in urban SSA women can, at once, be both similar and unique. The contextual drivers of the differences and similarities in the correlates of the prevalence rates of IPV sub-types among the region's urban women need further interrogation.