Multi-level analysis of intimate partner violence and its determinants among reproductive age group women in Ethiopia: evidence from Ethiopian Demographic Health Survey, 2016

Abstract

Abstract Background Intimate partner violence (IPV) is recognized as a main public health challenge, with serious consequences for women’s physical, mental, sexual, and reproductive health. Despite its public health importance, most studies of IPV in Ethiopia mainly focused on individual characteristics and didn’t identify how factors operating at different levels affect IPV. Thus, there is limited evidence regarding the hierarchical-level factors of IPV and the effect of individual and community-level determinants of IPV. The aim of this study is to assess the individual and community-level factors associated with violence against women among ever-married reproductive-age women in Ethiopia. Methods A retrospective analysis of secondary data retrieved from the Ethiopia Demographic and Health Survey was conducted among reproductive age group women (15–49 years of age) who reported ever being married within the available data set for the domestic violence module. STATA 14 was used to conduct the analysis. A two-level mixed-effects logistic regression analysis was used to determine associations between IPV and individual- and community-level factors. IPV variability across the community was assessed using ICC and PCV. The model’s fitness was assessed using the Akaike information criterion (AIC), the Bayesian information criterion (BIC), and the likelihood ratio test. Result The life time prevalence of IPV in this study was 33% [95% CI: 30.74, 34.25]. Women’s age 20–24 (AOR = 5.85, 95% CI: 201 3.10, 11.04), 25–29 age group (AOR = 6.41, 95% CI; 3.34, 12.32), 30–34 age group (AOR = 9.48, 95% CI: 4.71, 19.06), 35–39 age group (AOR = 9.88, 95% CI: 4.79, 20.39), 40–44 age group (AOR = 11.10, 95% CI: 5.16, 23.89), and 45–49, (AOR = 14.15, 95% CI: 6.01, 32.80), early marriage (AOR = 1.21, 95% CI: 1.08, 1.47), witnessing inter-parental violence during childhood (AOR = 2.80, 95% CI: 2.16, 3.96), having a lot of living children (AOR = 0.45, 95% CI: 0.26, 0.74), having a partner who drank alcohol (AOR = 3.00, 95% CI: 2.42–3.67), decision-making autonomy of the women (AOR = 0.77, 95% CI: 0.62, 0.97), Poor wealth index (AOR = 1.64, 95% CI: 1.23, 2.18), middle wealth index (AOR = 1.86, 95% CI: 1.36, 2.54) and exposure to media (AOR = 1.47, 95% CI: 1.06, 2.00) were all significantly associated with IPV. Conclusion and recommendation This study showed that one-third of the women experienced IPV in their lifetime. The finding suggested that community based interventions and multi-sectorial collaborations are needed to reduce the IPV and its adverse consequences.