Wealth Inequality and Intimate Partner Violence: An Individual and Ecological Level Analysis Across 20 Countries.

2021 
Intimate Partner Violence (IPV) has been linked to poor health. Economic position may be an important risk factor for IPV. We examined the association between economic position and IPV at country and individual levels. We analyzed Demographic and Health Surveys data of 187,716 ever-partnered women between ages 10 and 59 from 20 low- and middle-income countries. We calculated direct age-standardized 12-month prevalence of physical IPV and performed ecological analysis using Gini coefficients and Concentration indexes to assess correlation with 12-month prevalence of physical IPV. We conducted multivariable logistic regression for each country to assess the association between wealth status and physical IPV and a meta-analysis of the regression model to present results across countries. Compared to the Poorest quintile, odds of IPV among wealthier quintiles varied by country. In the Middle quintile, India had significantly reduced IPV (OR 0.75, 95%CI: 0.68-0.83). In the Richer and Richest quintiles, 4 and 6 countries had significant reductions in IPV, respectively. Only Mozambique was found to have significant increased IPV in the wealthiest quintile (OR 2.51, 95%CI: 1.45-4.38). Gini coefficient and physical IPV had a correlation coefficient of 0.502 (p value 0.033), while Concentration index had -0.276 (p value .253). Standardized prevalence for physical IPV ranged from 1.58% to 18.91%. Findings suggest that the relationship between wealth and IPV vary considerably in the included low- and middle-income countries, and that risk of IPV may not necessarily be higher among women in lower wealth brackets. Mozambique was the only country with increased odds of IPV among the Richest group as compared to the Poorest group. This study provides evidence IPV may transcend economic boundaries, and that studies looking at the link between inequality and IPV are paramount for designing adequate preventative policies.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    11
    References
    1
    Citations
    NaN
    KQI
    []