Prevalence and predictors of spousal violence against women in Afghanistan: evidence from Demographic and Health Survey data.

Abstract

Spousal violence against women is a serious public health problem that is prevalent in all societies, with one in three women around the world experiencing violence in their lifetime. This study examined the prevalence of spousal violence, and its determinants, in Afghanistan using data from the 2015 Afghanistan Demographic and Health Survey. Univariate, bivariate and logistic regression statistical techniques were used to assess the association of socioeconomic variables with spousal violence. The study sample comprised 20,827 currently married women aged 15-49. Fifty-two per cent of women reported experiencing some form of violence by their husband. A significant association was found between women's justification of violence, women's participation in decision-making in their household (COR=0.476; CI=0.446-0.509) and lower risk of experiencing spousal violence. After adjustment for demographic and socioeconomic factors, women's participation in all of four household decisions, either alone or jointly, was found to be associated with a lower risk of experiencing spousal violence (AOR=0.472; CI=0.431-0.516). In both the crude and adjusted models, the risk of experiencing spousal violence was high if the husband's desire for children was different from that of his wife. In the case of inequality in property ownership, the risk of spousal violence was significantly higher (COR=1.263; CI=1.178-1.353; AOR=1.159; CI=1.051-1.278) when women were joint owners of property compared with when they did not own any property. The findings point to an immediate need for legal and social interventions to prevent spousal violence against women, or at least reduce its prevalence, in Afghanistan.