Agreement on Reporting Intimate Partner Violence Among Nepalese Couples: A Cross-Sectional Study

When intimate partner violence (IPV) data are collected from only one partner, they are often subject to considerable reporting bias. However, it is not easy to collect such data from couples, and inaccuracies might result in discrepancies, which needs a resolution. We assessed the concordance on reports of lifetime and previous year physical, sexual, and emotional IPV against wives, as reported by both Nepalese wives and husbands. The association of possible risk factors with discordant reporting of IPV was also analyzed. We conducted a cross-sectional study in two areas in Nepal between August and September 2011. We collected data from 717 randomly selected couples on lifetime and previous year experience of physical, sexual, and emotional IPV against wives, as well as their sociodemographic characteristics. We calculated the kappa coefficients and agreement percentage to assess the concordance on wives' reports of IPV victimization and husbands' reports of IPV perpetration. We also performed multiple logistic regressions to identify the factors associated with discordant reporting of IPV among couples. Levels of concordance between wives' and husbands' reports of IPV were significantly low, as indicated by kappa coefficients, ranging from .20 (sexual and emotional IPV) to .24 (physical IPV) in lifetime experience and from .15(sexual IPV) to .18 (physical IPV) in previous year experience. Wives' caste, husbands' age and education, household income, and place of residence were significantly associated with discordance in IPV reports among Nepalese couples. Discordant reporting about IPV is common among Nepalese couples. Collecting information from both partners might be important to obtain more reliable data on IPV in the Nepalese context.