Variations in the association between polygyny and experience of intimate partner violence by husband's alcohol consumption: a cross-sectional study among postpartum women in Thailand.

BACKGROUND

Intimate partner violence is more common in polygynous couples than monogamous couples, but the extent that this association is modified by the husband's alcohol consumption is unknown. The objectives of this study are: (1) To describe the extent to which polygyny is associated with self-reported experience of intimate partner violence among women receiving postpartum care; (2) To describe the extent to which the mentioned association is modified by the husband's alcohol consumption.

METHODS

We conducted a hospital-based cross-sectional study among women age 18 years or older receiving postpartum care at 8 public hospitals in 3 provinces in Northern and Northeastern Thailand using self-administered questionnaires. We analyzed data using descriptive statistics, logistic regression, and stratified analyses.

RESULTS

A total of 1207 women agreed to participate in the study, 8% of whom reported that their husbands practiced polygyny. Women in a polygynous relationship were more likely than women in monogamous relationships to experience intimate partner violence (11.7% vs. 3.6%, Adjusted OR = 2.23; 95% CI = 0.94, 5.26). The prevalence of intimate partner violence was relatively low in both groups among women whose husbands did not drink (2.9% vs. 0%, Adjusted OR = N/A), and very high in both groups among those whose husbands binge-drank (46.2% vs. 20.8%, Adjusted OR = 9.54; 95% CI = 1.10, 82.54). However, the Breslow-Day Test of Homogeneity suggested that there was no statistically significant effect modification (p-value = 0.259).

CONCLUSION

Stakeholders in intimate partner violence should consider both alcohol use (particularly binge-drinking) and polygyny as risk factors for intimate partner violence. However, caveats regarding study design, misclassification and potential information bias, and lack of generalizability should be considered in the interpretation of the study findings.