Interactions among alcohol dependence, perinatal common mental disorders and violence in couples in rural Vietnam: a cross-sectional study using structural equation modeling

Abstract

Background: There is increasing recognition that perinatal common mental disorders (PCMDs) are prevalent in women in low and lower-middle income countries and emerging evidence that PCMDs and alcohol abuse occur in men in these settings. Domestic violence is associated with PCMDs in both women and men. The aim of this study was to examine the relationships among PCMDs, alcohol abuse and domestic violence in couples in a rural, low-income setting.

Methods: A cross-sectional, population-based study was undertaken in randomly selected communes in Ha Nam and Hanoi, Vietnam. All women in the selected study sites who were at least 28 weeks pregnant or were mothers of 4 - 6 week old babies in the recruitment period were eligible. The husbands of the women who consented to join the study were also invited to participate. Data sources were study-specific questions and standardised measures: PCMDs were assessed by psychiatrist-administered Structured Clinical Interviews for DSM IV disorders, and alcohol dependence (AD) by the CAGE questionnaire (cut-off of ≥ 2). Structural Equation Modeling was used to test direct, indirect and mutual relationships simultaneously in the hypothesised model.

Results: In total 364/392 (93%) eligible women agreed to participate. Of these, 360 were married, and 230 (64%) of their husbands also participated to yield a sample of 230 couples for analyses. Overall, in 7.4% (95% CI: 4.6-11.6) of couples both wife and husband were diagnosed with a PCMD; and 41.2% (95% CI: 35.1-47.8) of couples at least one member had a PCMD. Comorbid PCMD and AD were observed in 6.9% (95% CI: 4.3-11.0) of men, but did not occur in women. After controlling for other psychosocial risk factors comorbid PCMD and AD in husbands increased by 4.7 times the probability of PCMDS in their wives via intimate partner violence. PCMDS in wives did not increase the probability of PCMDS or AD in husbands.

Conclusions: These data provide evidence that comorbid PCMD and AD in husbands have a significant adverse effect on the mental health of their wives in rural areas of Vietnam. This indicates that strategies to prevent and treat PCMDs in women will be more effective if paired with initiatives to reduce alcohol dependence and violent behaviours in men.

Figure 1

Indirect effect of comorbid PCMD and AD in husbands on probability of PCMD in wives via intimate partner violence. Only significant paths were included (see Table 2 for details). All of the variables in the diagram (presented in rectangular boxes) are observed except for the unmeasured (latent) variable ‘Intimate partner violence’ (represented as an ellipse).Single-headed solid arrows represent statistically significant directional paths, whereas dashed lines indicate factor loadings of the latent variable.