Mental Health, Social Support, and Problem Drinking: A Population-Based Study

Background and Purpose: Problem alcohol consumption such as binge and heavy drinking causes serious health risks and increases mortalities. Understanding mechanisms that help prevent or reduce problem drinking is critical for social work practice. Self-medicating hypothesis and past studies have shown that people engage in problem drinking to cope with stress and mental health concerns. While some studies suggest that social support can help moderate the relationship between mental health needs on problem drinking, others indicate no or adverse effect. Using a population-based sample, the study aims to understand: 1) the association between mental health needs and problem drinking; and 2) the moderating role of social support on the relationship between mental health needs and problem drinking. Given that past studies suggest that relationship between mental health needs and drinking can differ by gender, we conducted gender-specific analyses.

Methods: We analyzed the 2016 Behavioral Risk Factor Surveillance System (BRFSS) data from four states that included the social support module. Drawn from multi-stage probability sampling, the total sample (N=33,705) consists of 14,796 men and 18,909 women. We first conducted descriptive analyses, followed by gender-specific logistic regressions on weighted data examining the associations between mental health needs and heavy and binge drinking. We then created an interaction term and assessed the moderation effect of social support by including social support and the interaction term of mental health and social support in the subsequent steps of the logistic regression. All models controlled for demographic factors.

Results: Descriptive data show that compared to women, male engage with significantly more binge drinking (21.5% versus 12.3%) and slightly greater heavy drinking (7% versus 6%). Women have more mental health needs while enjoying greater social support than men. For study aim 1, mental health needs is associated with heavy and binge drinking for both men (heavy drinking: OR=1.42, 95%CI=1.18-1.72, p<.0001; binge drinking: OR=1.31, 95%CI=1.16-1.48, p<.0001) and women (heavy drinking: OR=1.42, 95%CI=1.18-1.71, p<.0001; binge drinking: OR=1.45, 95%CI=1.27-1.67, p<.0001). For study aim 2, we find no association between social support and problem drinking across both genders. However, the mental health X social support interaction term is significant in men’s heavy drinking model (OR=0.65; 95%CI=0.49-0.85, p<.01). Moderation graph shows that those who have low social support and greater mental health needs are most likely to heavy drink compared to those with similar mental health needs but have medium or high social support.

Conclusions and Implications: Study findings provide support to the self-medicating hypothesis and show that those who have health needs are more likely to engage in hazardous drinking. Contrary to some of the previous empirical literature, our study suggests that the relationship between mental health needs and problem drinking are comparable between men and women. Surprisingly, social support only moderates the association between mental health needs and heavy drinking for men, but not for women. Future studies should seek to identify ways to increase social support for men while continuing to understand mechanisms that can help restrain the positive relationship between mental health needs and problem drinking.