Bridging Disciplinary Boundaries (January 11 - 14, 2007)

The role, if any, of the workplace in employee problem drinking has implications for social work practice in terms of opportunities for workplace alcohol abuse prevention. Therefore, based on two conceptual frameworks of work stress and employee problem drinking, we tested a moderated mediation model of stress-induced problem drinking. Specifically, we hypothesized that work stressors would lead to increased job dissatisfaction, thus facilitating problematic drinking among socially vulnerable individuals. In the model, vulnerable individuals were hypothesized to report lack of participation in nonwork-related activities and/or having met most friends at work. We tested our model using a stratified random sample of union-represented employees from two university occupation categories, Administrative Support and Blue Collar. Data were collected by means of a mailed survey using systematic survey procedures that yielded an overall response rate of 64%. Data analyzed were from 299 surveys completed from a (weighted) respondent sample of 409. Data were analyzed using SPSS Complex Samples 13.0. Both direct and indirect effects of the work stressor and job dissatisfaction variables on employee problem drinking were estimated in two-steps by employing path analysis techniques. Each estimated equation controlled for the sociodemographic variables of gender, occupation category, and education. The work stressors of job stress, work alienation, and workplace social support were found to have significant direct effects on job dissatisfaction (R² = .546, p < .001). In addition, job dissatisfaction was found to predict significantly employee problem drinking (Nagelkerke R² = .136). Therefore, support was found for a simple stress-induced model of employee problem drinking in response to work stressors and resultant job dissatisfaction. Next, we tested the hypothesized moderator effects of the individual social vulnerability variables on the job dissatisfaction-employee problem drinking relationship by employing hierarchical moderated regression techniques. The main effects for all sociodemographic, individual social vulnerability, work stressor, and job dissatisfaction variables were entered on the first step, followed by a block of two-way job dissatisfaction-by-individual social vulnerability interaction terms. No evidence was found for moderating effects on the job dissatisfaction-employee problem drinking relationship; however, the main effects of job dissatisfaction (OR = 1.03; p = .005), lack of participation in nonwork-related activities (OR = 1.03; p = .004), and having met most friends at work (OR = 6.24; p < .001) were found to be associated with employee problem drinking as compared to employees who drink non-problematically. These findings suggest that alcohol abuse prevention programs developed and implemented by social workers should focus on reducing work stressors that lead to job dissatisfaction, which in turn is associated with employee problem drinking. In addition, these programs should consider the lack of employee participation in groups outside of work, and in particular, the dynamics of employee-coworker friendships as potential risk factors for the development and maintenance of employee problem drinking.