The association between unemployment trajectories and alcohol consumption patterns. Evidence from a large prospective cohort in The Netherlands.

Unemployment is expected to influence alcohol consumption, but studies show mixed results, partly because most studies concentrate on current employment status. However, unemployment could be particularly consequential if it is part of a trajectory of employment precariousness. Moreover, the association between unemployment and alcohol consumption may not be homogeneous across the population, but differ by subgroups (e.g. socioeconomic status). This study longitudinally analyses the association between different employment trajectories and alcohol consumption, and examines if the association is moderated by socioeconomic status (SES), partner status, age and gender. Four waves of data of the Lifelines Cohort study are used. Sample consists of individuals from 18-50 years old, active in the labor market (n = 104,766) from the northern provinces of the Netherlands. Employment trajectories are defined by employment status in each wave, duration of unemployment, and number of exposures to unemployment. Drinking patterns are divided into "abstainers", "moderate drinking" (<1.5 drinks/day), "heavy drinking" (≥1.5 drinks/day) and "binge drinking" (≥5 drinks/occasion; 4 for women). The associations are estimated with multinomial logistic regression models. Results show that recent, long-term unemployment (≥ 6 months) is associated with higher rates of heavy drinking (RRR = 1.26 [95 % CI 1.03-1.54]), whereas short-term unemployment does not show any association with the outcome. Being continuously unemployed throughout the observation period shows a strong association with binge drinking (RRR = 1.43 [95 % CI 1.06-1.93]), as well as reporting 2 or more long unemployment spells (RRR = 1.49 [95 % CI 1.21-1.83]). The group of abstainers (77.95 % women) have significantly lower SES, and poorer health than their peers. For some individuals, recent unemployment increases the likelihood for abstinence (RRR = 1.23 [95 % CI 1.00-1.51]). Evidence suggests that length of unemployment is key in order to grasp its effects in terms of changing drinking patterns.