Objectives We posited mutually-reinforcing longitudinal pathways between occupation and alcohol use disorders (AUD).

Method Longitudinal trajectories of work substantive complexity were constructed by growth mixture modelling (GMM) of occupational data from the National Longitudinal Survey of Youth 1979 and O*NET work variables. AUD was determined using a set of 25 questions that map onto 9 criteria for alcohol dependence. Prevalent AUD was one occurred at or before the first alcohol questionnanire in 1989; an incident AUD was a new case between 1989 and the 1994 survey. The association between work trajectories and AUD was modelled using Poisson regression in a generalised linear model adjusting for covariates.

Results Lower work trajectories were associated with higher AUD prevalence (prevalence ratio 1.41; 95% confidence interval 1.04–1.91 for lowest versus highest class). Incident AUDs were associated with risk of a decline in work trajectory class in both high (OR=2.68; 95% CI 1.34–5.35) and low (OR=1.62 95% CI 1.01–2.60) initial classes. Interval educational attainment was not associated with AUD within a specific occupaitonal trajectory class

Conclusions Low occupational trajectories are associated with increased AUD prevalence. Incident AUDs may be followed by a decline in work SC. These findings suggest self-reinforcing relations between the development of AUD and both prior and subsequent work.