Abstract
Background: To examine associations between measures of stressful life events exposure and alcohol abuse/dependence (AAD) from ages 18 to 30 using data from a longitudinal birth cohort (n= 987 to 1011). Methods: Outcome measures included DSM-IV (American Psychiatric Association, 1994) AAD symptoms and AAD, at ages 20-21, 24-25, and 29-30 years. Exposure to a range of stressful life events was measured during the periods 18-21, 21-25, and 25-30 years using items adapted from the social readjustment rating scale (Holmes and Rahe, 1967). Data were analysed using Generalised Estimating Equation models, adjusted for non-observed sources of confounding using conditional fixed effects regression. Further analyses examined: gender. ×. life events exposure interactions, structural equation modelling of possible reciprocal causal pathways linking stressful life events and AAD symptoms, and an alternative conceptualization of the stressful life events measure. Results: After adjustment, those with the highest exposure to stressful life events had rates of AAD symptoms that were 2.24 (p<. .0001) times higher, and odds of AAD that were 2.24 times higher(p<. .01), than those at the lowest level of exposure. Associations between life events exposure and AAD symptoms were stronger for females than for males (p<. .05), with results consistent using a count measure of stressful life events. Structural equation modelling showed that the best-fitting model was one in which life events influenced AAD symptoms. Conclusions: The results suggest that there were persistent linkages between stressful life events and AAD, providing support for a stress-reduction model of alcohol consumption.