Maturing out of alcohol dependence: the impact of transitional life events.

OBJECTIVE: The purpose of this study was to investigate the effects of transitional life events related to education, employment, and family formation on the likelihood of recovery from alcohol dependence as defined by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), distinguishing the short- and long-term effects of these events and potential effect modification by treatment history, gender, and severity of dependence. METHOD: This analysis is based on data from the Wave 1 2001-2002 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), a cross-sectional, retrospective survey of a nationally representative sample of U.S. adults 18 years of age and older. The analytic sample consisted of 4,422 individuals with prior-to-past-year (PPY) onset of DSM-IV alcohol dependence. Time-dependent proportional hazards models were used to estimate the effects of completing school, starting full-time work, getting married, becoming separated/divorced/widowed, and becoming a parent on the outcomes of nonabstinent recovery (NR; e.g., low-risk asymptomatic drinking) and abstinent recovery (AR). RESULTS: Entry into and exit from a first marriage each increased the likelihood of NR during the first 3 years after those events occurred (hazard rate ratio HRR = 1.37 and 1.76, respectively). However, individuals who were still dependent 3 or more years after those events occurred had a decreased likelihood of subsequent NR (HRR = 0.70 for both events), as did those who were still dependent 3 or more years after completing schooling (HRR = 0.54). The likelihood of AR was more than doubled in the 3 years after first becoming a parent (HRR = 2.22) but was decreased among individuals still dependent 3 or more years after starting full-time work. For the outcome of NR, all of the negative effects associated with still being dependent 3 or more years after the occurrence of key life events were more strongly negative among individuals with less severe cases of dependence. CONCLUSIONS: Transitional life events demonstrate many effects on recovery, including both direct effects consistent with role socialization and associations more reflective of selectivity than causation. Taken as a whole, these events appear to contribute to (but by no means fully explain) the high rates of recovery from alcohol dependence that have been observed even in the absence of treatment.