The association between age at onset of drinking and subsequent alcohol-related consequences, including alcohol dependence, is well established in the literature. Although the vast majority of these studies are based on cross-sectional analyses, the association finds support in longitudinal studies (Andersen et al. 2003). The follow-up periods in longitudinal studies, however, tend to be of short duration, limited in number and include large proportions of respondents who have not passed through the risk period for developing alcohol use disorders. Alcohol use tends to increase during the adolescent years until about 21–23 years of age before leveling off and, for the majority, decreasing thereafter. The normative decline in heavy drinking with increasing age has been associated with the assumption of adult behaviors (e.g. employment, marriage, parenting, etc.). Comparable age trajectories of alcohol-related consequences and other problem behaviors suggest a categorization of substance abuse as ‘adolescent-limited’ and ‘life-course persistent’ (Moffitt 1993). Irrespective of these categorizations, early onset of drinking provides a powerful behavioral context in risk exposure among adolescents and young adults, prompting many in the health field to suggest that delaying the onset of alcohol and other drug use is a critical component in the development of preventive and educational efforts.

While early users have greater potential for problems, Robins (1984) cautions that early onset may be an indicator of other psychiatric disorders and is not a direct factor in subsequent problem manifestation. Other studies indicate that the question is indeed much more complicated. Aside from psychiatric disorders, precursors of early onset have been related to personality (i.e. impulsivity), social class (i.e. early detachment from educational pursuits), differential association (i.e. street subculture) and social structure (alcohol availability). Donovan & Jessor (1985) proposed that the associations between problem drinking with other problem behaviors such as poor school performance, precocious sexual behavior, illicit drug use, etc. comprise a single behavioral syndrome which has ties to the broader social environment. Elucidation of the role of precursors in problem behavior must also address the nature of the mediation between precursors and behavioral outcomes. For example, Bates & Labouvie (1997) demonstrate that early risk behaviors have indirect effects on later alcohol-related problem consequences (i.e. the association between early risk factors and subsequent consequences are mediated by the intensity of current use). The paper from Andersen et al. (2003) reveals that intensity of alcohol use at age 15 predicts excessive consumption at age 19 and directs attention to identifying factors which sustain the increase in drinking behavior.

Prescott & Kendler (1999) offer evidence which suggests that early onset and alcohol dependence are both manifestations of a general vulnerability to problem behavior and argue that attempts to prevent the development of alcohol dependence by delaying drinking onset are unlikely to be successful. Because much of the literature on earlier onset is based on secondary analyses in which many critical variables are absent, the complete picture awaits further study. This by no means implies that preventive and educational efforts related to early onset of drinking must remain on hold. While educational efforts appear to have limited impact, social control policies which curb alcohol availability have been shown to be effective interventions. Of course, it remains to be seen whether fuller understanding of the aforementioned associations will lead to more effective prevention.


  1. Top of page