AIMS: To examine if (1) there is a positive association between drinking volume in young men and life-time risk of alcohol dependence (AD) and (2) there are other associations between young adulthood factors and life-time risk of AD.

DESIGN: Prospective cohort study of sons of fathers with alcohol use disorder (AUD) and matched low-risk controls without paternal AUD. Setting and participants A total of 204 men, who were assessed at baseline in 1979 at age 19-20 years, were followed through record linkage with Danish registers and consecutive psychiatric interviews at the ages of 33, 43 and 53 years.

MEASUREMENTS: AD diagnoses were interview-based according to the Diagnostic and Statistical Manual of Mental Disorders, 3rd edition, or made by treating clinicians according to the International Classification of Diseases (ICD) revision 8 (ICD-8) until 1993 and revision 10 (ICD-10) from 1994.We estimated odds ratios (ORs) with 95% confidence intervals (CI) for the development of AD after adjustment for confounders including smoking, social status and paternal AUD.

FINDINGS: The following variables from the examination at age 19-20 independently predicted life-time AD: alcohol consumption > 21 beverages/week versus 0-21 [odds ratio (OR) = 2.46, 95% confidence interval (CI) = 1.22-4.97], police contact (OR = 2.60, 95% CI = 1.28-5.28) and institutionalization related to the individual (OR = 2.90, 95% CI = 1.39-6.02). Compared with < 1 beverages/week, the risk for AD did not increase significantly for drinking volume categories: 1-7, 8-14 or 15-21 beverages/week.

CONCLUSION: Independently of other risk factors in young adulthood, young Danish men's risk for life-time alcohol dependence appears to be predicted by a drinking volume at age 19-20 years exceeding 21 beverages per week.