Data from a representative sample of US adults revealed that 24% of ma
le life-time drinkers and 15% of female life-time drinkers met the DSM
-IV criteria for life-time alcohol dependence, i.e. dependence during
the year preceding interview or in any 12-month period prior to that y
ear. The median interval from first drink to onset of dependence was 3
.6 years for men and 3.0 years for women. After using survival techniq
ues to adjust for potential gender differences in the exposure to risk
of developing alcohol dependence, the cumulative conditional probabil
ity of having experienced onset of dependence was 35.1% for men and 24
.6% for women. The conditional probability of onset of dependence was
equal for men and women in the first year after initiation of drinking
, about 30% higher for men in the period 1-4 years after the first dri
nk, and about 45% higher for men thereafter. After using proportional
hazard; models to adjust for the effects of age cohort, race and ethni
city, family history of alcoholism and age at first drink, these perio
d-specific risk ratios remained virtually unchanged. Including a measu
re of average daily ethanol intake during periods of heaviest consumpt
ion rendered most of the gender differences statistically insignifican
t, revealing a slight excess risk of female dependence within the firs
t year after initiation of drinking among the heaviest drinkers and le
aving an excess male risk of dependence mostly among individuals with
average daily intakes of less than one ounce of ethanol. The results s
uggest that different frequencies of binge drinking might help to acco
unt for these remaining gender differences and that men's and women's
relative risk of developing alcohol dependence may vary as a function
of life cycle stage, with men's excess risk greatest in the college/yo
ung adult years.