Although recent data from animal models indicate that adolescent ethanol exposure increases self-administered ethanol intake in adult rats, the impact of age at first drink on the association between stress and drinking has not been studied in humans. Data collected in the 2001 to 2002 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) were used to estimate the extent to which age at first drink modified the association between stress and average daily volume (ADV) of ethanol intake in a sample of 26,946 past-year drinkers. Successive models estimated the magnitude and significance of the interaction between age at first drink (ages 14 or younger, 15-17, and 18 or older) and number of stressors (out of 12 past-year negative life events) after (1) adjusting for sociodemographic characteristics, (2) additionally adjusting for family history of alcoholism, comorbid psychopathology, adolescent, and past-year tobacco and illicit drug use, and (3) additionally adjusting for all other significant interactions with number of stressors. Even after adjusting for a wide range of confounders and their interactions with stress, initiation of drinking at ages 14 and younger increased the association between the number of stressors and ADV of ethanol consumption by 8% (p=0.014), when considering the full range of 12 potential stressors. In fact, the positive association between stress and consumption was significant only for this group of drinkers with early adolescent exposure to ethanol. Within this group, ADV of consumption increased by an average of 7% with each additional stressor experienced, although the exact percentage increase varied as a function of other covariates that had significant interactions with stress. When a reduced set of 4 stressors was considered, the magnitudes of the associations were mostly unchanged, but the modifying effect of age at first drink fell short of statistical significance (p=0.309) in the fully adjusted model. The findings of this study are consistent with the argument that early-onset drinking may increase stress-reactive ethanol consumption; however, these findings need to be replicated in an experimental human study in order to control fully the direction of the relationship between stress and consumption.