Data from the 1992 National Longitudinal Alcohol Epidemiologic Study (NLAES) revealed that 44% of U.S. adults 18 years of age and older were current drinkers who had consumed at least 12 drinks in the year preceding the interview. Twenty-two percent were former drinkers, and 34% were lifetime abstainers. These figures represent an 8% decrease in the prevalence of current drinking relative to 1988. The proportion of current drinkers decreased with age, was higher for men than women, increased with education and income, was lower than average among Black and Hispanic adults, was highest among never-married adults and lowest among those who were widowed, was lower in the South than in other regions, and was lower in rural than urban areas. The probabilities of ever having consumed five or more (5+) drinks or having been intoxicated in the past year revealed similar patterns to those already noted, but the probabilities of heavy drinking or intoxication on a weekly or more frequent basis showed no variation by race or ethnicity. Average daily consumption of more than 1 ounce of ethanol differed from the preceding measure of heavy drinking in its variation across population subgroups, declining less sharply with age and exhibiting a U-shaped pattern with respect to income. Examination of the prevalence of heavy drinking among current drinkers rather than within the total population revealed several differences, the most striking reversal being that the probability of heavy drinking decreased with education and income. Multiple logistic regression models predicting the various drinking outcomes indicated that most of the differentials held true after adjusting for intercorrelation among the sociodemographic variables.