Alcohol use has been linked to unintentional and intentional injuries, the leading cause of death among young people. Over 50,000 persons die annually from alcohol-related injury deaths. Analyses of the National Longitudinal Alcohol Epidemiologic Survey (NLAES) reveal that starting to drink at a younger age is associated with a greater likelihood of developing alcohol dependence and patterns of frequent heavy drinking. Moreover, even after controlling for personal history of alcohol dependence and f requency of heavy drinking, persons who began drinking at age 14 or younger relative to those who wait until they are age 21 or older are 2-4 times more likely to be unintentionally injured, in motor vehicle crashes, or be in physical fights after drinking both as adolescents and adults. We plan to conduct additional secondary data analyses of the 1992 NLAES (N=42,862), the 2002 National Epidemiologic Survey of Alcohol-Related Conditions (NESARC, N=43,093) and the 1999 College Alcohol Survey (N=14,138)to explore whether age of first drinking and intoxication is related to antisocial behaviors that produce injury and other alcohol problems. We also propose a random digit dial survey of 3,000 18-39 year olds to further explore whether and why starting to drink at an earlier age is related to: 1. development of alcohol dependence and frequent heavy drinking later in life even among non dependent persons. 2. persons placing themselves and others in situations that increase their risk of motor vehicle, and other unintentional and intentional injuries after drinking, even when history of alcohol dependence and measures of frequent heavy drinking are controlled analytically. 3. engaging in other risky behaviors (e.g. not wearing seat belts, carrying guns or other weapons, getting into physical fights, having unplanned and unprotected sex, smoking, using drugs, attempting suicide) and whether these behaviors are more likely to occur than when respondents have been drinking. Secondary data analyses will extend over the entire project. Piloting the new survey will occur in year 1, data collection years 2-4 and analyses year 5.