Objective: The purpose of this study is to examine the association between conduct disorder before age 15 and subsequent alcohol dependence, and to describe the lifetime prevalence of alcohol dependence among Navajo Indian women and men. Method: This was a case-control design which included both men (n = 735) and women (n = 351) and in which the Diagnostic Interview Schedule was used for the diagnosis of the lifetime history of alcohol dependence and conduct disorder. Alcohol dependent cases were selected from inpatient and outpatient treatment programs (204 men, 148 women). Whenever possible, controls were matched for age, sex and community of residence and were randomly selected and interviewed until a nonalcohol dependent individual was found. Among the men, there were 374 alcohol dependent controls and 157 nonalcohol dependent controls. Among the women, the figures were 60 and 143, respectively. When combined, the controls comprise samples of the adult male and female populations from which estimates of lifetime prevalence of alcohol dependence, and of the amount of alcohol dependence in the population attributable to conduct disorder, may be inferred. Results: Conduct disorder is a risk factor for alcohol dependence among both men and women. Lifetime prevalence of alcohol dependence in this population is high (70.4% for men and 29.6% for women), but the amount of alcohol dependence in the population attributable to conduct disorder is low. On the other hand, among the alcohol dependent, those with conduct disorder had the most severe alcohol- and nonalcohol-related problems. Conclusions: The potential limitations of the study are those common to case-control designs, especially biased recall by cases. There are also potential sampling biases among the controls. It is shown that none of the potential biases invalidate the findings, which support the hypothesis that in this population conduct disorder is a risk for alcohol dependence. The implications for primary prevention of alcohol dependence are discussed.