Incidence rates of alcohol and drug use disorders (AUDs and DUDs) are consistently higher in men than women, but information on whether sociodemographic and psychiatric diagnostic predictors of AUD and DUD incidence differ by sex is limited. Using data from Waves 1 and 2 of the National Epidemiologic Survey on Alcohol and Related Conditions, sex-specific 3-year incidence rates of AUDs and DUDs among United States adults were compared by sociodemographic variables and baseline psychiatric disorders. Sex-specific logistic regression models estimated odds ratios for prediction of incident AUDs and DUDs, adjusting for potentially confounding baseline sociodemographic and diagnostic variables. Few statistically significant sex differences in predictive relationships were identified and those observed were generally modest. Prospective research is needed to identify predictors of incident DSM-5 AUDs and DUDs and their underlying mechanisms, including whether there is sex specificity by developmental phase, in the role of additional comorbidity in etiology and course, and in outcomes of prevention and treatment.