The longitudinal course of cocaine dependence is characterized by alternating periods of abstinence and relapse. Although gender has emerged as an important predictor of relapse, previous studies have examined mean differences in use by gender. Focusing strictly on differences in averages between men and women does not address potential gender differences in transitions between use and abstinence. Transition rates for men and women were compared using data from the National Institute on Drug Abuse Collaborative Cocaine Treatment Study. Abstinence and nonabstinence for each of the 6 months of active treatment was determined by using a composite measure of use that incorporated information from weekly and monthly self-reports and urine toxicology screenings. Random effects were introduced to describe intersubject heterogeneity in transition rates. In this sample of 454 patients, rates of transition between abstinence and use were significantly different between men and women, with men showing twice the rate of transition between states despite similar average levels of use. These data may have important implications for both treatment planning and the types of outcomes considered in clinical practice and research.