The primary aim of this study was to describe the movement of adolescents and young adults into and out of drug use and to predict heavy drug use. The data source is the Department of Labor's National Longitudinal Survey of Youth, which began in 1979 with a sample of 12,686 adolescents aged 14-21. After 17 rounds and 19 years, the response rate in 1998 was 87 percent. A majority of the sample reported no drug use in five index rounds during the 15-year period between 1984 and 1998. Most of the persons using any drug in a given round never used the drug again or used the drug for one additional round. Very few individuals reported drug use in all five index rounds. Reported findings include: (1) Youth who wait longer before their first use of alcohol, cigarettes, marijuana, cocaine, or crack are less likely to become heavy cocaine users; (2) Youth who first used cigarettes, alcohol, and marijuana late more closely resemble non-users with regard to their heavy cocaine use; (3) Youth who started smoking cigarettes daily, using cocaine, or using crack late have heavy cocaine use percentages more like early users than non-users; (4) Those who smoked marijuana more than 50 times as adolescents are more than six times as likely to become heavy cocaine users as those who did not smoke marijuana as adolescents; (5) Young male drug users are almost twice as likely as female drug users to become heavy cocaine users; (6) Those who were suspended from school are one and one-half times more likely to become heavy cocaine users as those who were not suspended from school; (7) Individuals reporting a significant amount of illegal income as adolescents are more than two times more likely to become heavy cocaine users than those who had no illegal income as adolescents; (8) Those who attended religious services at least twice a month are one third less likely to become heavy cocaine users as those who did not attend religious services; and (9) Those selling hard rugs during adolescence are twice as likely to become heavy cocaine users as those who did not sell drugs as adolescents. Selected variables are appended. (Contains 14 footnotes and 30 exhibits.)