Adolescent drug use initiation and transition into other drugs: A retrospective longitudinal examination across race/ethnicity

Saijun Zhang, Shiyou Wu, Qi Wu, Daniel W. Durkin, Flavio F. Marsiglia

Research output: Contribution to journalArticlepeer-review

18 Scopus citations


Background: Understanding adolescent drug use mechanisms is critical for drug use prevention. Although some theories such as the gateway theory suggest that drug users gradually transition into using more addictive drugs, there is no consensus about such a hypothesis. One important factor that hinders the advancement of knowledge in this area is the scarcity of longitudinal studies examining the type of drugs adolescents initially use and the different pathways adolescents take to transition into using other drugs as they grow older. Methods: Using the pooled sample of adolescent dug users (14–17 years old; n = 10,644) from the National Survey on Drug Use and Health (2015–2018), we constructed longitudinal data on adolescents’ illicit drug use history other than the use of tobacco and alcohol based on the age of drug initiation. This allowed us to investigate what drugs were initially used by adolescents, how the use of these drugs may have progressed into a new drug, and whether there were racial/ethnic differences in the initiation and progression. The retrospective longitudinal data analyses applied life table method and Cox regression models. Results: Two thirds of the adolescent drug users initiated their drug use trajectories with marijuana, one quarter with inhalants, and the remaining with hallucinogens, prescription drugs, and hard drugs. Adolescent drug users who initiated with different drugs showed unique trajectories to the use of a new drug. By year 8, the probability of using a new drug was about 40% and 70% to 80% for adolescents who initiated with inhalants and other drugs, respectively. The probability of using a new drug for adolescents who initiated with marijuana and inhalants accumulated stably over time, and its difference with that of other drug users diminished over time. The multivariate Cox regression models suggest the observed discrepancies generally held after controlling for covariates. There were also racial/ethnic differences in adolescent drug use initiation and progression, with Black/African American adolescents being the least likely to switch to the use of a new drug. Conclusion: Adolescents’ initial use of marijuana and inhalants may lead to substantial risks of using other drugs over time. It is therefore important to screen adolescent drug use comprehensively and provide early interventions to prevent an escalation to more detrimental drugs. The findings provide new evidence to support aspects of both the gateway and generalized risk drug use theories.

Original languageEnglish (US)
Article number106679
JournalAddictive Behaviors
StatePublished - Feb 2021


  • Adolescent
  • Drug use initiation
  • Drug use progression
  • Gateway theory
  • Generalized risk theory
  • Longitudinal data analysis

ASJC Scopus subject areas

  • Medicine (miscellaneous)
  • Clinical Psychology
  • Toxicology
  • Psychiatry and Mental health


Dive into the research topics of 'Adolescent drug use initiation and transition into other drugs: A retrospective longitudinal examination across race/ethnicity'. Together they form a unique fingerprint.

Cite this