Citation

Fuller CM, Vlahov D, Arria AM, Ompad DC, Garfein R, Strathdee SA. Public Health Rep. (1974) 2001; 116(Suppl 1): 136-145.

Affiliation

Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, NY 10032, USA. cf317@columbia.edu

Copyright

(Copyright © 2001, Association of Schools of Public Health)

DOI

unavailable

PMID

11889281

PMCID

PMC1913680

Abstract

OBJECTIVE: The purpose of this study was to evaluate the extent to which demographic, sexual, and non-injection drug use practices predict adolescent initiation of injection drug use. METHODS: Street recruited injection drug users 15-30 years of age in Baltimore, Maryland, who initiated injection within five years of study enrollment, completed a questionnaire that included a year-by-year history regarding the five years prior to initiation of injection. Factors associated with initiation during adolescence (< or = 21 years of age) versus young adulthood (>21 ) were determined using logistic regression. RESULTS: Of 226 participants, most were female (61%) and African American (64%). Median age of participants was 25; median age at initiation of injection was 23. Factors significantly associated with adolescent initiation in multivariate analysis included race other than African American, and practices prior to initiating injection including condom use, lack of cocaine use, exclusive crack smoking just prior to initiation, and smoking marijuana. Adolescent initiates also had shorter durations of illicit drug use prior to initiating injection. CONCLUSION: Short-term non-injection drug use, particularly exclusive crack smoking, was associated with adolescent initiation of injection drug use. Early prevention efforts targeting this high-risk group of younger drug users are warranted in order to delay or prevent onset of injection drug use.

Language: en