Injection drug use and crack cocaine smoking: Independent and dual risk behaviors for HIV infection

Clyde B. McCoy, Shenghan Lai, Lisa R. Metsch, Sarah E. Messiah, Wei Zhao

Research output: Contribution to journalArticlepeer-review

65 Scopus citations


Purpose Previous studies have examined the practices of injecting drugs or smoking crack cocaine as high-risk, but independent, factors for HIV transmission. To explore the independent and dual risks of injection practices and crack smoking, this study examined HIV seroprevalence rates among distinct drug user groups, based on patterns of daily administration. Methods A sample of 3555 drug users and neighborhood controls in urban Miami, FL and rural Belle Glade and Immokalee, FL were partitioned into four mutually-exclusive groups: 1) injection drug users (IDUs); 2) crack-cocaine smokers; 3) dual users who both smoked crack and injected drugs; and 4) non-drug-user controls. Results HIV seroprevalence rates were 45.1% for IDUs, 30.5% for dual users, 20.1% for crack smokers and 7.3% for controls. Multivariate logistic regression analysis found that when compared with controls odds ratios for HIV seropositivity were 9.81 for IDUs, 5.27 for dual users, and 2.24 for crack smokers. Conclusions These findings provide evidence of: 1) behavioral and structural co-factors that influence HIV exposure patterns among drug users; and 2) the substantially higher risk of HIV infection among IDUs compared with other drug users. Intervention strategies must be tailored for the specific drug use subpopulations to optimize efficacy.

Original languageEnglish (US)
Pages (from-to)535-542
Number of pages8
JournalAnnals of epidemiology
Issue number8
StatePublished - Sep 2004
Externally publishedYes


  • Crack Cocaine
  • HIV Infection
  • Injection Drug Users

ASJC Scopus subject areas

  • Epidemiology


Dive into the research topics of 'Injection drug use and crack cocaine smoking: Independent and dual risk behaviors for HIV infection'. Together they form a unique fingerprint.

Cite this