Prevalence and predictors of sexual risks among homeless youth

Linda L. Halcón, Alan R. Lifson

Research output: Contribution to journalArticlepeer-review

153 Scopus citations


This study examined prevalence of sexual risks among homeless adolescents and described factors associated with those risks. Community-based outreach methods were used successfully to access this difficult-to-reach population. The sample included 203 homeless youth aged 15-22 recruited from community sites. Questionnaire items addressed demographics, sexual behaviors, alcohol/drug use, STI history and testing, and pregnancy history. In cross-sectional analysis, 58.7% of males and 75.6% of females reported recent sexual intercourse. Of those, one third of males and half of the females used no barrier method with at least one partner. Both gender (female) and race/ethnicity (non-Black) were associated with having intercourse without a barrier contraceptive method. Over one fifth reported a history of "survival sex" or receiving money, drugs, clothing, shelter, or food for sex. These results show disturbingly high rates of a number of sexual risks, reinforcing the need for targeted interventions with this highly vulnerable population.

Original languageEnglish (US)
Pages (from-to)71-80
Number of pages10
JournalJournal of Youth and Adolescence
Issue number1
StatePublished - Feb 2004

Bibliographical note

Funding Information:
This project was supported by Grant NIH 1-RO3CA79388-01 through the National Cancer Institute. During preparation of the manuscript, Dr. Halcón was supported in part by a nursing training grant, T80 MC 00021 (Center for Adolescent Nursing, Director: Bearinger) from the Maternal and Child Health Bureau (Title V, Social Security Act) Health Resources and Services Administration, Department of Health and Human Services.


  • Adolescents
  • Homeless
  • Sexual behavior
  • Youth


Dive into the research topics of 'Prevalence and predictors of sexual risks among homeless youth'. Together they form a unique fingerprint.

Cite this