Gender Differences in the Pathways from Sexual Abuse to Risky Sexual Behavior in Homeless Youth

Background:  Sexual abuse (SA) prior to becoming homeless is commonly reported among homeless youth and can lead to adverse health outcomes, such as risky sexual behavior. While the mechanisms that underlie the association between SA and risky sexual behavior are well-documented, little is known about how this may be different for men and women. Interventions targeting sexual risk reduction for men and women may benefit from an understanding of gender-specific mediating mechanisms between SA and sexual risk behaviors. This study seeks to identify the ways in which gender moderates the relationship between SA and risky sexual behavior among a sample of both men and women homeless youth, and how mediating mechanisms may differ by gender.

Methods: These data draw from the YouthNet study, which included 4 waves of recruitment at three drop-in centers within the Los Angeles area. Participants (n= 1,046) consist of homeless youth who were recruited between 2011 and 2013. Respondents completed a computerized self-administered interview in English or Spanish.  Participants also had the option of listening to the questions and response options through headphones. Questions included participants’ sexual abuse history, prior to and after becoming homeless were included, along with items assessing post-traumatic stress disorder, gang involvement, intimate partner violence, suicidal ideation, and substance use frequency. These mechanisms were used in logistic regression to assess differences in paths from SA to risky sexual behavior, by gender.

Results: Participants ranged in age from 13-29 (m= 21.34, sd=2.16), were predominantly comprised of men (71.57%), and the majority identified as heterosexual (75.15%). Findings indicate a high presence of self-reported SA (20.16%) and 36.69% were re-victimized while homeless. Risky sexual behavior was assessed in the past year and included measures of concurrent sex (37.18%), in the past 3 months and included exchange sex (17.46%), and the past month and included unprotected anal sex (9.94%), unprotected vaginal sex (42.73%) and having 2 or more partners (20.2%). Men with SA who reported using cocaine in the past 30 days were 1.85 times as likely than non-using men without past SA to have multiple partners (SE= .399, p= .032) and 1.96 times as likely to engage in concurrent sex (SE=.438, p=.025).  Men with SA who used heroin were 1.66 times as likely to engage in concurrent sex (SE= .642, p= .048) and those who used meth were 2.76 times as likely to report exchange sex (SE= .551, p< .001). Reporting suicidal ideation significantly increased the likelihood of engaging in unprotected vaginal sex by 1.92 (SE= .703, p= .027) for men. For women, those with SA and PTSD were 2.24 times as likely to engage in exchange sex (SE= 1.725, p= .012).

Conclusion: Results indicate paths in the association between SA and risky sexual behavior differ between men and women homeless youth. Trends in substance use as a mechanism in this relationship warrants its inclusion when addressing HIV risk behavior with men. These findings also indicate a need to incorporate mental health treatment of traumatic symptoms when targeting sexual risk prevention and intervention for women.