The Influence of Online Sex Seeking Behaviors on Sexual Risk Outcomes Among Homeless Youth

BACKGROUND AND PURPOSE: The majority of homeless youth report access to the internet. While other studies have reported on protective factors associated with internet utilization among homeless youth, not much is known about the impact of online sex seeking behaviors on sexual risk outcomes for homeless youth. Research has shown that homeless youth report engaging in more sexual risk taking behaviors, such as concurrent partners and no condom use, than their non-homeless peers. These sexual risk taking behaviors can expose homeless youth to negative health outcomes. The increased access to sexual partners using online methods may compound sexual risk taking behavior, creating the potential for additional exposure to sexually transmitted diseases, including HIV for these youth.

METHODS: This study used self-report data from the NIMH funded YouthNet project, which collected responses from 1,046 youth recruited from three homeless service agencies in Los Angeles. Questions regarding internet usage included measures for online sex seeking, internet and social media use patterns and internet access methods. Sexual risk outcomes included measures for concurrent partners within a 12 months period and unprotected anal and vaginal sex at last sexual encounter. Offline risk factors associated with sexual risk taking were controlled for, including history of physical abuse, sexual abuse and drug and alcohol use prior to sex. Logistic regression with a mediated path analysis was conducted to test for mediation of internet use on sexual risk outcomes.

RESULTS: Of the 1,046 respondents, 448 (44.18%) reported daily internet use, 342 (33.89%) reported daily social media use and 323 (32.11%) reported internet access through a cell phone. Online sex seeking was reported by 344 (36.91%) of youth, 335(37.18%) reported concurrent partners, 502(47.99%) reported no condom use at their last sexual encounter. Online sex seeking was significantly associated with sexual risk taking outcomes including concurrent partners (OR=3.715, 95% C.I. 2.626, 5.255) and unprotected vaginal sex (OR=1.619, 95% C.I. 1.104, 2.375). Daily internet use was negatively associated with concurrent partners (OR=.402, 95% C.I. 264, .615) and daily social networking was associated with unprotected vaginal sex (OR=1.578, 95% C.I. 1.006, 2.476). Sobel test results indicated that online sex seeking mediated the relationship with concurrent partners for LGBT youth, youth with histories of sexual trauma and youth reporting drugs and alcohol prior to sex. Daily internet use mediated the relationship with concurrent partners for African American youth and online sex seeking mediated the relationship with unprotected vaginal sex for males.

CONCLUSIONS AND IMPLICATIONS: While prior research has shown how connections to pro-social networks via the internet can reduce sexual risk-taking for homeless youth, the current study findings indicate that sexual risk taking behaviors for homeless youth may be further compounded by online sex seeking. Service agencies that provide internet access to homeless youth should be aware of how internet use may impact risk taking behaviors, alongside the previously identified benefits of internet access. In addition, new interventions that target homeless youth who are engaged in online sex seeking are needed, as these youth are engaging in greater HIV risk-taking behaviors.