227. Utilization of Social Network Analysis To Understand Birth Control Uptake In Youth Who Experience Homelessness

Abstract

Within the U.S., 4.3% of 13-17-year old’s and 12.5% of 18-25-year old’s experienced homelessness during the past year. Female homeless youth become pregnant at a rate five times greater than that of the general U.S. population, with nearly half becoming pregnant at least once during adolescence. Developing trusted, healthy relationships is a challenge for youth who are homeless. Networks can represent a potentially diverse groups of people who provide tangible or intangible resources to the individual, but also may encourage engagement in unhealthy or risky behavior. This exploratory study examines implementation data from a teen pregnancy prevention program to better understand how homeless youth social networks share information and how that information can impact the uptake of birth control. To measure interpersonal relationships, we used a social network analysis (SNA) to assess the quality and strength of peer relationships. SNA is an ideal technique for this study because the study population is part of a bounded network defined by program participation. Eleven 14 to 18 years old participants took part in this study. Most participants were sexually active (n=9) and had first engaged in consensual sex at 13 to 18 years (M=15 years) of age. None of the participants were using birth control before the program. Data were collected through interviews and participants were asked which peers in the program they knew, who they trusted, and if they shared (or received) information on birth control. Data was symmetrized and density and centrality statistics were analyzed. At the end of the program, 73% of the network participants were on some type of birth control at one-month follow-up. However, network density showed that only 14.3% of the network shared information about birth control. To understand if one or a few people controlled the information, centralization was analyzed on birth control information sharing. The whole network centralization was 0.339. This means that the sharing information network was centralized (> than 0.25). Degree centrality was then examined to understand who controlled information flow. One participant had a degree of 8 ties and four others had a degree of 5 ties each. Four of these 5 participants adopted some form of long-acting reversible contraception (LARC) by the end of the program. One person was using condoms. While overall network density was not very high, centralization was significant which means that a few key players controlled the information flow. The majority of these individuals (4 of 5) adopted LARC by the end of the program, which may indicate that the information they shared with their peers about birth control influenced other youth’s decisions on adoption. This exploratory study expands the limited research on the impact of social support on homeless youth and pregnancy prevention. 73% birth control utilization, one-month post-intervention, among a high-risk subpopulation that was not previously on birth control represents important findings for research on homeless youth.