Research That Matters (January 17 - 20, 2008)

Purpose: The number of HIV/AIDS infections acquired during adolescence is on the rise. At least half of the approximately 40,000 new HIV infections per year in the US are among youths and young adults under 25 years old. The primary risk for acquiring HIV/AIDS during adolescence comes from high-risk sexual behaviors as well as drug use. Violence exposure, traumatic experiences and mental health problems have also been shown to increase risk, and individuals with service histories in child welfare, juvenile justice or mental health are considered particularly vulnerable for infection. However, little is known about how factors combine to increase risk. Understanding the constellation of factors that contribute to increased rates of health-risking sexual behaviors can aid in the development of tailored prevention interventions. The current analysis examines risk factor profiles in a high-risk sample of young injection drug users (IDU). Particular attention is paid to respondents' service histories.

Methods: This research uses data from a NIDA-funded cross-sectional social network study on risk and protective behaviors among 320 young female IDUs and their partners. Using a multi-method sampling strategy, persons between the ages of 15-23 years and reporting injecting drug use and/or having sex with an IDU in the last 30 days were recruited. Respondents were 40% female and 60% male with an average age of 20.7 (SD=3.6). Two-thirds were Caucasian. Almost three-quarters reported being currently homeless. Data were collected via 60-90 minute interviews and captured sociodemographic information as well as abuse and violence history, alcohol and drug use history, mental health and service history using population adapted standardized measures. For this analysis, Latent Class Analysis (LCA) was used to identify different risk factor profiles within this high risk group.

Results: Descriptive data depict a population with high-risk current behaviors and histories of multiple risk factors. Over one-half reported a history of child maltreatment with one-fifth reporting having been sexually abused. Over 90% reported having been victimized by violence and 37.9% reported having perpetrated violence. The average age of onset of consensual sex was reported to be 14.4 (SD=14.2). Condom use was reportedly low (19.1%) and one-third reported having received money or drugs for sex. Results of the Latent Class Analysis, using M-Plus, suggest a 2-class model with 47% and 53% falling into each class, respectively. Class 2 respondents had higher rates of sexual risk behaviors (e.g, earlier onset of sexual behavior, more partners, less condom use, etc.), and overwhelmingly had service use histories in child welfare (81.3%), mental health (91.9%) and/or juvenile justice (70.0%). Mental health indicators (e.g., inpatient psychiatric hospitalization, suicide attempt, reported mental health problems) and episodes in foster care and group homes were the most distinguishing features between the two classes.

Conclusions: Study findings have important implications for the development of HIV prevention interventions. They indicate that such interventions need to go beyond skill-based prevention approaches to explicitly address mental health needs. They further suggest that prevention interventions need to be tailored for delivery in the service systems that serve those most vulnerable to infection.