Past Exposure to Stressful Life Events and Sexual Risk Behaviors Among Justice-Involved Couples

Background & Purpose: Men under community supervision have disproportionally high rates of stress and HIV/STIs. Research shows that overall stress is associated with sexual risk behaviors and communication skills on sexual risk behaviors. Associations between types of stressful life events (SLEs) and sexual risk behaviors and communication skills to reduce risk are unknown. Impacts of partner-experienced stress on sexual risk behavior and communication about sexual risk also remain unclear. Examining the association between a sex partner’s stress on sexual risk in the partnership is critical to designing interventions to reduce sexual risk behaviors and increase communication skills to reduce risk. This paper focuses on the following research questions: 1) What is the prevalence/pattern of stressful life events among males under community supervision and their female sexual partners? 2) Are there differences by gender, race, and ethnicity? 3) How do individual and partner exposures to stress impact sexual risk behaviors?

Methods: Data came from the baseline survey of a randomized clinical trial of an HIV/STI prevention intervention (2013-2016). Participants included males with substance-use history recruited from community supervision in New York City and their primary female sexual partner (n=460 individuals within 230 couples). Measures were self-reported sexual behaviors, SLEs, and demographic information. Latent class analysis identified patterns of stress using indicators from the Stressful Life Events Questionnaire. Individual assignment to stress class was based on highest probability of class membership. Multinomial logistic regression identified associations between gender, race, ethnicity, and stress class. Individual and partner stress classes served as independent variables, with sexual risk behaviors as dependent variables, in multilevel mixed-effects Poisson regression models with random effects for couples (p<0.05). Models controlled for gender and marital status. Analysis used Stata 16.

Results: We found four stress classes: Class-1 (low exposure 46.7%), Class-2 (high exposure to street-based violence/deprivation 17.0%), Class-3 (high exposure to household-based violence/deprivation 25.0%) and Class-4 (high exposure 11.3%). Males had higher odds of being in Class-2 (RRR = 5.3) and lower odds of being in Class-3 (RRR = 0.3). Black individuals had lower odds of being in Class-4 (RRR = 0.3). Black/Latino people had higher odds of being in Class-4 than in Class-1 (RRR = 6.5). Individual effects of Class-2, 3 and 4 on certain sexual risk behaviors, as compared to Class-1, included: Class-2 and 3 discussed condoms less with their partners (aIRR = 0.3, aIRR = 0.3); Class-3 had lower condom use self-efficacy (aIRR = 0.7) and discussed HIV less (aIRR = 0.3); Class-4 had higher numbers of sexual partners (aIRR = 1.6). Having a partner with stress Class-4 was also associated with higher numbers of sexual partners (aIRR = 1.5).

Conclusions & Implications: Our study suggests that SLEs of males under community supervision and their primary female sexual partners occur in distinct patterns. We describe associations between stress classes and sexual risk behaviors and distinctions by gender, race, and ethnicity. Our study has implications for tailoring services to improve sexual health outcomes for an already vulnerable population—individuals of color disproportionately impacted by SLEs and justice involvement.