ECPN Student Poster Contestant: Delinquency, Alcohol Acceptability, and Unprotected Sex Among LGBTQ Adolescents Living in the Midwest

Introduction

Condom use is effective for the prevention of HIV/ bacterial STI’s and pregnancy. According to the CDC, adolescents and young adults account for over 50% of new STI cases. LGBT sexual minority youth are disproportionately represented among new STI cases. The objective of this study is to explain the contextual effects of peer microsystems that include delinquency and alcohol acceptability on individual risk behaviors of LGBTQ sexual minority youth. Of particular focus is the co-presence of alcohol or drugs and condom non-use during coitus.

Methods

The Social Ecological Model guides the analysis of the behavioral health data from the CDC’s 2015 Youth Risk Behavior Survey (YRBS). The outcomes of interest are individual level condom non-use and alcohol or drug use. Individual level covariates include: sex, grade, race, sexual orientation, sexual preference, missing school because feeling unsafe, dating violence, cigarette use, alcohol use, marijuana use, forced to have sex, and number of sexual partners. Strata level variables capture the peer risk microsystems of delinquent behavior and substance abuse to assess the effects on individual level outcomes. Delinquency included: being threatened or injured with a weapon at school and being in a physical fight at school. Substance abuse included: alcohol use, marijuana use, cocaine use, ever bullied, and ever offered, sold, or possessed drugs at school. The analysis consisted of a 3-level hierarchical linear model to assess the social ecological effects that group/strata and state level factors have on individual level predictors and the outcome. Analysis relied on HLM v7.02 software.

Results

Results from the analysis of 2,790 students across 129 strata and 4 states (Illinois, Indiana, Kentucky, and Michigan) indicate that the probability of the co-presence of alcohol and/or drug use, and condom non-use at the last sexual encounter is a function of coercive sex (β = 0.094892, p = 0.013), the number of sexual partners in the last 6 months (β = 0.037057, p = 0.006), and being an LGBTQ sexual minority adolescent (β = 0.139097, p = <0.001). The alcohol acceptability between the states is significantly different (β = 0.670726, p = <0.001) and increases the probability of the co-presence of alcohol or drug use and condom non-use at the individual level (β = 0.045652, p = 0.02). Of particular note, LGBTQ sexual minority adolescents reported higher rates of the co-presence of alcohol, or drug and condom non-use compared to heterosexual adolescents (68.35% vs 44.29%)

Conclusions

Alcohol and/or drug use, and condom non-use are nested in a complex causal web of substance abuse, drug acceptability in the environments, individual experiences of sexual coercion, the number of recent sex partners, and identifying as an LGBTQ sexual minority adolescent. The tendency to conflate Midwestern states rather than considering them separately often results in overlooking the unique environmental alcohol and drug acceptability effects that may contribute to the individual models of risk. While the literature shows that alcohol and drug use do not predict condom non-use, alcohol use is often a precursor to a sexual event. Promotional efforts to decrease condom non-use should address alcohol use prior to coitus.