186. Predictors of Polysubstance Use in Rural High School Youth: A Latent Class Analysis


      Although polysubstance use is a serious health concern for adolescents, little research has explored this issue in rural youth. Study objectives were to: 1) use latent class analysis to identify patterns of polysubstance use in a sample or rural youth; 2) examine risk and protective factors associated with polysubstance use class membership.


      Data were baseline surveys collected prior to a substance use prevention program implemented in rural high school health classes in 2019. High school students aged 14-19 living in rural Indiana areas with high rates of HIV, HCV, and opioid use completed a baseline paper survey. Latent class analysis (LCA; MPlus v.1.8.6) was used to identify and classify patterns (i.e., classes) of lifetime and past 30-day use both trichotomized: Never [0 days], 1-9 times [days], 10+ times [days]) in tobacco, electronic vapor products, alcohol, binge drinking and marijuana. We then used multinomial logistic regression (SPSS, v. 28.0; all p<.05) to evaluate the odds of polysubstance use class membership as a function of demographic (gender [male/female], sexual minority [no/yes] and ethnicity [White/LatinX]), risk (adverse childhood experiences (ACEs) [8 items; alpha=0797], peer pressure susceptibility [5 items; alpha=0.929] and positive attitudes about substance use [16 items; alpha=0.951]) and protective (decision making capability [14 items; alpha=0.545], healthy relationship attitudes [5 items; alpha=0.881], parental communication [3 items; alpha=0.886], parental connection [5 items; alpha=0.952] and school connection [5 items; alpha=0.867]) factors.


      The 311 participants were (49% male, 1% non-binary, 87% heterosexual, 37% 9th graders, 54% 10th graders, 9% 11th and 12th graders). 32% reported ever using tobacco, 46% vapes, 57% alcohol, 34% binge drank and 25% marijuana. Use in the past 30-day reports were lower (14%: tobacco, 26%: vapes, 21%: alcohol, 9%: marijuana). LCA results supported a 4-class solution of polysubstance use (i.e., 4 distinct patterns of substance use): Regular Polysubstance Users (14%), Lifetime Alcohol, Vape & Tobacco Users (17%%), Moderate Lifetime Alcohol & Vape Users (19%), and Abstainers (50%). In multinomial regression models, older grade level significantly predicted membership in the Lifetime Alcohol, Vape & Tobacco class compared to Moderate Lifetime Vape & Alcohol class (OR=3.07) or the Abstainers class (OR=3.77). Higher prevalence of ACEs predicted greater odds of being a Regular Polysubstance User vs. a Moderate Lifetime Alcohol & Vape User (OR=1.47) or an Abstainer (OR=1.81). Higher peer pressure susceptibility was associated with being a Regular Polysubstance as compared to a Moderate Lifetime Alcohol & Vape User (OR=1.15) or an Abstainers (OR=1.15). More positive attitudes about substance use predicted being a Regular Polysubstance User (OR=1.19), a Lifetime Alcohol, Vape & Tobacco User (OR=1.16) or a Moderate Lifetime Alcohol & Vape User (OR=1.09) as compared to being an Abstainer.


      Rural youth exhibit distinct patterns of lifetime and recent polysubstance use that can be used to tailor interventions. Our findings highlight the importance of trauma-informed interventions that target factors like peer influence and substance use attitudes to mitigate polysubstance use.

      Sources of Support

      Support for this project is provided to Health Care Education and Training, Inc. (HCET) through Grant Number 1TP1AH000201-01-00 from the HHS Office of Population Affairs.