Abstract
PURPOSE: To identify associations of past-year witnessing violence with expectancies (anticipated effects) for alcohol and cannabis use in Black, Latinx, and White youth, including possible variations by level of neighborhood advantage and/or race/ethnicity. METHODS: Data were drawn from Follow-up 3 of the Adolescent Brain Cognitive Development Study (n=7,332; weighted distributions: 45.53% girl, 52.33% boy, 2.14% other gender; 11.80% Black, 25.13% Latinx, 63.07% White; weighted mean age=12.94 [SE=0.01]). Positive and negative expectancies were measured using the AEQ-AB for alcohol and the MEEQ-B for cannabis. Neighborhood disadvantage was captured via the Area Deprivation Index (ADI) and analyzed as quartiles. General linear models were fitted to data for each of the four expectancies scores, adjusting for socioeconomic status indicators and testing for witnessing violence by race/ethnicity interactions. Quartile-specific regression coefficients were produced. RESULTS: Witnessing violence was most prevalent in the highest (most disadvantaged) ADI quartile. Across quartiles, positive alcohol expectancies (βs:0.12-0.26) and positive cannabis expectancies (βs:0.20-0.38) were elevated in youth who witnessed violence; associations were weakest in the lowest quartile. Associations with negative expectancies were non-significant for alcohol and lower only in the second highest quartile for cannabis. All race/ethnicity interactions were non-significant. CONCLUSIONS: Risk conferred by witnessing violence manifests early in the development of alcohol and cannabis use, shaping anticipated positive effects even before many youth initiate use. In addition to lower exposure, residing in an advantaged neighborhood may modestly mitigate risk associated with witnessing violence for developing positive expectancies, underscoring the importance of intervening early with youth in disadvantaged neighborhoods.