HIV/sexually transmitted infection (STI) prevention interventions are only modestly successful among youth, particularly for young people of colour and sexual and gender minority (SGM) youth. Even among disparate intervention modalities delivered with high fidelity, differences between intervention types have been minimal. One consistent theme has emerged: the role of the youth:provider relationship in predicting intervention response. In line with calls for examination of relational factors, the next essential step is a harmonised analysis to evaluate connections between the youth:provider relationship and co-occurrence of alcohol and cannabis use, in youth HIV/STI prevention intervention response.

Our team has completed six sizeable HIV/STI behavioural prevention studies, generating n=1136 independent youth (baseline M_age=17, range=13–24; 43% female; 21% SGM; 54% Hispanic; 9% African American; 7% Native American/Alaska Native) who received prevention programming and were followed at 1-, 3-, 6-, and/or 12 months. We will harmonise these studies and build a longitudinal mixed-effects machine learning model, with youth:provider relationship as a predictor of intervention response. Participant factors, provider factors and their interaction will be included in the model. Given high rates of alcohol and cannabis comorbidity, we will also examine syndemic outcomes (co-occurring HIV/STI risk behaviours, alcohol use and cannabis use). These data are crucial to informing next step HIV/STI and syndemic intervention programming with this age group.

This secondary analysis study is exempt from human subjects regulations under category 4(iii) as determined by the Institutional Review Board at UConn Health. Results will be disseminated via presentations at annual scientific conferences, submissions to peer-reviewed journals, to mental health and substance use providers, as well as community programmes for youth at high risk for HIV/STI and substance use.