Abstract for: Representing lifetime drinking pattern diversity in a quantitative model of alcohol use: a literature-informed approach.

Although alcohol consumption is a quantitative construct, it is discussed in theoretical and empirical alcohol research as having qualitative, categorical levels of use (e.g., stages) with corresponding thresholds associated with levels of risk for harm, and substantial diversity in transitions between stages. Current system dynamics (SD) alcohol models often fail to utilize alcohol research in model development and parameterization, which may impact model validity. To address this, we developed a literature-guided quantitative alcohol SD model and tested prevention/intervention strategies at different stages of the alcohol use continuum. Stocks represented non-risky drinking (NRD), risky drinking (RD), alcohol use disorder (AUD), treatment, and sobriety, and flows allowed for transitions as represented within the alcohol use literature. Simulated model projections were compared to historical data from a target community from 2016-19 to evaluate simulation model accuracy and validity. Results indicated transitioning from NRD to RD was an important area for prevention. Primary prevention of initial transitions from NRD to RD or RD to AUD provided the strongest change compared to intervention (promoting transitions out of RD or AUD) or secondary prevention (preventing transitions back to RD or AUD). This study highlights the strength of using literature to guide alcohol use models.