PROJECT SUMMARY Alcohol use disorder (AUD) is a significant public health problem, yet treatments demonstrate only modest efficacy, likely due to the the profound phenotypic heterogeneity of AUD. In order to improve the efficacy of AUD treatments, it is imperative to better characterize this heterogeneity which may, in turn, elucidate clearer treatment targets for precision medicine approaches. This likely requires shifting conceptualizations of AUD away from clinical description and towards etiologic mechanisms, an approach embodied by the goals of some modern conceputalizations of AUD, such as the Addiction Research Domain Criteria (AARDoC) and Addictions Neuroclinical Assessment (ANA). However, current research suggests that AARDoC and ANA suffer from important shortcomings, including limited clinical efficiency, and may therefore benefit from further development, refinement, and validation. To address these shortcomings, the proposed project aims to (a) empirically test the models articulated by modern conceptual AUD etiological frameworks, including the ANA, and (b) derive a mechanism-based computerized adaptive test (CAT) assessment of AUD developed using principles of objective test construction and community-based participatory research strategies. First, a candidate set of self-report items indexing 13 etiologic domains articulated by the Etiologic, Theory-Based, Ontogenetic Hierarchical (ETOH) framework of AUD mechanisms, which serves as a recent extension of AARDoC/ANA, will be derived from the literature and two rounds of cognitive interviews will be used to refine the item set among a diverse group of participants (N = 50) with hazardous or harmful alcohol use. Next, items will be administered to a combined community and clinical sample (N = 1,200) to empirically test the structure of items and determine the best-fitting model. Item response theory will then be used to calibrate the items for the purpose of building a CAT for each of the domains identified (e.g., reward, cognitive control, negative emotionality). Using the refined and calibrated item set and domain-specific CATs, data will be collected from an additional independent sample of heavy drinkers (N = 100). Ecological momentary assessment over 14 days and a follow-up assessment will also be conducted with the goal of evaluating the psychometric properties of the CATs in ecologically valid contexts and over time. Specifically, to determine if the domain-specific CATs demonstrate validity (e.g., convergent, discriminant, predictive) and reliability (e.g., test-retest) across diverse patient populations and sex/gender groups. All research aims will be conducted alongside and in consultation with individuals with lived experience of AUD to ensure the measure is acceptable, feasible, and adequately contextualized. This project is consistent with NIAAA's Strategic Plan, specifically Goal 1 (Identify mechanisms of alcohol-related pathology) and Goal 2 (Improve diagnosis and tracking of AUD). The resulting measure also has the potential to support progress towards Goal 4 (Develop and improve treatments for AUD) by facilitating the assessment of mechanisms that may serve as viable targets in AUD treatments, including behavioral and pharmacological interventions.

PROJECT NARRATIVE To improve alcohol use disorder treatment, particularly through precision medicine approaches, it is imperative that alcohol use disorder etiologic mechanisms are identified and assessed in a way that is reliable and valid across diverse groups of individuals. The proposed study will use a multimethod approach to develop and validate a computerized adaptive test of alcohol use disorder etiologic mechanisms. Ultimately, the goal of this work is to establish an efficient and comprehensive assessment of mechanisms that can be used in service of determining the optimal treatment, whether behavioral, pharmacological, or both, for a given person with alcohol use disorder.