Deep Phenotyping of Heavy Drinking in Young Adults with Behavioral Scales, Neuropsychological Tasks, and Smartphone Sensing Technology

摘要 酒精使用障碍(AUD)在青壮年达到高峰，这使得我们必须了解年轻人特定人的酒精风险概况，以预防AUD或在这种疾病变得慢性之前进行干预。国家酒精滥用和酒精中毒研究所(NIAAA)的神经生物学框架--成瘾神经临床评估(ANA)--为了解这一人群中的AUD提供了一种创新的方法。ANA认为，AUD在3个神经功能领域的个体差异有助于区分AUD的实质性临床异质性。ANA建立在NIMH研究领域标准(RDoC)的基础上，这是一个维度框架，用于根据6个核心生物/心理系统中不同程度的功能障碍来调查精神障碍。作为起点，NIAAA认可了最初的3域ANA模型，该模型与大多数RDoC系统保持一致。两个RDoC系统，不在最初的ANA中，包括睡眠/昼夜节律和社交过程，与年轻人饮酒风险高度相关。三域ANA模型已经在成人研究中得到验证，并预测治疗结果，包括我们团队的工作。它还没有在年轻人身上进行调查。我们建议研究ANA模型，扩展到包括睡眠/昼夜节律和社交过程，在年轻人(非大学/大学，年龄18-25岁)(N=350)中，他们报告最近中度到重度饮酒。具体来说，年轻人将参与一项为期12个月的纵向研究，包括完成自我报告问卷、神经心理任务，以及从事被动和主动智能手机数据收集。这些评估包括推荐/类似的ANA措施、与RDoC相关的睡眠/昼夜节律和社交措施，以及提高ANA可扩展性的新型智能手机措施。考虑到年轻人对智能手机的广泛使用，智能手机数据收集严格、不引人注目、可扩展，并且与年轻人高度相关。智能手机可以通过嵌入式传感器和手机使用日志被动地生成丰富的每时每刻的神经行为数据(例如，移动性、社会性)，并通过调查提示主动生成。这些数字行为指标显示了预测精神障碍症状、病程、治疗反应和大脑功能活动的希望。我们将使用研究参与者的数据来实现以下目标：对于目标1，我们将使用基线自我报告和与三个ANA领域和RDoC睡眠/昼夜节律和社会过程相关的神经心理学测量来验证年轻人ANA模型(ANA-YA)。然后，我们将检查ANA-YA模型和基线饮酒措施之间的基线关联。我们还将探索ANA-YA表型的纵向变化，并测试这些变化是否可以预测12个月的酒精结果。对于目标2，我们将检查智能手机数据与ANA-YA域的基线关联，然后检查智能手机数据的纵向变化，以及这些变化是否可以预测12个月的ANA-YA表型。我们的结果将推动青年AUD神经生物学的科学发展，并确定有效、有效的评估来区分这一群体中的酒精风险。

ABSTRACT Alcohol use disorder (AUD) reaches peak levels during young adulthood, making it critical that we understand person-specific alcohol risk profiles in young adults to prevent AUD or intervene before this disorder becomes chronic. The National Institute on Alcohol Abuse & Alcoholism’s (NIAAA) neurobiological framework, the Addictions Neuroclinical Assessment (ANA), offers an innovative approach for understanding AUD in this population. The ANA posits that individual differences in 3 neurofunctional domains can help differentiate the substantial clinical heterogeneity in AUD. The ANA builds upon the NIMH Research Domain Criteria (RDoC), a dimensional framework for investigating mental disorders in terms of varying degrees of dysfunction in 6 core biological/psychological systems. As a starting point, NIAAA endorsed an initial 3-domain ANA model, which aligns with most RDoC systems. Two RDoC systems, not in the initial ANA, include sleep/circadian and social processes and are highly relevant to young adult alcohol risk. The 3-domain ANA model has been validated in research with adults and predicts treatment outcomes, including work by our team. It has yet to be investigated in young adults. We propose to study the ANA model, expanded to include sleep/circadian and social processes, in young adults (non-college/college, ages 18-25) (N=350), who report recent moderate to heavy drinking. Specifically, young adults will participate in a 12-month longitudinal study, which involves completing self-report questionnaires, neuropsychological tasks, and engaging in passive and active smartphone data collection. These assessments include recommended/similar ANA measures, RDoC-relevant sleep/circadian and social measures, and novel smartphone measures to improve ANA scalability. Smartphone data collection is rigorous, unobtrusive, scalable, and highly relevant for young adults given their extensive smartphone use. Smartphones can generate rich moment-by-moment neurobehavioral data (e.g., mobility, sociality) passively through embedded sensors and phone usage logs and actively through survey prompts. These digital behavioral indicators show promise for predicting psychiatric disorder symptoms, course, treatment response, and functional brain activity. We will use data from study participants to achieve the following aims: For Aim 1, we will validate an ANA model for young adults (ANA-YA) using baseline self-report and neuropsychological measures related to the 3 ANA domains and RDoC sleep/circadian and social processes. We will then examine baseline associations between the ANA-YA model and baseline drinking measures. We will also explore longitudinal change in ANA-YA phenotypes and test whether these changes predict 12-month alcohol outcomes. For Aim 2, we will examine the baseline associations of smartphone data to ANA-YA domains and then examine longitudinal change in smartphone data and whether these changes predict 12-month ANA-YA phenotypes. Our results will advance the science of young adult AUD neurobiology and identify efficient, valid assessments for distinguishing alcohol risk in this group.