PROJECT ABSTRACT Binge drinking (defined as drinking five or more drinks on one occasion for men, and four or more drinks for women), is highly prevalent in the US. Binge drinking accounts for more than half of the 80,000 annual deaths attributed to excessive alcohol consumption and its economic costs exceeds $191 billion in the US. National HIV Behavioral Surveillance data indicate that 48% of heterosexual men, 58% of men who have sex with men, and 40% of heterosexual women reported binge drinking (past 30 days). Binge drinking has been independently associated with condomless sex and HIV infection. Binge drinking is by far the most prevalent exposure linked to HIV infections. Current interventions for binge drinking and binge-drinking associated sexual behaviors remain limited. Most binge drinkers have never participated in alcohol or drug treatment programs and the vast majority of binge drinkers (90%) do not meet criteria for severe AUD. Real-time data on binge drinking and sexual episodes around binge drinking can inform interventions and there is a great need to examine the psychosocial factors that precede and predict these behaviors—i.e., the antecedents of binge alcohol and binge-drinking associated sexual risk behaviors. Ecological momentary assessments (EMA) is a method of collecting real-time data about an individual’s behaviors and experiences, often administered via cell phones. EMA can capture the contextual factors and circumstances preceding drinking behaviors, thereby providing a more nuanced understanding of the immediate antecedents of binge drinking. Recent advances in biosensor technology include the availability of wrist-worn monitors that can collect data on alcohol use, estimate number of drinks, and identify binge drinking episode using transdermal alcohol concentration (TAC). Hence, real-time data from EMA and biosensors for TAC can identifying high-risk situations and opportunities for personalized interventions. This study will develop risk prediction models for binge drinking and binge-drinking associated sexual behavior, which can ultimately help inform the development ecological momentary interventions (EMI) that will target high risk periods for binge drinking and binge-drinking associated sexual behaviors. Research Design Summary: This study, entitled “BEATS: Binge drinking Ecological Antecedents with Transdermal alcohol monitoring Study,” will enroll 100 sexually-active adults who binge drink alcohol and assess real-time antecedents for binge-drinking and binge-drinking associated sexual behaviors using EMA during a 30- day follow-up. Participants will also wear a small wrist biosensor that will track alcohol consumption and measure TAC continuously for 30 days. Data from the EMA and TAC will be analyzed using machine learning approaches to develop algorithms that predict risk for binge drinking (Aim 1) and binge-drinking associated sexual behaviors (Aim 2). We will also examine the feasibility and acceptabilit...