Mechanisms of Change for an Effective Alcohol Text Message Intervention

点击翻译按钮获取中文摘要

 DESCRIPTION (provided by applicant): Young adults ages 18-25 have high rates of hazardous alcohol use and alcohol-related consequences, but are rarely exposed to alcohol interventions. The Emergency Department (ED) setting provides an important opportunity to identify young adult hazardous drinkers who could benefit from an alcohol intervention. Among young adult hazardous drinkers seen in the ED, a text message (TM) intervention has demonstrated effectiveness in reducing hazardous alcohol use (binge drinking) through 9-month follow-up. However, the "active ingredients" (TM strategies that facilitate reductions in binge drinking) and "mechanisms of change" (processes occurring in the individual that lead to reduced binge drinking) of TM intervention for binge drinking remain unclear, and impede progress in constructing mobile interventions with maximum impact and durability. To address this gap, a dismantling design is needed to determine the TM intervention's component effects, and to identify the mechanisms through which components operate, which will have broad implications for the design of mobile interventions and informing theories of behavior change. This project will determine whether TM intervention "active ingredients" involve the critical combination of "goal prompt" (i.e., TM prompt to commit to a low drinking quantity goal), "feedback" (e.g., real-time TM feedback on willingness to set a drinking goal), and "cued self-monitoring" or each of the three components separately. TM intervention is a methodological innovation in research on mechanisms of behavior change because of high fidelity of intervention delivery via computerized algorithm, and collection of real-time data on response to specific intervention components. Young adult (ages 18-25) ED patients who screen positive for hazardous drinking/AUD (N=1,875) will be recruited to participate in a randomized trial of a 12-week TM intervention to reduce binge drinking. Patients will be randomized to one of five conditions: (1) "Cued Self-Monitoring" (TM control condition), (2) "Goal Prompt", (3) "Feedback", (4) all 3 TM components combined, and (5) Assessment Only (control condition). All participants complete assessments at baseline (ED visit), 3- and 6-months post-randomization. Participants in the four TM arms complete weekly TM queries during the intervention phase to track changes in drinking behavior and proposed mediators of intervention effects (intention, motivation, implicit cognitions) in order to determine component-specific change mechanisms. We also will examine individual differences (e.g., baseline alcohol severity) associated with intervention effects, to identify "for whom" treatment has effects. Determining "how" and "for whom" TM intervention has effects will be used to revise TM intervention content to maximally impact identified mechanisms, and has broader implications in terms of determining key components to include in mobile interventions more generally. Further, fine-grained data on response to intervention collected in real time will enable a paradigm shift based on the development of a more dynamic model of behavior change.