The goal of the Brigham and Women’s Hospital (BWH)/Harvard Medical School Roybal Center for Therapeutic Optimization using Behavioral Science is to develop principle-driven interventions to enhance the evidence-based use of prescription medications. The objective of the Behavioral Intervention Development (BID) Core is to conduct studies that employ the Center’s principles of multi-disciplinarity, testing principle- driven interventions in real-world settings, explicit testing of mechanisms of action, and using novel analytic and technological approaches to facilitate behavioral interventions at scale. For the first year of this renewal application, we selected two trials which are highly aligned with the goals of the BWH Roybal Center from a pool of innovative ideas. These trials were selected because they address medication issues that impose substantial morbidity, evaluate principle-driven behavioral approaches, leverage advanced analytic and technological approaches, test mechanisms behind behavior change, and are led by multidisciplinary teams. To ensure the successful completion of selected trials, the BID Core will provide research services for every trial funded by the Center, which will improve efficiency of conducting trials and support less experienced investigators. Specific services will include Core Services provided to all trials and Additional (“Ad-hoc”) Core Services accessed by trial investigators as-needed. The BID Core will be led by experienced investigators of multiple Roybal-funded trials who will oversee the conduct of trials and BID Core research staff and facilitate communication with study teams and the Administrative Core. Thus, the overall aim of the BID Core itself is to provide infrastructure for the successful and efficient conduct of pragmatic trials. The first trial for this first renewal year will evaluate the effectiveness of personalizing electronic health record (EHR) tools using reinforcement learning, a machine learning approach derived from behavioral science, for PCPs to support deprescribing of high-risk medications in a large healthcare system. This project is a two-arm randomized trial of 60 PCPs, following their eligible patients over 7 months for deprescribing outcomes measured using EHR data. In addition to evaluating whether personalization is a mechanism that can influence prescribing, we will also identify provider characteristics maximize behavior change. The second trial for this first renewal year will evaluate the effectiveness of a gamification intervention on oral diabetes and cholesterol medication adherence. This project is a two-arm trial of 506 Medicare patients with poor adherence in a large and diverse healthcare system, comparing a gamification intervention that provides behavioral tactics based on patients’ gaining or losing points on their adherence over time. Differences in adherence is evaluated based on prescription fill data over a 6-month follow-up. At the end of the study,...