Project Summary/Abstract Substance misuse is inconsistently addressed in routine medical practice. Primary care physicians frequently fail to diagnose substance abuse when presented with early symptoms of alcohol or drug use. Patients report that less than one-third of primary care physicians carefully screen them for substance abuse, in spite of available and effective screening protocols. Screening, Brief Intervention, and Referral to Treatment (SBIRT) is an integrated, evidence-based approach aimed at delivering early intervention for drug and alcohol misuse in medical settings and has been demonstrated to reduce substance misuse. Motivational Interviewing (MI), an underlying component of SBIRT, is an evidence-based, brief interventional approach that has been demonstrated to be highly effective in triggering change in high-risk lifestyle behaviors. Unfortunately, the MI approach is very difficult to teach to practitioners as it is counterintuitive to most providers who commonly take a directive approach with patients. Because of ineffective training, MI and SBIRT are under-utilized. The primary goal of this project is to transform MI and SBIRT training with a software-based training solution that analyzes practitioner responses and gives immediate feedback to improve MI skills. Our team has been developing and testing a training tool that uses natural language processing (NLP) to provide Real-time Assessment of Dialogue in Motivational Interviewing (ReadMI). ReadMI is a low cost, ultra-portable solution for accurate and automatic real-time assessment of dialog during the course of MI training. It makes use of the latest advances in deep-learning-based speech recognition and mobile and cloud computing technologies to enable instantaneous MI training assessment and analysis. Currently functioning with 92% accuracy, ReadMI provides immediate feedback to MI and SBIRT trainees regarding the number of open- and closed-ended questions asked, provider-vs-patient conversation time, and other integral components of MI. ReadMI will advance the skill development of physicians and other healthcare professionals in their use of MI and SBIRT, minimizing the need for time-consuming reviews of recorded training sessions. Our aims are: 1) develop ReadMI to provide MI training metrics with 95% accuracy and 2) test ReadMI in a randomized controlled trial (RCT) with 200 medical students and residents, 100 who will receive SBIRT/MI training with ReadMI and 100 who will serve as the control group, receiving customary SBIRT/MI training. The prospective applications of ReadMI extend far beyond substance use disorder (SUD) prevention, detection, and intervention. The effectiveness of all chronic disease management is largely dependent on patient behavior. MI is a patient-focused conversation between the provider and the patient that reinforces the patient's motivation to make positive changes in health-related behavior. By making efficient MI training available to...