Abstract Significance: Trichotillomania (TTM), characterized by repeated pulling of one’s hair to the point of hair loss or thinning, produces impairment in physical, social, psychological and occupational functioning. While evidenced-based treatments (EBTs, i.e. self-monitoring, stimulus control, and Habit Reversal Training (HRT)) do exist, access is limited due to lack of trained treatment providers as well as lack of willingness to seek treatment due to the shame and guilt associated with TTM. This project builds on the success of our Phase I project, where we validated a discreet wearable device’s ability to detect the subtle motions associated with TTM, thereby improving awareness of the behavior. We further incorporated HRT into the device and app system. We also found a significant reduction of trichotillomania symptom severity in a pilot trial. We plan to further incorporate the EBTs for TTM into the system and test for efficacy in a 70-person RCT in this Phase II. Hypothesis: It is hypothesized that a device and app system can deliver personalized and relevant EBTs at the right time. It is further hypothesized this system would be clinically effective in reducing TTM severity relative to an active control condition. The system will identify possible TTM episodes using passively detected contextual factors, in addition to the patient’s historical pulling episode record from the device. To build this predictive model, we will use Ecological Momentary Assessment (EMA) to record high fidelity TTM episode frequency and context across 40 participants with TTM. Thus, for example, if a patient typically pulls at a given known place and time (e.g., the library in evenings), the system will deliver an EBT reminder when the user is at that location (e.g., reminding the user to sit in a crowded area and clench fists). Specific Aim 1: We will develop foundational elements to deliver EBTs via the device and app system through iterative feedback with 10 clinicians and 10 patients. We will refine concept materials to deploy a functional assessment of the user’s TTM, determine a comprehensive list of EBT reminders (i.e. stimulus control and Habit Reversal Training) for common situations, and also develop a provider directory and data portal. Specific Aim 2: We will develop the back-end predictive model based on EMA data from 20 patients over 6 weeks. EBT reminders will be tested for accuracy on a new set of 20 patients over 6 weeks. The HabitAware app and device will then be modified to deploy the predictive EBT reminder model. Specific Aim 3: We will first conduct an Open Trial of ten participants with the device and app system to ensure system functionality. We will then conduct a 70-participant RCT with an active control “reminder bracelet with app journaling” condition to establish efficacy based on change in TTM symptom severity. Long-Term Goal: The overall system will be commercialized and allow patients access to a clinically effective system for ...