ABSTRACT As rates of adolescent depression and suicidality continue to trend upwards, the healthcare system struggles to address the need for and lack of mental health service use. The pediatric patient-centered medical home model may improve adolescent depression outcomes by enhancing access to and coordinating care, as well as providing ongoing monitoring. Unfortunately, despite guideline recommendations, over 2/3 of adolescents identified with depression symptoms in primary care do not receive symptom monitoring and 19% do not re- ceive symptom reassessment. This lack of symptom monitoring and reassessment can result in untoward health outcomes including a decrease in functioning, increased use of acute and crisis services, and hospitali- zations due to suicidality. Current technologies which incorporate data passively collected from smartphones offer an opportunity for intercurrent monitoring between patient visits which limits burden on the patient to self- report and limits burden on the healthcare system, allowing primary care teams to triage contacting and as- sessing patients a system identifies with an increase in disease severity. This formative study will demonstrate the usability and potential clinical utility of MoodRing, a technology intervention which will collect passive mo- bile phone sensor data on aspects of adolescent phone use related to depressive symptom severity (e.g. com- munication patterns, social media use, travel) and integrate this data into a multi-user (adolescent, parent, pri- mary care provider/care manager) platform from which symptoms can be viewed and secure communication can occur. MoodRing, as supported by Health Belief Model, may lead to improved quality of depression man- agement (increased symptom reassessment, therapy/medication adherence) through increasing self-efficacy, social support from parent and care team, as well as encouraging application of self-management skills through increased self-management knowledge, skills, and symptom feedback. MoodRing builds on a solid foundation of investigators experienced in design of technology interventions to increase adolescent initiation of depression treatment, who have already developed machine algorithms for passive sensing and a small business partner with vast experience in working with health researchers to develop multi-user web/mobile platforms. This STTR Phase I study seeks to accomplish two aims. The first is to apply a machine learning pipeline developed for college-aged youth to adolescents with depression and determine whether self-reported depressive symptoms can be reliably predicted from passive data with at least 85% accuracy. The second is the user design and system architecture of MoodRing. If milestones are achieved that models are successful at predicting depressive symptoms and the proposed MoodRing intervention is acceptable to adolescents, par- ents, and primary care providers/care managers, then we will pursue the STTR Phase II stud...