Improving longitudinal mental health monitoring to generate personalized guides and timely referrals

This proposal addresses the need for a scalable and engaging method for patients to track their mental health so they can receive high-quality, effective, and patient-centered care recommendations and referrals. It aims to decrease the ~50% gap of untreated mental illness3 and shorten the average decade-long time to care2 from symptom onset to days-weeks. Tracking a patient’s progress can help assess the effectiveness of treatment, lead to timely adjustments, facilitate early identification of adverse events or complications, improve communication between patient and provider, and provide a comprehensive view to understand long-term patterns and outcomes of treatment. Until now, there has not been a good way to simultaneously track, triage, and tailor appropriate mental health recommendations and referrals cohesively and automatically. The for-profit MiResource will partner with the non-profit Child Mind Institute (CMI) to address this gap by integrating separate software solutions, curating comprehensive guides on mental health topics, and using AI to generate personalized guides. CMI has clinical experts to help select appropriate assessments, edit guides, and evaluate the relevance of guides and referrals. For over a decade, CMI’s editorial staff has composed over 1,000 highly accessed, online, freely available mental health articles and guides that will serve as one of the sources of guides for the project. CMI has developed a configurable mobile mental health platform called MindLogger, which has already been integrated into MiResource’s platform through an Application Programmer Interface to administer assessments. CMI is also developing a mental health database and recommender system called Mosaic and a multimodal journaling app called Mirror, all of which are well-suited to the assessment and recommendation components of the project. MiResource provides the referral infrastructure, a software-as-a-service solution designed to help connect people to local mental health providers and programs, and a network of 70 US colleges that use MiResource for their students. Goal: Build a scalable, personalized, user-friendly, and expert-informed digital ecosystem for continuous journaling and assessment, guidance, and referral for adolescents and college-age students. A combination of mobile apps, an online provider directory, and AI-personalized guides will automate triage, tracking, and tailored recommendations and referrals. Phase I will (Aim I-1) build longitudinal assessments to monitor therapeutic progress and automate referrals; (Aim I-2) create a standard format for generating comprehensive mental health guides; and (Aim I-3) integrate assessment, journaling, recommendation, and referral software as a unified mental health platform. Phase II will (Aim II-1) build comprehensive mental health guides in Aim I-2’s format, manually curated by content experts; (Aim II-2) generate personalized guides in Aim I-2’s format using AI and based on journal entries; and (Aim II-3) clinical experts and participants evaluate AI-generated, personalized guides and referrals to ensure clinical relevance of Aim I-3’s integrated platform.