This award will advance the national health and prosperity by enabling effective monitoring strategies for chronic depression and suicide ideation. Mitigating depression has become a national health priority, as it affects 1 out of 10 American adults. Depression related mortality and morbidity result in significantly degraded emotional and physical well-being as well as loss of productivity. This project addresses two interwoven challenges to effective depression monitoring: modeling the complex dynamics of a depression trajectory, and understanding the heterogeneity of the population susceptible to depression as a whole. By analyzing population-level electronic health records, the project will lead to a data-driven decision framework to enable adaptive care strategies for treatment of chronic depression and prevention of depression-related suicide. It will improve the quality of life of American adults who are living with chronic depression. The project represents an interdisciplinary collaborations among researchers in engineering, psychiatry, and health services. The accompanying educational plan aims to broaden STEM interest in high-impact public health problems, and to provide opportunities for women and underrepresented minorities. This research will develop a decision framework and associated computational platform to design adaptive depression and suicide monitoring strategies, informed by knowledge learned from data regarding depression trajectories and suicide ideation patterns. This research will also develop a scientific foundation to evaluate the cost-effective monitoring strategies to provide population-level policy recommendations. Methods include statistical modeling, optimization, and decision analysis to improve mental health services. The framework will be able to 1) effectively analyze the collected heterogeneous depression trajectories of a patient population and proactively probe new trajectories, 2) adequately characterize the disease dynamic process that govern these trajectories and design adaptive monitoring schedules, and 3) adopt rigorous cost-effectiveness analysis to evaluate how effectively the monitoring strategies can detect adverse depression outcomes.