Rapid innovation of precision psychiatry interventions using dynamic systems modeling and ecological quasi-experiments - PROJECT SUMMARY/ABSTRACT The US faces a mental health crisis; nearly 83 million Americans experience mental illness, but fewer than 50% of these individuals accessed past year mental health services. There is a critical need for highly disseminable, potent mental health interventions. Standalone digital interventions offer promise for improving clinical outcomes at scale; however, efficacy of these interventions to date is modest, likely due in part to insufficient personalization to patient heterogeneity and to momentary changes in mental health symptoms. The present study seeks to rapidly improve the efficacy of precision psychiatry digital health interventions by developing real-world, person-centered maintenance models for psychopathology using ecological (i.e., in patient’s daily lives) data and dynamic systems modeling. Dynamic systems modeling of repeated time series data yielded by ecological momentary assessment (EMA) and smartphone sensors will be used to evaluate the individual and interactive effects of empirical maintenance factors to understand each individual’s maintenance system for psychopathology in their daily life. Just-in-time adaptive interventions (JITAIs) are prompts delivered in-the-moment via smartphone at identified instances of risk for maladaptive behaviors. Although often evaluated in aggregate across months, JITAIs could be conceptualized as ecological quasi-experiments that induce use of specific therapeutic skills at observable times. Accordingly, this study will use micro-randomized (in which the content of the JITAI is randomly assigned in the moment) JITAIs delivered at times of elevated risk for maladaptive behavior based on the individual’s developed dynamic system to evaluate the effects and mechanisms of specific therapeutic skills on outcomes of interest, to provide granular insight into treatment mechanisms. Eating disorders are the ideal population in which to test proof-of-concept for these methods to improve efficacy of precision psychiatry digital interventions, as they are characterized by easily measurable maladaptive behaviors (i.e., dietary restriction, binge eating, compensatory behaviors), are maintained by a complex intersection of biological, psychology, and social factors, and their frontline treatment, enhanced cognitive behavioral therapy (CBT-E), is primarily comprised of behavioral skills. The aims are: 1) test the hypothesis that person-centered dynamic models will describe risk for dietary restriction, binge eating, add compensatory behaviors with ≥ 60% average goodness-of-fit, 2) test hypothesized effects and mechanisms of CBT-E skills on eating disorder behaviors and primary maintenance factors using JITAIs, and 3) characterize feasibility and acceptability of the standalone JITAIs. The study will enroll N = 170 adults with eating disorders who will complete four months of ecological data collection and receive JITAIs to use CBT-E skills at times of elevated risk for 10 weeks of the data collection period. The study will set the stage for future randomized controlled trail to evaluate the optimized, standalone JITAI system, which will have the potential to substantially improve access to evidence-based mental healthcare.
