The purpose of the new funding opportunity announcement, RFA-OD-17-004 for Intensive Longitudinal Analysis of Health Behaviors: Leveraging New Technologies To Understand Health Behaviors (U01), is to establish a cooperative agreement network to collaboratively study factors that influence key health behaviors in the dynamic environment of individuals, using intensive longitudinal data collection and analytic methods. Importantly, progress has been slow and frustrating in translating knowledge of the brain to new and more effective treatments for human brain diseases such as severe mental disorders. In fact, severe mental disorders, which include psychotic disorders, are brain diseases that are not only devastating because they result in severe disruptions that occur early in life, but, for many, the course of illness is progressive, leading to chronic debilitation and early mortality. Thus the need to accelerate knowledge about the factors that trigger (or increase or decrease the likelihood) of manic and psychotic episodes, and to translate this knowledge to more effective treatment interventions, is critical. The primary goal of the proposed ?Robust Predictors of Mania and Psychosis? is to identify biological, environmental, and social factors that trigger dangerous mental states, particularly mania and psychosis, in individuals known to be at risk for these conditions. The eventual goal of this work is to provide quantifiable and predictable information that can be used to scaffold biological observations and tailor intervention strategies to maximize efficacy at the individual level. We first develop models to predict conventional clinical measures specific to psychosis and mania using (1) digital, low- to-minimal burden interactions through smartphones and wearables (Aim 1), and (2) measures extracted from face and voice during in-person clinical interactions (Aim 2), work which leverages existing data we have already collected. We will next collect one hundred person-years of pseudo-continuous multivariate behavioral data from one hundred individuals with a psychotic disorder, to further test and validate our early observations in a wider array of individuals with affective and non-affective psychotic disorders, who are likely to experience illness fluctuations within a one-year timeframe, employing several strategies to optimize participant engagement (Aim 3). We will also perform, as a representative example, a study comparing sleep, energy expenditure, and mania symptoms over time, using data obtained in the first three aims, to quantify how the relationship between energy expenditure and energy perception varies across our study population in ways that could have important consequences for health behaviors (Aim 4). The main goals of this project are thus to acquire high quality, temporally dense behavioral, cognitive, and clinical data on an important cohort of young adult patients, not only to facilitate future investigations linking these behavioral change points to neurobiological processes but also as a precursor to more effective, targeted therapeutics, such as real-time interventions that could be delivered based on dynamic factors in an individual's environment.
The main goal of the proposed ?Robust predictors of mania and psychosis? is to identify biological, environmental, and social factors that trigger dangerous mental states, particularly mania and psychosis, in individuals known to be at risk for these conditions by acquiring high quality, temporally dense, longitudinal behavioral data, and using machine learning approaches to iteratively identify factors that increase or decrease the likelihood of these conditions. We focus on early psychosis (both affective and non-affective psychosis), which is a critical time period when there are fewer confounds such as prolonged medication exposure and chronicity, and when early intervention strategies will be most effective, prior to the progression that often leads to debilitating and chronic illnesses, to great suffering, and to an enormous public health problem and economic burden. By developing predictors for mania and psychosis from smartphone, wearable, and audio- video data and then testing these predictors over 100 person-years in individuals with active psychotic disorders, the data collected will make it possible to identify specific factors in the physical, social, and/or built environment that facilitate or hinder healthy behaviors in early psychosis and bipolar disorder and lay the groundwork for tailored intervention strategies that fully account for these factors to maximize efficacy at the individual level.
