Ecological Assessment of Proximal Risk Factors for Suicide During Care Transitions

Project Summary/Abstract Suicidal ideation and behavior are growing public health problems in the United States. This is particularly concerning during the high-risk period following discharge from psychiatric hospitalization when risk for suicide is among the highest rates observed. Unfortunately, our current ability to predict suicide is only slightly above chance. This poor predictive ability may be due to an overreliance on distal/cross-sectional suicide risk factors alone, rather than an emphasis on proximal risk factors (the dynamic cognitive, affective, and behavioral experiences occurring in the days, hours, and minutes prior to suicide). The proposed study builds upon our team's existing infrastructure and expertise in digital phenotyping and ecologically valid models of suicide risk. We will examine how affective (e.g., negative affect) and contextual (e.g., social interactions, cognitions, & sleep disruption) proximal risk factors interact to influence suicidal ideation and confer risk for suicidal behavior, particularly during the high-risk period following hospitalization. We will recruit and follow-up with 240 psychiatric inpatients hospitalized for suicidal ideation or behavior over a five-year study period. Participants will complete baseline assessments including clinical interviews, laboratory-based behavioral assessments, and self-reports of distal risk factors associated with suicide. During hospitalization, participants will begin completing ecological momentary assessments (EMAs) of affect, cognition, behavior, and social/environmental context. Following discharge, participants will continue EMA monitoring for 4-weeks at baseline and again for two 3-week EMA “bursts” at 3- and 6-month follow-up assessments. During each EMA monitoring period, we will collect digital phenotyping data (e.g., sleep, location, activity level/movement) using wearables and smartphone sensors. We will conduct 3- and 6-month follow-up assessments in our lab to re-administer lab-based behavioral assessments and collect outcome data including suicidal ideation, suicidal behavior, and re-hospitalization. The proposed study aims to identify proximal risk factors for suicide and use intensive longitudinal methods to characterize the relationships among affective and contextual proximal risk factors while also examining how distal risk factors moderate the relationships among proximal risk factors and between proximal risk factors and suicidal ideation and behavior. We will examine these relationships both before and after discharge from psychiatric hospitalization. The results of this study have the potential to substantially enhance our understanding of suicide phenomenology as it exists in the real world, particularly across the high-risk period following psychiatric hospitalization, with the potential to improve our ability to predict, prevent, and ultimately treat suicidal thoughts and behaviors using traditional and technology-enhanced interventions.