Creating and Evaluating the Predictive Utility of Risk Phenotypes for Bipolar Spectrum Disorders in Adolescence

About This Grant

PROJECT SUMMARY/ABSTRACT: Bipolar spectrum disorders (BSDs) are associated with major personal and public health burdens. Despite this heavy burden, the etiology of BSD is not fully understood. Further research on risk factors for BSD during adolescence, when likelihood of first onset of a BSD is highest, is needed to understand how BSD onset and symptoms can be better predicted and interventions delivered earlier. Determining the degree of risk for BSD conferred by various predictors is a vital step toward creating intervention and prevention programs that can identify individuals most at risk in order to reduce the likelihood of BSD onset, delay onset, or lessen course severity. Extant research has established several person-level factors that confer risk and influence dysregulation throughout the course of BSDs. The social and circadian rhythm model of BSDs posits that social and circadian rhythm dysregulation can result in mood symptoms and episodes. In another separate line of research, evidence suggests that hypersensitivity to rewards confers risk for BSDs. Researchers have suggested that the reward and circadian models of BSD risk and course can be combined into a joint, bidirectional model, such that disturbance in one of these systems, through a feedback loop, may promote dysregulation in both systems, contributing to mood symptoms and episodes. Additional theoretically and empirically supported predictors can be combined statistically with reward and circadian factors to better predict risk of bipolar symptoms. These factors include family history of BSDs, hypomanic personality, higher trait impulsivity, exposure to childhood adversity, affective lability, and substance use. However, the means by which predictive factors may be combined to better inform risk for bipolar symptoms is poorly understood. Although myriad risk factors for BSDs have been identified, little work has been done to statistically integrate information obtained through a multimodal approach to determine which individuals are most at risk. Thus, the proposed project seeks to evaluate empirically derived risk groups based on multimodal assessment of multiple risk factors for BSD during adolescence, a critical developmental period in which onset of BSDs is most likely. I will use participants from my sponsor's R01 study, which aims to examine the interplay of reward and circadian factors longitudinally to predict first onset of BSDs, add measures of additional risk factors, and statistically integrate these multimodal risk indicators with latent class analysis to evaluate the predictive utility of empirically-derived risk groups. My sponsors and I have designed a training plan involving coursework, workshops, experiential learning, and mentorship that will allow me to develop greater expertise in the development of mood pathology, learn advanced statistical methods required for this project, and gain the skills necessary for my future career as an independent clinical scientist. The proposed study will take place in Temple University's clinical psychology Ph.D. program, which has a successful track record of conducting impactful NIH-funded research and training clinical research scientists.