This is a request for an administrative supplement for years 03 through 05 of the Predicting Risk Imaging Suicidal Minds (PRISM) study (MH116652). In the ?parent? grant, we proposed to study 300 young adult (aged 18-30) suicidal ideators (SI) (about half of whom will have made a suicide attempt (SA), 100 never-suicidal psychiatric controls (PCs), and 100 healthy controls (HCs), use fMRI to assess neurosemantic signatures (NSSs) in response to 28 words related either to suicide, positive emotion, or negative emotion at intake and 3 months, and assess for SI and SA at intake, 3, 6, and 9 months thereafter. The goals are to determine if: (1) NSSs are sensitive to changes in level of SI when repeated at 3 months; and (2) whether NSS can predict trajectories of SI and SA upon prospective follow-up. The overarching goal is to establish reliable neurocognitive markers of SI and SA in individual participants, and to assess these measures? ability to predict future SI and SA. Thus far, the project has recruited 86% of target and has a retention rate of 86%. In a pilot study of 17 SI and 17 HCs, we were able to discriminate SI from HC with 91% accuracy and identify those SI who had a history of a SA with 94% accuracy. We request an administrative supplement to add an EEG measure of NSSs, to be administered after the fMRI (within 2 days) in 60 SI, 30 PC, and 30 HC, in order to see if a machine learning algorithm can correctly classify those with SI and SA, and determine how this classification compares to that obtained via fMRI. Our pilot work for this EEG study comes from an R21 led by Dr. Just that showed that EEG could accurately classify which concept participants were thinking with an accuracy of 0.73, around 80% of the accuracy obtained via fMRI.
The aims of this supplement are to determine if: (1) EEG-based NSSs evoked by specific words can discriminate SI from PC and from HC; (2) the extent of EEG-based NSSs evoked by specific words will correlate with the severity of current SI; (3) EEG-based NSS?s evoked by specific words can identify which participants with SI have had a history of an SA; and (4) whether EEG-NSSs can predict trajectories of SI. The costs of recruitment and assessment are already covered by the parent grant, so that we are only requesting funds for participant payments and the materials and personnel costs to administer, process, and analyze the EEG NSS and equipment. Because EEG is available in most hospital-based clinics and much cheaper than fMRI, if this approach can be demonstrated to be accurate, it may enhance clinically applicability of NSSs to assessment and treatment.
This project takes a novel approach to the study of suicidal risk by examining the differences in brain activation patterns between suicidal and non-suicidal young adults when thinking about concepts related to suicide, negative, and positive emotions while undergoing neuroimaging. In this administrative supplement, we request to add an EEG measure of neural activity in response to words related to suicide, positive, and negative emotions to learn if this more readily available assessment approaches the accuracy of fMRI. If this project is as successful as our preliminary work, it will advance clinical practice by improving clinicians? ability to: (1) detect and monitor suicidal risk, (2) understand alterations in thinking and feelings related to suicide in their patients; and (3) develop personalized treatment strategies for their suicidal patients based on their altered patterns of thinking and feeling that can more precisely and effectively reduce suicidal risk.
