Dominic Denning was awarded 2 years of funding from the National Institute of Mental Health's competitive F31 initiative ($90,639; 3rd percentile) to develop predictive algorithms for suicidal thoughts and behaviors among socially marginalized college students. His project will leverage ecological momentary assessment and passively derived data from people's phone usage to generate digital phenotypes during a 28-day study period. A subset of the data will be used to train machine learning algorithms to predict suicidal thoughts and behaviors and then test the utility of these algorithms on another subset of participant data to determine the viability of these methods in accurately forecasting suicide risk.
This work is particularly important as traditional top down/theory driven approaches are barely better than chance at predicting suicidal thoughts and behaviors. Thus, bottom up/data driven approaches may yield more accurate suicide prediction as people may be motivated to underreport suicide risk for social desirability or to avoid crisis intervention. Findings from this study may alter how clinical and research professionals monitor suicide risk and augment access to crisis interventions in the moments when people need it most.