Predicting SUD after Trauma: Leveraging Audiovisual Data in the Emergency Department
Principal Investigators: Dr. Shaddy K. Saba, Assistant Professor, NYU Silver School of Social Work; Dr. Katharina Schultebraucks, Co-Director of the Computational Psychiatry Program and Associate Professor, Department of Psychiatry and Department of Population Health, NYU Grossman School of Medicine
Dates of award: 4/1/2025–8/31/2025
Amount of award: $25,000
Patients treated in emergency departments after traumatic events are at increased risk of developing substance use disorder (SUD), yet identifying who is most at risk remains challenging. This newly funded pilot study will investigate whether audiovisual data capturing speech, facial expressions, and movement can improve the prediction of new-onset SUD among trauma survivors.
The research team will leverage data from an ongoing study of 350 trauma survivors admitted to emergency departments. Participants are recorded discussing the event that brought them to the hospital, and SUD outcomes are assessed through clinician diagnoses in electronic health records and, for a subset of participants, self-report assessments. The team will first develop a machine learning model using electronic health record data to predict SUD during the six months following an emergency department visit. They will then test whether adding audiovisual data improves the model’s performance.
Findings from this pilot study will inform larger efforts to validate predictive algorithms for trauma-exposed patients. Ultimately, this work could support scalable risk-stratification tools integrated into emergency department workflows, enabling more timely and targeted SUD prevention following trauma.