Federal Project Grant Award Summary

Yale University received a $336,250 project grant from the National Institute on Drug Abuse (NIDA) under the Drug Use and Addiction Research Programs (CFDA 93.279), effective April 1, 2026 through March 31, 2028. The award supports the development of a behavioral foundation model that leverages advanced natural language processing and large language models to transform qualitative epidemiological survey responses from the National Epidemiologic Survey on Alcohol and Related Conditions III (NESARC-III), National Survey on Drug Use and Health (NSDUH), and UK Biobank into quantitative, machine-interpretable representations of substance use behaviors. The research team will harmonize structured and free-text survey data into unified narratives, fine-tune state-of-the-art open-source large language models with integrated demographic tokens, and rigorously validate model performance using established metrics including area under the ROC curve, calibration analysis, and cross-dataset generalizability assessments.

The project will deliver open-access tools and analytical capabilities designed to enable precision public health initiatives for substance use disorder prevention and intervention. Specific deliverables include risk stratification models for substance use disorder outcomes, latent clustering methodologies to identify distinct risk and resilience profiles, and data-driven optimization strategies for survey instrument design. These products will enhance early identification of high-risk populations and support targeted interventions to reduce the public health burden of substance use disorders.