Project Grant R01GM170992

The National Institute of General Medical Sciences, a component of the Department of Health and Human Services National Institutes of Health, awarded $259,270 to the University of Wisconsin–Madison on August 20, 2026, under the Biomedical Research and Research Training program (CFDA 93.859). This Project Grant (R01GM170992) supports development of data infrastructure, analytic methods, and forecasting models to measure how federally funded biomedical research translates into improved health outcomes. The project reframes research impact assessment as a decision science problem and builds tools to trace research investments through the pathway from initial funding to therapeutic application.

The award funds three specific aims. Aim 1 constructs an enriched NIH knowledge graph linking grants to downstream outputs including FDA approvals, clinical practice guidelines, and real-world health data, quantifying semantic relatedness across document types. Aim 2 applies econometric and network analysis to this knowledge graph to identify features explaining success or failure at three pivotal decision points: targeted funding initiatives at shifting research direction, patent portfolios at subsequent FDA approval, and clinical guidelines' evidence base contributing to real-world prescribing. Aim 3 develops machine learning models to forecast successive stages of therapeutic development from FDA approval through guideline recommendation to real-world prescription uptake.

Work is performed in Madison, Wisconsin. The period of performance runs through March 31, 2030.