Yanfei Wang
I am an Assistant Scientist at the University of Florida, working on causal inference, target trial emulation, and longitudinal clinical modeling using electronic health records and other real-world healthcare data.
My research focuses on how to turn routine clinical data into decision-relevant evidence without pretending that observational data behave like clean randomized trials. I develop methods that make treatment strategies, patient histories, and empirical support more explicit in longitudinal analysis.
What I work on
Target trial emulation
I design frameworks for reconstructing clinically meaningful treatment comparisons from routine EHR data, including treatment initiation, switching, add-on therapy, and delayed treatment effects.
Longitudinal treatment strategies
I study how to define and evaluate treatment histories when care evolves over time, with emphasis on decision-time alignment, sustained exposure, and support-aware comparisons.
Bias-aware real-world evidence
My work also focuses on confounding, censoring, irregular observation, and support limitations, with the goal of producing evidence that is more transparent about what the data can and cannot support.
Featured work
Operational target trial emulation using EHR data
A recent line of work develops an operational framework for reconstructing treatment action from clinical records, aligning follow-up with decision time, and restricting evaluation to empirically supported regions of the data.