Postdoctoral Scholar, Department of Biomedical Data Science
University of Chicago
Dr. Rachel Melamed is a postdoctoral scholar at the University of Chicago’s department of biomedical data science and computational biomedicine. She has a background in cancer genomics but most recently is developing methodology related to population level causal effect estimation.
In particular, Dr. Melamed is building on methods developed in the OHDSI community that use high-dimensional propensity scores as a way to control for many confounding influences in causal inference. To complement this work, she is exploring methods for automatic construction of patient cohorts that are more comparable, require less expert input, and are better suited for high dimensional propensity score estimation. Another area of interest explores how “missing not at random” relationships in health care data can act as confounders for causal inference, and how to control for these biases.