Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 320 - Electronic Health Records, Causal Inference and Miscellaneous
Type: Contributed
Date/Time: Wednesday, August 11, 2021 : 3:30 PM to 5:20 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #319082
Title: Profile Matching for the Generalization and Personalization of Causal Inferences
Author(s): Eric Cohn* and Jose Zubizarreta
Companies: Harvard University and Harvard University
Keywords: causal inference; generalization; matching methods; observational studies; propensity score

In this talk, we introduce profile matching, a new multivariate matching method for covariate adjustment in randomized experiments and observational studies that finds the largest possible self-weighted samples across multiple treatments groups that are balanced relative to a covariate profile. This covariate profile is flexible and can represent a population or individual, facilitating the generalization, transportation, and personalization of causal inferences. Profile matching achieves covariate balance by construction, but unlike existing approaches to matching, it does not require specifying a matching ratio, as this is implicitly optimized for the data. Also, profile matching does not require accessing individual-level data of the target population; instead, the target population can be characterized by summary statistics in the covariate profile. We evaluate the performance of profile matching in a simulation study generalizing results from a randomized trial to a target population. We illustrate the utility of this method in an exploratory observational study of opioid use and mental health.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program