Generalizing treatment effects with incomplete covariates - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Preprints, Working Papers, ... Year : 2022

Generalizing treatment effects with incomplete covariates


We focus on the problem of generalizing a causal effect estimated on a randomized controlled trial (RCT) to a target population described by a set of covariates from observational data. Available methods such as inverse propensity weighting are not designed to handle missing values, which are however common in both data sources. In addition to coupling the assumptions for causal effect identifiability and for the missing values mechanism and to defining appropriate estimation strategies, one difficulty is to consider the specific structure of the data with two sources and treatment and outcome only available in the RCT. We propose and compare three multiple imputation strategies (separate imputation, joint imputation with fixed effect, joint imputation without source information), as well as a technique that uses estimators that can handle missing values directly without imputing them. These methods are assessed in an extensive simulation study, showing the empirical superiority of fixed effect multiple imputation followed with any complete data generalizing estimators. This work is motivated by the analysis of a large registry of over 20,000 major trauma patients and a RCT studying the effect of tranexamic acid administration on mortality. The analysis illustrates how the missing values handling can impact the conclusion about the effect generalized from the RCT to the target population.
Fichier principal
Vignette du fichier
main.pdf (2.56 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03517373 , version 1 (07-01-2022)


  • HAL Id : hal-03517373 , version 1


Imke Mayer, Julie Josse, Traumabase Group. Generalizing treatment effects with incomplete covariates. 2022. ⟨hal-03517373⟩
96 View
86 Download


Gmail Facebook X LinkedIn More