EUVIMEDEUVIMED entdecken
Crossref

Bias reduction in g-computation for covariate adjustment in randomized clinical trials

Xin Zhang iD, Lin Liu iD, Haitao Chu

Biometrics · 01.07.2026 · 82 (3)

DOI: 10.1093/biomtc/ujag129 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biometrics, online erschienen am 13.07.2026. Verfasst von Xin Zhang und 2 weiteren Autor:innen. Beteiligte Einrichtung: Data Sciences and Analytics, Pfizer Inc., Shanghai, 201203, China. Hauptthemen laut MEDLINE: Randomized Controlled Trials as Topic, Models, Statistical. Die Arbeit nennt 34 Literaturangaben. Gefördert von: SFC, Science and Technology Talent and Platform Program of Yunnan Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

G-computation is a powerful covariate adjustment method for analyzing randomized clinical trials. It typically relies on fitting canonical generalized linear models. However, this could be problematic when the sample size or event number is small relative to the number of covariates. Common issues include the underestimation of the variance and the potential nonexistence of maximum likelihood estimators. Bias reduction methods are commonly employed to address these issues, including Firth correction, which guarantees the existence of corresponding estimates. Yet, their application within g-computation remains underexplored. In this article, we analyze the asymptotic bias of g-computation estimators and propose a novel bias-reduction method that improves both estimation and inference. Our approach performs bias correction via generalized Oaxaca–Blinder estimators, and thus the resulting estimators are guaranteed to be bounded. The proposed debiased estimators use slightly modified versions of maximum likelihood or Firth correction estimators for nuisance parameters. We also introduce a simple small-sample bias adjustment for variance estimation to improve finite-sample inference validity. Through extensive simulations, we demonstrate that our method offers superior finite-sample performance, effectively addressing the bias-efficiency tradeoff. Finally, we illustrate its practical utility by reanalyzing a completed randomized clinical trial. In this example, our method improves precision in a small subgroup analysis for which the standard method fails to fit the regression model.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Randomized Controlled Trials as TopicModels, StatisticalBiasComputer SimulationHumansLikelihood FunctionsData Interpretation, StatisticalSample SizeBiometryLinear Models

Stichwörter der Autor:innen

Firth correctionbias reductioncovariate adjustmentg-computationgeneralized Oaxaca–Blinder estimatorsmall-sample bias adjustment

Autor:innen und Institutionen

  • Data Sciences and Analytics, Pfizer Inc., Shanghai, 201203, China.
  • Institute of Natural Sciences, MOE-LSC, School of Mathematical Sciences, CMA-Shanghai, and SJTU-Yale Joint Center for Biostatistics and Data Science, Shanghai Jiao Tong University, Shanghai, 200240, China.
  • Data Sciences and Analytics, Pfizer Inc., New York, NY 10001, United States.
  • Division of Biostatistics and Health Data Science, School of Public Health, University of Minnesota, Minneapolis, MN 55455, United States.

ORCID-Profile: Xin Zhang, Lin Liu

Förderung

  • SFC
  • Science and Technology Talent and Platform Program of Yunnan Province

Zitieren

APA

Zhang, X., Liu, L., & Chu, H. (2026). Bias reduction in g-computation for covariate adjustment in randomized clinical trials. Biometrics, 82(3). https://doi.org/10.1093/biomtc/ujag129

Vancouver

Zhang X, Liu L, Chu H. Bias reduction in g-computation for covariate adjustment in randomized clinical trials. Biometrics. 2026;82(3). doi:10.1093/biomtc/ujag129

BibTeX
@article{Zhang2026,
  author = {Zhang, Xin and Liu, Lin and Chu, Haitao},
  title = {Bias reduction in g-computation for covariate adjustment in randomized clinical trials},
  journal = {Biometrics},
  year = {2026},
  volume = {82},
  number = {3},
  doi = {10.1093/biomtc/ujag129}
}
Kostenlos starten →