EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Causal machine learning for heterogeneous treatment effects in the presence of missing outcome data

Matthew Pryce iD, Karla Diaz-Ordaz iD, Ruth H Keogh iD, Stijn Vansteelandt iD

Biometrics · 2025

DOI: 10.1093/biomtc/ujaf098 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biometrics, 2025. Verfasst von Matthew Pryce und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medical Statistics, London School of Hygiene & Tropical Medicine, London WC1E 7HT, United Kingdom. Hauptthemen laut MEDLINE: Machine Learning. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Medical Research Council, UK Research and Innovation, ERC.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

When estimating heterogeneous treatment effects, missing outcome data can complicate treatment effect estimation, causing certain subgroups of the population to be poorly represented. In this work, we discuss this commonly overlooked problem and consider the impact that missing at random outcome data has on causal machine learning estimators for the conditional average treatment effect (CATE). We propose 2 de-biased machine learning estimators for the CATE, the mDR-learner, and mEP-learner, which address the issue of under-representation by integrating inverse probability of censoring weights into the DR-learner and EP-learner, respectively. We show that under reasonable conditions, these estimators are oracle efficient and illustrate their favorable performance through simulated data settings, comparing them to existing CATE estimators, including comparison to estimators that use common missing data techniques. We present an example of their application using the GBSG2 trial, exploring treatment effect heterogeneity when comparing hormonal therapies to non-hormonal therapies among breast cancer patients post surgery, and offer guidance on the decisions a practitioner must make when implementing these estimators.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningHumansBreast NeoplasmsTreatment OutcomeComputer SimulationFemaleData Interpretation, StatisticalCausalityBiometryBiasModels, StatisticalTreatment Effect Heterogeneity

Stichwörter der Autor:innen

causal machine learningheterogeneous treatment effectsinfluence functionsmissing outcome data

Autor:innen und Institutionen

  • Department of Medical Statistics, London School of Hygiene & Tropical Medicine, London WC1E 7HT, United Kingdom.
  • Department of Statistical Science, University College London, London WC1E7HB, United Kingdom.
  • Department of Applied Mathematics, Computer Science, and Statistics, Ghent University, Ghent, 9000, Belgium.

ORCID-Profile: Matthew Pryce, Karla Diaz-Ordaz, Ruth H Keogh, Stijn Vansteelandt

Förderung

  • Medical Research Council
  • UK Research and Innovation
  • ERC

Zitieren

APA

Pryce, M., Diaz-Ordaz, K., Keogh, R. H., & Vansteelandt, S. (2025). Causal machine learning for heterogeneous treatment effects in the presence of missing outcome data. Biometrics. https://doi.org/10.1093/biomtc/ujaf098

Vancouver

Pryce M, Diaz-Ordaz K, Keogh RH, Vansteelandt S. Causal machine learning for heterogeneous treatment effects in the presence of missing outcome data. Biometrics. 2025. doi:10.1093/biomtc/ujaf098

BibTeX
@article{Pryce2025,
  author = {Pryce, Matthew and Diaz-Ordaz, Karla and Keogh, Ruth H and Vansteelandt, Stijn},
  title = {Causal machine learning for heterogeneous treatment effects in the presence of missing outcome data},
  journal = {Biometrics},
  year = {2025},
  doi = {10.1093/biomtc/ujaf098}
}
Kostenlos starten →