EUVIMEDDécouvrir EUVIMED
Open access · CC BY 4.0

Illustrating Implications of Misaligned Causal Questions and Statistics in Settings With Competing Events and Interest in Treatment Mechanisms

Takuya Kawahara iD, Sean McGrath iD, Jessica G Young

Statistics in Medicine · 2026-05 · 45 (10-12)

DOI: 10.1002/sim.70535 ↗

At a glance

Journal article in Statistics in Medicine, published online on 2026-04-24. Written by Takuya Kawahara and 2 co-authors. Institution: Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, Massachusetts, USA. Major MEDLINE topics: Causality. Cited 3 times so far (Crossref); 26 references. The full text is freely available (open access, CC BY 4.0). Funded by: Japan Society for the Promotion of Science, Patient-Centered Outcomes Research Institute.

RésuméSubjectsRéférences bibliographiquesCiteRelated studies

Abstract (original language)

In the presence of competing events, many investigators are interested in a direct treatment effect on the event of interest that does not capture treatment effects on competing events. Classical survival analysis methods that treat competing events like censoring events, at best, target a controlled direct effect: the effect of the treatment under a difficult to imagine and typically clinically irrelevant scenario where competing events are somehow eliminated. A separable direct effect, quantifying the effect of a future modified version of the treatment, is an alternative direct effect notion that may better align with an investigator's underlying causal question. In this paper, we provide insights into the implications of naively applying an estimator constructed for a controlled direct effect (i.e., “censoring by competing events”) when the actual causal effect of interest is a separable direct effect. We illustrate the degree to which controlled and separable direct effects may take different values, possibly even different signs, and the degree to which these two different effects may be differentially impacted by violation and/or near violation of their respective identifying conditions under a range of data generating scenarios. Finally, we provide an empirical comparison of inverse probability of censoring weighting to an alternative weighted estimator specifically structured for a separable effect using data from a randomized trial of estrogen therapy and prostate cancer mortality.

Subjects

MeSH terms (MEDLINE)

CausalityHumansSurvival AnalysisModels, StatisticalData Interpretation, StatisticalComputer SimulationTreatment OutcomeProstatic NeoplasmsRandomized Controlled Trials as Topic

Author keywords

competing eventestimandinverse‐probability weightingpositivityseparable effects

Authors and institutions

  • Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, Massachusetts, USA.
  • Clinical Research Promotion Center, The University of Tokyo Hospital, Tokyo, Japan.
  • Department of Biostatistics, Yale School of Public Health, New Haven, Connecticut, USA.
  • Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, USA.

ORCID profiles: Takuya Kawahara, Sean McGrath

Funding

  • Japan Society for the Promotion of Science
  • Patient-Centered Outcomes Research Institute

Cite

APA

Kawahara, T., McGrath, S., & Young, J. G. (2026). Illustrating Implications of Misaligned Causal Questions and Statistics in Settings With Competing Events and Interest in Treatment Mechanisms. Statistics in Medicine, 45(10-12). https://doi.org/10.1002/sim.70535

Vancouver

Kawahara T, McGrath S, Young JG. Illustrating Implications of Misaligned Causal Questions and Statistics in Settings With Competing Events and Interest in Treatment Mechanisms. Statistics in Medicine. 2026;45(10-12). doi:10.1002/sim.70535

BibTeX
@article{Kawahara2026,
  author = {Kawahara, Takuya and McGrath, Sean and Young, Jessica G},
  title = {Illustrating Implications of Misaligned Causal Questions and Statistics in Settings With Competing Events and Interest in Treatment Mechanisms},
  journal = {Statistics in Medicine},
  year = {2026},
  volume = {45},
  number = {10-12},
  doi = {10.1002/sim.70535}
}
Start for free →