Skip to content
DraconiticResearch
Search Sources API
190,332 studies31 sourcesUpdated 53 min ago
Peer-reviewedCC BYInternational journal of epidemiology

Hazard ratios and their causal interpretation: challenges and guidance.

  • Jonathan W Bartlett1
  • Dominic Magirr2
  • Tim P Morris3
  1. 1Department of Medical Statistics, London School of Hygiene & Tropical Medicine, London, United Kingdom.
  2. 2Statistical Methodology, Novartis Pharmaceuticals AG, Basel, Switzerland.
  3. 3Statistical Methodology, Novartis Pharmaceuticals UK Ltd, London, United Kingdom.
Read full textPDFDOICite

Abstract

The hazard ratio, typically estimated using Cox's famous proportional hazards model, is the most common effect measure used to describe the association or effect of a covariate on a time-to-event outcome. In recent years, the hazard ratio has been argued by some to lack a causal interpretation, even in randomized trials, and even if the proportional hazards assumption holds. This is concerning, not least due to the ubiquity of hazard ratios in analyses of time-to-event data. We review these critiques and describe how we think hazard ratios should be interpreted. In light of the issues involved, alternative measures may be preferable to describe effects of exposures or treatments on time-to-event outcomes.

Related

  1. Epidemiological Causal Graph Identification: Challenges, Identifiability and Algorithms
    Sambit Mishra, Yingying Wang, Christine K. Johnson +1
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  2. Federated Causal Discovery in Medicine: Trends, Opportunities, and Challenges
    Niccolò Rocchi, Marco Scutari, Alessio Zanga +2
    Frontiers in Digital HealthPeer-reviewedCC BY-SA 4.0Full textGeneral biologyStatisticsHAL
  3. The Challenges of Hyperparameter Tuning for Accurate Causal Effect Estimation.
    Damian Machlanski, Spyridon Samothrakis, Paul Clarke
    Machine learningPeer-reviewedCC BYFull text1 citedStatisticsEurope PMC
  4. Nonparametric hazard rate estimation with associated kernels and minimax bandwidth choice
    Luce Breuil, Sarah Kaakai
    HALPreprintLicence Ouverte 2.0Full textStatisticsHAL
  5. A new characterization of the hazard rate and reversed hazard rate orders with applications
    Carmen Sangüesa
    arXivPreprintarXiv licenceFull textProbabilityarXiv

Cite

BibTeXRIS
BibTeX
@article{bartlett2026hazard,
  title = {{Hazard ratios and their causal interpretation: challenges and guidance.}},
  author = {Jonathan W Bartlett and Dominic Magirr and Tim P Morris},
  journal = {International journal of epidemiology},
  year = {2026},
  month = oct,
  doi = {10.1093/ije/dyag228},
  url = {https://europepmc.org/article/MED/42925542},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - Hazard ratios and their causal interpretation: challenges and guidance.
AU  - Jonathan W Bartlett
AU  - Dominic Magirr
AU  - Tim P Morris
PY  - 2026
DA  - 2026/10/01
JO  - International journal of epidemiology
DO  - 10.1093/ije/dyag228
UR  - https://europepmc.org/article/MED/42925542
AB  - The hazard ratio, typically estimated using Cox's famous proportional hazards model, is the most common effect measure used to describe the association or effect of a covariate on a time-to-event outcome. In recent years, the hazard ratio has been argued by some to lack a causal interpretation, even in randomized trials, and even if the proportional hazards assumption holds. This is concerning, not least due to the ubiquity of hazard ratios in analyses of time-to-event data. We review these critiques and describe how we think hazard ratios should be interpreted. In light of the issues involved, alternative measures may be preferable to describe effects of exposures or treatments on time-to-event outcomes.
SN  - 0300-5771
SN  - 1464-3685
ER  -