Analysis of time-to-event for observational studies: Guidance to the use of intensity models

Per Kragh Andersen, Maja Pohar Perme*, Hans C. van Houwelingen, Richard J. Cook, Pierre Joly, Torben Martinussen, Jeremy M.G. Taylor, Michal Abrahamowicz, Terry M. Therneau

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

27 Citations (Scopus)
34 Downloads (Pure)

Abstract

This paper provides guidance for researchers with some mathematical background on the conduct of time-to-event analysis in observational studies based on intensity (hazard) models. Discussions of basic concepts like time axis, event definition and censoring are given. Hazard models are introduced, with special emphasis on the Cox proportional hazards regression model. We provide check lists that may be useful both when fitting the model and assessing its goodness of fit and when interpreting the results. Special attention is paid to how to avoid problems with immortal time bias by introducing time-dependent covariates. We discuss prediction based on hazard models and difficulties when attempting to draw proper causal conclusions from such models. Finally, we present a series of examples where the methods and check lists are exemplified. Computational details and implementation using the freely available R software are documented in Supplementary Material. The paper was prepared as part of the STRATOS initiative.

Original languageEnglish
JournalStatistics in Medicine
Volume40
Issue number1
Pages (from-to)185-211
Number of pages27
ISSN0277-6715
DOIs
Publication statusPublished - 2021

Keywords

  • censoring
  • Cox regression model
  • hazard function
  • immortal time bias
  • multistate model
  • prediction
  • STRATOS initiative
  • survival analysis
  • time-dependent covariates

Cite this