On collapsibility and confounding bias in Cox and Aalen regression models

Lifetime Data Anal. 2013 Jul;19(3):279-96. doi: 10.1007/s10985-013-9242-z. Epub 2013 Jan 18.


We study the situation where it is of interest to estimate the effect of an exposure variable [Formula: see text] on a survival time response [Formula: see text] in the presence of confounding by measured variables [Formula: see text]. Quantifying the amount of confounding is complicated by the non-collapsibility or non-linearity of typical effect measures in survival analysis: survival analyses with or without adjustment for [Formula: see text] typically infer different effect estimands of a different magnitude, even when [Formula: see text] is not associated with the exposure, and henceforth not a confounder of the association between exposure and survival time. We show that, interestingly, the exposure coefficient indexing the Aalen additive hazards model is not subject to such non-collapsibility, unlike the corresponding coefficient indexing the Cox model, so that simple measures of the amount of confounding bias are obtainable for the Aalen hazards model, but not for the Cox model. We argue that various other desirable properties can be ascribed to the Aalen model as a result of this collapsibility. This work generalizes recent work by Janes et al. (Biostatistics 11:572-582, 2010).

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Aspirin / therapeutic use
  • Bias
  • Biostatistics
  • Humans
  • Infant
  • Infant, Newborn
  • Kaplan-Meier Estimate
  • Life Tables
  • Middle Aged
  • Myocardial Infarction / drug therapy
  • Myocardial Infarction / mortality
  • Pneumonia / etiology
  • Pneumonia / prevention & control
  • Proportional Hazards Models*
  • Regression Analysis
  • Young Adult


  • Aspirin