Skip to main content

Table 2 Power comparisons of Cox regression, Time-dependent Cox regression and Landmark methods on simulated data under the alternative hypothesis that cumulative drug dose is significantly related to disease outcome

From: Statistical methods for elimination of guarantee-time bias in cohort studies: a simulation study

β 0 Cumulative dose Cox regression Time-dependent
Cox regression
Landmark analyses
τ = 5 τ = 7
log(0.015) Low 214 58 119 74
Moderate 1000 109 348 181
High 1000 627 186 396
log(50.015) Low 723 47 289 103
Moderate 1000 296 792 471
High 1000 977 452 827
log(100.015) Low 957 52 288 94
Moderate 1000 322 836 523
High 1000 989 538 888
  1. β 0 values represent disease incidence rate; τ= Landmark time
\