Skip to main content

Table 4 Assessment of proportional hazards for each variable in the multivariate Cox model for the example case-cohort study of colorectal cancer risk, using waist circumference as the primary variable

From: Testing the proportional hazards assumption in case-cohort analysis

Covariates in the model Corr with event time Corr with rank of time Corr with KM estimates
Waist 0.030 0.024 0.026
METs of physical activity: (3.75,10) 0.365 0.361 0.373
(10,20) 0.055 0.049 0.053
> = 20 0.629 0.688 0.645
  1. *Note: Model description is the same as above except that in this model waist circumference is the primary exposure variable. For simplicity, we only listed the variables that do not satisfy proportionality assumption in the multivariate Cox model.