Skip to main content

Table 5 exp(γ) estimations and their 95% confidence interval according to the HP and LWA models and to the matching methods, with the p-value of the Wald’s Test

From: Matching methods to create paired survival data based on an exposure occurring over time: a simulation study with application to breast cancer

Models

Method 1

Method 2

Holt and Prentice model

exp γ = 0.90 CI 95 = [ 0.36 2.21 ] p value = 0.82 se = 0.46

exp γ = 2.00 CI 95 = [ 0.75 5.33 ] p value = 0.17 se = 0.50

Lee, Wei and Amato model

  

L W A u

 

exp γ = 1.22 CI 95 = [ 0.61 2.42 ] p value = 0.58 se = 0.35

L W A a

 

exp γ = 1.24 CI 95 = [ 0.62 2.45 ] p value = 0.54 se = 0.35

L W A i :R H− and good cPP

 

exp γ = 5.70 CI 95 = [ 0.80 40.50 ] p value = 0.08 se = 0.98

L W A i :R H− and poor cPP

 

exp γ = 1.77 CI 95 = [ 0.34 9.24 ] p value = 0.49 se = 0.83

L W A i :R H+ not treated and good cPP

 

exp γ = 2.87 CI 95 = [ 0.63 13.10 ] p value = 0.17 se = 0.76

L W A i :R H+ not treated and poor cPP

 

exp γ = 0.89 CI 95 = [ 0.30 2.69 ] p value = 0.83 se = 0.55

L W A i :R H+ treated and good cPP

 

exp γ = 0.95 CI 95 = [ 0.07 13.66 ] p value = 0.97 se = 1.33

L W A i :R H+ treated and poor cPP

 

exp γ = 0.30 CI 95 = [ 0.04 2.42 ] p value = 0.25 se = 1.05

  1. The estimation of the pregnancy effect standard error (se) by the HP and LWA models are given.