Current exposure‐risk model (current effect of the current value of the time‐dependent variable) 1‐day delay exposure‐risk model (current effect of the current value of the time‐dependent variable) 2‐day delay exposure‐risk model (current effect of the current value of the time‐dependent variable) WCE exposure‐risk model (current effect of the current and past values of the time‐dependent variable for the best‐ fitting model) Exposure ‐risk model selected for final model LOCF venti‐ lator days (%) 1 HR 95% CI p‐value AIC HR 95% CI p‐value AIC HR 95% CI p‐ value AIC Relevant exposure window & number of knots2 HR & 95%C I p‐ value 3 AIC Intestinal prophylaxis 970 970 975 12 day Fig 3E 0 965 WCE Yes 2395 (30.4) 0.12 (0.04, 0.41) 0.0007 0.13 (0.04, 0.42) 0.0008 0.19 (0.06, 0.54) 0.002 2 knots No 5477 (69.6) Ref Ref Ref Oropharyngeal prophylaxis 967 968 973 21 days Fig 3D 0 965 Current Yes 4784 (60.8) 0.14 (0.05, 0.38) 0.0001 0.14 (0.05, 0.39) 0.0002 0.20 (0.08, 0.51) 0.0006 1 knot No 3088 (39.2) Ref Ref Ref (unconstrained) 1 The number of ventilation days after reducing the numbers of missings for time‐dependent covariates with the ‘last observation carried forward (LOCF) approach’. The total number of ventilation days was 7872 2 for best‐fitting model 3 The p‐value was estimated using 1000 bootstrapped data sets to account for multiple testing when selecting the best‐fitting WCE model. 84 Chapter 4
RkJQdWJsaXNoZXIy MTk4NDMw