Skip to main content

Table 3 Average and range of c statistic and calibration slopes in the prostate cancer case-study

From: Improving prediction models with new markers: a comparison of updating strategies

Measure Original Model Model Revision Model Revision with Shrinkage Recalibration with extension CLR CLR simple Imputation
c-statistic 0.69 [0.68–0.69] 0.73 [0.69–0.75] 0.74 [0.73–0.75] 0.74 [0.73–0.75] 0.74 [0.73–0.75] 0.74 [0.73–0.75] 0.73 [0.73–0.74]
Calibration Slope 0.77 [0.71–0.82] 0.76 [0.42–1.08] 0.75 [0.41–1.06] 0.96 [0.55–1.72] 0.61 [0.54–0.72] 0.74 [0.59–0.91] 0.78 [0.66–0.93]
  1. A previously developed prediction model (RC3) was extended with a marker (PHI) using data from one cohort and validated in four cohorts not used at model development