Table 2

Discrimination and calibration statistics from recalibrated predicted probability algorithms

 All 22–27 weeks 28–32 weeks CRIB CRIB II CRIB II(−T) CRIB CRIB II CRIB II(−T) CRIB CRIB II CRIB II(−T) Observed deaths, n (%) 317 (9.7) 259 (35.8) 58 (2.3) Predicted deaths, n (%) 317.0 (9.7) 317.0 (9.7) 317.0 (9.7) 208.0 (28.7) 256.1 (35.4) 259.7 (35.9) 109.0 (4.3) 60.9 (2.4) 57.3 (2.3) Brier score 0.052 0.053 0.053 0.163 0.166 0.164 0.020 0.021 0.021 Spiegelhalter’s z-score p = 0.2 p = 0.5 p = 0.5 p<0.001 p = 0.4 p = 0.6 p = 1.00 p = 0.6 p = 0.4 AUC 0.920 0.921 0.923 0.830 0.807 0.811 0.839 0.826 0.833 Calibration-in-the-small 0.0002 0.0001 0.0002 0.0085 0.0040 0.0043 0.0013 <0.0001 <0.0001 Calibration-in-the-large 0 0 0 0.0050 <0.0001 <0.0001 0.0004 <0.0001 <0.0001 Cox’s regression: Intercept α (SE) 0 0 0 0.28 (0.11) 0.02 (0.10) 0.01 (0.10) −0.93 (0.22) −0.24 (0.34) −0.08 (0.35) Slope β (SE) 1 1 1 0.72 (0.06) 0.99 (0.08) 1.03 (0.08) 0.95 (0.08) 0.94 (0.10) 0.97 (0.11) Test: α = 0, β = 1 – – – p<0.001 p = 0.9 p = 0.9 p<0.001 p = 0.8 p = 0.9 Test: α = 0 | β = 1 – – – p<0.001 p = 0.8 p = 0.9 p<0.001 p = 0.7 p = 0.9 Test: β = 1 | α – – – p<0.001 p = 0.9 p = 0.7 p = 0.5 p = 0.5 p = 0.8
• AUC, area under the curve; CRIB, Clinical Risk Index for Babies; CRIB II(−T), CRIB II without temperature at admission.