Skip to main content

Table 3 Prediction performance of the four prediction models

From: A prediction model for good neurological outcome in successfully resuscitated out-of-hospital cardiac arrest patients

Cut-off probability

Misclassification rate

Sensitivity

Specificity

H1

H12

H24

H48

H1

H12

H24

H48

H1

H12

H24

H48

0.45

26.2 (9.1)

22.9 (8.0)

21.8 (8.2)

–

75.2 (12.5)

78.4 (12.2)

79.8 (12.4)

–

70.8 (14.9)

76.2 (11.8)

77.4 (13.2)

–

0.50

25.3 (9.2)

22.5 (8.2)

21.5 (8.2)

–

72.9 (12.8)

76.5 (12.8)

77.6 (12.9)

–

77.4 (13.7)

78.9 (11.6)

79.9 (12.6)

–

0.55

24.8 (9.2)

22.3 (8.3)

21.5 (8.4)

23.7 (9.6)

70.5 (13.1)

74.1 (13.5)

75.3 (13.6)

78.6 (14.2)

74.3 (14.4)

81.5 (11.3)

82.2 (12.3)

74.6 (15.6)

0.60

–

–

–

23.4 (9.5)

–

–

–

76.8 (14.4)

–

–

–

77.2 (15.0)

0.65

–

–

–

23.3 (9.4)

–

–

–

74.6 (14.6)

–

–

–

77.4 (13.2)

  1. Misclassification rate is the percentage of cases misclassified. The optimal cut-off probability yielding the smallest misclassification rate is indicated in bold for each time point. Misclassification rate, sensitivity and specificity are presented in percentage (standard errors)