Skip to main content

Table 2 Performance measures for the complete and abridged models in the training and test samples

From: Development of caesarean section prediction models: secondary analysis of a prospective cohort study in two sub-Saharan African countries

  

Cut-off point

SP

SE

DA

NPV

PPV

AUC

AUC CI (95%)

Training sample

Admission

Model 1

16.12

78.72

63.84

76.84

93.79

30.18

77.76

76.01

79.50

 2 h

Model 2

18.91

71.42

74.43

72.00

92.04

38.61

78.76

75.98

81.54

 

Model 2A

18.96

66.43

76.44

68.59

91.10

38.55

78.07

75.44

80.70

 4 h

Model 3

21.76

77.83

76.44

77.53

92.45

48.19

85.92

83.03

88.81

 

Model 3A

32.02

85.04

71.28

82.04

91.40

57.02

85.65

82.86

88.44

 6 h

Model 4

23.57

77.78

79.75

78.26

92.20

53.85

85.14

80.46

89.83

 

Model 4A

35.96

86.40

62.82

80.79

88.16

59.04

84.31

79.60

89.02

Max. score

Model 5

13.66

87.90

85.20

87.56

97.61

50.60

93.80

92.88

94.71

 

Model 5A

10.69

84.69

87.96

85.12

97.91

46.30

93.66

92.77

94.54

Test sample

Admission

Model 1

16.12

78.68

63.20

76.62

93.30

31.28

78.70

76.19

81.20

 2 h

Model 2

18.91

70.15

67.65

69.64

89.52

36.51

75.88

71.50

80.27

 

Model 2A

18.96

66.38

68.42

66.71

91.41

28.68

74.73

69.74

79.72

 4 h

Model 3

21.76

77.17

76.00

76.92

92.21

47.50

85.29

80.79

89.78

 

Model 3A

32.02

88.47

64.00

83.51

90.63

58.54

84.39

79.43

89.36

 6 h

Model 4

23.57

76.58

77.78

76.81

93.41

44.68

85.19

77.42

92.95

 

Model 4A

35.96

82.88

70.00

80.14

91.09

52.50

86.52

80.07

92.96

 Max. score

Model 5

13.66

89.26

80.12

88.03

96.66

53.68

93.66

92.36

94.96

 

Model 5A

10.69

84.69

86.81

84.95

97.81

44.92

93.72

92.33

95.11

  1. *SP: Specificity, SE: Sensitivity, DA: diagnostic accuracy, NPV: negative predictive value, PPV: positive predictive value, AUC: area under the ROC curve, CI: Confidence interval (CI)