Skip to main content

Table 2 Performance of the WS final model at different risk thresholds, entire study sample (N = 12,395)

From: Prediction of pre-eclampsia in nulliparous women using routinely collected maternal characteristics: a model development and validation study

Risk threshold

PE/n

Sensitivity (95% CI)

Specificity (95% CI)

PPV (95% CI)

NPV (95% CI)

Positive LR (95% CI)

Negative LR (95% CI)

≥threshold

<threshold

2%

196/5180

97/7215

67% (61–72%)

59% (58–60%)

4% (3–4%)

99.0% (98.0–99.0%)

1.62 (1.49–1.77)

0.56 (0.48–0.66)

3%

138/1731

155/10664

47% (41–53%)

87% (86–87%)

8% (7–9%)

99.0% (98.0–99.0%)

3.58 (3.14–4.07)

0.61 (0.55–0.68)

4%

105/1098

188/11297

36% (31–41%)

92% (91–92%)

10% (8–11%)

98.0% (98.0–99.0%)

4.37 (3.71–5.15)

0.70 (0.64–0.76)

5%

92/776

201/11619

31% (26–37%)

94% (94–95%)

12% (10–14%)

98.3% (98.0–98.5%)

5.56 (4.62–6.68)

0.73 (0.67–0.79)

8%

54/374

239/12021

18% (14–23%)

97% (97–98%)

14% (11–18%)

98.0% (97.8–98.3%)

6.97 (5.35–9.08)

0.84 (0.79–0.89)

  1. CI Confidence interval, LR Likelihood ratio, NPV Negative predictive value, PE Pre-eclampsia, PPV Positive predictive value, WS Western Sydney