Skip to main content

Table 3 Summary measures of the regression models considered

From: Predicting the prevalence of chronic kidney disease in the English population: a cross-sectional study

Model/Statistic

Main-effects

Clinical

Parsimonious

Clinical; no age2

AIC

248 207

247 771

256 757

248 246

BIC

248 403

248 001

256 803

248 465

DoF

17

18

4

19

In-sample classification measures

Sensitivity

22.05%

17.79%

10.78%

18.91%

Specificity

98.77%

99.07%

99.24%

98.97%

AUROC

0.899

0.899

0.890

0.899

PPV

58.07%

58.29%

50.98%

57.21%

NPV

94.27%

94.30%

93.85%

94.36%

Out-of-sample classification measures

Sensitivity

22.15%

17.76%

10.94%

18.86%

Specificity

98.74%

99.06%

99.24%

98.96%

AUROC

0.898

0.898

0.889

0.898

PPV

57.23%

57.75%

50.88%

56.76%

NPV

94.33%

94.35%

93.91%

94.41%

  1. AIC: Akaike’s Information criteria. BIC: Bayesian information criteria. DoF: Degrees of Freedom. AUROC: Area under the receiver operating characteristic curve. PPV: Positive predictive value. NPV: Negative predictive value.