Skip to main content

Table 3 Performance of CARSS models for predicting the risk of COVID-19

From: Accuracy of automated computer-aided risk scoring systems to estimate the risk of COVID-19: a retrospective cohort study

Model

Mean risk without adverse outcome

Mean risk with adverse outcome

Absolute risk difference

Scaled brier score

Discrimination

AUC (95% CI)

Calibration

Slope (95% CI)

CARM_N

0.09

0.15

0.06

-0.02

(-0.03 to -0.01)

0.68

(0.66 to 0.70)

0.47

(0.41 to 0.54)

CARM_NB

0.09

0.17

0.08

-0.05

(-0.06 to -0.04)

0.68

(0.65 to 0.70)

0.37

(0.31 to 0.43)

CARS_N

0.09

0.17

0.08

0.05

(0.04 to 0.06)

0.73

(0.71 to 0.75)

0.81

(0.72 to 0.89)

CARS_NB

0.09

0.16

0.07

0.01

(0.00 to 0.02)

0.68

(0.66 to 0.70)

0.56

(0.47 to 0.64)

  1. CARM_N: for predicting mortality with NEWS data only; CARM_NB: for predicting mortality with NEWS and Blood test results data; CARS_N: for predicting sepsis with NEWS data only; CARS_NB: for predicting sepsis with NEWS and Blood test results data