Table 4.

Model performance for varying outcome predictions

Model	AUC (95% CI)	Sensitivity (95% CI)	Specificity (95% CI)	Gap*	PPV (95% CI)	NPV (95% CI)	MCC (95% CI)
Child models
ED only	0.86 (0.85 to 0.87)	0.78 (0.77 to 0.79)	0.78 (0.77 to 0.79)	0.44	0.09 (0.08 to 0.10)	0.992 (0.990 to 0.994)	0.626 (0.613 to 0.639)
Hospital and ED	0.91 (0.91 to 0.91)	0.80 (0.80 to 0.80)	0.92 (0.92 to 0.92)	0.28	0.09 (0.09 to 0.09)	0.998 (0.998 to 0.998)	0.746 (0.742 to 0.750)
Adult models
ED only	0.85 (0.85 to 0.85)	0.76 (0.76 to 0.76)	0.80 (0.80 to 0.80)	0.44	0.11 (0.11 to 0.11)	0.990 (0.989 to 0.991)	0.619 (0.614 to 0.624)
Hospital & ED	0.89 (0.89 to 0.89)	0.79 (0.79 to 0.79)	0.84 (0.84 to 0.84)	0.37	0.12 (0.12 to 0.12)	0.993 (0.993 to 0.993)	0.689 (0.687 to 0.691)
All ages models
ED only	0.85 (0.85 to 0.85)	0.74 (0.74 to 0.74)	0.81 (0.81 to 0.81)	0.45	0.12 (0.12 to 0.12)	0.989 (0.988 to 0.990)	0.602 (0.597 to 0.607)
Hospital & ED	0.90 (0.90 to 0.90)	0.84 (0.84 to 0.84)	0.80 (0.80 to 0.80)	0.36	0.10 (0.10 to 0.10)	0.995 (0.995 to 0.995)	0.711 (0.709 to 0.713)

*The gap between sensitivity and specificity. Calculated as follows: Gap = (1−Sensitivity) + (1−Specificity).

AUC, area under curve; ED, emergency department; MCC, Matthews correlation coefficient; NPV, negative predictive value; PPV, positive predictive value.