Table 3.

Predictive performance statistics of the newly developed SEAL model (apparent and optimism-corrected measures) and the ACS-NSQIP model (external validation) when applied to the HELAS cohort to predict the risk of serious 30-day postoperative complications

Performance measure	SEAL model			ACS-NSQIP model
	Apparent (95% CI)a	Mean optimism	Optimism-corrected (95% CI)	Unadjusted (95% CI)	Recalibrated (95% CI)
C-statistic	0.81 (0.80,0.83)	0.01	0.80 (0.79, 0.81)	0.71 (0.69,0.72)	0.71 (0.69,0.72)
Intercept	0.00 (-0.06,0.06)	0.00	0.00 (-0.01, 0.01)	0.27 (0.21,0.33)	0.00 (0.00,0.00)
Slope	1.10 (1.08,1.13)	0.09	1.01 (0.99, 1.03)	1.03 (0.97,1.11)	1.00 (0.98,1.03)
Brier scaled (%)	28.1 (27.6,29.0)	3.0	25.1 (24.1,26.1)	9.3 (7.9,10.4)	10.8 (9.2,12.2)

CI confidence interval, HELAS Hellenic Emergency Laparotomy Study, SEAL SErious complications After Laparotomy, ACS-NSQIP American College of Surgeons National Surgical Quality Improvement Program

^aApparent predictive performance implies assessment directly in the derivation cohort that yields an optimistic estimate of model performance, because the regression coefficients are optimized for the derivation cohort. The degree of optimism was estimated using Bootstrap resampling and the estimated mean optimism was subtracted from the apparent performance to indicate the expected model performance for future patients similar to the derivation cohort