Key result
The additive and logistic EuroSCORE models overestimated mortality in Australian cardiac surgery patients (predicted 5.31% and 8.76% vs observed 3.20%), despite good discrimination (AUC 0.83).
Why the study?
Does the EuroSCORE model accurately predict mortality in Australian cardiac surgery patients?
Population
8,331 patients undergoing cardiac surgery at six institutions in Victoria, Australia between July 2001 and…
Comparison
Additive and logistic EuroSCORE risk prediction… vs Observed mortality
Design
Cohort
Authors
Loading...
May overestimate operative risk in Australia; leaves open need for local model recalibration before routine use.
Observational (n=8,331)
Yes
Does the EuroSCORE model accurately predict mortality in Australian cardiac surgery patients?
Effect estimate: AUC 0.83
p-value: p=<0.05
The EuroSCORE models significantly overestimate mortality in Australian cardiac surgery patients, indicating poor calibration and suggesting the need for a locally derived risk prediction model.
Yap et al. (2006) conducted an observational in Cardiac surgery (n=8,331). EuroSCORE model vs. Observed outcomes was evaluated on Mortality (AUC 0.83, p=<0.05). The additive and logistic EuroSCORE models overestimated mortality in Australian cardiac surgery patients (predicted 5.31% and 8.76% vs observed 3.20%), despite good discrimination (AUC 0.83).