The early prediction of mortality in acute pancreatitis: a large population-based study

Identification of patients at risk for mortality early in the course of acute pancreatitis (AP) is an important step in improving outcome.

Using Classification and Regression Tree (CART) analysis, a clinical scoring system was developed for prediction of in-hospital mortality in AP. The scoring system was derived on data collected from 17,992 cases of AP from 212 hospitals in 2000-2001. The new scoring system was validated on data collected from 18,256 AP cases from 177 hospitals in 2004-2005. The accuracy of the scoring system for prediction of mortality was measured by the area under the receiver operating characteristic curve (AUC). The performance of the new scoring system was further validated by comparing its predictive accuracy with that of Acute Physiology and Chronic Health Examination (APACHE) II.