Stratification of the severity of critically ill patients with classification trees
Rodríguez Pozo, Ángel
MetadataShow full item record
Background: Development of three classification trees (CT) based on the CART (Classification and Regression Trees), CHAID (Chi-Square Automatic Interaction Detection) and C4.5 methodologies for the calculation of probability of hospital mortality; the comparison of the results with the APACHE II, SAPS
II and MPM II-24 scores, and with a model based on multiple logistic regression (LR). Methods: Retrospective study of 2864 patients. Random partition (70:30) into a Development Set (DS) n = 1808 and Validation Set (VS) n = 808. Their properties of discrimination are compared with the ROC curve (AUC CI 95%), Percent of correct classification (PCC CI 95%); and the calibration with the Calibration Curve and the Standardized Mortality Ratio (SMR CI 95%). Results: CTs are produced with a different selection of variables and decision rules: CART (5 variables and 8 decision rules), CHAID (7 variables and 15 rules) and C4.5 (6 variables and 10 rules). The common variables were: inotropic therapy, Glasgow, age, (A-a)O2 gradient and antecedent of chronic illness. In VS: all the models achieved acceptable discrimination with AUC above 0.7. CT: CART (0.75(0.71-0.81)), CHAID (0.76(0.72-0.79)) and C4.5 (0.76(0.73-0.80)). PCC: CART (72(69- 75)), CHAID (72(69-75)) and C4.5 (76(73-79)). Calibration (SMR) better in the CT: CART (1.04(0.95-1.31)), CHAID (1.06(0.97-1.15) and C4.5 (1.08(0.98-1.16)). Conclusion: With different methodologies of CTs, trees are generated with different selection of variables and decision rules. The CTs are easy to interpret, and they stratify the risk of hospital mortality. The CTs should be taken into account for the classification of the prognosis of critically ill patients.
Is part ofBMC Medical Research Methodology, 2009, vol. 9, núm. 83, p. 1-12
The following license files are associated with this item:
Except where otherwise noted, this item's license is described as cc-by, (c) Trujillano et al., 2009
Showing items related by title, author, creator and subject.
Classification of dermatological disorders in critical care patients: a prospective observational study Badia Castello, Mariona; Servià Goixart, Lluís; Casanova i Seuma, Josep M. (Josep Manel); Montserrat, Neus; Vilanova, Judit; Vicario, Esther; Rodriguez, Angel; Trujillano Cabello, Javier (Elsevier, 2013)Purpose: The objective of this study was to identify dermatological disorders detected in the intensive care unit (ICU), to analyze their specific characteristics, and to define a useful classification for intensive care ...
Plasma antioxidant capacity in critical polytraumatized patients?: methods, severity, and anatomic location Servià Goixart, Lluís; Trujillano Cabello, Javier; Serrano Casasola, José Carlos Enrique; Pamplona Gras, Reinald; Badia Castello, Mariona; Jové Font, Mariona; Justes, Margarida; Domingo, Joana; Portero Otín, Manuel (BioMed Central, 2014)
A prospective pilot study using metabolomics discloses specific fatty acid, catecholamine and tryptophan metabolic pathways as possible predictors for a negative outcome after severe trauma Servià Goixart, Lluís; Jové Font, Mariona; Sol, Joaquim; Pamplona Gras, Reinald; Badia Castello, Mariona; Montserrat, Neus; Portero Otín, Manuel; Trujillano Cabello, Javier (BMC (part of Springer Nature), 2019-05-22)Background: We wanted to define metabolomic patterns in plasma to predict a negative outcome in severe trauma patients. Methods: A prospective pilot study was designed to evaluate plasma metabolomic patterns, established ...