• Ramona Haslinger (Speaker)

Activity: Talk or presentationOral presentation


Objectives: The aim of this study was to compare different classification methods for the retrospective detection of patients with one or more surgical and/or medical complications. Methods: Data from hospital data bases of eight clinics in Upper Austria were combined with data collected by data nurses from handwritten patient documentation. Complication (yes/no) was selected as the target variable. Patients undergoing acute appendectomy (n= 1341), acute and elective cholecystectomy (n= 1307) and elective inguinal hernia repair (n = 1574) were included into the study. Data on services received by the patients was combined with 45 parameters concerning patient risk. For each of the three datasets three different methods, logistic regression, boosting for generalized linear models (boosting) and classification trees (ctree) were applied to detect patients with complications. To assess the classification performance of the models, positive predictive value (PPR), rate of positive predictions (RPP) and sensitivity were compared after the application of the models on additional test data. In addition to predictive performance the number variables in the final model is also an important criterion for choice of the final model as more variables need more efforts on data collection and preparation. Results: A comparison of sensitivity, PPV and RPP showed that models selected by ctree perform better than the classical logistic regression and boosting models. With PPVs between 16 % and 30 %, sensitivities from 80 % to 95 % and RPPs around 30 %. Additionally the ctree involved the smallest number of variables going from 6 to 9. Conclusion: Application of classification tree models on existing data can save time and costs for the identification of patients with a history of surgical and medical complications in surgery.
Period8 Sept 2017
Event title6th World Congress of Clinical Safety
Event typeWorkshop
LocationRom, ItalyShow on map