| Popis: |
1Abstract1.1RationaleThe multidisciplinary mortality and morbidity conference is the core of programs that aim to improve the quality of trauma care and is used to identify and address opportunities for improvement based on reviewing patient cases. Current systems rely on audit filters for review selection, a process that is hampered by high frequencies of false positives.1.2ObjectivesTo develop, validate, and compare the performance of different machine learning models for predicting opportunities for improvement.1.3MethodsWe conducted a registry based study using all patients from the Karolinska university hospital that had been reviewed regarding the presence of opportunity for improvement, a binary consensus decision from the mortality and morbidity conference. We developed eight binary classification models using 45 predictors. Training used an 80%-20% train-test split and 1000 resamples without replacement estimated confidence intervals. Performance (sensitivity, specificity, integrated calibration index, Area under the receiver operating characteristics curve) was also compared to current audit filters.1.4Measurements and Main ResultsThe dataset included 6310 patients where opportunities for improvement were present among 431 (7%) patients. The audit filters (Area under the receiver operating characteristics curve: 0.624) was outperformed by all machine learning models. The best performing model was LightGBM (Area under the receiver operating characteristics curve: 0.789).1.5ConclusionsMachine learning models outperform the currently used audit filters and could prove to be valuable additions in the screening for opportunities for improvement. More research is needed on how to increase model performance and how to incorporate these models into trauma quality improvement programs.ImpactOur research explores a novel system that predicts opportunities for improvement in trauma patients using machine learning and outperforms the established approach using audit filters. This could allow for reallocation and optimization of review resources as well as a means of possibly identifying new types of opportunities for improvement. The methodology uses state-of-the-art machine learning largely unexplored in a medical context adding to a broader general scientific knowledge reaching outside this project. |
