No abstract

BackgroundTrauma is predicted to become the third leading cause of death in India by 2020, which indicate the need for urgent action. Trauma scores such as the international classification of diseases injury severity score (ICISS) have been used with great success in trauma research and in quality programmes to improve trauma care. To this date no valid trauma score has been developed for the Indian population.Study designThis retrospective cohort study used a dataset of 16047 trauma-patients from four public university hospitals in urban India, which was divided into derivation and validation subsets. All injuries in the dataset were assigned an international classification of disease (ICD) code. Survival Risk Ratios (SRRs), for mortality within 24 hours and 30 days were then calculated for each ICD-code and used to calculate the corresponding ICISS. Score performance was measured using discrimination by calculating the area under the receiver operating characteristics curve (AUROCC) and calibration by calculating the calibration slope and intercept to plot a calibration curve.ResultsPredictions of 30-day mortality showed an AUROCC of 0.618, calibration slope of 0.269 and calibration intercept of 0.071. Estimates of 24-hour mortality consistently showed low AUROCCs and negative calibration slopes.ConclusionsWe attempted to derive and validate a version of the ICISS using SRRs calculated from an Indian population. However, the developed ICISS-scores overestimate mortality and implementing these scores in clinical or policy contexts is not recommended. This study, as well as previous reports, suggest that other scoring systems might be better suited for India and other Low- and middle-income countries until more data are available.

No abstract

No abstract

Rationale The 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 when selecting patients for review, a process that is hampered by high frequencies of false positives. Objectives To develop, validate, and compare the performance of different machine learning models for predicting opportunities for improvement. Methods We 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 was also compared to current audit filters. Measurements and Main Results The 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). Conclusions Machine 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.