Prognostic models in traumatic brain injury

Background Prognostic models are important tools for heterogeneity adjustment in traumatic brain injury ( TBI ). Prognoses after TBI have been particularly challenging to predict, with limited availability of robust prognostic models. TBI patients are by definition trauma patients, and often treated in the intensive care unit ( ICU ). Several prognostic models for ICU and trauma patients have been developed, although their applicability in patients with TBI is uncertain. Recently, however, some new prognostic models specifically designed for patients with TBI were introduced. Still, the optimal type of prognostic model in TBI remains unknown. Aim This aims to investigate the applicability of different types of prognostic models in patients with TBI and to develop novel models with enhanced performance to previous models, focusing on long‐ term outcome prediction. Methods Four patient databases of patients with TBI treated in the ICU were used to validate three TBI specific models, two computerized tomography ( CT ) scoring systems, one trauma scoring system, and three intensive care scoring systems. Models were validated by assessing their discrimination using area under the curve ( AUC ), calibration, and explanatory variation. Logistic regression was used for model customization and development. Models were internally validated using a resample bootstrap technique or a split‐sample technique. Primary outcome was 6‐month mortality and unfavorable neurological outcome by the G lasgow O utcome S cale. 30‐day in‐hospital mortality was used for the trauma scoring system. Results Study populations ranged from 342 to 9915 patients. The TBI models showed the best performance with AUC s between 0.80 and 0.85, followed by the intensive care scoring systems and the CT scores with AUC s between 0.68 to 0.80 and 0.63 to 0.70, respectively. Most models showed poor calibration, although good calibration was achieved following customization. The trauma scoring system exhibited modest to good discrimination ( AUC 0.76–0.89) for short‐term mortality prediction but poor calibration. Several new prognostic models, with statistically significant superior performance to previous models were created, among them a combined TBI ‐ ICU model (‘ IMPACT ‐ APACHE ’) and a novel CT scoring system (‘ T he H elsinki CT score’). Using a TBI ‐specific model, based on admission characteristics, up to 40% of the patient's final long‐term outcome could be predicted. Conclusion The TBI models showed superior predictive performance to the intensive care and trauma scoring systems, showing that TBI patients are a highly specific population in the trauma and ICU setting. Thus, the use of a TBI ‐specific model is advocated in the setting of TBI . The newly proposed models were found to be significant improvements over previous models, but require external validation to show generalizability.