School of Medicine Publications and Presentations

Document Type

Conference Proceeding

Publication Date

Summer 4-30-2021


Introduction: Traumatic brain injury (TBI) is a significant cause of morbidity and mortality in children. While most TBI-related admissions are mild, identifying early predictors of poor outcome may assist clinicians with timely medical decisions and effective triage utilization.

Objective: We sought to develop and validate a clinical tool for predicting in-hospital death in children after a traumatic brain injury.

Methods: Data was collected for children (≤18 years) from the National Trauma Data Bank between the years of 2007 to 2015. We included children who sustained any TBI, defined as: (i) open and closed skull fractures, (ii) cerebellar, cortical, or brain stem contusions, and (iii) subarachnoid, subdural, or epidural hemorrhages. We excluded studies that did not report patient age or individuals who were dead on arrival or died in the emergency room (ER). Our interest was in clinical variables that can be readily measured upon admission to the ER. As such, the predictors included patient demographics, mechanism and intent of injury, vital signs in the ER, mode of transportation, respiratory status, time from injury to ER arrival, Glasgow Coma Score (GCS), and Injury Severity Score (ISS). Multivariable logistic regression, with forward selection, was used to investigate associations between predictive variable and mortality. We randomly split the samples into a training set (70%) and a test set (30%). Model performance was measured via the C-statistic and accuracy [(true positive + true negative) / patient population].

Results: A total of 124,078 children were included in the study (69% male; median [IQR] age, 13.0 [6.0, 16.0.] years; 69% White). The rate of death was 5.5% (n=6,862). Children more likely to die were older (16 vs. 12 years, p

Conclusion: Herein, we provide an accurate early prediction model for mortality in children after a TBI. Translation of our findings has led to the development of a web application that can be used by emergency healthcare providers in trauma centers.

Academic Level

medical student



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.