Implementation of a Smartphone-based Decision Tool to Determine Level of Care for Pediatric Head Trauma
Seth D Goldstein, Marcelo Cerullo, Corina Noje, Eric Henderson, Philomena Costabile, *Susan Ziegfeld, Lisa Puett, Katherine Hoops, Lindsey Rasmussen, Courtney Robertson, Eric Jackson, *Isam Nasr
Johns Hopkins Hospital, Baltimore, MD
Background (issue): The determination of hospital level of care for pediatric patients with mild traumatic brain injury (mTBI) is not currently evidence based nor standardized. The objectives of this study were to develop a prognostic scoring tool to determine need for admission to an intensive care unit (ICU) for children with mTBI, followed by construction of a smartphone app to use as a clinical decision making tool.
Methods: Patients <5 years old with mTBI and Glasgow Coma Score (GCS) ≥13 in a pediatric trauma center 2013-2015 were identified. ICU-level monitoring requirement was determined by eventual need for surgery, changes on repeat head imaging, in-hospital status epilepticus, or intubation. A predictive model was derived using logistic regression of baseline clinical risk factors.
Findings: A clinical decision nomogram was created after analysis of 200 patient records. Testing of the model resulted in a 10-fold cross-validation C-statistic of 0.87. Negative predictive value for ICU admission was 98%. 62.4% had a GCS of 15. A smartphone app was created and can be accessed via web browser at https://s3.amazonaws.com/jhdctr/main.html
Conclusions (implications for practice): In children with mTBI, objective factors can be used to predict need for ICU admission. We now present a robust decision support tool that is easily used by clinicians via smartphone app, which we believe will lead to better use of limited resources.
Back to 2017 Program and Abstracts