Decision Support for Prolonged, and Tactical Combat Casualty Care

Abstract

AbstractIn Tactical Combat Casualty Care (TCCC), evacuation from a battlefield may not be immediately available, resulting in a medic providing Prolonged Field Care (PFC) for one or more casualties over hours or days.We report on the development and evaluation of the Trauma Triage Treatment and Training Decision Support (4TDS) system, an Android phone and tablet-based application, that is designed to support clinician casualty care decisions, to detect the probability of a casualty experiencing shock, and to provide refresher training in life-critical skills. Refresher training scenarios ensure retention of crucial knowledge and skills needed to perform life-critical skills during extended deployments. As part of 4TDS, we also used machine learning to develop a model that scans vital signs data to detect the probability of a casualty developing shock, which is a life-threatening condition that is more likely to occur during PFC.Interface development methods included a literature review, rapid prototyping, and agile software development, while evaluation methods included design requirements reviews, subject matter expert reviews, and usability assessments. We developed a shock model using a logistical regression approach, trained the shock model on public Intensive Care Unit (ICU) patient data, then trained it on de-identified ICU patient data provided by our collaborators at Mayo Clinic. We evaluated the model in a silent test at Mayo Clinic over six months to compare model identification of shock with clinician decisions on the same patients.More timely, accurate decisions will improve battlefield casualty care outcomes and reduce the potential for misadventures.KeywordsDesign requirements reviewUsability assessmentDecision supportShock assessmentTactical Combat Casualty CareProlonged Field Care