Design and development of an evidence-based cognitive aid for out-of-hospital treatment of pediatric patients - prehospital pediatric care is a low-frequency, high-stake event

Medication errors and delayed interventions are common in this domain, which could lead to severe consequences. The occurrence of many safety events can be attributed to the relative lack of prehospital provider experience and knowledge in the care of acutely ILL and injured children in the out-of-hospital setting. Despite some efforts, there is still a lack of effective cognitive support for prehospital providers to promote recognition of clinical conditions, reinforce protocol adherence, and support a shared mental model in the care of children by EMS. In this project, our goal is to design and develop a novel evidence-based, context-sensitive cognitive aid for prehospital providers to support their treatment of pediatric patients in the out-of-hospital setting while considering three prominent research gaps: first, prior work primarily relied on simulation to assess medical errors in prehospital children care. Even though simulation is a promising approach for identifying medical errors and areas for improvement, the assessed medical scenarios and identified medical errors are often limited. Second, limited research has focused on developing digital cognitive aids for prehospital pediatric care that can account for and alert various prehospital pediatric patient safety events. Third, the challenges of implementing cognitive support interventions in the dynamic, fast-paced, and cognitively consuming prehospital context still persist and remain unaddressed, such as misalignment between system design and clinical workflow, and cognitive overload of using the system under time pressure. Considering these research gaps, we propose three compelling specific aims: 1) characterize medical errors in prehospital care of children through a "human-in-the-loop computational approach. 2) design and develop cognitive aids for prehospital providers. 3) implement and evaluate the proposed technology solution in simulated environments. To carry out these research activities, we formed an interdisciplinary team consisting of experts from pediatric emergency care, human-computer interaction, natural language processing (NLP), and artificial intelligence (AI). Our work will contribute novel knowledge, including a comprehensive assessment and categorization of medical errors occurred during prehospital children care through a large-scale EMS EHR data analysis, approaches to generate decision recommendations, and design insights regarding how to present cognitive supports and suggestions to fast-paced medical teams in an easy-to-absorb, less distractive manner to avoid alert fatigue and disruptions to workflow. This proposal aligns with the priorities of NIH in developing novel digital health technologies to improve healthcare services delivery at the point of care.