Spanish Fork

Researchers are evaluating a clinical decision support (CDS) tool called ePNa, originally designed for emergency departments, to improve pneumonia diagnosis and treatment in urgent care clinics in Utah. This study focuses on adapting ePNa for use in urgent care centers (UCCs), where pneumonia patients are frequently seen, and combining it with Stanford's CheXED artificial intelligence model to enhance chest image analysis. The goal is to implement and test this adapted tool to support clinicians in making accurate and timely pneumonia care decisions, especially as diagnostic and treatment methods evolve during the COVID-19 pandemic. The study involves adapting ePNa to fit the data limitations and workflow of urgent care clinics, incorporating AI-based chest image classification that provides results in less than one second. The adapted tool will be piloted with selected "super user" clinicians and then deployed to one of two randomly chosen UCC clusters, while the other cluster continues usual care. The implementation process will follow the CFIR framework for best practices in integrating new clinical tools, including identifying barriers and facilitators through focus groups, interviews, and workflow observation. Participants include pneumonia patients aged 12 years and older with specific diagnostic codes, and clinicians working in the selected urgent care clinics. The study will collect data on pneumonia diagnosis accuracy, patient transfers to emergency departments, and safety outcomes like unplanned hospital visits and mortality within 30 days. Physician surveys will assess user experience with ePNa. The study is planned to last up to three years, with ongoing monitoring of how ePNa impacts urgent care clinical environments.