This protocol will collect real-world data retrospectively from the electronic health record (EHR) as data obtained from the delivery of routine medical care to develop a machine learning (ML)-based Clinical Decision Support (CDS) system for severe sepsis prediction and detection.
The purpose of this study is to gather data for the clinical development of the Sepsis Onset Warning System (SOWS) Software as Medical Device (SaMD) product to support a De Novo FDA submission and commercialization in the United States. Product development of SOWS is funded in part with federal funds from the Department of Health and Human Services; Office of the Assistant Secretary for Preparedness and Response; Biomedical Advanced Research and Development Authority.
Data will be obtained from passive prospective collection of patient encounter data throughout the duration of the planned study to support the product development life cycle activities associated with developing the Sepsis Onset Warning System (SOWS) for severe sepsis risk detection. Inputs from patient health records in combination with proprietary hematology parameters developed by Beckman Coulter, such as Monocyte Distribution Width (MDW), will be used. The SOWS tool will look to use clinical measurements which are commonly and reliably available in the EHR as structured data elements, such as heart rate, temperature, blood pressure, and laboratory results and account for changes in these values over time.
