CU Anschutz-developed tools consistently and accurately apply the new sepsis criteria to patients' electronic health records
Pediatric sepsis, a major inflammatory response to trauma that can cause organ failure and tissue death, kills about 3.3 million children each year worldwide. The catastrophic overreactions to infection or injury often hit lower-resource clinics the hardest.
In January, the diagnostic criteria for sepsis in children received its first update in two decades when a University of Colorado Anschutz Medical Campus-led international research team unveiled a novel, data-driven system for diagnosis. The modernized criteria, based on numeric scores of organ dysfunction, showed improved diagnostic performance compared with the prior pediatric sepsis criteria, which was based primarily on inflammatory response.
The new Phoenix criteria, led by Tell Bennett, MD, MS, professor of biomedical informatics and pediatric critical care at the CU School of Medicine, marks an important shift in the definition of pediatric sepsis, and it is useful in all types of healthcare settings around the world.
"That was the big picture - to get a diagnostic criteria for pediatric sepsis that was in line with the adult definition that was focused on life-threatening organ dysfunction and that would also be applicable across the globe," said Peter DeWitt, PhD, assistant research professor in the Department of Biomedical Informatics.
To create the diagnostic scoring system, the research team collected data from 10 hospitals, including in the United States, Bangladesh, China, Colombia and Kenya. The CU Anschutz research team of DeWitt, Research Assistant Professor Seth Russell, MS, and Research Service Program Manager Meg Rebull worked in collaboration with the Society for Critical Care Medicine (SCCM).
Tools offer accurate information for research
The next step was to provide researchers worldwide with the means to apply the Phoenix criteria to electronic health record data without needing to implement the criteria themselves. DeWitt said use of the tools - an R package and Python module - will allow researchers to calculate an accurate and consistent pediatric sepsis score. He is the first author on a just-published paper about the new tools in JAMIA Open.
DeWitt said he created the tools "to provide us a way to make sure that if somebody says, 'Based on this patient's record, they have sepsis' it shows that, yes, they do. We can trust that this diagnostic is accurate as opposed to having the question, 'Are you sure?' … We know that if this tool is used, then we can trust those conclusions in the research. It'll give us a good foundation for comparing results, not only within published papers, but across published papers."
Collecting pediatric-specific data was essential to the project, because sepsis and septic shock, as well as many other conditions, present differently in children, DeWitt said. While working to create the Phoenix criteria, "we needed to get a lot of samples and a really large data set," DeWitt said.
Next step: bringing the science to the bedside
"And that's what made this possible was the hard work of Tell Bennett, Nelson (Sanchez-Pinto, MD, MBI, FAMIA of Lurie Children's Hospital in Chicago) and the SCCM task force in getting shareholders interested in participating to give us data from all over the globe," DeWitt said. "We ended up with more than 3 million (patient) encounters in our data set."
Importantly, the two freely available tools - R package and Python module - will reduce errors in research and provide a common implementation for various research groups, across multiple data formats and programming paradigms.
While the tools are geared toward researchers, the CU Anschutz team is also collaborating with a Canadian research team on a tool for clinicians to diagnose pediatric sepsis patients at the bedside.
That's the necessary next step so that children who are at high risk of becoming septic can be quickly diagnosed, DeWitt said. "And then two different tracks of research can start from there - on interventions that reduce mortality and damage and those that improve patient outcomes."
