Our collective expertise at Lurie Children’s and Manne Research Institute drives considerable advances in pediatric medicine to improve the lives of children. One area where our physician-scientists and nurse-scientists are making an impact is sepsis. Earlier this year, three of our researchers with extensive experience in critical care and emergency medicine published two papers in JAMA that provide much-needed evidence to update the definition of sepsis in children. As members of an international effort to develop and validate new clinical criteria for pediatric sepsis, Elizabeth Alpern, MD, Division Head of Emergency Medicine and Vice Chair Department of Pediatrics, L. Nelson Sanchez-Pinto, MD, an attending physician in critical care, and Lauren Sorce, PhD, APRN-NP, a senior nurse scientist in critical care, applied a data-driven approach to developing new diagnostic criteria that stand to have a substantial impact on defining sepsis and septic shock in children.
The pediatric sepsis diagnostic criteria were last updated in 2005, and those were based on expert opinion, not data. Expert opinion is important, but it can introduce biases. That is why the work of the Pediatric Sepsis Definition Task Force, convened by the Society of Critical Care Medicine, is important as we transition from using expert opinion to data-driven evidence when developing and validating clinical diagnostic criteria for pediatric sepsis. Lauren notes that the Pediatric Sepsis Definition Task Force's work represents the first data-driven, machine learning-based approach of this magnitude in critical care. The task force’s data group, co-led by Nelson, based the updated criteria on data from electronic health records and an analysis of more than 3 million pediatric hospital encounters at 10 international hospitals. Nelson’s expertise as a critical care physician, clinical informatics specialist, and data scientist informs his application of data science and artificial intelligence to improve the care of children with critical care conditions; the National Institutes of Health funded this research on the new sepsis criteria. Nelson emphasizes the importance of incorporating data and machine learning into projects like this, explaining that using data from electronic health records gave the team a more objective foundation to create diagnostic criteria that are both credible and practical for use worldwide. Libby adds that the strength of data gives physicians a comprehensive approach to treating all patients that may be affected by a disease or condition—in this case, sepsis—and not overlook some patients or give too much weight to the experiences of other patients.
In addition to the importance of using data to shape the criteria, the member composition of the task force is significant. The Society emphasized a multiprofessional approach to patient care and prioritized representation to include experts from across the care continuum and specialties at sites worldwide. As co-vice chair of the task force and a nurse scientist with research and clinical experience in critical care medicine, Lauren not only guided the process of convening the task force but also ensured that the nursing voice was represented. For Libby, a pediatric emergency medicine physician-scientist, participating in the task force allowed her to represent her care setting and share the unique components of emergency care with other members. This included drawing on her research experience using large databases and evidence-based practices to improve the quality of emergent care.
While Libby, Nelson, and Lauren have different roles in pediatric research, they share the goal of improving how we treat critically ill children. Many of their Manne Research Institute colleagues are also valuable collaborators on national and international guideline panels, especially as these panels increasingly incorporate data into the guideline development process. Please share in my gratitude to our research experts who have represented the research institute on panels and contributed their expertise in guideline development processes.
