Michael C. Spaeder, MD, MS
Michael Spaeder MD, MS
Professor of Pediatrics
Program Director, Fellowship in Pediatric Critical Care Medicine
Dr. Spaeder is an academic pediatrician and a pediatric critical care specialist. Before his medical training, he completed graduate studies in mathematical statistics at George Washington University and was employed as a statistical consultant with the Economics Consulting and Quantitative Analysis group at Ernst & Young, LLP in Washington, DC. In his medical research career, he has sought to bridge his foundation in mathematics and statistics with his clinical focus, caring for critically ill and injured infants and children in the pediatric intensive care unit (PICU). His research, based in the Center for Advanced Medical Analytics at UVA, focuses on the use of physiologic monitoring data to identify patients at risk for clinical deterioration.
Critically ill and injured infants and children in the PICU often deteriorate suddenly due to sub-acute potentially catastrophic illness. Standard around-the-clock bedside monitoring of physiologic parameters, which might give clues about incipient deterioration, often flies by unnoticed. To solve this problem of undetected changes, predictive monitoring integrates these data to give clinicians early warning of events, allowing earlier, more effective intervention.
The Center for Advanced Medical Analytics at UVA is a national leader in the creation and implementation of predictive modeling algorithms for clinical deterioration in critically ill patients. These algorithms often work through the analysis of the behavior of certain physiologic parameters (e.g. heart rate variability). Often coupled with data extracted from the electronic health record, individualized patient stability indices can be calculated, allowing for early identification of at-risk patients.
Dr. Spaeder has developed predictive algorithms for sepsis and respiratory failure for infants and children in the PICU using machine learning techniques, harnessing data from the bedside monitor and electronic health record. These algorithms are now incorporated into the Continuous Monitoring of Event Trajectories (CoMET) predictive analytics software (Nihon Kohden Digital Health Solutions, Irvine, CA). Implemented in the UVA Health Children’s PICU since July of 2022, Dr. Spaeder and his team recently published on improved cardiac arrest outcomes observed following implementation.*
Dr. Spaeder is working with colleagues at Boston Children’s Hospital and the Children’s Hospital of Atlanta to externally validate the algorithms developed at UVA as well as develop new algorithms for clinical deterioration. He is also engaged in efforts to evaluate how well his algorithms, originally developed to predict sepsis and respiratory failure, perform in predicting other clinical outcomes such as necrotizing enterocolitis and cardiac arrest.
* Spaeder MC, Lee L, Miller C, Keim-Malpass J, Harmon WG, Kausch SL. Incidence of cardiac arrest following implementation of a predictive analytics display in a pediatric intensive care unit. Resusc Plus. 2025 Jan 2;21:100862.
Visit for more Publications https://www.ncbi.nlm.nih.gov/myncbi/michael.spaeder.2/bibliography/public/