Article Summary
Body
- New research shows the potential for widely available wearable devices, like Fitbits, to predict postoperative complications in children.
- Using a novel algorithm, scientists were able to retrospectively predict postoperative complications in children up to three days before formal diagnosis with 91% sensitivity and 74% specificity.
- By more quickly identifying complications, clinicians will be better positioned to administer appropriate treatment before an issue becomes more serious.
Body
An estimated 4 million children undergo surgical procedures in hospitals across the United States each year. Although postoperative complications, such as infections, can pose significant health risks to kids, timely detection following hospital discharge can prove challenging.
A new study published in Science Advances — and led by researchers at Shirley Ryan AbilityLab, the Ann and Robert H. Lurie Children’s Hospital of Chicago and University of Alabama at Birmingham — is the first to use consumer wearables to quickly and precisely predict postoperative complications in children and shows potential for facilitating faster treatment and care.
“Today, consumer wearables are ubiquitous, with many of us relying on them to count our steps, measure our sleep and more,” said Arun Jayaraman, PT, PhD, the study’s senior author, a scientist at Shirley Ryan AbilityLab and professor at Northwestern Medicine. “Our study is the first to take this widely available technology and train the algorithm using new metrics that are more sensitive in detecting complications. Our results suggest great promise for better patient outcomes and have broad implications for pediatric health monitoring across various care settings.”
As part of the study, commercially available Fitbit devices were given to 103 children for 21 days immediately after appendectomy — the most common surgery in children, which results in complications up to 38% of the time. Rather than just using the metrics automatically captured by the Fitbit to identify signs of complications (e.g., low activity, high heart rate, etc.), Shirley Ryan AbilityLab scientists trained the algorithm using new metrics related to the circadian rhythms of a child's activity and heart rate patterns.
In the process, they found such metrics were more sensitive to picking up complications than the traditional metrics. In fact, in analyzing the data, scientists were able to retrospectively predict postoperative complications up to three days before formal diagnosis with 91% sensitivity and 74% specificity.
“Historically, we have been reliant upon subjective reporting from children — who often have greater difficulty articulating their symptoms — and their caregivers following hospital discharge. As a result, complications are not always caught right away,” said Fizan Abdullah, MD, PhD, who conducted the research while serving as an attending physician of pediatric surgery at Lurie Children’s and a professor at Northwestern Medicine. “By using widely available wearables, coupled with this novel algorithm, we have an opportunity to change the paradigm of postoperative monitoring and care — and improve outcomes for kids in the process.”
This research is part of a four-year, National Institutes of Health (NIH)-funded project. As a next step, scientists plan to transition this approach into a real-time (vs. retrospective) system that analyzes data automatically and sends alerts to children’s clinical teams.
“This study reinforces wearables’ potential to complement clinical care for better patient recoveries,” said Hassan M.K. Ghomrawi, PhD, MPH, vice chair of research and innovation, Department of Orthopaedic Surgery at University of Alabama at Birmingham. “Our team is eager to enter the next phase of research exploration.”
Journal: Science Advances
Title: Biorhythms Derived from Consumer Wearables Predict Postoperative Complications in Children
Authors: Rui Hua, Michela Carter, Megan K. O’Brien, J. Benjamin Pitt, Soyang Kwon, Renee C. B. Manworren, Gia Oscherwitz, Arianna Edobor, Austin Chen, Hassan MK Ghomrawi, Fizan Abdullah, and Arun Jayaraman
