Machine learning improves identification of asthma risk in children

A machine learning tool that analyzes information already captured in a child’s electronic health record helped pediatricians more accurately assess asthma risk in standardized clinical case scenarios, according to a pilot randomized clinical trial led by a Regenstrief Institute researcher. The study was published in Scientific Reports.

The study evaluated a machine learning-enabled clinical decision support tool called the Passive Digital Marker, which uses routinely collected EHR data to classify young children as having a high or low risk of developing persistent asthma.

Asthma is one of the most common chronic childhood diseases, but predicting which young children with wheezing or other respiratory symptoms will go on to develop persistent asthma remains difficult. While some children outgrow these symptoms, others require ongoing treatment, making early risk assessment an important but challenging part of pediatric care.

“This tool doesn’t replace a pediatrician’s clinical judgment,” said Arthur H. Owora, Ph.D., Regenstrief Institute research scientist and lead author of the study. “It helps bring together years of clinical information that’s already in the electronic health record, giving clinicians another source of information when making decisions about a child’s asthma risk.”

Unlike many prediction tools, the Passive Digital Marker requires no additional testing or questionnaires. Instead, it analyzes information already documented in the EHR, including respiratory symptoms, allergies, medication history, respiratory infections and family history, then provides clinicians with a simple high- or low-risk assessment.

Pediatricians using the tool correctly predicted future asthma more often than those using standard assessment alone, achieving an average accuracy of 83% compared with 61%. The improvement was largely driven by better identification of children who later developed persistent asthma.

The researchers emphasize that the tool is designed to support clinical decision-making, not replace it, by helping clinicians quickly synthesize years of patient information into an easy-to-interpret risk assessment.

Because the study was conducted using standardized patient cases rather than real-world clinical encounters, additional research is needed to determine whether the tool improves patient outcomes in everyday pediatric practice.

Who’s behind this story?

Gaby Clark

MA in English, copy editor since 2021 with experience in higher education and health content. Dedicated to trustworthy science news.

Full profile →

Advertisements

Andrew Zinin

Master’s in physics with research experience. Long-time science news enthusiast. Plays key role in Science X’s editorial success.

Full profile →