Search reports, ask a question, or jump to a page.
The problem
Intervention: Electronic health record-integrated machine learning-based Passive Digital Marker (PDM) that classifies children as high or low risk for persistent school-age asthma using routinely collected clinical data
Who it was tested with: 34 practicing pediatricians in Indiana evaluating 10 standardized patient vignettes; children aged birth to 3 years with follow-up to ages 6-11 years.
Children or communities are compared with a group that did not receive the intervention, or followed over years. This is the design that best separates what the intervention did from everything else that changed.
"Clinicians with access to the PDM demonstrated higher prognostic accuracy than clinicians without access (mean [95% CI] 0.83 [0.69-0.97] vs 0.61 [0.52-0.70]; P = .008)." The randomized design with control and intervention groups and standardized assessment timing provides experimental evidence that the machine learning tool causally improves clinician accuracy.
Small pilot sample from single state may limit generalizability; standardized vignettes may not reflect real clinical decision-making; limited to one region's healthcare system
Across countries and designs
1 study · 1 country
