'Trending' clinicians' notes in EHRs may help predict real-time COVID-19 surges

Tracking "trending phrases" in EHR notes may help hospitals forecast surges in infectious disease outbreaks, according to a Feb. 24 report in Nature Digital Medicine.

The study analyzed keywords in EHRs to produce real-time maps of trending symptoms. Researchers monitored symptoms like "pneumonia" and "dry cough." The spikes resembled positive COVID-19 test result spikes reported by their respective hospital, indicating that tracking keywords could provide up to four days' notice for hospitals to prepare for surges in COVID-19 admissions.

Researchers said that tracking trending words is easy and cost-effective.

While this study operated at the single-hospital level, the study says it can be scaled at the local and national levels.

James Teo, PhD, clinical director of artificial intelligence at King's College Hospital and Guy's and St. Thomas' Hospital in London, said in a news release on the study, "By teaching computers how to read and understand doctors' notes, we hope to reveal important patterns and trends that could help in the fight against COVID-19 and other diseases."

More articles on EHRs:
Carnegie Mellon teams up with Amazon, Microsoft to build clinical forecasting AI tool
Allscripts posts $728M in net earnings for Q4: 4 notes
16 hospitals, health systems seeking Allscripts, Cerner, Epic, Meditech talent

 

© Copyright ASC COMMUNICATIONS 2021. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 

Featured Whitepapers

Featured Webinars