A common symptom of hypertension, an often underrecognized condition, can be found “hidden” within existing electronic health records, according to a study published March 31 in JAMA Cardiology.
Researchers from Boston-based Harvard Medical School and New York City-based NYU Grossman School of Medicine created an AI-algorithm to analyze patient EHRs within the Somerville, Mass.-based Mass General Brigham health system between March 2023 and June 2024.
The algorithm was used to identify patients with imaging that showed thickening of the heart muscle, frequently caused by hypertension, but who were not being treated for the condition, according to a March 31 news release from Mass General Brigham.
Here are five things to know from the study:
- Using existing EHR data, the AI-algorithm identified 648 patients who had not been diagnosed or treated for hypertension, but who had shown thickening of the heart walls on an echocardiogram.
- The physicians for 326 of those patients were notified of the finding by a population health coordinator, who also provided additional follow-up services. The physicians of the remaining 622 patients were not notified and continued to provide standard care.
- Over a 12-month period, 15.6% of patients whose physicians had been notified received hypertension diagnoses, compared to 4% of patients whose physicians were not notified.
Similarly, 16.3% of patients in the intervention group were prescribed antihypertensive medication, compared to 5% in the control group. - Of the notified physicians, 82% responded to the notification and 72% had a positive reaction to the intervention.
- “This work highlights the potential benefit of leveraging preexisting, but potentially underutilized, cardiovascular data to improve health care delivery,” the study authors wrote.
Read the full study here.
Leave a Reply