Mayo analyzes 1.1M EHR notes linking preexisting conditions to long-term COVID-19

Rochester, Minn.-based Mayo Clinic teamed up with artificial intelligence firm Nference to analyze 1.1 million clinical notes and find the connection between preexisting conditions and the severity of COVID-19 symptoms, according to a July 27 report published in NPJ Digital Medicine.

The study analyzed EHR notes of 1,803 hospitalized patients who were seen at the health system from March 12, 2020, to Sept. 15, 2020. Researchers characterized the links between 21 preexisting conditions and the development of 20 complications. Then, the study looked at the relationship between preexisting conditions and the length of infection (0-30 days, 31-60 days and 61-90 days).

Four key findings:

  1. Hypertension was the most significant risk factor, associated with 10 different complications, such as pleural effusion, acute respiratory distress syndrome, cardiac arrhythmia and anemia.

  2. Eighty-nine patients experienced pleural effusion in the months following a positive COVID-19 test, followed by 45 patients who had cardiac arrhythmias, 33 patients who had heart failure and 29 patients who had respiratory failure.

  3. Pleural effusion was strongly linked to preexisting conditions, with the largest number of patients who had it noting hypertension, Type 2 diabetes, chronic kidney disease, anemia and steroid use as preexisting conditions.

  4. Of patients who experienced symptoms for more than two months after their first positive COVID-19 test, pleural effusion was most commonly associated with patients who had cancer, chronic kidney disease and Type 2 diabetes as preexisting conditions.

Copyright © 2022 Becker's Healthcare. All Rights Reserved. Privacy Policy. Cookie Policy. Linking and Reprinting Policy.


Featured Whitepapers

Featured Webinars