A natural language processing (NLP) algorithm to identify patients with chronic cough, created and validated by Regenstrief Institute, Indiana University and Merck & Co. research scientists, is being recognized for its ingenuity and its potential.
The paper describing the algorithm’s development, Identifying and Characterizing a Chronic Cough Cohort Through Electronic Health Records, was published in the June 2021 edition of Chest Journal, along with an editorial written by Akio Niimi, M.D., PhD.
In the editorial, Dr. Niimi discussed the promise of the algorithm to improve chronic cough research. The condition is challenging to identify through electronic health records because it does not have an International Classification of Disease (ICD) code, therefore most indicators of the condition exist in unstructured data within the EHR. Cough is the most common complaint that leads patients to seek medical attention. Up to 40 percent of patients referred to specialist clinics do not learn the cause of their chronic cough, so this condition is an important research target.
In the study, led by Regenstrief research scientist and IU School of Medicine professor Michael Weiner, M.D., MPH, the algorithm improved detection of chronic cough patients nearly seven-fold and created the largest chronic cough cohort ever recorded.
Dr. Niimi states that chronic cough characterization from the EHR by NLP shows great potential as a research tool, enabling large cohort studies that were not possible before.
Funding for this research came from Merck & Co., Inc. This is part of a partnership between Regenstrief Institute and Merck to collaborate on projects using clinical data to inform delivery of healthcare.