Researchers Use Big Data to Establish Long COVID Subtypes – UBNow: News and Views for UB Faculty and Staff

Announced earlier this year by the National Academies of Sciences, Engineering, and Medicine Consensus Definition Long COVID researchers said it's not the final word on the condition, but that the definition will likely be revised as new research emerges.

Earlier this year, researchers from UB and the Department of Veterans Affairs who were working on a data-driven long COVID-19 definition using big data were asked to report their findings in testimony before the National Academies.

They published the results of their study in April. JMIR Public Health Surveillance It's based on more than 250,000 patients who have tested positive for COVID at the Veterans Health Administration.

“The VA was a great place to do this study because we had a large number of patients,” says corresponding author Peter L. Elkin, dean of biomedical informatics at UB's Jacobs School of Medicine and Biomedical Sciences. “We were able to look at all of the patients' electronic medical records and analyze all of the different symptoms of this complex disease and find all of the different subtypes that exist.”

One goal is to help clinicians better recognize cases of long COVID. “The idea is to get more clinicians to recognize who has long COVID and who hasn't come out of long COVID yet,” he explains. “Now that we have a definition, they can tell who in their specialty has long COVID and who doesn't.”

The findings could help researchers develop a more scientific approach to understanding how COVID infection causes certain symptoms seen in long COVID.

The study was conducted using electronic medical records of more than 2.3 million patients seen at VA medical facilities between Jan. 1, 2020 and Aug. 18, 2022. There were 367,148 patients who tested positive for COVID-19 at a VA facility, of which 268,320 were considered to have long COVID because they received a new diagnosis within one to seven months of their positive COVID-19 test.

The researchers assigned a total of 324 ICD (International Classification of Diseases) codes to Long COVID patients based on their symptoms. They identified 180 clinical scenarios and 17 clinical subtypes that were increased in Long COVID patients.

Cardiology saw the highest number of long COVID cases, with diagnoses including low blood pressure, heart failure, arrhythmia and atrial fibrillation, followed by neurology (symptoms include lower back pain, severe muscle weakness and cognitive impairment), ophthalmology and pulmonology.

The most commonly cited symptoms were fatigue and acute shortness of breath. Respiratory problems are among the most common symptoms of long COVID, including chronic cough, respiratory failure and dependence on supplemental oxygen. Cognitive problems, such as a feeling of brain fog, were also frequently mentioned.

Factors that increase a patient's risk of developing long COVID include older age, other health conditions before infection, severity of COVID, decreased oxygen saturation during COVID, etc. Patients who did not receive the COVID-19 vaccine were 1.3 times more likely to develop long COVID.

The study provides definitions for each subtype of long COVID and odds ratios that define the risk for each.

“These data will enable us to more accurately identify patients with long COVID-19 and will help inform future public health research and policy initiatives,” Elkin said.

The authors note that a limitation of this study is that 84% of study participants were men.

Elkin's co-authors are postdoctoral researchers Skylar Resendez, Hugo Sebastian Ruiz Ayala and Prahalad Rangan in the Jacobs School's Department of Biomedical Informatics, and Steven H. Brown, Jonathan Nebeker and Diana Montera of the Department of Health Informatics at the Department of Veterans Affairs.

The research was supported by the National Library of Medicine, the National Institute on Alcohol Abuse and Alcoholism, the National Center for Advancing Translational Sciences, the National Institutes of Health, and the Department of Veterans Affairs.