Long COVID has myriad symptoms, ranging from fatigue and headaches to cardiovascular and respiratory complications. Despite the high variability, a team of researchers from multiple U.S. institutions have been able to categorize Long COVID into four distinct subtypes based on clusters of symptoms that frequently appear. Their findings have implications for disease management and outlook.*

What the researchers did

Until recently, most Long COVID studies investigated symptoms individually. To remedy this oversight, Rainu Kaushal, M.D., MPH, chair of Weill Cornell Medicine’s Department of Population Health Sciences, spearheaded the largest Long COVID analysis of its kind in a search for symptom patterns. The study, published in the journal Nature Medicine, was funded by a $9.8 million grant from the National Institutes of Health’s RECOVER (Researching COVID to Enhance Recovery) initiative. For this study:

• Electronic health records (EHRs) of 34,605 Americans with documented COVID-19 were derived from the National Patient-Centered Clinical Research Network (PCORnet).
• Two datasets were compared: INSIGHT, with 20,881 people from New York, versus OneFlorida+, with 13,724 people from Florida, Georgia, and Alabama.
• Using a machine learning algorithm, records were analyzed for patterns among more than 137 symptoms and conditions experienced 30-180 days after initial infection.

What they found

The researchers identified four patterns of Long COVID symptoms, or four (4) post-COVID subtypes, among INSIGHT patients linked to their demographics, pre-existing health conditions, and COVID-19 severity. These included:

Heart, kidney, and circulation issues

• 34% of patients
• 49% male
• Older average age (median age 65 years)
• Relatively greater number of pre-existing conditions
• 37% infected during the pandemic’s first months in the U.S.
• 61% of those hospitalized for COVID-19

Respiratory and sleep disorders, anxiety, headache, and chest pains

• 33% of patients
• 63% women
• Median age 51 years
• Pre-existing conditions mostly respiratory (asthma, for example)
• Most infected during later pandemic waves
• 31% of those hospitalized for COVID-19

Musculoskeletal and nervous system issues

• 23% of patients
• 61% women
• Median age 57 years
• 38% of those hospitalized for COVID-19

Digestive and respiratory system issues

• 10% of patients
• 62% women
• Median age 54 years
• 33% of those hospitalized for COVID-19

White patients made up the predominant race/ethnicity for all groups, followed by Hispanic patients, except for the digestive and respiratory issues subtype in which Hispanics dominated. Women dominated all subtypes except for the first (heart-kidney-circulation). This gender difference in Long COVID risk aligns with previous findings, but no one yet knows why this is.

When these findings were compared to the OneFlorida+ group, similar results were discovered. Participants with negative COVID tests, who developed symptoms during the 30- to 180-day study period, didn’t produce similar patterns.

What it means

The authors say their results will support other research into the causes of Long COVID and new treatments, and hopefully expedite specialized treatments for patients in each group. They’re following up on their results in several ways.

• Defining Long COVID symptom patterns so they can be easily identified from EHRs
• Identifying risk factors for these symptom patterns
• Finding existing therapies that can be repurposed for Long COVID

“Understanding the epidemiology of Long COVID,” Kaushal said, “allows clinicians to help patients understand their symptoms and prognoses, and facilitates multi-specialty treatment for patients.”

*Weill Cornell Medicine. (2022, December 21). Study Identifies Four Major Subtypes of Long COVID. https://news.weill.cornell.edu/news/2022/12/study-identifies-four-major-subtypes-of-long-covid