• Posted May 29, 2026

Long COVID Might Be Twice As Common As Previously Thought

The number of people suffering with long COVID could be double current estimates, a new study suggests.

An AI tool found that about 16% of nearly 460,000 patients with COVID-19 had developed long COVID, researchers reported May 27 in JAMA Network Open.

Applied across the United States, those rates translate to more than 18 million Americans with long COVID, which is twice as high as current estimates, researchers said.

“Over 10 million people with long COVID would go entirely undetected by the diagnostic code that health systems and policymakers rely on to track the disease burden,” said senior researcher Hossein Estiri, an associate professor in the Mass General Brigham Department of Medicine in Boston.

And it’s likely the picture is even worse than these estimates, researchers said.

“The figures we uncovered are almost certainly an undercount,” Estiri said in a news release.

The current diagnostic coding used for post-COVID conditions captures fewer than 7% of patients with long COVID, researchers said in background notes.

For the new study, researchers developed an AI algorithm that can detect long COVID by identifying conditions that appeared after a person’s infection but can’t be explained by any preexisting conditions in their medical history.

The team then ran the AI through medical records from COVID patients treated at 58 U.S. hospitals.

Results showed that roughly 16% of patients overall had long COVID, with rates ranging from nearly 14% to as high as almost 23% across regions.

Researchers also found that long COVID cases continued to increase through mid-2024 across all regions studied, contrary to the assumption that long COVID is a legacy of the pandemic’s early waves.

The AI found significant quarterly increases in New England, Southern California and Western Pennsylvania, researchers said.

Worse, trends point to continued growth over the next decade if current patterns persist.

“These patients are not absent from clinical care; they are absent from the diagnostic code that would identify them as long COVID patients,” said lead researcher Jiazi Tian, a data scientist in the Clinical Augmented Intelligence Group at Mass General Brigham.

“The cardiologist seeing new dysautonomia, the endocrinologist seeing new metabolic disease, the neurologist seeing unexplained cognitive complaints — some of these presentations are long COVID arriving without the label that would connect them to a COVID-19 infection,” Tian said in the release.

More information

The U.S. Centers for Disease Control and Prevention has more about long COVID.

SOURCE: Mass General Brigham, news release, May 27, 2026