Over 10 Million Americans Have Long COVID and Don't Know It, Groundbreaking AI Study Reveals

Byline: CM News Health & Science Desk Date: June 5, 2026

A landmark new study has revealed that more than 10 million Americans are living with long COVID without ever having received a diagnosis — meaning they are suffering from a recognised chronic illness that their healthcare providers have not identified, and that government surveillance systems are entirely missing. The research, conducted by Mass General Brigham and published in JAMA Network Open, used a novel artificial intelligence algorithm to comb through electronic medical records of nearly 460,000 COVID-19 patients at 58 hospitals across the United States. The findings suggest that the actual scale of long COVID in America may be twice as large as current estimates — and that the condition continues to grow with no plateau in sight.

Key Facts

- The AI-powered study identified over 10 million long COVID cases that would go entirely undetected by the diagnostic codes that health systems and policymakers rely on to track the disease burden [un]

- The team identified 74,560 long COVID patients in the study sample, representing 16.3% of all COVID patients reviewed — translating to roughly 15 million Americans when extrapolated nationally [un]

- The study analysed nearly 460,000 patients with COVID-19 across 58 US hospitals, finding approximately 1-in-6 developed long COVID [un]

- The research was led by Mass General Brigham, one of the nation's leading academic health systems, and published in JAMA Network Open [aol]

- Prevalence of long COVID varied significantly by region, ranging from 13.6% in Western Pennsylvania to 22.7% in Southern California [un]

- The CDC confirms that more than 200 long COVID symptoms have been identified, with fatigue, brain fog, and post-exertional malaise among the most commonly reported [Wikipedia]

The Hidden Epidemic

Six years after the start of the COVID-19 pandemic, a new crisis is quietly unfolding in doctor's offices, urgent care centres, and hospital waiting rooms across America. Millions of Americans have suffered long-term symptoms of the virus — but exactly how many people developed long COVID, defined as symptoms lasting at least three months, has been hard to pin down. [aol]

The Mass General Brigham study has now provided the most comprehensive answer yet — and the numbers are alarming. Investigators used a novel AI algorithm to comb through medical records of nearly 460,000 patients with COVID-19 across 58 US hospitals, finding approximately 1-in-6, or roughly 16%, developed long COVID. [un] These rates are two-fold higher than current government estimates and reflect what researchers describe as the growing cumulative burden of chronic conditions following COVID-19 infection.

The core problem, the study found, is one of detection. "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 corresponding author Hossein Estiri, PhD, of Mass General Brigham. "The figures we uncovered are almost certainly an undercount." [uber]

Why So Many Cases Go Undiagnosed

The scale of undiagnosed long COVID is not accidental — it is the product of structural failures in how the condition is identified and recorded across the American healthcare system.

Long COVID can be elusive because it causes a range of different symptoms: fatigue, difficulty breathing, joint pain, heart palpitations, dizziness, depression, menstrual changes, and more. [aol](https://www.aol.com/news/solo-traveler-drove-london-lagos-160208516.html) Because these symptoms span multiple body systems, patients frequently present to different specialists — a cardiologist for heart palpitations, a neurologist for cognitive symptoms, an endocrinologist for metabolic changes — without any of those practitioners connecting the dots back to a prior COVID-19 infection.

Lead author and data scientist Jiazi Tian described what that invisibility looks like from inside a clinical setting: "The cardiologist seeing new dysautonomia, the endocrinologist seeing new metabolic disease, the neurologist confronting unexplained cognitive complaints — some of these presentations are long COVID arriving without the label that would connect them to a COVID-19 infection." [uber]

The AI algorithm developed by the research team was specifically designed to bridge that gap — scanning medical records for patterns of new conditions emerging after COVID infection, rather than relying on a single long COVID diagnostic code being applied by a clinician.

What Are the Symptoms?

Fatigue, brain fog, and post-exertional malaise — the worsening of symptoms after physical or mental exertion — are the most commonly reported long COVID symptoms, though researchers have now identified more than 200 distinct symptoms associated with the condition. [Wikipedia]

Other frequently reported symptoms include cognitive problems such as difficulty concentrating and memory lapses, shortness of breath, heart palpitations, headache, stomach pain, and joint pain. These symptoms often fluctuate and may worsen significantly after physical or mental exertion — a pattern clinicians now describe using the term post-exertional malaise, which is also well-known from other chronic conditions such as ME/CFS. [TikTok]

Research from Northwestern University found that among US long COVID patients, brain fog, fatigue, muscle pain, headache, dizziness, and sensory disturbances such as numbness or tingling were the most common neurological symptoms reported. Insomnia was also prevalent, with nearly 60% of non-hospitalised American patients reporting sleep problems [Facebook] — a rate significantly higher than that seen in patients from lower-income countries such as Nigeria and India.

For many patients, the cruel reality is that symptoms do not simply fade. Some people continue to report the same debilitating experiences months or even years after their initial infection — and some develop entirely new symptoms over time, suggesting that long COVID may trigger ongoing biological processes rather than representing a fixed post-infection state.

Who Is Most at Risk?

Several analyses suggest a higher burden of long COVID among middle-aged women, older adults, and people with pre-existing conditions such as diabetes, heart disease, or autoimmune disorders. However, younger adults and previously healthy individuals can also develop disabling symptoms — underscoring the need for broad awareness and early medical assessment. [TikTok]

Reinfection also substantially increases risk. People who have had COVID reinfections are more likely to experience severe fatigue, post-exertional malaise, perceived immune dysfunction, physical function limitation, and worsened existing long COVID. [euronews] This finding carries significant implications for public health messaging around the importance of preventing repeat infections, even in individuals who have already recovered from a prior bout of COVID-19.

Children are not immune either. An estimated 6 million children in the US have long COVID — exceeding the number of children with asthma [euronews] — a figure that carries serious implications for educational outcomes, childhood development, and long-term public health planning.

The Broader Research Landscape

The Mass General Brigham study is the most significant recent entry in a rapidly expanding body of research on long COVID's true prevalence. Work by Dr. Ziyad Al-Aly and colleagues at the Department of Veterans Affairs, drawing on millions of patient records with large uninfected control groups, established clear causal links between COVID-19 and long-term organ damage — even after mild or asymptomatic infection — including elevated risks of heart disease, stroke, diabetes, kidney disease, and neurological disorders lasting years after the acute phase. [uber]

New biomarker tests introduced in 2025 can now predict which patients are likely to develop chronic symptoms, and targeted treatments developed since 2024 have shown 60% improvement rates for specific symptom clusters [Wikipedia] — representing genuine progress in a field that, until recently, had very few evidence-based treatment options.

A separate study published in May 2026 by the University of Turku in Finland challenged a previously held assumption that brain inflammation was the primary driver of long COVID cognitive symptoms. Based on their findings, researchers suggested that some patients with persistent symptoms may benefit more from treatments focused on stress management and emotional regulation, rather than therapies aimed solely at reducing inflammation [YouTube] — a finding that could shift clinical approaches to managing the condition.

Analysis: A Public Health System Failing Its Patients

The Mass General Brigham findings lay bare a failure of the healthcare and public health infrastructure to keep pace with the scale of long COVID. When more than 10 million Americans are living with a chronic, debilitating condition that the diagnostic systems designed to track it are simply missing, the consequences extend far beyond individual patients.

Undiagnosed long COVID means workers who are quietly losing productivity are not accessing appropriate care. It means healthcare costs that are being attributed to other conditions rather than their true cause. It means policymakers who are making funding and resource decisions based on data that may be half the size of the actual problem.

The CDC has indicated that updated prevalence estimates by state and jurisdiction are anticipated by the end of 2026 [The Standard] — but the Mass General Brigham study suggests that any estimate relying on traditional diagnostic codes will continue to significantly undercount the true burden.

What Happens Next

For individual Americans experiencing unexplained fatigue, persistent headaches, cognitive difficulties, or other symptoms that emerged or worsened following a COVID-19 infection, the clear message from researchers is to seek evaluation from a healthcare provider familiar with long COVID — even if prior medical visits have not resulted in a diagnosis.

For the healthcare system and policymakers, the study's findings represent an urgent call to invest in improved diagnostic tools, dedicated long COVID clinical pathways, and expanded surveillance systems that go beyond diagnostic code tracking.

With over 65 million Americans now living with documented long COVID symptoms, and the condition's true burden potentially double that recorded in official figures [Wikipedia] , the scale of the public health challenge ahead is significant — and the window for early intervention for millions of currently undiagnosed patients may be narrowing with each passing month.

Conclusion

The Mass General Brigham study is a watershed moment in the understanding of long COVID's true scale in the United States. By deploying AI to look beyond the diagnostic codes that have long underpinned the official picture of the disease, researchers have revealed a hidden population of over 10 million Americans living with long COVID without a diagnosis — and without the care, support, or recognition their condition demands. As science continues to advance on the causes and treatments of long COVID, the most immediate priority must be ensuring that the millions already living with its effects are found, diagnosed, and given access to the help they need.

Sources: Mass General Brigham, JAMA Network Open, WBUR, Boston Globe, CIDRAP, Northwestern University Feinberg School of Medicine, CDC, ScienceDaily, Patient-Led Research Collaborative