Long COVID belongs to an old family of post-infection illness

Say the infection was mild. Maybe you never tested, or the home test came back negative. Three months later your heart pounds on the stairs at work, and one busy Saturday leaves you flattened until Tuesday.

People in that spot tend to hear one of two explanations. The first is that nothing is physically wrong, so it must be stress. The second is that the virus is still hiding in your body and the right antiviral will flush it out.

The evidence gathered since 2020 fits neither story well. A 2024 national definition placed long COVID in an old family of illnesses that follow infections, a family that includes lingering symptoms after Lyme disease, irritable bowel syndrome after food poisoning and weakness that appears decades after polio. Those older relatives carry lessons. So do the first large drug trials for long COVID, which came back empty.

What the 2024 definition requires, and what it doesn’t

In 2024 the National Academies of Sciences, Engineering, and Medicine published a definition requested by two offices within the Department of Health and Human Services. Three of the committee members and staff who produced it set out the core wording in The New England Journal of Medicine (NEJM) in July 2024. Long COVID, they wrote, is “an infection-associated chronic condition that occurs after SARS-CoV-2 infection and is present for at least 3 months…” It can run continuously, come and go, or worsen over time, in one organ system or several.

Four features of that definition run against common assumptions:

• No positive test required. Early rapid tests missed many infections, people throw away home test strips, and antibody levels fade and are muddied by vaccination. Demanding proof, the committee judged, would wrongly exclude people. The Centers for Disease Control and Prevention (CDC) says the same on its consumer page.
• Any severity of infection. Long COVID can follow a severe, mild or entirely silent infection, and it can begin weeks or months after what looked like full recovery.
• Diagnosed in the clinic. No biomarker available today settles the question, and CDC states that no approved laboratory test can tell whether symptoms are due to long COVID.
• Not a diagnosis of exclusion. Conditions a clinician can diagnose in their own right, such as postural orthostatic tachycardia syndrome (POTS, a racing heart on standing), myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) and autoimmune diseases like lupus, can be part of long COVID rather than alternatives to it.

The committee made the definition broad on purpose, and the NEJM authors are candid about the cost. An inclusive definition misses few people who have the condition but also sweeps in some who don’t, so clinicians still have to weigh other causes. The authors also warn that the three-month line is no reason to ignore earlier symptoms. The National Academies’ project page notes that clinicians may use the diagnostic code for post-COVID condition before the three months are up.

Even the vocabulary was chosen with care. The committee kept the patients’ own term, long COVID, and called it a disease state because patients said the word syndrome got them dismissed.

One of the authors, Ely, a physician who had studied intensive care survivors for 25 years, admits he began as a skeptic. He suspected long COVID was the familiar aftermath of critical illness. That idea fell apart when similar problems turned up in tens of thousands of people, most of whom had never been hospitalized and had only mild first symptoms.

How many Americans have it, and which way the trend runs

The National Center for Health Statistics (NCHS) asks about long COVID in its yearly household survey. Its measure is self-reported: new symptoms lasting three months or more after a COVID-19 infection confirmed by a test or a doctor. In 2022, 6.9% of adults said they had ever had long COVID and 3.4% said they had it then. Women reported it more than men, 8.5% against 5.2%, and adults aged 35 to 49 most of all.

The 2023 survey, analyzed by CDC researchers in JAMA Network Open, put ever-had at 8.4% and current at 3.6%. About 2.3% of all adults had current long COVID that limited their activities. Roughly two in three people with current long COVID reported some limitation, counting “a little” as well as “a lot.” Rates ran higher in rural areas and among lower-income adults.

A 2026 reanalysis of the 2022 to 2024 surveys, by researchers at Xi’an Jiaotong University in China and published in Frontiers in Public Health, tracks the trend. Ever-had levelled off at 8.3% in 2024, and current cases held at 3.3%. Among adults who had been infected, though, the share reporting long COVID fell: ever-had dropped from 17.7% to 13.7%, and current from 8.6% to 5.5%.

That corrects a common belief. Long COVID is not climbing year after year. The national total has stopped rising even as more people have caught the virus, because a smaller share of infected people report it. CDC likewise says new cases are fewer than early in the pandemic.

Thresholds matter when you read these figures. The same 2026 analysis found that 19.8% of people with current long COVID reported significant limitation. That is a stricter bar than the any-limitation count, so the two numbers measure different things rather than contradicting each other.

CDC lists groups more likely to develop long COVID: women, Hispanic and Latino people, people who had more severe COVID-19, especially those who needed hospital or intensive care, people with underlying health conditions and people who did not get a COVID-19 vaccine. Every new infection carries its own risk, the agency says, and it calls vaccination the best available tool for prevention. The NEJM authors add that CDC counted about 5,000 US deaths through 2023 in which long COVID was a cause or contributor, a count they consider far too low.

The older siblings

The 2024 definition files long COVID under infection-associated chronic conditions, illnesses that can follow infection by viruses, bacteria, fungi or parasites. The NEJM authors point out that people with post-treatment Lyme disease and ME/CFS lived with infection-linked conditions for decades without a pandemic to draw attention to them. CDC says it isn’t fully known why some people develop chronic symptoms after an infection, and that in many cases no specific treatment or cure exists. Four better-studied relatives show what the family can look like. How tiredness after mono and similar infections usually fades is traced in our explainer on fatigue duration.

Irritable bowel syndrome after a gut infection

Irritable bowel syndrome (IBS) means recurring belly pain with changes in bowel habits. A 2017 meta-analysis by Mayo Clinic gastroenterologists, published in Gastroenterology, pooled 45 studies of 21,421 people who had infectious enteritis, an infection of the intestines often caught from food or water. Twelve months later, 10.1% had IBS. In the first year after infection, their risk was 4.2 times that of people without one.

The type of germ mattered. After protozoal or parasitic infections, 41.9% developed IBS; after bacterial infections, 13.8%. Women had about twice the odds, and more severe infections raised the risk. People given antibiotics during the illness also had higher odds, an observational link that doesn’t show the drugs caused anything; whether a gut infection needs an antibiotic remains the prescriber’s call. The studies differed considerably, so the pooled figures are approximate.

Post-polio syndrome

The National Institute of Neurological Disorders and Stroke (NINDS) describes post-polio syndrome as muscle weakness, fatigue, muscle wasting and scoliosis that usually appear 15 to 40 years after polio. It is thought to come from motor nerve cells deteriorating over many years. Only polio survivors get it, and not all of them do. It is rarely life-threatening, though it can erode a person’s independence. NINDS gives no figure for how many survivors are affected, so treat unsourced percentages with suspicion. The lesson here is about time: an infection’s after-effects can surface decades later.

Life after Ebola

Ebola may be the most instructive relative, because researchers compared survivors with people who were never infected. A study in Liberia, published in NEJM in 2019 and funded by two National Institutes of Health (NIH) institutes, followed 966 survivors and 2,350 of their close contacts who had no antibodies to the virus.

About a year after illness, survivors reported memory loss far more often than contacts (29.2% vs 4.8%), along with joint pain (47.5% vs 17.5%) and fatigue (18.4% vs 6.3%). Headache told a different story: 47.6% of survivors had it, but so did 35.6% of contacts who never caught the virus. Without that comparison group, most of the headache burden would have been blamed on Ebola.

Most problems eased over the following year in both groups. An eye inflammation called uveitis did not. The virus’s genetic material also turned up in survivors’ semen as long as 40 months after illness, evidence that some viruses can linger in parts of the body.

Lyme disease

Lyme disease teaches the same lesson in a different form. After the standard antibiotic course, some people go on having fatigue, aches and trouble thinking, a pattern CDC calls post-treatment Lyme disease syndrome, with no known cause. CDC states its size as a comparison rather than a rate: half a year after treatment, the symptoms turn up 5% to 10% more often in people who had Lyme disease than in a comparison group that never had it. Without the comparison, tiredness and aches that many people have anyway would be counted as Lyme’s doing. Our brain fog article looks at what the condition means for memory and concentration.

The hidden-infection theory, put to the test

Ebola shows that viral remnants can persist, and viral persistence is among the leading explanations researchers have proposed for long COVID. If leftover virus drives the symptoms, an antiviral ought to help. At least three placebo-controlled trials have tested that idea with nirmatrelvir plus ritonavir, an oral antiviral used against acute COVID-19.

The first, STOP-PASC, ran at Stanford University and appeared in JAMA Internal Medicine in 2024. It randomly assigned 155 adults, ill for an average of 17.5 months and nearly all vaccinated, to 15 days of the drug or a placebo. At 10 weeks, overall symptom severity, fatigue, breathlessness, thinking and physical function were no better with the drug.

A 2025 Yale-led trial in The Lancet Infectious Diseases, run remotely with 100 adults across the contiguous US, likewise found that 15 days of the drug did not significantly improve physical health scores at four weeks.

The largest test was RECOVER-VITAL, part of NIH’s long COVID research program, published online in The Lancet Infectious Diseases on August 31, 2026. At 69 US sites, 959 adults with at least 12 weeks of symptoms took the antiviral for 15 or 25 days, or a placebo. Each fell into one of three patterns: thinking problems, autonomic symptoms such as lightheadedness on standing, or exercise intolerance. The authors’ conclusion: “Nirmatrelvir-ritonavir for 15 days or 25 days showed no evidence of benefit in long COVID in any of the three phenotypes studied.” Secondary measures showed no differences either. No safety signals emerged, and serious adverse events occurred in 4% of participants.

Lyme disease went through the same arc decades earlier. Placebo-controlled trials that gave people with persistent symptoms after treated Lyme disease months of extra antibiotics found no added benefit, and CDC links long-term antibiotic use to serious and sometimes deadly complications. Those trials, and the clinics that still sell such treatment, get a full hearing in our piece on the diagnostic odyssey.

In this library’s terms, the antiviral regimens tested for long COVID so far are tested and not supported. That falls short of disproving viral persistence. These trials tried one drug at two durations, in people who had been sick for months or years, and the idea can be tested in other ways.

Another RECOVER trial also disappointed. For thinking problems, computerized brain training, alone or paired with cognitive rehabilitation or brain stimulation, failed to beat its comparison, a trial our brain-training article details. As of July 2026, a newer RECOVER program was enrolling at 17 sites to test the drug baricitinib, with trials of low-dose naltrexone, semaglutide and a nerve block in the neck still being prepared. Early results from RECOVER’s trial of ivabradine for autonomic symptoms had been presented at a cardiology meeting in March 2026, with the full paper still pending.

“All in your head” doesn’t fit either

The opposite explanation fares no better. The definition’s authors reject treating long COVID as whatever is left when tests come back normal. Their list of conditions that can be part of it includes measurable ones, from blood clots and heart rhythm problems to diabetes and lung disease. In the Liberian study, survivors also had more abnormal findings on physical examination than uninfected contacts living through the same epidemic.

Psychology still belongs in the picture, and leaving it out would be its own distortion. In the IBS meta-analysis, anxiety, depression and somatization, a tendency to experience distress as physical symptoms, all raised the odds of IBS after infection; for somatization, the odds were about four times higher. Yet the kind of germ changed the risk roughly threefold too. Both can be true at once: an infection starts the process, and the state of a person’s mind and body shapes how it unfolds. How symptoms can persist without visible damage is explored in our article on functional disorders.

Questions to ask about an illness that lingers after infection

Bring these to a primary care visit, or to a specialist if you have one:

• Could any of my symptoms be a condition you can test for separately? POTS, a new autoimmune disease or a heart rhythm problem can each be checked and treated, whatever umbrella label sits on top.
• What else could explain this? The definition is broad by design, so its authors expect clinicians to consider alternatives. Asking keeps that step on the table.
• Do my symptoms get worse a day or two after effort? Post-exertional malaise, a delayed crash after physical or mental exertion, is on the definition’s symptom list and changes how activity advice should be built; planning activity around it is what our pacing guide covers.
• Has a condition I already had gotten worse since the infection? The definition says long COVID can aggravate existing illness, which makes it a question for whoever manages that condition.
• If someone offers a drug, infusion or supplement to clear a hidden infection, what trial supports it? For antivirals in long COVID and extended antibiotics after treated Lyme disease, the placebo-controlled answers so far are negative.
• Is there a research study I might qualify for? NIH’s RECOVER program was still enrolling in 2026.

When it can’t wait

• Severe or worsening symptoms before three months. The definition’s authors say earlier symptoms deserve careful assessment, so there is no need to sit out the clock.
• Signs of a heart attack or stroke. Call 911 for chest pain or pressure, shortness of breath at rest, sudden numbness or weakness on one side of the body, sudden confusion or trouble speaking, or a sudden severe headache. The National Heart, Lung, and Blood Institute and CDC list these as warning signs.
• Lightheadedness or a racing heart every time you stand. Raise it soon; a clinician can check how your heart rate and blood pressure change when you stand.
• Hopelessness. If living with the illness has led to thoughts of suicide, call or text 988 to reach the 988 Suicide & Crisis Lifeline, which is free, confidential and open around the clock.

The Takeaway

Long COVID is the newest and most studied member of an old family, and the family’s history offers two lessons that hold up. Symptoms after infection are real and measurable, but only a comparison group shows how much of them the infection caused. And the intuitive fix, more antimicrobial drugs to clear a hidden germ, has not beaten placebo in the major trials of it so far, after Lyme disease and now after COVID-19.

What helps most is a specific description of your symptoms, paired with one question: which pieces can be identified and treated as conditions in their own right? A long COVID label should start that search, not end it.

Sources

1. National Academies of Sciences, Engineering, and Medicine. “A Long COVID Definition: A Chronic, Systemic Disease State with Profound Consequences (Consensus Study Report, 2024).” nationalacademies.org
2. The New England Journal of Medicine. “Ely EW, Brown LM, Fineberg HV. "Long Covid Defined." N Engl J Med, 2024;391(18):1746-1753. PubMed. DOI: 10.1056/NEJMsb2408466.” pubmed.ncbi.nlm.nih.gov
3. Centers for Disease Control and Prevention. “Long COVID Basics (last reviewed December 16, 2025; updated May 6, 2026).” cdc.gov
4. National Center for Health Statistics, CDC. “Adjaye-Gbewonyo D, Vahratian A, Perrine CG, Bertolli J. Long COVID in Adults: United States, 2022. NCHS Data Brief No. 480 (September 2023).” cdc.gov
5. JAMA Network Open. “Vahratian A, Saydah S, Bertolli J, Unger ER, Gregory CO. "Prevalence of Post-COVID-19 Condition and Activity-Limiting Post-COVID-19 Condition Among Adults." JAMA Netw Open, 2024;7(12):e2451151. DOI: 10.1001/jamanetworkopen.2024.51151.” doi.org
6. Frontiers in Public Health. “Jia X, Wang J, Cui X, et al. "Trends in long COVID among US adults, 2022-2024." Front Public Health, 2026;14:1809635. PubMed. DOI: 10.3389/fpubh.2026.1809635.” pubmed.ncbi.nlm.nih.gov
7. Centers for Disease Control and Prevention. “About Chronic Symptoms Following Infections (last reviewed November 24, 2025).” cdc.gov
8. Gastroenterology. “Klem F, Wadhwa A, Prokop LJ, et al. "Prevalence, Risk Factors, and Outcomes of Irritable Bowel Syndrome After Infectious Enteritis: A Systematic Review and Meta-analysis." Gastroenterology, 2017;152(5):1042-1054.e1. PubMed. DOI: 10.1053/j.gastro.2016.12.039.” pubmed.ncbi.nlm.nih.gov
9. National Institute of Neurological Disorders and Stroke. “Motor Neuron Diseases (includes post-polio syndrome; last reviewed March 13, 2026).” ninds.nih.gov
10. The New England Journal of Medicine. “PREVAIL III Study Group; Sneller MC, Reilly C, Badio M, et al. "A Longitudinal Study of Ebola Sequelae in Liberia." N Engl J Med, 2019;380(10):924-934. PubMed. DOI: 10.1056/NEJMoa1805435.” pubmed.ncbi.nlm.nih.gov
11. Centers for Disease Control and Prevention. “Chronic Symptoms and Lyme Disease (last reviewed June 26, 2026).” cdc.gov
12. JAMA Internal Medicine. “Geng LN, Bonilla H, Hedlin H, et al. "Nirmatrelvir-Ritonavir and Symptoms in Adults With Postacute Sequelae of SARS-CoV-2 Infection: The STOP-PASC Randomized Clinical Trial." JAMA Intern Med, 2024;184(9):1024-1034. PubMed. DOI: 10.1001/jamainternmed.2024.2007.” pubmed.ncbi.nlm.nih.gov
13. The Lancet Infectious Diseases. “Sawano M, Bhattacharjee B, Caraballo C, et al. "Nirmatrelvir-ritonavir versus placebo-ritonavir in individuals with long COVID in the USA (PAX LC): a double-blind, randomised, placebo-controlled, phase 2, decentralised trial." Lancet Infect Dis, 2025;25(8):936-946. PubMed. DOI: 10.1016/S1473-3099(25)00073-8.” pubmed.ncbi.nlm.nih.gov
14. The Lancet Infectious Diseases. “Baden LR, Shah NS, Liu STH, et al. "Nirmatrelvir-ritonavir targeting viral persistence in post-COVID-19 condition (long COVID) in the USA (RECOVER-VITAL): a randomised, double-blind, placebo-controlled, phase 2 trial." Lancet Infect Dis, published online August 31, 2026. PubMed. DOI: 10.1016/S1473-3099(26)00406-8.” pubmed.ncbi.nlm.nih.gov
15. The New England Journal of Medicine. “Klempner MS, Hu LT, Evans J, et al. "Two controlled trials of antibiotic treatment in patients with persistent symptoms and a history of Lyme disease." N Engl J Med, 2001;345(2):85-92. PubMed. DOI: 10.1056/NEJM200107123450202.” pubmed.ncbi.nlm.nih.gov
16. JAMA Neurology. “Knopman DS, Koltai D, Laskowitz DT, et al. "Evaluation of Interventions for Cognitive Symptoms in Long COVID: A Randomized Clinical Trial." JAMA Neurol, 2026;83(1):49-59. PubMed. DOI: 10.1001/jamaneurol.2025.4415.” pubmed.ncbi.nlm.nih.gov
17. RECOVER Initiative (National Institutes of Health). “RECOVER at mid-year: 2026 progress and what's ahead (July 1, 2026).” recovercovid.org
18. National Heart, Lung, and Blood Institute. “Heart Attack: Symptoms.” nhlbi.nih.gov
19. Centers for Disease Control and Prevention. “Signs and Symptoms of Stroke (last reviewed May 19, 2026).” cdc.gov
20. 988 Suicide & Crisis Lifeline. “988 Suicide & Crisis Lifeline.” 988lifeline.org