Health information to help you understand symptoms. It isn't a diagnosis.
Check symptoms

Research deep dives

Long COVID: what the research has found so far

Last reviewed November 15, 2026. OwieClear editorial team. Waiting for clinician review. Not medical advice. Talk with your doctor about your own care.

Long COVID means symptoms that last or start months after a COVID-19 infection, such as tiredness that gets worse after activity, brain fog, dizziness, and a racing heart.

The research is young and moving fast. Below we read the key studies in order: how common it is, how to define it, what may cause it, and what has and hasn't helped.

Related pages: Long COVID · COVID-19

The bottom line

What we know
  • Long COVID is real and can affect many organs, even after a mild infection.
  • Vaccination and fewer infections lower the risk.
  • There is no proven cure yet. A 15-day course of Paxlovid didn't help people who already had it.
What is still open
  • What causes it. Leftover virus, an immune system stuck in the 'on' position, tiny clots, and reactivated viruses are all being studied. It may be several diseases.
  • Whether early metformin should be used to prevent it. One trial looked promising, but it hasn't been confirmed.
  • A blood test to diagnose it doesn't exist yet.

The studies, in order

  1. 1. How big is it (Al-Aly and others, 2021)
  2. 2. Defining it (Thaweethai and others (RECOVER), 2023)
  3. 3. A clue in the blood (Cervia-Hasler and others, 2024)
  4. 4. An antiviral trial (Geng and others (STOP-PASC), 2024)
  5. 5. Can it be prevented (Bramante and others (COVID-OUT), 2023)
What the evidence labels mean
GuidelineAn official recommendation from an expert group after weighing all the evidence.
Review of many studiesCombines the results of many studies (a systematic review or meta-analysis).
Randomized trialPeople were randomly given a treatment or a comparison, the best way to test cause and effect.
Large observational studyFollows many people over time. Shows strong links, but not always cause and effect.
Small or early studyA small study, case series, or first results. Needs to be confirmed.
Lab or animal studyDone in cells or animals. May not apply to people.
Study 1 of 5

COVID raised the risk of new illnesses in many organs

Al-Aly and others, 2021Nature (US veterans' health records)Large observational study

What they did

Researchers compared 73,435 veterans who had COVID-19 but weren't hospitalized with almost 5 million veterans who didn't have COVID-19, and tracked new diagnoses for 6 months.

What they found

  • After the first 30 days, people who had COVID-19 had a higher risk of death and of new problems.
  • New problems showed up in the lungs, nerves and brain, mental health, heart, metabolism (including diabetes), gut, kidneys, blood clotting, and skin.
  • The sicker someone was at first, the higher the risk.

What it means

Long COVID isn't only a lung problem. It can affect almost any part of the body, even after a mild case.

What it can't tell us

  • Veterans are mostly older men, so results may differ for other groups.
  • Health record studies can miss things or count them wrong.

Read it yourself

Next: Defining it

Study 2 of 5

12 symptoms that set long COVID apart

Thaweethai and others (RECOVER), 2023JAMA (NIH RECOVER study)Large observational study

What they did

The NIH's RECOVER study surveyed 9,764 adults at 85 sites, 8,646 who had COVID-19 and 1,118 who hadn't, about more than 30 symptoms at least 6 months later.

What they found

  • 12 symptoms best separated long COVID: worse tiredness after activity, fatigue, brain fog, dizziness, gut symptoms, heart palpitations, changes in sexual desire, loss of smell or taste, thirst, chronic cough, chest pain, and abnormal movements.
  • About 1 in 4 (23%) of people who'd had COVID-19 scored high enough to count as long COVID.
  • Unvaccinated people, people infected before Omicron, and people infected more than once were more likely to have it.

What it means

Researchers now have a shared definition to study it. It also showed long COVID comes in several symptom patterns.

What it can't tell us

  • The definition is for research, not for diagnosing one person.
  • People who join a long COVID study may be more likely to have symptoms, so 23% may be higher than in everyone.

Read it yourself

Next: A clue in the blood

Study 3 of 5

Part of the immune system stays switched on

Cervia-Hasler and others, 2024Science (Zurich, Switzerland)Small or early study

What they did

Researchers followed 113 people with COVID-19 and 39 healthy people for up to a year. They measured more than 6,500 proteins in the blood. 40 people had long COVID at 6 months.

What they found

  • In active long COVID, the complement system, a part of the immune system that attacks germs, stayed switched on and didn't settle down.
  • There were signs of damage to red blood cells, platelets, and blood vessels, and signs of tiny clotting problems.
  • In people who recovered, these blood levels returned to normal.

What it means

This points to a possible cause and to possible blood tests and treatments aimed at the complement system.

What it can't tell us

  • It was a small study and needs to be repeated in larger groups.
  • It doesn't explain why symptoms differ so much between people.

Read it yourself

Next: An antiviral trial

Study 4 of 5

15 days of Paxlovid didn't help

Geng and others (STOP-PASC), 2024JAMA Internal Medicine (randomized trial, Stanford)Randomized trial

What they did

155 adults who'd had long COVID for at least 3 months were randomly given Paxlovid or a placebo twice a day for 15 days, then followed for 15 weeks. The idea was that leftover virus might be driving symptoms.

What they found

  • Paxlovid didn't improve the main long COVID symptoms more than placebo.
  • Taking it for 15 days was generally safe. Diarrhea and a bad taste were common.

What it means

Simply treating leftover virus late doesn't seem to fix long COVID for most people. Antivirals might need to be given earlier, or combined with other treatments.

What it can't tell us

  • It was small, and almost everyone (98%) was vaccinated.
  • Some subgroups might still benefit. The trial was funded by Pfizer, which makes Paxlovid.

Read it yourself

Next: Can it be prevented

Study 5 of 5

Early metformin lowered long COVID by 41% in one trial

Bramante and others (COVID-OUT), 2023Lancet Infectious Diseases (randomized trial, US)Randomized trial

What they did

Adults with overweight or obesity who had just caught COVID-19 were randomly given metformin (a common diabetes pill), ivermectin, fluvoxamine, or a placebo for about 2 weeks. They were asked about a long COVID diagnosis over 10 months.

What they found

  • 6.3% of people on metformin got a long COVID diagnosis, compared with 10.4% on placebo, about 41% less.
  • Starting within 3 days of symptoms seemed to work even better.
  • Ivermectin and fluvoxamine didn't lower long COVID.

What it means

A cheap, safe, widely available pill taken early might prevent some long COVID. It needs to be confirmed before it becomes standard.

What it can't tell us

  • Long COVID was based on people reporting a doctor's diagnosis, not a set test.
  • Long COVID wasn't part of the original plan for this trial, and it only included adults with extra weight.

Read it yourself

More plain-language reading

See something wrong or unclear on this page? Suggest a correction. OwieClear is for education only and isn't medical advice.