Metformin During Acute COVID-19 Cuts Long COVID Risk by Half, Study Confirms

A new analysis from the ACTIV-6 clinical trial confirms that taking the common diabetes drug metformin during an acute COVID-19 infection can cut the risk of developing long COVID by roughly 50 percent. The findings, reported by Medical Xpress, provide important confirmation of earlier trial results and suggest that a widely available, inexpensive medication with a long safety record may help prevent a condition that affects millions.

That last part is what makes this worth paying attention to. Most of the headlines about long COVID over the past few years have been about how little we can do once it sets in. A result showing that a pill most primary care doctors already keep on their prescription pad can halve the odds of ever getting there is a different kind of news. It moves the conversation from managing a chronic problem to preventing it, and it does so with a drug that has been in human use for well over sixty years.

Key takeaways

• Metformin started during acute COVID-19 reduced the risk of clinician-diagnosed long COVID by about 50 percent.
• The results come from the ACTIV-6 trial, a large randomized study that confirms earlier findings.
• Metformin is a cheap, widely available drug with a well-established safety profile.
• The medication was most effective when taken soon after symptoms began, within the first few days of illness.
• The benefit held across age groups, sexes, and vaccination status, which suggests the effect is real and not an artifact of one subgroup.
• Metformin is not a vaccine substitute and is not a treatment for existing long COVID. It is a preventive signal, and any use should go through a clinician.

What the study found

The ACTIV-6 trial was designed to test several repurposed medications for COVID-19. In this latest analysis, researchers looked specifically at metformin, a drug typically used to treat type 2 diabetes. Participants who took metformin within a few days of developing COVID-19 symptoms were significantly less likely to later receive a diagnosis of long COVID from their doctor.

The 50 percent reduction in risk is a substantial effect, especially when compared to other interventions tested during the pandemic. The study authors noted that the benefit was consistent across different age groups, sexes, and vaccination statuses. Importantly, the drug appeared to work best when started early, within the first few days of symptom onset.

It helps to understand what a 50 percent relative reduction does and does not mean. It does not mean half of everyone who takes metformin avoids long COVID, because most people who catch COVID-19 never develop the long form in the first place. What it means is that among the people who would have gone on to develop it, roughly half were spared. In a condition with no approved preventive drug, cutting the case count by half at a population level is the sort of effect that changes guidelines, not just headlines. The catch is timing. The window is measured in days, not weeks, which is a recurring theme in COVID-19 pharmacology and one of the harder parts to act on in the real world.

The other detail worth holding onto is that the diagnosis was made by a clinician rather than pulled from a questionnaire. Long COVID research has been plagued by soft endpoints, where people self-report lingering symptoms that may or may not trace back to the infection. Anchoring the outcome to a doctor’s assessment raises the bar and makes the result harder to dismiss as noise or expectation.

What is long COVID and why prevention matters

Long COVID, also known as post-acute sequelae of SARS-CoV-2 infection, remains a major public health challenge. Symptoms can include fatigue, brain fog, shortness of breath, and joint pain, and they can last for months or years. Currently, there are no approved treatments specifically for preventing long COVID.

The defining feature of long COVID is that it does not follow the tidy arc people expect from a respiratory infection. You are supposed to feel bad for a week or two and then get better. In long COVID, the acute illness resolves but the person does not return to their old baseline. Weeks later they are still winded climbing stairs, still losing words mid-sentence, still exhausted after a normal day. Some cases involve a distinctive crash after physical or mental effort, called post-exertional malaise, where doing too much on a good day triggers a setback that lasts for days. That single feature is one of the cruelest parts of the condition, because the usual advice to push through and exercise your way back to health can actively make it worse.

Prevention matters here more than in almost any other setting because the treatment side of the equation is so weak. Once long COVID is established, care is mostly supportive: pacing, symptom management, rehabilitation, and time. There is no drug you can take to reverse it reliably. That imbalance, a hard condition to treat but potentially preventable, is exactly why a preventive signal from a cheap, familiar drug lands with weight. Every case avoided at the front end is a case that never has to be managed, imperfectly, for months at the back end.

Why this matters

Metformin is an attractive candidate because it is already widely used, costs pennies per dose, and has a well-known safety profile. Unlike some other COVID-19 treatments, it does not require special storage or administration, making it accessible in low-resource settings. The new results suggest it could be a practical tool for reducing the burden of long COVID.

The economics are hard to overstate. A course of metformin costs less than a cup of coffee. It does not need a cold chain, an infusion center, or a specialist to prescribe. Compare that to the newer antiviral options, which are more expensive, have their own drug interaction headaches, and are gated to higher-risk patients. A generic pill that any clinician anywhere can prescribe is the kind of intervention that actually scales to the parts of the world where most COVID-19 infections happen. That practicality is a large part of why researchers keep circling back to it.

There is also a quieter reason this matters. Metformin is one of the most studied drugs in the pharmacy, with decades of data on millions of patients. When a drug with that kind of track record shows a new benefit, the safety questions that usually slow down adoption are mostly already answered. We are not waiting years to find out whether it causes rare harms, because we already know its risk profile in fine detail. That shortens the distance between a promising trial and a real-world recommendation.

How might metformin work against long COVID?

Researchers do not yet have a settled answer, but there are several plausible mechanisms, and they are not mutually exclusive. The most direct is antiviral. Laboratory work has shown that metformin can interfere with the machinery cells use to replicate the virus, partly through its effect on a pathway called mTOR, which the virus hijacks to reproduce. Slow viral replication early, the thinking goes, and you reduce the total viral burden that drives the downstream damage.

The second mechanism is anti-inflammatory. A large part of the harm in both severe COVID-19 and long COVID appears to come not from the virus itself but from an overheated immune response. Metformin activates an enzyme called AMPK, a kind of cellular energy sensor, and through it dampens some of the inflammatory signaling that runs out of control in these patients. A calmer immune response during the acute phase may leave less lingering dysfunction behind.

A third strand involves clotting and the lining of blood vessels. Long COVID has been linked to tiny clots and to dysfunction in the endothelium, the single-cell layer that lines every blood vessel in the body. Metformin has modest effects on both, which could plausibly protect the vascular system during a period when it is under attack. The honest summary is that we have several reasonable stories and not yet a proven one. What the ACTIV-6 result does is turn these mechanistic hunches into something worth investigating hard, because now there is a clinical effect to explain.

How the trial worked

The ACTIV-6 trial was a large, randomized, placebo-controlled study conducted across multiple sites in the United States. Participants were adults with confirmed COVID-19 who were not hospitalized. They were randomly assigned to receive either metformin or a placebo, and they were followed for several months to track outcomes, including the development of long COVID.

The diagnosis of long COVID was made by clinicians based on symptoms and medical history, not just self-report. This adds credibility to the findings, as clinician-diagnosed cases are generally more reliable than patient-reported symptoms alone. The study was funded by the National Institutes of Health as part of its rapid-response research program.

The randomized, placebo-controlled design is the part that gives this weight. When people are assigned to metformin or placebo by chance, the two groups end up similar in every other respect, from age to vaccination to underlying health. That means the difference in long COVID rates is far more likely to be caused by the drug than by some hidden difference between the groups. This is the difference between a study that shows a drug works and one that merely shows the kind of people who happened to take a drug did better. In ACTIV-6, the participants took metformin for a defined course starting soon after their symptoms began, which is why the early-treatment window keeps coming up as the operative detail.

This is not the first signal

What makes the ACTIV-6 analysis notable is the word confirms. This is not the first trial to point in this direction. An earlier randomized study, often referred to as the COVID-OUT trial, tested metformin in non-hospitalized adults with COVID-19 and reported that people who received the drug were meaningfully less likely to be diagnosed with long COVID over the following months. That result surprised a lot of people at the time, and the natural scientific reaction to a surprising single trial is caution. One study can be a fluke.

The value of ACTIV-6 is that it is a second, independent look at the same question in a different population, and it landed in the same place. When two separate randomized trials, run by different teams, point the same direction, the finding stops looking like a statistical accident and starts looking like a real biological effect. That is how evidence accumulates in medicine. Individual trials rarely settle anything on their own. Consistency across trials is what moves a claim from interesting to credible, and it is what eventually gets a recommendation into a guideline.

The metabolic thread: why a diabetes drug, and what your baseline has to do with it

It is not an accident that the drug showing up in these results is a metabolic one. Metformin’s day job is improving how the body handles insulin and blood sugar, and metabolic health turns out to be one of the strongest predictors of how badly someone fares with COVID-19 in the first place. People with insulin resistance, elevated blood sugar, and excess visceral fat tend to run higher levels of chronic, low-grade inflammation, and that background state appears to make both severe acute illness and lingering post-viral symptoms more likely.

This is the same reason metformin has quietly become a fixture beyond diabetes. Clinicians use it off-label for insulin resistance, for polycystic ovary syndrome, and as part of weight and metabolic management, precisely because it nudges the underlying machinery in a healthier direction. The long COVID result fits that larger picture. A drug that improves the metabolic terrain may be helping the body weather a viral hit that lands hardest on people whose terrain is already inflamed.

The practical takeaway is not that everyone should chase a metformin prescription. It is that your metabolic baseline is not a side issue. The markers that describe it, things like fasting insulin, hemoglobin A1c, fasting glucose, and triglycerides, are the same markers that describe your resilience against a whole category of stressors, infection included. Most people have never seen these numbers laid out together, which is a shame, because they are far more informative as a set than any one of them alone. If a viral infection is going to expose the weak points in your metabolism, it is worth knowing where those weak points are before the infection, not after.

Who does this realistically apply to?

The honest answer is that the research is still catching up to the enthusiasm. What the trials studied were non-hospitalized adults who started the drug early in a confirmed infection. That is the population where the evidence lives. It does not automatically extend to people who are already weeks into an infection, to those already living with long COVID, or to children, who were not the focus of these studies.

Within that adult, early-infection group, the people with the most to gain are logically those at higher baseline risk: older adults, people with metabolic conditions, and those with a history of severe or prolonged COVID-19 episodes. But logically at higher risk is not the same as proven to benefit more, and this is where an individual conversation with a clinician does real work. A doctor can weigh your infection timing, your kidney function, your other medications, and your personal risk factors in a way that a trial average cannot.

Safety, side effects, and who should be careful

Metformin’s long track record is reassuring, but familiar does not mean harmless for everyone. The most common issue is gastrointestinal: nausea, diarrhea, cramping, and a metallic taste, especially in the first days on the drug. These effects are usually dose-related and often ease over time or with the extended-release form, but they are common enough that they are worth expecting rather than being surprised by, particularly when you are also feeling rotten from an acute infection.

The more serious considerations are about the kidneys and, rarely, about a condition called lactic acidosis. Metformin is cleared by the kidneys, so it is generally avoided or dose-adjusted in people with significantly reduced kidney function, and it is often paused around situations that can stress the kidneys, such as severe dehydration or certain imaging procedures. Lactic acidosis is a rare but serious buildup of acid in the blood, and the people most at risk are those with poor kidney function, liver disease, or heavy alcohol use. Long-term use can also lower vitamin B12 levels, which is why clinicians sometimes monitor it. None of this makes metformin a dangerous drug. It makes it a drug that belongs under a clinician’s supervision rather than something to source and self-dose based on a headline.

A realistic scenario

Consider a 52-year-old with a bit of extra weight around the middle, blood sugar that has crept into the pre-diabetic range, and an otherwise busy, healthy-enough life. She tests positive for COVID-19 on day two of symptoms. Under the older playbook, unless she was sick enough to need an antiviral, the advice was mostly rest and fluids and hope for the best on the long-tail risk. The new evidence gives her clinician something concrete to consider: she is an adult, early in a confirmed infection, and carries a metabolic profile that puts her at elevated risk. That is close to the exact scenario the trials were built around.

Notice what makes the scenario work. She caught it early, she has a reason to be higher risk, and she has a clinician who can check her kidney function and her medication list before deciding. Change any of those, catch it late, ignore the kidney check, self-prescribe from an online pharmacy, and the neat picture falls apart. The lesson is not the drug in isolation. It is the drug used in the specific way the evidence supports.

Limitations and next steps

While the results are promising, the study has limitations. The trial was conducted during a period when the Omicron variant was dominant, so it is unclear whether the same benefit would apply to other variants. Additionally, the study did not test metformin in combination with other drugs, such as antivirals or monoclonal antibodies.

Researchers say more work is needed to understand exactly how metformin reduces long COVID risk. It may work by lowering viral load, reducing inflammation, or affecting the immune response. Future studies could also explore whether the drug is effective in people who are already experiencing long COVID symptoms.

There are open questions worth naming plainly. We do not yet know the ideal dose and duration for this specific purpose, since the drug was designed and dosed for blood sugar, not for long COVID prevention. We do not know how the benefit stacks with antivirals, which many higher-risk patients are already taking. And because the trials enrolled adults, the findings say nothing reliable about children or adolescents. These are not reasons to dismiss the result. They are the map of what the next round of research has to fill in before metformin becomes a formal recommendation rather than a reasonable, evidence-backed option to discuss.

What to actually do with this

Do not buy metformin online and start dosing yourself off the back of a news story. That is the one clear wrong move, both because the drug needs a kidney check and a medication review and because self-treatment tends to get the timing and dose wrong anyway. If you test positive for COVID-19 and you are early in the illness, especially if you carry metabolic risk factors, this is a reasonable thing to raise with your clinician promptly, while the early-treatment window is still open.

The broader move is to know your metabolic baseline before you ever get sick. The same numbers that make someone a candidate for benefit here, blood sugar, insulin sensitivity, inflammation, are the numbers that quietly shape your risk across a wide range of conditions. Getting them measured and understood is one of the highest-yield things a healthy-seeming adult can do, and it turns a vague sense of I should probably take care of myself into a specific, trackable set of targets.

Frequently Asked Questions

Who should consider taking metformin for COVID-19?

Based on the study results, adults who test positive for COVID-19 and are at risk for long COVID may benefit from taking metformin early in their illness. However, individuals should always consult a healthcare provider before starting any new medication, as metformin can cause side effects and may not be safe for everyone, particularly those with kidney or liver problems.

Is metformin a substitute for vaccination or other treatments?

No. Metformin is not a replacement for COVID-19 vaccines, which remain the best way to prevent severe illness and death. It is also not a substitute for antiviral drugs like Paxlovid, which are recommended for high-risk individuals. Metformin may be used as an additional preventive measure, but it should not replace standard care.

How long should metformin be taken after a COVID-19 diagnosis?

In the ACTIV-6 trial, participants took metformin for 14 days starting within a few days of symptom onset. The exact duration that provides the best protection is still being studied. Patients should follow their doctor’s advice regarding dosing and duration, as individual factors may influence the optimal regimen.

Does it matter how quickly I start the drug?

Timing appears to be one of the most important factors. In both the ACTIV-6 analysis and earlier research, the benefit was strongest when metformin was started within the first few days of symptoms. This mirrors what we see with COVID-19 antivirals, where early treatment matters far more than late treatment. If you are going to raise this with a clinician, do it early rather than waiting to see how the infection unfolds.

If I already take metformin for diabetes, am I protected?

This is an open question the trials were not designed to answer. The studies tested starting metformin during an acute infection, not the effect of being on it beforehand. It is reasonable to hope that ongoing use offers some protection, but that has not been established, and it is not a reason to change your regimen without talking to your doctor. If you are already on the drug, keep taking it as prescribed and discuss any COVID-19 illness with your clinician like anyone else would.

Can metformin treat long COVID once I already have it?

The evidence here is about prevention, not treatment. The trials measured whether starting metformin during the acute infection reduced the chance of developing long COVID later. They did not test whether it helps people who are already living with the condition. Whether it has any role once symptoms are established is a separate question that future research will need to address.

This is an original report by Vital Signs Today, informed by reporting from Medical Xpress. Read the original source.

This article is for information only and is not medical advice. See our Medical Disclaimer.