By Theo Loxley
Published 27 March 2026  |  TherapyInsights

A widely used diabetes drug that costs a few dollars a month is now at the centre of a growing body of evidence suggesting it could reduce the risk of developing long COVID, one of the most persistent and poorly understood consequences of the pandemic.

Metformin, prescribed to more than 150 million people worldwide for type 2 diabetes, has been tested in multiple clinical trials and large-scale observational studies since 2022. The findings have been striking enough to prompt some researchers to call for its inclusion in COVID-19 treatment guidelines and contentious enough that others urge continued caution.

Long COVID remains a significant public health problem. The World Health Organization estimates that 10 to 20 per cent of people infected with SARS-CoV-2 experience prolonged symptoms. In Australia alone, modelling published in the Medical Journal of Australia estimated the economic cost of long COVID at up to $9.6 billion in 2022, with working-age adults bearing the greatest burden.

So what does the evidence actually show? And how close is metformin to becoming a standard part of COVID-19 care?

Long COVID also referred to as post-COVID condition or post-acute sequelae of SARS-CoV-2 describes a range of symptoms that persist for weeks, months, or in some cases years after the initial infection.

The most commonly reported symptoms include chronic fatigue, cognitive dysfunction often described as “brain fog,” breathlessness, chest pain, joint and muscle pain, sleep disturbances, and anxiety or depression. The condition has been associated with more than 200 distinct symptoms, and its severity can range from mildly disruptive to completely disabling.

Despite the end of pandemic-era emergency measures, long COVID continues to affect hundreds of millions of people globally. The Australian Centre for Disease Control has described it as an emerging public health problem that limits daily activities and affects people’s ability to work. It is not rare. It is not mild. And for many, it is not over.

There is currently no approved cure for long COVID. Treatments are largely symptomatic managing fatigue, pain, or cognitive difficulties as they arise, rather than addressing the underlying condition.

This is what makes prevention so important. If a safe, affordable intervention could reduce the likelihood of developing long COVID in the first place, the implications would be enormous not just for individual patients, but for healthcare systems, employers, and economies.

In Australia, a modelling study in the Medical Journal of Australia estimated that long COVID caused the loss of more than 102 million working hours in 2022 alone. The burden fell hardest on adults aged 30 to 49 the most economically active segment of the population. The annual cost per severely affected individual was estimated at more than $77,000.

Against that backdrop, even a modest reduction in long COVID incidence from an inexpensive, widely available medication would represent a significant public health return.

The strongest evidence for metformin’s role in long COVID prevention comes from two large randomised controlled trials the gold standard in clinical research.

The first was the COVID-OUT trial, a phase 3, quadruple-blind study conducted across six US sites and published in The Lancet Infectious Diseases in 2023. It enrolled nearly 1,300 adults with overweight or obesity who had been diagnosed with COVID-19 within the previous few days. Those who received a 14-day course of metformin had a 41 per cent lower incidence of long COVID over 10 months compared with those receiving a placebo. When treatment was started within three days of symptom onset, the reduction was even larger approximately 63 per cent.

The second was the ACTIV-6 trial, a large US government-funded study that tested multiple repurposed drugs for COVID-19. Results published in 2025 found that participants receiving metformin were approximately 50 per cent less likely to be told by a medical provider that they had long COVID at six months.

These trial results have been supported by large observational studies. A UK retrospective analysis published in Clinical Infectious Diseases in September 2025 drew on data from more than 624,000 patients in the Clinical Practice Research Datalink. It found that adults with overweight or obesity who started metformin within 90 days of a COVID-19 diagnosis had a 64 per cent lower risk of developing long COVID at one year.

Separately, a study from the NIH’s RECOVER Initiative the largest US research programme on long COVID found that adults already taking metformin for diabetes were consistently less likely to develop long COVID than those on other diabetes medications. The findings, published in Diabetes Care in 2024, aligned with the trial data.

In March 2026, an editorial published in Clinical Infectious Diseases brought the evidence together. Its authors argued that the combined weight of two randomised trials and two electronic health record analyses now supports incorporating metformin into treatment guidelines for non-hospitalised adults with acute SARS-CoV-2 infection. They noted that there are no drug interactions preventing patients from taking metformin alongside currently recommended COVID-19 therapies.

This editorial represents the strongest call to date from researchers for metformin to move from experimental finding to clinical practice in the context of long COVID.

Scientists do not yet have a complete picture of why metformin appears to reduce the risk of long COVID, but several mechanisms are under investigation.

The most prominent theory centres on metformin’s well-documented anti-inflammatory properties. Long COVID is believed to involve persistent inflammation and immune dysregulation following the initial infection. Metformin may help by dampening that inflammatory response during the critical early phase of illness.

Laboratory research has also shown that metformin can restore cellular processes disrupted by SARS-CoV-2 specifically, it appears to inhibit certain inflammatory pathways activated by the virus and prevent virus-induced premature aging of neurons, which could be relevant to the cognitive symptoms many long COVID patients experience.

Notably, a separate line of research has revealed that metformin acts directly on the brain through a pathway scientists missed for six decades. As explored in depth in our analysis of why metformin is being reconsidered as a multi-disease drug, the drug’s effects appear to extend well beyond simple blood sugar regulation.

The intersection of these findings anti-inflammatory effects, cellular protection, and now brain-based mechanisms may help explain why a diabetes drug appears to have an impact on a post-viral condition.

The existing clinical trial evidence is strongest for a specific population: adults with overweight or obesity who receive metformin early after a COVID-19 diagnosis.

Both the COVID-OUT and ACTIV-6 trials enrolled participants who were overweight or obese, and the UK observational study focused on the same group. This is significant because obesity is itself a risk factor for both severe COVID-19 and long COVID, potentially making the anti-inflammatory and metabolic effects of metformin particularly relevant.

The RECOVER Initiative data suggested a protective effect in people already taking metformin for diabetes raising the question of whether ongoing use confers different benefits than short-term treatment during acute infection.

Importantly, the evidence does not yet extend to all populations. Whether metformin offers the same protection to people with a healthy BMI, younger adults, children, or those with certain pre-existing conditions remains an open question. Researchers have emphasised that further trials are needed before any broad recommendations can be made.

Despite the encouraging findings, there are important caveats that any honest assessment of the evidence must address.

First, the number of high-quality randomised trials remains small. While the COVID-OUT and ACTIV-6 results are promising, they represent a limited evidence base from which to draw universal conclusions. A systematic review and meta-analysis published by the BMJ in 2025 described the evidence for metformin reducing long COVID as “low certainty” not because the findings were negative, but because they were drawn from a small number of trials.

Second, the definition of long COVID itself varies across studies, making direct comparisons difficult. Some trials relied on self-reported diagnoses, while others used medical provider confirmation. This inconsistency complicates the picture.

Third, most trial participants were enrolled during the Omicron era. Whether the findings apply equally to future variants, or to people with different vaccination histories, is unknown.

Fourth, metformin has not been tested as a treatment for people who already have long COVID only as a preventive measure during acute infection. A separate clinical trial registered on ClinicalTrials.gov is investigating whether metformin can reduce fatigue in people with established long COVID, but results are not yet available.

None of these limitations invalidate the existing evidence. But they do mean that the story is still evolving, and that certainty remains some distance away.

Even with the caveats, the potential implications for healthcare systems are difficult to ignore.

Metformin is one of the cheapest medications in existence. It is off-patent, widely manufactured, and already stocked in pharmacies worldwide. If it proves effective at reducing long COVID in even a subset of the population, the cost-benefit ratio would be extraordinary compared with the billions of dollars in lost productivity and healthcare spending that long COVID currently generates.

Long COVID is not just a clinical problem it is an economic and workforce problem. Research published in the American Journal of Managed Care in 2026 described long COVID as both an occupational health issue and an economic burden, noting that persistent symptoms including fatigue and cognitive dysfunction reduce work capacity and increase reliance on medical leave and disability accommodations.

For publicly funded healthcare systems, a cheap preventive intervention that reduces the downstream need for rehabilitation, disability support, and chronic disease management represents the kind of upstream investment that health economists routinely advocate for but rarely find.

In Australia, the intersection of long COVID with existing health system pressures makes this research particularly relevant.

The Australian Centre for Disease Control has identified long COVID as an emerging public health problem. Between 2020 and 2024, more than 12 million COVID-19 cases were notified in the country, and serosurveillance data suggests at least two-thirds of Australians had been infected by the end of 2022.

Modelling in the Medical Journal of Australia found that long COVID’s burden falls disproportionately on working-age adults, with those aged 30 to 49 accounting for roughly half of estimated cases. For a country already grappling with workforce shortages across healthcare, aged care, and disability services, any condition that pushes productive workers out of the labour force compounds existing pressures.

The National Disability Insurance Scheme is directly affected. Long COVID can produce functional impairments that meet the threshold for NDIS support including cognitive impairment, reduced mobility, and chronic fatigue. As NDIS costs continue to rise, understanding what is and isn’t funded under the scheme becomes increasingly important for participants navigating post-viral conditions.

Aged care is similarly exposed. Older Australians who develop long COVID may experience accelerated functional decline, increasing demand for residential and home-based care services at a time when the system is already under strain.

If metformin or similar interventions can meaningfully reduce the flow of patients into these systems, the long-term savings could be substantial. But the policy conversation has barely begun.

Metformin is a prescription medication. It is not available over the counter and should not be taken without medical supervision.

While the drug has a strong safety profile built over six decades of use, it is not without side effects. The most common are gastrointestinal nausea, diarrhoea, stomach cramps, and loss of appetite particularly during the first few weeks of use. Long-term use can lead to vitamin B12 deficiency, which requires monitoring. In rare cases, metformin can cause lactic acidosis, a serious condition, though this risk is considered very low when the drug is prescribed appropriately.

Metformin is not suitable for everyone. It should be used with caution in people with kidney impairment, and is generally not recommended for those with advanced liver disease.

Critically, metformin is not currently approved in Australia or anywhere for the prevention of long COVID. Any use for this purpose would be off-label, meaning it falls outside the drug’s approved indications. Off-label prescribing is legal, but the decision must be made by a qualified healthcare provider based on the individual patient’s circumstances.

Patients who are interested in whether metformin might be appropriate for them should raise the question with their doctor. Self-prescribing based on research headlines is not recommended.

Metformin shows genuine promise as a tool for reducing the risk of long COVID. The evidence from two randomised controlled trials and multiple observational studies is more robust than for almost any other proposed intervention. The March 2026 editorial calling for its inclusion in treatment guidelines reflects a growing consensus among some researchers that the data is strong enough to act on.

But “promise” is not “proof.” The evidence base, while encouraging, remains limited in scope. It applies most clearly to adults with overweight or obesity, and uncertainty persists around optimal timing, dosing, and which populations benefit most. More trials are underway, and the coming years will be critical in determining whether metformin becomes a standard part of COVID-19 care or remains an intriguing lead that requires further investigation.

What is clear is that long COVID is not going away, and the search for effective prevention strategies is more urgent than ever. For a healthcare system already stretched by chronic disease, rising out-of-pocket costs, and an aging population, the prospect of a cheap, safe, widely available drug that could reduce the burden of post-viral illness deserves serious attention and serious scrutiny.

Patients should discuss any concerns about long COVID prevention with their healthcare provider.

About the Author

Theo Loxley is a healthcare journalist at TherapyInsights covering NDIS, aged care, and the real-world impact of policy on Australian health services.