- Long COVID remains a major burden because it can affect people across risk groups and symptoms can persist for months. E
- vidence from multiple randomised trials and large health record analyses suggests that starting metformin during acute COVID-19 or soon after infection can reduce the risk of developing long COVID.
- The reported average impact across studies was roughly one prevented case for every fifty people treated with a two-week course, which is meaningful at population scale if the finding holds broadly.
Metformin is best known as a long-established medicine for type 2 diabetes. It is inexpensive, widely available and familiar to clinicians.
In recent years it has also been studied for effects beyond glucose control, including impacts on inflammation, immune signalling and cellular energy pathways.
That made it a plausible candidate to test in COVID-19, not as a replacement for antiviral therapy, but as an adjunct that might reduce downstream complications.
The new research synthesis pulled together evidence from several randomised clinical trials across different adult risk categories, along with high-quality observational analyses using electronic health records.
The consistent theme was timing. Benefits were observed when metformin was started during the acute infection window or shortly after, rather than months later.
This is biologically plausible because long COVID risk may be influenced by factors early in infection such as viral replication dynamics, immune overactivation and the degree of tissue inflammation.
Across the included studies, metformin was associated with a lower probability of meeting definitions of long COVID compared with placebo or usual care.
The summary also highlighted related findings reported in some trials, including reduced viral load and less viral rebound.
If true, that could matter because prolonged viral persistence or repeated immune triggering is one proposed contributor to persistent symptoms.
An important practical claim was safety. In the trials described, metformin during acute COVID-19 did not appear to cause low blood sugar and serious side effects were uncommon.
That matches what clinicians already know: metformin rarely causes hypoglycaemia on its own.
The most common issues are gastrointestinal, such as nausea, abdominal discomfort and diarrhoea, which can be dose-related and sometimes limit adherence.
However, “safe for many” is not the same as “safe for everyone”.
Metformin has well-known contraindications and cautions.
People with significantly reduced kidney function may not be able to take it because the drug is cleared by the kidneys and accumulation increases the risk of a rare but serious complication called lactic acidosis.
Severe liver disease, heavy alcohol use and conditions that cause dehydration or low oxygen can also increase risk.
During an acute infection, dehydration is not uncommon, so clinical judgement matters. The right conclusion is not that people should self-medicate, but that metformin may be a viable preventive option that clinicians can consider for appropriate patients.
The synthesis also noted what is not yet known.
Evidence in children and adolescents is lacking, so the findings should not be assumed to apply to paediatric populations.
It is also not established that metformin treats long COVID once it is already present.
Prevention and treatment are different problems, and timing is likely to be crucial.
You may be interested in:
- Metformin may work in the brain as well as the liver and gut
- Skin cancer risk lowered by metformin
- New recommendations on metformin and antipsychotic-induced weight gain published
- Reassuring new data on the long-term safety of metformin in pregnancy
Another useful point is that metformin does not have to be framed as a competitor to antiviral drugs. Some of the studies included people who also received approved COVID-19 treatments. That matters because real-world care involves combinations, not isolated interventions.
If metformin’s effect is partly mediated through metabolic and immune pathways, it could theoretically complement antivirals that directly suppress viral replication. Whether combination strategies produce additive benefits is still a question for further research, but the concept is coherent.
For readers, the most grounded interpretation is cautious optimism. Metformin is not a miracle cure and it is not appropriate for everyone.
Still, it is unusual to see a familiar, low-cost medicine show a consistent preventive signal across multiple trials and datasets for a condition as frustrating and heterogeneous as long COVID.
If future studies continue to support the effect, the main challenge will shift from “does it work” to “who benefits most, when should it be started and how should it be integrated into standard care”. Until then, the practical message is simple: if you get COVID-19 and are concerned about long COVID risk, speak to a clinician early.
The window for prevention appears to be during acute infection, not weeks later.
