Table of Contents
The New Study
Arthritis is one of the most common reasons adults seek medical care, and yet effective treatment options remain frustratingly limited — particularly for the most prevalent form. A new study published in JAMA has drawn attention by testing whether a decades-old, widely used, extremely affordable medication could offer meaningful relief for knee osteoarthritis [1].
Osteoarthritis is the breakdown of cartilage within joints. It is the most common type of arthritis and a major cause of disability worldwide. Knee osteoarthritis alone affects an estimated 365 million people globally [1]. Current treatments largely focus on managing symptoms — pain relievers, physical therapy, injections — rather than slowing or reversing the underlying disease process.
The condition is strongly associated with being overweight or obese [1]. Some of this connection is mechanical: greater body mass places additional compressive load on the knee joint, accelerating wear. But the link runs deeper than mechanics. Obesity promotes systemic low-grade inflammation, and excess adipose tissue disrupts insulin sensitivity and blood sugar regulation. Both of these metabolic effects further damage articular cartilage, creating a compounding cycle of deterioration [1].
Researchers behind the new study identified a compelling hypothesis: metformin, a medication that has been prescribed for type 2 diabetes for over 60 years, simultaneously targets several of these drivers. It reduces systemic inflammation, improves insulin sensitivity and blood sugar regulation, and promotes modest weight loss. If the main pathological mechanisms of knee osteoarthritis in obese individuals overlap with the known pharmacological actions of metformin, could repurposing the drug offer a new treatment approach?
To test this, researchers designed a randomised, double-blind, placebo-controlled trial. Participants — overweight or obese adults with knee osteoarthritis — received either metformin or a matching placebo for 6 months. Before and after the intervention, participants rated their knee pain on a 100-point visual analogue scale.
At the 6-month mark, pain changes were compared between groups [1]. In the metformin group, the mean pain score dropped by just over 31 points. In the placebo group, pain dropped by 19 points. The placebo group's substantial improvement is a reminder of how potent the placebo response can be in pain trials — but the metformin group showed significantly greater improvement above and beyond this effect.
Researchers also applied the WOMAC (Western Ontario and McMaster Universities Osteoarthritis Index), a validated questionnaire assessing pain, stiffness, and physical function in people with osteoarthritis [1]. This secondary outcome also favoured metformin over placebo, adding confidence to the primary pain measure.
However, several caveats are important for contextualising these results.
First, the absolute difference between groups matters more than the within-group change. Before the trial, the average pain score was approximately 60 out of 100. Metformin reduced this to around 28 — effectively halving the pain score. But the placebo group also fell substantially, from about 60 to 40 [1]. The between-group difference attributable to metformin was therefore just over 11 points on the 100-point scale.
The trial's pre-specified minimum clinically important difference — the threshold researchers had set for what would constitute a meaningful benefit — was 15 points between groups. The 11-point result fell short of this target. It is important to note, however, that metformin outperformed anti-inflammatory medications such as ibuprofen and Celebrex in a separate comparison study, which places the magnitude of its effect in a more positive light [1].
The second caveat is that some of the arthritic benefit may be driven by metformin's weight loss effect — a question examined in more detail below.
The study concludes on a cautiously optimistic note. The results point to a meaningful potential benefit from metformin, but the trial was relatively small. A larger and longer clinical trial will be needed before the magnitude of effect can be stated with confidence [1].
Timing also provides a clue about mechanism. Researchers reported no detectable benefit at the 3-month interim assessment, but a clear benefit by 6 months [1]. This time course is consistent with the biological mechanisms by which metformin operates — its effects on systemic inflammation and metabolic parameters accumulate gradually, suggesting that benefits may continue to grow with longer treatment durations. The 6-month window captures what may be only the early phase of a longer-term trajectory.
Supporting evidence from a longer observational study reinforces this interpretation. A 4-year cohort study followed obese patients with knee osteoarthritis and measured knee cartilage volume by MRI at the start and end of follow-up. Metformin users experienced approximately half the rate of cartilage loss compared to non-users [2]. While observational data cannot establish causation, the structural finding — actual cartilage preservation, not just pain relief — is biologically meaningful and consistent with the intervention trial.
Taken together, both studies suggest that metformin may offer relief for knee arthritis in people who are overweight, and possibly protect cartilage over time — but the evidence base is still early-stage, and the mechanism is not fully understood.
Metformin's anti-inflammatory properties are well established [3], and this is a plausible primary mechanism for its arthritic benefit. But weight loss also plays a role in osteoarthritis. To disentangle these effects, the trial measured weight change in both groups. The metformin group lost an average of 1.8 kg over 6 months, compared to 1.2 kg in the placebo group — a difference of just 0.6 kg [1].
Research on obesity and osteoarthritis has established that a body weight reduction of approximately 5% is needed to produce meaningful symptom relief in obese patients with knee arthritis, while 10% reduction is needed for substantial benefit [4]. The 0.6 kg differential between the groups falls far below these thresholds, suggesting that weight loss alone does not account for the observed pain improvement — anti-inflammatory or other direct effects of metformin on the joint are likely the primary drivers.
For patients with type 2 diabetes who are already prescribed metformin for glucose management, the knee arthritis data suggests a potential additional benefit at no additional cost or risk. For non-diabetic individuals, the benefit–risk assessment is more nuanced, as outlined in the following sections.
Other potential benefits
For overweight individuals without diabetes who are experiencing knee arthritis, the case for metformin rests on two additional potential benefits alongside the arthritis evidence — and must also weigh two important potential concerns.
Benefit 1: Weight management
The first additional benefit is weight management. Excess body weight increases mechanical stress on joints and significantly raises all-cause mortality risk. A large recent cohort study found that a BMI above 30 was associated with a substantially higher risk of death from all causes [5].
Most metformin weight loss research comes from populations with diabetes, but the landmark Diabetes Prevention Program (DPP) trial specifically enrolled people at high risk of developing diabetes — without yet having it. The trial assigned participants to one of three groups: intensive lifestyle modification, metformin 1,700 mg daily, or placebo. Over approximately 3 years, the metformin group lost an average of 2.1 kg (4.6 pounds) [6].
A 10-year follow-up of the DPP cohort revealed an important pattern: the lifestyle intervention group, despite achieving larger initial weight losses, slowly regained weight over the decade of follow-up. The metformin group, by contrast, maintained their weight loss at the 10-year mark [7]. This durability of effect is clinically notable — sustainable weight management, even modest, has health benefits that compound over time.
A more recent meta-analysis pooling data from 21 different trials of metformin across various populations found a modest but consistent average reduction in BMI of approximately 1 point [8]. Taken together, the evidence indicates that metformin produces real but modest weight reduction — not the dramatic losses seen with GLP-1 receptor agonists like semaglutide, but a meaningful and sustained effect.
The mechanisms behind metformin's weight effects include a reduction in hepatic glucose output (the liver releases less glucose into the bloodstream) [9] and appetite suppression. An early study found that people taking metformin voluntarily consumed less food and reported lower levels of hunger [10]. Metformin also stimulates production of glucagon-like peptide 1 (GLP-1), the same intestinal hormone that medications such as Ozempic are designed to mimic [9]. GLP-1 slows gastric emptying and increases satiety, reducing overall caloric intake.
This mechanistic overlap with GLP-1 makes metformin a relevant lower-cost option for people for whom GLP-1 receptor agonist medications are financially inaccessible. The weight loss effects are considerably smaller than those produced by modern GLP-1 drugs, but they are real, well-documented, and durable. Clinically meaningful weight loss effects from metformin appear at doses of 1,500 mg or higher — consistent with the 2,000 mg daily dose used in the new arthritis trial.
Benefit 2: Potential lifespan effects (preliminary)
The second potential benefit is more controversial. Claims that metformin can slow biological aging and extend human lifespan have circulated widely online [11]. Initial scientific interest in this hypothesis was driven by metformin's known effects on processes considered central to aging: it reduces chronic low-grade inflammation, helps regulate insulin and glucose metabolism, and counteracts oxidative stress [12].
Some observational evidence has linked metformin use with reduced cancer incidence and slower age-related functional decline [13]. These associations prompted researchers to test metformin directly for life-extending effects in the Interventions Testing Program, which runs rigorous preclinical longevity testing in rodent models. Metformin was found to have no significant effect on lifespan in these animal studies [14].
The human evidence is similarly cautious. A systematic review of existing studies found no compelling evidence that metformin extends lifespan in humans, though it does reduce early mortality attributable to diseases such as type 2 diabetes [15]. A large prospective study that followed adults at high diabetes risk — without diabetes at enrolment — for 21 years concluded that metformin did not significantly affect all-cause mortality, cancer mortality, or cardiovascular mortality [16].
The picture may become clearer in coming years. The Targeting Aging with Metformin (TAME) trial is currently enrolling 3,000 older adults who will be followed for 6 years to assess whether metformin can delay the onset or progression of age-related chronic diseases — a broader endpoint than simple mortality [17]. TAME is the most rigorous human test of metformin's potential as an aging intervention, and its results will be closely scrutinised by researchers in this field.
For now, the available evidence does not support the use of metformin specifically for extending lifespan in people without metabolic disease.
Should everyone be taking this?
Alongside the potential benefits outlined above, two important concerns have emerged from the research that must be factored into any individual assessment.
Concern 1: Blunted exercise adaptation
Regular aerobic exercise is one of the most well-evidenced interventions for long-term cardiovascular and metabolic health. A 2019 trial in which both groups exercised found that participants taking metformin improved their cardiovascular fitness — measured as VO2 max — by approximately half as much as participants taking a placebo [18]. A 2022 replication study confirmed this finding, again demonstrating that metformin attenuated VO2 max improvements from exercise training by roughly 50% [19].
This is a clinically meaningful trade-off. If a medication significantly reduces the cardiovascular benefits of exercise, it undermines one of the most powerful tools available for long-term health. This concern is particularly relevant for people who are physically active or who are undertaking an exercise programme for weight management — which is precisely the population most likely to consider metformin for arthritis management.
Research has explored nutritional strategies that may support exercise adaptation. Trimethylglycine (TMG) — a methyl donor found in foods such as beetroot and wheat bran — has been studied in combination with exercise for its potential effects on strength and body composition [18, 19].
From the MicroVitamin range
MicroVitamin includes TMG (trimethylglycine) 500 mg — a methyl donor studied in combination with exercise for effects on strength and body composition. MicroVitamin.
Concern 2: Testosterone reduction in men
A second potential concern relates to male sex hormones. A study found that metformin use was associated with lower testosterone levels in men [20]. This matters because adequate testosterone is important for muscle mass, bone density, energy levels, and metabolic health in older men. Lower testosterone levels in men have been associated with a higher risk of mortality, particularly among older adults [21].
The clinical significance of the testosterone reduction attributable to metformin — and whether it is large enough to produce meaningful downstream effects in the absence of other risk factors — remains an area of ongoing research. But it is a consideration for men, particularly those who are already at the lower end of the normal testosterone range.
The overall picture
Weighing the evidence: metformin has well-established benefits in reducing systemic inflammation and supporting modest weight loss. The new JAMA trial adds preliminary evidence for pain reduction in knee osteoarthritis among overweight adults. The drug has been in clinical use for over six decades and carries a well-characterised safety profile. Adverse effects — primarily gastrointestinal symptoms such as nausea and diarrhoea — are usually temporary and can be minimised by starting at a low dose (for example, 500 mg daily) and titrating gradually to the therapeutic range [21].
The cost profile is unusually favourable. At the 2,000 mg daily dose used in the arthritis study, a month's supply can be obtained for approximately $6.14 at CostPlus Drugs [22] — making it accessible regardless of insurance status.
Against these potential advantages, the evidence that metformin significantly blunts cardiovascular adaptations to exercise is robust across two independent trials. For physically active individuals, this is a meaningful trade-off that deserves careful consideration. The observed reduction in testosterone levels in men adds a further consideration, particularly for older men or those with existing testosterone concerns.
For people with type 2 diabetes, metformin remains a well-established first-line treatment with a strongly favourable benefit–risk balance. For non-diabetics who are overweight and experiencing knee arthritis pain, the evidence base is promising but preliminary. The arthritis trial's key results — an 11-point between-group difference on a 100-point pain scale over 6 months — fall below the pre-specified minimum clinically important threshold of 15 points, though the biological plausibility and the supporting cohort data on cartilage preservation are encouraging.
Any decision about metformin for arthritis management outside of a diabetes diagnosis should be made in consultation with a treating clinician, who can assess individual circumstances including activity level, weight, hormone profile, and the current state of evidence — particularly as larger and longer trials, including TAME, report their findings in coming years.
Reference List
1. https://jamanetwork.com/journals/jama/article-abstract/2833338
2. https://arthritis-research.biomedcentral.com/articles/10.1186/s13075-019-1915-x
3. https://pmc.ncbi.nlm.nih.gov/articles/PMC10680465/
4. https://pmc.ncbi.nlm.nih.gov/articles/PMC4238740/
5. https://pmc.ncbi.nlm.nih.gov/articles/PMC10321632/
6. https://pmc.ncbi.nlm.nih.gov/articles/PMC1370926/
7. https://pubmed.ncbi.nlm.nih.gov/19878986/
8. https://journals.sagepub.com/doi/10.1177/2042018820926000
9. https://pmc.ncbi.nlm.nih.gov/articles/PMC6520185/
10. https://onlinelibrary.wiley.com/doi/10.1002/j.1550-8528.1998.tb00314.x
11. https://fortune.com/well/2023/05/04/metformin-anti-aging-longevity-risks-side-effects/
12. https://www.frontiersin.org/journals/pharmacology/articles/10.3389/fphar.2024.1330797/full
13. https://pmc.ncbi.nlm.nih.gov/articles/PMC6779524/
14. https://pmc.ncbi.nlm.nih.gov/articles/PMC5013015/
15. https://pubmed.ncbi.nlm.nih.gov/34421827/
17. https://www.afar.org/tame-trial
18. https://pmc.ncbi.nlm.nih.gov/articles/PMC6351883/
19. https://pmc.ncbi.nlm.nih.gov/articles/PMC9321693/
20. https://pmc.ncbi.nlm.nih.gov/articles/PMC8740051/
21. https://pmc.ncbi.nlm.nih.gov/articles/PMC9938530/
22. https://www.costplusdrugs.com/medications/metformin-1000mg-tablet/
