Metformin: Independent Evidence on Benefits, B12, Kidneys & Ageing

Table of contents

• Evidence summary
• What metformin is
• How it works
• What it is prescribed for
• What works and what does not
• Benefits by claim
• Risks and all side effects
• All interactions
• Who should avoid metformin
• Dosage and how to take it
• Follow the money: who makes it and who funded the evidence
• Related research
• Frequently asked questions
• Sources and funding notes

Evidence summary

Independent evidence and credibility scorecard

Independence tiers: 1 = regulator or government body; 2 = academic or Cochrane team with no declared commercial funding for the work; 3 = independent authors or bodies with declared industry funding, grants or ties; 4 = manufacturer-funded or manufacturer-authored. Credibility runs from A (highest) to D.

Our sources cluster in the UK and the US. NICE (cost-focused, tier 1) and the ADA (industry-supported, tier 3) face opposite pressures, so their agreement that metformin is still a sensible starting drug is meaningful.

What metformin is

Metformin (dimethylbiguanide) is a tablet medicine for type 2 diabetes, derived from guanidine compounds in the herb goat's rue. The French doctor Jean Sterne first reported it for diabetes in 1957. Its relatives phenformin and buformin were mostly withdrawn in the late 1970s because of lactic acidosis, which delayed metformin's acceptance; it reached the US only in 1995 (Bailey 2017).

It comes in two main forms:

• Standard (immediate) release, taken two or three times a day with meals. In the UK the original brand is Glucophage 500 mg and 850 mg tablets (UK SmPC).
• Modified or extended release (often labelled SR, XR or MR), usually taken once a day with the evening meal (FDA label).

Metformin is also sold in fixed combinations with other diabetes drugs. It is prescription-only in the UK, EU and US. It is off-patent and made by many generic companies.

How it works

Metformin does not make the pancreas release more insulin. That is why, used alone, it does not usually cause low blood sugar (UK SmPC). The UK product information lists several mechanisms:

• It lowers glucose production by the liver.
• It helps muscle and other tissues take up and use glucose, partly by improving insulin action.
• It changes how the gut handles glucose, raises release of the gut hormone GLP-1, reduces bile-acid reabsorption and alters the gut microbiome.
• It activates an energy-sensing enzyme called AMPK.

The glucose-lowering effect in humans is well proven. Exactly which pathway matters most is still debated. The AMPK and "energy-sensing" effects are the basis of anti-ageing interest, but much of that work is from cells and animals. The gut actions help explain why diarrhoea and stomach upset are so common.

Pharmacokinetics that matter in practice

Metformin is not broken down by the liver; the kidneys remove it unchanged using the OCT2 and MATE transporters (FDA label). When the kidneys fail, levels rise, which is the main route to lactic acidosis. It crosses the placenta and passes into breast milk (UK SmPC).

What it is prescribed for

Licensed uses

In the UK, Glucophage is licensed for type 2 diabetes when diet and exercise are not enough, alone or with other diabetes drugs or insulin, in adults and children from 10 years (UK SmPC). The US label covers adults and children aged 10 to 16; extended release is not recommended under 17 (FDA label). Off-label uses include prediabetes, PCOS and gestational diabetes.

Where guidelines place it

• NICE (England, February 2026): for adults with type 2 diabetes and no other relevant conditions, offer modified-release metformin and an SGLT-2 inhibitor. Start metformin first, check it is tolerated, then add the SGLT-2 inhibitor once metformin is at its highest tolerated dose. If metformin is unsuitable, offer an SGLT-2 inhibitor alone (NG28 1.13; NG28 1.20). The same pairing is advised in heart failure and in kidney disease with an eGFR of 30 or more (NG28 1.14 and 1.18).
• American Diabetes Association (2026): describes metformin as "historically" the first-line treatment and still a common choice when glucose lowering is the only consideration. It notes metformin is inexpensive and widely available, and that drug choice in people with heart or kidney disease is now driven by organ protection (ADA Standards 2026).

So metformin remains a foundation drug, but newer classes now join it early for heart and kidney protection.

What works and what does not

Benefits by claim

Type 2 diabetes: the UKPDS 34 result

UKPDS was a UK trial that recruited people with newly diagnosed type 2 diabetes from 1977 to 1991. In the metformin part, 753 overweight people (mean age 53) were randomised to diet alone (411) or an intensive policy using metformin (342) and followed for a median of 10.7 years (UKPDS 34). Compared with diet, metformin was linked to:

• 32% fewer diabetes-related events of any kind (95% CI 13% to 47%).
• 42% fewer diabetes-related deaths (9% to 63%).
• 36% fewer deaths from any cause (9% to 55%).

Metformin also caused less weight gain and fewer hypos than sulphonylureas or insulin. The caveats: the metformin group was small, the trial was open-label, and a supplementary trial in the same paper linked early addition of metformin to sulphonylureas with 96% more diabetes-related deaths (95% CI 2% to 275%), a signal a wider analysis did not confirm (UKPDS 34).

Does the benefit last?

Glucose differences vanished within a year of the trial ending, yet 10 years later the metformin group still had 33% fewer heart attacks and 27% fewer deaths (UKPDS 80). Linked NHS records showed the effect persisting 24 years after the trial: deaths 20% lower (5% to 32%) and heart attacks 31% lower (12% to 46%), or 4.9 fewer deaths and 6.2 fewer heart attacks per 100 people. There was no significant effect on stroke or small-vessel disease (UKPDS 91).

What newer trials say

A Cochrane review of 18 trials (10,680 people) found none at low risk of bias in every area and none against placebo. It concluded there is "no clear evidence" either way on death, heart attack or stroke versus other drugs (Gnesin 2020). Few companies have reason to fund outcome trials of a cheap generic.

Preventing diabetes

The US Diabetes Prevention Program randomised 3,234 adults with raised blood glucose and an average BMI of 34. Over 2.8 years, new diabetes per 100 person-years was 11.0 on placebo, 7.8 on metformin 850 mg twice daily and 4.8 with an intensive lifestyle programme. Metformin cut risk by 31% (17% to 43%); lifestyle cut it by 58% (48% to 66%). About 14 people needed metformin for three years to prevent one case, compared with about 7 for the lifestyle programme (DPP 2002).

Polycystic ovary syndrome (PCOS)

The 2023 international PCOS guideline says metformin alone should be considered in adults with PCOS and a BMI of 25 or more for weight and metabolic outcomes, and may be considered at lower BMI, "acknowledging limited evidence". It advises telling women that metformin and active lifestyle change have similar effects. For hirsutism and irregular periods, the combined pill is preferred; for metabolic problems, metformin is preferred (Teede 2023).

For fertility, letrozole is first line, with clomiphene alone or plus metformin as second line (guideline summary). Cochrane found metformin may increase live births versus placebo (OR 1.59, 1.00 to 2.51; low quality), but clomiphene did better in women with obesity, and gut side effects were nearly five times as likely (OR 4.76) (Morley 2017).

Gestational diabetes and diabetes in pregnancy

In the MiG trial, 751 women with gestational diabetes in New Zealand and Australia were randomised to metformin or insulin. The rate of newborn complications was almost identical: 32.0% with metformin and 32.2% with insulin (RR 0.99, 0.80 to 1.23). Almost half the metformin group (46.3%) also needed some insulin. Women strongly preferred metformin: 76.6% would choose it again, versus 27.2% for insulin (MiG 2008).

In MiTy, adding metformin to insulin in 502 pregnant women with type 2 diabetes did not change the main newborn outcome (40% vs 40%). Mothers had better glucose control, needed less insulin and had fewer caesareans (53% vs 63%), but more babies were small for their gestational age (13% vs 7%; RR 1.96, 1.10 to 3.64) (MiTy 2020).

The UK product information now says data from more than 1,000 exposed pregnancies show no increase in birth defects, that evidence on children's long-term weight is "limited and inconclusive", and that metformin can be considered in pregnancy as an addition or alternative to insulin (UK SmPC). The PCOS guideline adds that long-term effects on offspring "remain unclear", with a suggestion of higher childhood weight (Teede 2023).

Anti-ageing: the TAME hype versus the evidence

Interest grew partly from a 2014 UK registry study in which people with diabetes starting metformin outlived matched people without diabetes (Bannister 2014). A Danish national-register and twin study found the opposite: higher death rates on metformin after adjustment (IRR 1.32 in singletons; 1.64 in twin pairs) (Keys 2022). Observational data cannot separate the drug from the people who take it.

The Targeting Aging with Metformin (TAME) trial was proposed in 2016 as the first trial to treat ageing itself (Barzilai 2016). It aims to enrol 3,000 people aged 65 to 79 for six years. In 2024 NPR reported that funding had been slow to arrive: the US National Institute on Aging had set aside about $5 million, against an estimated cost of $45–70 million. Its lead investigator said the main obstacle is that metformin is generic, so no drug company stands to profit (NPR, 2024). We found no published TAME results.

Two small randomised trials in healthy older adults point the other way. In 53 people doing 12 weeks of aerobic training, metformin weakened the gains in insulin sensitivity and fitness (VO2max) (Konopka 2019). In 94 people doing 14 weeks of strength training, those on placebo gained more lean body mass than those on 1,700 mg a day of metformin (Walton 2019). Until TAME or a similar trial reports, metformin for ageing in healthy people is an unproven experiment.

Cancer

Pooled randomised trials found no effect on cancer (RR 1.02, 0.82 to 1.26), despite earlier observational hopes of a one-third reduction (Stevens 2012).

Risks and all side effects

Common side effects

The UK product information lists these by frequency (UK SmPC):

• Very common (more than 1 in 10): nausea, vomiting, diarrhoea, stomach pain and loss of appetite. These are worst when starting and usually settle.
• Common (up to 1 in 10): vitamin B12 decrease or deficiency; taste disturbance.
• Very rare (fewer than 1 in 10,000): lactic acidosis; abnormal liver tests or hepatitis that resolve on stopping; skin reactions such as redness, itching and hives.

In the US licensing trial of standard-release Glucophage, diarrhoea affected 53.2% versus 11.7% on placebo and nausea or vomiting 25.5% versus 8.3%; 6% stopped because of diarrhoea. In separate extended-release trials, diarrhoea affected 9.6% versus 2.6% (FDA label). These are different trials, not a direct comparison.

Does modified release really help the gut?

When trials compared the two forms directly at the same dose (9 trials, 2,609 adults), extended release only clearly reduced indigestion (RR 0.58, 0.34 to 0.98). Diarrhoea, nausea and stomach pain were similar. Standard release lowered HbA1c very slightly more (by 0.09 percentage points), a difference unlikely to matter (Abrilla 2021). NICE acknowledged "limited evidence" but chose modified release first, citing fewer gut effects in evidence outside its formal review and committee experience, plus once-daily dosing (NG28). For people already on standard release, NICE says continue, or switch if it is not tolerated or if they prefer (NG28 1.23).

Vitamin B12 deficiency

In the Diabetes Prevention Program follow-up, low or borderline B12 was more common on metformin at 5 years (19.1% vs 9.5%) and 13 years (20.3% vs 15.6%), and each year of use raised the odds of deficiency by about 13% (Aroda 2016). The ADA also suggests periodic B12 testing (ADA 2026).

Lactic acidosis

How rare is it? A Cochrane review of 347 studies found no cases in 70,490 patient-years on metformin; the statistical upper limit was 4.3 per 100,000 patient-years, versus 5.4 on other drugs (Salpeter 2010). Those studies mostly excluded high-risk people, so the warning still matters when someone is acutely unwell.

Low blood sugar

Metformin alone does not normally cause hypoglycaemia. It can occur when food intake is low, after heavy exercise without enough calories, or when metformin is combined with insulin or sulphonylureas (FDA label; UK SmPC).

Contamination: the NDMA recalls

In 2020 the FDA found the probable carcinogen NDMA in some metformin. It asked five firms to recall extended-release products above the acceptable limit: Apotex and Amneal (all lots), Marksans/Time-Cap and Lupin (one lot each) and Teva/Actavis (several lots); more recalls followed into 2021. Immediate-release metformin tested clean, and patients were told to keep taking their medicine until replaced (FDA NDMA record).

Overdose

No hypoglycaemia was seen with doses up to 85 g, but lactic acidosis occurred; overdose needs hospital care (UK SmPC).

All interactions

The main interactions come from the label and UK product information (FDA label; UK SmPC):

• Alcohol: heavy or binge drinking raises lactic acidosis risk, especially with fasting, poor nutrition or liver disease. Not recommended.
• Iodinated contrast dye (some CT scans and angiograms): can cause sudden kidney injury and metformin build-up. See the next section.
• Drugs that can reduce kidney function: anti-inflammatory painkillers (NSAIDs, including COX-2 inhibitors), ACE inhibitors, angiotensin receptor blockers and diuretics, especially loop diuretics. Kidney function needs closer monitoring.
• Drugs that block metformin's kidney transporters (OCT2/MATE): cimetidine raised peak metformin levels by 60% and total exposure by 40%; ranolazine, vandetanib and dolutegravir act similarly.
• Carbonic anhydrase inhibitors (topiramate, zonisamide, acetazolamide): can cause acidosis and may add to lactic acidosis risk.
• Drugs that raise blood glucose: steroids, thiazide and other diuretics, some antipsychotics (phenothiazines), thyroid hormones, oestrogens and oral contraceptives, phenytoin, nicotinic acid, sympathomimetics, calcium-channel blockers and isoniazid. Glucose control needs watching when these start or stop.
• Insulin and sulphonylureas: raise the risk of low blood sugar when combined.
• Nifedipine: raised peak metformin levels by 20% in volunteers.

Contrast scans, surgery and sick-day rules

The UK label says to stop metformin before or at the time of any intravascular iodinated contrast, and before surgery under general, spinal or epidural anaesthesia, restarting no earlier than 48 hours later once kidney function is stable (UK SmPC). The US label limits the contrast rule to people with an eGFR of 30 to 60, liver disease, alcoholism, heart failure or contrast given into an artery (FDA label).

Sick-day rule: during dehydration (severe diarrhoea or vomiting, fever, not drinking) metformin is paused and a health professional contacted (UK SmPC).

Who should avoid metformin

Contraindications

The UK product information says metformin must not be used with (UK SmPC):

• allergy to metformin;
• any metabolic acidosis, including lactic acidosis and diabetic ketoacidosis, or diabetic pre-coma;
• severe kidney failure (eGFR below 30);
• acute conditions that can affect the kidneys, such as dehydration, severe infection or shock;
• conditions causing low oxygen, such as decompensated heart failure, respiratory failure, a recent heart attack or shock;
• liver failure, acute alcohol intoxication or alcoholism.

Kidney disease

NICE, the ADA and the US label agree metformin is not used below an eGFR of 30 (NICE; ADA). The US label adds that starting it at 30 to 45 is not recommended (FDA label). The UK label sets maximum doses by kidney function and asks for kidney tests at least yearly, or every 3 to 6 months in older and at-risk people (UK SmPC).

Source: Glucophage UK SmPC, section 4.2 (written by the brand holder, approved by the MHRA). Label figures, not personal dosing advice.

Use with care

Older adults (kidney checks more often), stable heart failure (with monitoring; not acute or unstable heart failure), liver disease (the US label advises avoiding it) and people at risk of low B12 such as vegans or those after weight-loss surgery (FDA label; UK SmPC). NICE still offers modified-release metformin to people with frailty (NG28 1.19).

Children and adolescents

Licensed from age 10 once type 2 diabetes is confirmed; long-term effects on growth and puberty are not known (UK SmPC).

Pregnancy and breastfeeding

The UK label allows metformin to be considered in pregnancy with or instead of insulin (UK SmPC); the older US Glucophage label says not to use it unless clearly needed (FDA label, 2017). In PCOS, the guideline advises stopping it at the positive pregnancy test unless otherwise needed (Teede 2023). It passes into breast milk; no harm has been seen, but data are limited and the UK label does not recommend breastfeeding during treatment (UK SmPC).

Dosage and how to take it

These are label doses, reported for information only. Your prescriber sets your dose.

• UK label (standard release, adults): usually 500 mg or 850 mg two or three times a day with or after meals, adjusted after 10 to 15 days. Maximum 3 g a day in three divided doses. A slow increase helps the gut (UK SmPC).
• US label (standard release, adults): usually 500 mg twice a day or 850 mg once a day, increased gradually. Maximum 2,550 mg a day; 2,000 mg a day for children aged 10 to 16 (FDA label).
• US label (extended release, adults): once a day with the evening meal; maximum 2,000 mg a day (FDA label).
• Doses studied: 850 mg twice a day in the Diabetes Prevention Program (DPP); 1,000 mg twice a day in MiTy (MiTy). The PCOS guideline suggests a maximum of 2.5 g a day in adults and 2 g in adolescents, increasing by 500 mg every one to two weeks (Teede 2023).

How to take it

With food, to reduce gut upset. Modified-release tablets are generally swallowed whole and taken with the evening meal. Kidney function and, in people at risk, vitamin B12 are checked over time.

Follow the money: who makes it and who funded the evidence

Who made it, and who sells it now

Metformin was developed for diabetes by Jean Sterne in France in the 1950s, and introduced to the US in 1995 (Bailey 2017). The Glucophage brand is a registered trademark of Merck Santé S.A.S., part of Merck KGaA of Darmstadt, Germany. In the US it was licensed to and distributed by Bristol-Myers Squibb (FDA label). In the UK, the marketing authorisation holder is Merck Serono Ltd (UK SmPC). We found no current Glucophage-branded label on the US DailyMed database; US patients mostly receive generics.

Despite generic competition, the brand is still a large business. Merck KGaA reported Glucophage net sales of €975 million in 2025, up from €954 million in 2024, with organic growth of 5.9%. Asia-Pacific accounted for 52% (€509 million), Latin America 23%, Europe 15% and the Middle East and Africa 10%, with no North American sales listed (Merck KGaA annual report 2025).

Who funded the evidence

The landmark trials were mainly publicly funded but not industry-free. UKPDS took research grants from 13 drug companies, including Lipha and Bristol-Myers Squibb (which later held the US Glucophage licence), alongside UK and US public money (UKPDS funding statement); its 24-year follow-up was funded by the University of Oxford (UKPDS 91). The Diabetes Prevention Program was led by the US NIH, with extra money from the ADA, Bristol-Myers Squibb and Parke-Davis, and drug supplied by Lipha (DPP funding). MiTy was funded by the Canadian Institutes of Health Research (MiTy). The ADA says no industry money funds its Standards (ADA methodology), but its top corporate supporters include Eli Lilly, Novo Nordisk and Boehringer Ingelheim, which sell newer diabetes drugs (ADA supporters). TAME has only a small NIA grant and philanthropic pledges (NPR 2024).

Who profits either way

Because metformin is generic, commercial pressure mostly runs against it: makers of newer patented drugs gain if guidelines move away from metformin-first. Merck KGaA gains if branded Glucophage stays first line in Asia and Latin America. Longevity clinics gain from anti-ageing claims that no completed trial supports. Health systems gain from a cheap first-line drug.

Documented integrity events

The main documented event is the 2020 NDMA contamination of some extended-release generic metformin, which led to FDA-requested recalls from five named manufacturers and later recalls from others (FDA). We found no FDA warning letters or legal actions concerning Glucophage marketing claims in the sources we reviewed.

Related research

For wider, evidence-graded context, see these Pure City Research guides:

• Type 2 diabetes: prevention and management evidence
• Diabetes and heart disease: cardiovascular risks and treatment purposes
• Diabetes supplements: benefits, limits and safety
• Berberine: glucose evidence, formulations and safety
• Myo-inositol: PCOS, fertility and pregnancy trials
• Chronic kidney disease and the heart: tests and medicine safety
• Atrophic and autoimmune gastritis: B12, iron and follow-up
• Heart failure: treatment evidence and safety
• Alcohol-related liver disease
• Ramipril: uses, kidney monitoring and safety
• Heart disease during pregnancy: planning and medicines

Frequently asked questions

Sources and funding notes

What we could not verify: the full funding statement of UKPDS 34 itself (we used the funding list from the same study programme's UKPDS 38 paper); the funding statements of the MiG trial, the Stevens 2012 cancer meta-analysis, the Bannister 2014 and Keys 2022 registry studies, the Konopka 2019 trial and the Abrilla 2021 meta-analysis; the original FDA 2016 drug safety communication on kidney function (the page has moved, so we used the 2017 label that reflects it); whether TAME has enrolled participants or secured full funding since 2024; current US ownership and sales of the Glucophage brand; and any metformin price or prescription-volume figures. We did not review the UK modified-release (Glucophage SR) product information separately.

1. NICE NG28: Type 2 diabetes in adults, initial medicines (last updated 18 February 2026); also how to introduce medicines, reviewing medicines and person-centred medicine — national guideline; UK public body.
2. MHRA Drug Safety Update: metformin and reduced vitamin B12 levels (20 June 2022) — UK regulator.
3. FDA updates and press announcements on NDMA in metformin — US regulator.
4. Glucophage and Glucophage XR US prescribing information (revised April 2017) — written by Bristol-Myers Squibb (licensee of Merck Santé), approved by FDA.
5. Glucophage 500 mg and 850 mg film-coated tablets, UK SmPC (updated 22 December 2025) — written by Merck Serono Ltd, approved by MHRA.
6. ADA. 9. Pharmacologic approaches to glycemic treatment: Standards of Care in Diabetes—2026 and Introduction and methodology — US professional body; funded from ADA general revenue.
7. ADA national corporate supporters — ADA's own disclosure.
8. UKPDS Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet 1998 — UK randomised trial; mixed public, charity and industry grants.
9. UKPDS 38. BMJ 1998 — used for the UKPDS programme's funding list (MRC, BDA, Department of Health, NIH, BHF, charities and grants from 13 drug companies including Lipha and Bristol-Myers Squibb).
10. Holman RR et al. 10-year follow-up of intensive glucose control in type 2 diabetes (UKPDS 80). NEJM 2008 — University of Oxford; funding statement not reviewed in full.
11. Adler AI et al. UKPDS 91. Lancet 2024 — University of Oxford pump-priming; senior author reports fees from AstraZeneca, Novartis and Anji.
12. Gnesin F et al. Metformin monotherapy for adults with type 2 diabetes. Cochrane 2020 — Cochrane; one author with multiple diabetes-industry ties.
13. Salpeter SR et al. Risk of fatal and nonfatal lactic acidosis with metformin use. Cochrane 2010 — Cochrane; no conflicts declared.
14. Knowler WC et al. Diabetes Prevention Program. NEJM 2002 and full text with funding — US NIH-led; also ADA, Bristol-Myers Squibb, Parke-Davis; Lipha supplied drug.
15. Aroda VR et al. Long-term metformin use and vitamin B12 deficiency in DPPOS. JCEM 2016 — DPP research group (US).
16. Abrilla AA et al. Metformin extended-release versus immediate-release: meta-analysis. Diabetes Res Clin Pract 2021 — University of the Philippines; no competing interests declared.
17. Rowan JA et al. Metformin versus insulin for gestational diabetes (MiG). NEJM 2008 — New Zealand/Australia; funding not verified.
18. Feig DS et al. Metformin in women with type 2 diabetes in pregnancy (MiTy). Lancet Diabetes Endocrinol 2020 — Canadian Institutes of Health Research; Lunenfeld-Tanenbaum Research Institute; University of Toronto.
19. Løvvik TS et al. Metformin in pregnant women with PCOS (PregMet2). Lancet Diabetes Endocrinol 2019 — Norway/Sweden/Iceland; funded by the Research Council of Norway, the Novo Nordisk Foundation, St Olav's University Hospital and the Norwegian University of Science and Technology.
20. Morley LC et al. Insulin-sensitising drugs for women with PCOS. Cochrane 2017 — Cochrane; one author consulted for Merck Serono.
21. Teede HJ et al. Recommendations from the 2023 international evidence-based guideline for PCOS. Hum Reprod 2023 and MJA summary 2024 — Monash University-led, NHMRC-funded; approved by NHMRC.
22. Stevens RJ et al. Cancer outcomes and all-cause mortality in adults allocated to metformin. Diabetologia 2012 — Oxford-led collaboration; funding not verified.
23. Bannister CA et al. Mortality with metformin or sulphonylurea vs non-diabetic controls. Diabetes Obes Metab 2014 — Cardiff University; funding not verified.
24. Keys MT et al. Reassessing the evidence of a survival advantage with metformin. Int J Epidemiol 2022 — University of Southern Denmark; funding not verified.
25. Barzilai N et al. Metformin as a tool to target aging. Cell Metab 2016 — TAME investigators (US academic).
26. Konopka AR et al. Metformin inhibits mitochondrial adaptations to aerobic exercise training in older adults. Aging Cell 2019 — US academic; funding not verified.
27. Walton RG et al. Metformin blunts muscle hypertrophy (MASTERS). Aging Cell 2019 — US academic; no conflicts declared.
28. Bailey CJ. Metformin: historical overview. Diabetologia 2017 — Aston University (UK); history.
29. NPR (via WFAE). A cheap drug may slow down aging. A study will determine if it works (22 April 2024) — US public-radio journalism.
30. Merck KGaA, Darmstadt, Germany. Annual Report 2025, Healthcare section — company report; Glucophage sales figures.

Educational research reviewed 6 October 2026. This article is not medical advice; talk to a qualified clinician or pharmacist before starting, stopping or combining treatments.