Metformin is one of the most prescribed medicines in the world, and for type 2 diabetes it earns that place. It lowers blood sugar reliably, it cut heart attacks and deaths in overweight patients over about a decade, it costs very little, and it has decades of safety data behind it. In people with prediabetes, about a third fewer went on to develop diabetes while taking it. It has also become the central drug in the question of whether a medicine can slow human aging, and here the evidence is thinner.
That interest rests on laboratory work, a mechanism metformin shares with fasting and exercise, and a records study in which metformin users lived slightly longer than people who did not have diabetes. That study cannot separate the drug from the people who receive it, and metformin has not been shown to extend human lifespan. For a healthy person the trade-off is sharper: metformin blunts the fitness and muscle gains from exercise, the best-supported longevity habit there is. Starting metformin is a prescription decision made with a clinician, and there is no dose on this page.
Findings & Outcomes
What It Is
Metformin is an oral medicine for type 2 diabetes and the most widely prescribed drug in its class, the biguanides. Its origin is a plant, Galega officinalis, known as French lilac or goat's rue, a European folk remedy for symptoms we would now call diabetes. Chemists traced that effect to a compound called galegine, and metformin is the refined version that came into wide clinical use from the late 1950s.
Metformin lowers blood sugar in type 2 diabetes mainly by reducing the glucose the liver releases into the blood, and it also makes muscle and other tissues respond better to insulin. It does not make the pancreas release more insulin, which is why on its own it rarely causes blood sugar to drop too low. It is approved as the first-line treatment for type 2 diabetes, it is used to help prevent diabetes in people with prediabetes, and it is inexpensive with a well-mapped safety record.
A second use for metformin comes from its cellular mechanism. It acts on the same nutrient-sensing pathways that fasting and exercise act on, and that overlap led aging researchers to ask whether a drug already taken by millions might slow some of the biology of aging. The proven diabetes use and the longevity question stay separate, because the human evidence supports the first and not yet the second. There is no dose here, because starting metformin is a prescription decision.
What It Does
Metformin's findings fall into three groups, ranked from strongest evidence to weakest. The diabetes evidence is the strongest. Metformin lowers HbA1c, the three-month average of blood sugar, by roughly one percentage point, and in UKPDS 34 the overweight adults assigned to it had about 36% lower all-cause mortality and about 39% fewer heart attacks over a median of nearly eleven years, against conventional treatment. Those are outcomes that accrue over years of treatment.
Prevention comes next. In people with prediabetes, metformin cut the rate of progressing to type 2 diabetes by 31% over about three years, while an intensive lifestyle program in the same trial cut it by 58%. The basics outperformed the drug in the study built to test the drug.
The longevity evidence sits lowest. The trial built to test whether metformin delays age-related disease in people without diabetes, TAME, has not reported. The most-cited human hint, a records study in which metformin users lived slightly longer than matched people without diabetes, is confounded. Set against that interest are two costs measured in older adults: metformin blunted both the fitness gains from aerobic training and the muscle growth from resistance training, and over years of use it lowers vitamin B12.
The Research & Studies
Everything here is based on the research we have collected and checked, sorted into groups and ordered with the strongest evidence first. Click any claim to open the studies behind it.
Blood Sugar
Metformin lowered HbA1c by about 1.1 percentage points in type 2 diabetes
Metformin lowers long-term blood sugar in type 2 diabetes by around one percentage point of HbA1c, which is a meaningful drop.
A meta-analysis of metformin trials found it lowered HbA1c by roughly 1.1 percentage points versus placebo, with a modest dose-response as the daily dose rose toward about 2000 mg. Measured in: Meta-analysis of 35 randomized controlled trials of metformin, 7,960 participants, in type 2 diabetes.. The size of the drop depends on the starting blood sugar and the dose, and the pooled studies varied in design and length.
The study · 1
Hirst et al., quantifying the effect of metformin treatment and dose on glycemic control · Diabetes Care 2012
Metformin cut progression from prediabetes to diabetes by 31%, lifestyle by 58%
In people with prediabetes, metformin cut the rate of developing diabetes by about a third, and a lifestyle program cut it by well over half.
In the Diabetes Prevention Program, metformin reduced the incidence of type 2 diabetes by 31% over an average 2.8 years versus placebo, while an intensive lifestyle intervention reduced it by 58%. Measured in: 3,234 adults with impaired glucose tolerance, 68% women, mean age 51, randomized to placebo, metformin or lifestyle, mean 2.8 years.. Lifestyle outperformed the drug in the same trial, and metformin's preventive effect was smaller in older and leaner participants.
The study · 1
Knowler et al., reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin (DPP) · N Engl J Med 2002
Heart And Vascular
Metformin cut all-cause mortality by 36% in overweight type 2 diabetes
In overweight people with type 2 diabetes, metformin lowered heart attacks and deaths over about a decade compared with usual care.
In UKPDS 34, overweight adults with type 2 diabetes assigned to metformin had about 32% fewer any diabetes-related endpoints, 39% fewer heart attacks and 36% lower all-cause mortality over a median 10.7 years, versus conventional treatment. Measured in: 753 overweight adults with newly diagnosed type 2 diabetes randomized to metformin (n=342) or conventional treatment, mainly diet (n=411); median follow-up 10.7 years.. This was an overweight type 2 diabetes population in a trial begun decades ago, and the mortality benefit was a secondary comparison that later trials have not reproduced as cleanly.
The study · 1
How it works
Metformin lowers the liver's glucose output and activates the energy sensor AMPK
Metformin lowers the sugar the liver makes and switches on a cellular low-fuel sensor called AMPK, the pathway that links it to aging biology.
Mechanistic reviews describe metformin's main action as reducing the liver's glucose production, partly through mild inhibition of mitochondrial complex I and activation of the energy sensor AMPK, alongside AMPK-independent and gut-mediated effects. Measured in: Synthesis of cell, animal and human mechanistic studies.. How much of the effect runs through AMPK versus other routes is still debated, and the concentrations that act on aging pathways in cells can exceed usual clinical exposure.
The study · 1
Foretz et al., metformin: from mechanisms of action to therapies · Cell Metab 2014
Cardiorespiratory Fitness
Metformin blunted the fitness gains from aerobic training in older adults
Older adults who took metformin while doing aerobic training got a smaller fitness gain than those on a placebo.
In a randomized, double-blind, placebo-controlled trial, older adults who took metformin during a 12-week aerobic training program gained less cardiorespiratory fitness and showed blunted skeletal-muscle mitochondrial adaptations than those on placebo. Measured in: 53 older adults, mean age 62, completing 12 weeks of supervised aerobic training, randomized to metformin (n=27) or placebo (n=26).. This ran 12 weeks in older adults, and whether the blunting persists long-term or applies to younger, non-diabetic trainees is not established.
The study · 1
Konopka et al., metformin inhibits mitochondrial adaptations to aerobic exercise training in older adults · Aging Cell 2019
Muscle And Strength
Metformin blunted muscle growth from resistance training in older adults
Older adults who took metformin while strength training built less muscle than those on a placebo, even though they got about as strong.
In the MASTERS randomized, double-blind, placebo-controlled trial, adults over 65 who took metformin during 14 weeks of progressive resistance training gained less muscle mass than those on placebo, though gains in strength were similar between groups. Measured in: 94 adults aged 65 and older completing 14 weeks of progressive resistance training, randomized to metformin or placebo.. Muscle size gains were blunted while strength gains were not, and the trial ran 14 weeks in adults over 65.
The study · 1
Walton et al., metformin blunts muscle hypertrophy in response to progressive resistance exercise training in older adults (MASTERS) · Aging Cell 2019
Longevity And Mortality
The trial to test metformin against aging has not reported results
Researchers proposed a large trial called TAME to test whether metformin can slow aging-related disease in people, because that has not yet been shown.
A 2016 perspective set out the rationale for the TAME trial, a proposed randomized trial testing whether metformin delays the onset of age-related diseases in people without diabetes, and stated that a human lifespan or healthspan benefit has not been demonstrated. Measured in: Perspective and design rationale, not a study of participants.. This is a rationale and design paper, not a result, and the trial itself had not reported an outcome as of writing.
The study · 1
Barzilai et al., metformin as a tool to target aging · Cell Metab 2016
Metformin users outlived matched non-diabetics in a records study, a confounded result
In one large records study, people on metformin sometimes outlived comparable people without diabetes, a striking result that is heavily confounded.
In a large retrospective cohort, people with type 2 diabetes started on metformin monotherapy had slightly lower all-cause mortality than matched people without diabetes, and lower mortality than diabetes patients started on a sulfonylurea. Measured in: About 78,000 metformin initiators and 12,000 sulfonylurea initiators with type 2 diabetes, each matched to non-diabetic controls, UK primary-care records.. What could explain it instead: Metformin initiators were compared with sulfonylurea initiators and with non-diabetic controls who differ in illness burden, and healthy-user, prescribing and survivor biases can make metformin users appear to fare better for reasons other than the drug.. This is observational and cannot establish that metformin caused the survival difference.
The study · 1
Bannister et al., can people with type 2 diabetes live longer than those without? Mortality in metformin or sulphonylurea initiators versus matched non-diabetic controls · Diabetes Obes Metab 2014
The Longevity Question
Online, metformin has shifted from a diabetes drug to something closer to a general longevity pill, a cheap tablet a healthy person might take to add years. That runs ahead of the human evidence.
TAME, short for Targeting Aging with Metformin, was designed as a large randomized trial to test whether metformin delays the onset of age-related diseases in people without diabetes. It was proposed on the strength of the drug's mechanism, animal data, and observational signals, and it has not reported an outcome. The central human test of the longevity idea is still pending.
The most-cited human signal is a records study of about 78,000 people started on metformin, matched against people without diabetes. The metformin group lived slightly longer than those non-diabetic controls. The result is confounded: an observational comparison cannot separate the drug from the people who receive it, and healthy-user, prescribing, and survivor effects can make metformin users appear to do better for reasons unrelated to the drug. Metformin has not been shown to extend lifespan or healthspan in humans. The interest rests on a plausible mechanism, and the claim is a hypothesis under test.
One trade-off is usually left out of the popular account, and it lands hardest on the person most drawn to metformin for aging. In two randomized, placebo-controlled trials in older adults, taking metformin during a training program produced smaller gains than the same training with a placebo: less cardiorespiratory fitness after aerobic training, and less muscle mass after resistance training. The effect appears to run through the same mitochondrial and AMPK actions that make the drug interesting for aging. Exercise is among the best-supported things anyone can do for healthy aging.
For a healthy person chasing longevity, exercise has far stronger support than metformin, and metformin can blunt part of what exercise builds.
How It Works
Metformin acts mainly in the liver, where it reduces the overproduction of glucose that causes high blood sugar in type 2 diabetes, and it makes muscle and other tissues respond better to insulin. At the cellular level it mildly and reversibly inhibits complex I of the mitochondria, one step in how cells produce energy. That lowers the cell's energy charge and activates AMPK, an enzyme that switches on when cellular fuel runs low and shifts the cell from building toward conserving and recycling. The gut is a further site of action, since much of the drug stays in the intestine and changes glucose handling there, which is part of why the common side effects are digestive.
That AMPK signal is why metformin draws interest for aging. AMPK and the nutrient-sensing pathways around it are among the recognized hallmarks of aging, and they are the same channels through which fasting and exercise are thought to do much of their good. Activating that sensor overlaps with the signaling of mild caloric restriction. How much of metformin's effect runs through AMPK, as against complex I directly and the gut-mediated routes, is still being worked out, and the concentrations that act on aging pathways in cells can exceed what usual clinical doses reach. The mechanism is established; the exact balance among these routes is not. For the system it acts on, see insulin and glucose handling, and for where AMPK sits among the drivers of aging, see biological aging.
Go Deeper
- Type 2 diabetes: the condition metformin treats first-line, where its evidence is strongest.
- Insulin and glucose handling: the metabolic system metformin acts on, and why lowering liver glucose output matters.
- Resistance training: the muscle-building habit metformin can blunt, and the best-supported way to hold strength with age.
- Walking: the low-barrier movement that underlies blood sugar and metabolic health.
- GLP-1 medications: the newer diabetes and weight drugs metformin is now weighed against.
- Blood sugar: the goal hub that pulls together food, movement, and the drugs that steady blood sugar.
The Chinese Medicine View
Metformin is a product of modern chemistry, refined in the twentieth century, so there is no classical Chinese entry for it. No historical text assigns it a channel, a temperature, or a flavor, and no traditional formula contains it. Even its plant ancestor, goat's rue, comes from European folk medicine, not the Chinese materia medica. What follows relates the drug to the tradition as an analogy, and it borrows no classical claim, because none exists. Nothing here suggests any herb or formula reproduces what metformin does.
The territory the drug acts on is one the tradition has long reasoned about. The Spleen, in this framework, governs the transformation and transport of food into usable substance. When that function is weak, the tradition describes food and fluid failing to move and gathering instead as Dampness and Phlegm, a picture often mapped onto insulin resistance and carrying excess weight. Read only as analogy, a drug that lowers the liver's sugar output and changes how the gut handles a meal is acting on the same ground the tradition assigns to the Spleen and Stomach.
The dietary-therapy counsel that fits here is plain and old: do not overload the Spleen with rich, sweet, heavy food, because constant abundance is the state the system handles worst. That principle stands on its own and needs no drug to support it. The tradition also holds that a strong intervention is warranted where there is a problem that calls for one, which is a fair way to think about a healthy person reaching for a medicine they do not medically need. The evidence for metformin rests on its own trials; this lens does not add to that evidence or replace it.
Cautions
Everything to be aware of is here, in one place. This practice suits most healthy people; a few situations call for real care.
Metformin lowered vitamin B12 over years of use
In long-term follow-up of the Diabetes Prevention Program, metformin use was associated with lower vitamin B12 levels and a higher rate of B12 deficiency over years compared with placebo. Deficiency accumulated over years of use and is correctable once detected, and the data come from a diabetes-prevention population.Aroda et al., long-term metformin use and vitamin B12 deficiency in the Diabetes Prevention Program Outcomes Study
Metformin did not raise lactic acidosis risk at normal kidney function
A Cochrane systematic review pooling prospective trials and cohort studies found no cases of fatal or nonfatal lactic acidosis attributable to metformin, and no difference in blood lactate, compared with other diabetes treatments. The included studies largely excluded people with significant kidney or liver impairment, so this population-level safety does not remove the need to respect kidney thresholds and pause the drug around acute illness or contrast scans.Salpeter et al., risk of fatal and nonfatal lactic acidosis with metformin use in type 2 diabetes mellitus
A prescription drug, and the decision sits with a clinician
Metformin is a prescription medicine. Whether it fits, at what dose, and how fast to build up depends on your blood sugar, kidney function, other conditions, and other medicines. Starting, changing, or stopping it belongs with a prescriber, and this page gives no dose because there is no safe general dose to give a healthy person outside diabetes care.
B12 depletion over time
Long-term metformin use lowers vitamin B12 and raises the chance of deficiency, which can contribute to anemia and to nerve symptoms if it goes unnoticed. In long-term follow-up of the Diabetes Prevention Program, low B12 was more common on metformin than placebo, and the risk rose with each year of use. Periodic B12 testing during long-term use is a reasonable step to raise with a prescriber, and deficiency is straightforward to correct once found.
Gastrointestinal side effects, and the extended-release option
Nausea, diarrhea, cramping, and a metallic taste are common, especially early and while the dose is being raised. For many people these gastrointestinal effects ease with time, with taking the drug alongside food, with a slower increase, or with an extended-release form. Severe or persistent symptoms are a reason to check back with the prescriber.
Lactic acidosis, kept in proportion, and when the drug is paused
The old fear attached to metformin is lactic acidosis, a rare buildup of acid in the blood. At the population level it is not borne out: a Cochrane review found no cases of lactic acidosis attributable to metformin and no rise in blood lactate compared with other diabetes drugs. That reassurance depends on respecting the boundaries, because the drug is cleared by the kidneys and can accumulate when they are impaired. It is paused around events that stress the kidneys, such as serious illness, dehydration, or scans using contrast dye, and it is dosed to kidney function. This is exactly the judgment a prescriber makes.
It can blunt some exercise adaptations
In randomized trials in older adults, metformin reduced the fitness gain from aerobic training and the muscle growth from resistance training that the same exercise produced without the drug. For a healthy person who trains, this is a meaningful trade-off, since exercise is among the best-supported things anyone can do for healthy aging, and the drug can offset part of that gain.
Not a do-it-yourself longevity drug
No completed human trial establishes a benefit or a dose for metformin as a way to slow aging in a healthy person. The mechanism is established and the research is ongoing, including the TAME trial. If the longevity question interests you, the step that fits the evidence is a conversation with a knowledgeable clinician who can weigh the trade-offs.
Start slow, be smart, read the research, and consult a professional if you have any concerns. This is here to inform your choice, not make it for you.
Common Questions
Does metformin extend lifespan?
In humans, it has not been shown to. No completed trial demonstrates that metformin extends lifespan or healthspan, which is the specific question the TAME trial was designed to test, and that trial has not reported. The interest comes from a plausible mechanism, its effect on the AMPK nutrient-sensing pathway, and from animal and observational data. The observational hint that metformin users sometimes outlive expectations is confounded, meaning it cannot separate the drug from the people who happen to receive it. The accurate summary is a promising hypothesis under active study.
Should a healthy person take metformin for aging?
The evidence does not support it yet. For someone without diabetes, taking metformin to slow aging runs ahead of what has been shown, there is no completed trial establishing a benefit or a dose for that use, and the drug carries trade-offs that land hardest on a fit person. The basics have far stronger support: exercise, whole foods, sleep, and holding on to muscle. In the Diabetes Prevention Program, a lifestyle program prevented nearly twice as much diabetes as metformin did. If the question interests you, weigh it with a clinician before starting on the strength of the idea.
Does metformin interfere with exercise or muscle?
It can, and this is one of the clearer trade-offs. In randomized, placebo-controlled trials in older adults, metformin blunted the fitness improvement from 12 weeks of aerobic training and the muscle growth from 14 weeks of resistance training that the same exercise produced without the drug. Strength gains held up better than muscle size in the resistance study. The effect appears to work through the same mitochondrial and AMPK actions that make metformin interesting for aging. For someone who trains and is already fit, that is a cost worth weighing.
How serious is the lactic acidosis risk?
It is rare, and the reputation is larger than the risk when the boundaries are respected. A Cochrane review pooling trials and cohorts found no cases of lactic acidosis attributable to metformin and no rise in blood lactate compared with other diabetes drugs. The safety depends on kidney function, since the drug is cleared by the kidneys and can build up when they are impaired, so it is dosed to kidney function and paused around events that stress the kidneys, such as serious illness, dehydration, or contrast scans.
Does metformin deplete B12?
Over years of use, yes. Long-term follow-up of the Diabetes Prevention Program found low vitamin B12 more often on metformin than placebo, with the risk rising for each year of use. Left unnoticed, low B12 can contribute to anemia and nerve symptoms. Periodic testing during long-term use is worth raising with a prescriber, and a deficiency is straightforward to correct once it is found.
Explore Related
Other pages this one connects to, by the evidence they share, the outcomes they touch, and the ground they cover.
All 10 sources on this page independently checked and cross-referenced.
Thomas Dehli, Founder & Editor, Sacred Lotus
Sacred Lotus has published Chinese medicine reference material since 2001. Integrative pages are held to the same standard as the herb and formula library: cite the source, grade the claim at its real strength, and say where the research has not looked. This page is educational and it is not medical advice. Last reviewed and updated August 9, 2026.
Evidence strength
How confidently the research supports a claim. Strength describes the evidence, not our endorsement.