Metformin: diabetes, longevity evidence and costs

Metformin has substantial diabetes evidence, but extending that evidence to healthy aging requires care. Observational survival comparisons do not establish causality. Newer randomized work measures particular biological clocks or frailty rather than proving longer life. Two older-adult exercise trials found smaller improvements with metformin; those findings do not establish the same effect for every patient.

How metformin works: less sugar from the liver, better use of insulin

The label lists three effects. Metformin "decreases hepatic glucose production, decreases intestinal absorption of glucose, and improves insulin sensitivity" ^[3]. In plain terms, your liver puts out less sugar, your gut takes in a little less, and your muscles pull more sugar from the blood.

It doesn't make your pancreas release more insulin. The label says insulin secretion "remains unchanged" ^[3]. That's one reason metformin on its own rarely drops blood sugar too low.

The longevity interest comes from what metformin does in cells. Lab and animal studies tie it to energy-sensing pathways linked to aging, and it calms inflammation in the lab. Whether that adds up to slower aging in people is the open question. For the inflammation side, see does metformin help with inflammation.

What the studies show: proven for diabetes, unproven for aging

Randomized diabetes trials, observational mortality comparisons and aging-marker trials answer different questions. Opposite observational results do not establish either a longevity benefit or that metformin caused excess deaths. The newer aging studies require attention to population, endpoint and publication status.

The prediabetes result is the one most people can use. In the Diabetes Prevention Program, about 14 people had to take metformin for three years to prevent one case of diabetes. For the lifestyle program, it was about 7 ^[8]. Metformin helps, and diet and activity help more.

Does metformin help you live longer? Markers are different from lifespan

The cited research does not establish lifespan extension in otherwise healthy people. It is too broad to say no randomized aging study exists: recent trials assess biological clocks or frailty. A clock estimate or frailty-index change is not interchangeable with longer survival or prevention of every age-related disease.

In 2014, a UK records study found that people with type 2 diabetes who started on metformin had a longer median survival than matched people without diabetes ^[9]. That result got a lot of attention. In 2022, a Danish study using national registers found the reverse. It compared metformin users with people without diabetes, including twin pairs where one twin had diabetes. Metformin users had higher death rates, and the authors said design differences were unlikely to explain the gap ^[10].

TAME (Targeting Aging with Metformin) was proposed as a broader aging-outcome trial ^[13]. A proposal is not a result. In the 2026 METFORAGING pilot, forty non-diabetic adults with controlled HIV were randomized for 96 weeks; thirty-five completed treatment. The primary PhenoAge comparison was not statistically significant (adjusted difference −1.02 years, 95% CI −5.30 to 3.26; P=.627). That small, specific population does not settle healthy-population longevity ^[1].

A 2026 preprint reports a two-year randomized study of 145 older adults with glucose intolerance. The deficit-accumulation frailty index and two epigenetic clocks favored metformin, but the Fried frailty phenotype did not differ after removing the weight-loss criterion. This is not yet peer-reviewed evidence and does not demonstrate lifespan extension; its population, several endpoints and mixed findings matter ^[2].

Side effects: stomach upset is common, lactic acidosis is rare but serious

Diarrhea, nausea and gas are the side effects most people notice. The serious risk is lactic acidosis, a buildup of acid in the blood, which carries the label's boxed warning.

• Boxed warning. Lactic acidosis has caused deaths. The risk rises with kidney problems, age 65 or older, contrast dye for scans, surgery, low oxygen states such as heart failure flares, heavy drinking and liver problems ^[3].
• Who shouldn't take it. People with severe kidney impairment (eGFR below 30), metformin allergy, or metabolic acidosis, including diabetic ketoacidosis ^[3]. Starting treatment is not recommended when eGFR is 30–45; existing therapy requires reassessment below 45.
• Common (more than 5%). Diarrhea, nausea and vomiting, gas, weakness, indigestion, belly discomfort and headache ^[3].
• Vitamin B12. Metformin can lower B12. The label calls for blood counts every year and a B12 check every 2 to 3 years ^[3].
• Low blood sugar. Mostly a risk when it's combined with insulin or drugs that push insulin release ^[3].

On dosing, the label sets the range. Immediate-release tablets start at 500 mg twice a day or 850 mg once a day, with meals, and the label allows up to 2,550 mg a day ^[3]. Those are diabetes-label regimens, not an established longevity dose; immediate- and extended-release schedules differ. Ask what the selected regimen is based on.

FDA status: approved for diabetes since 1995, used off-label for aging

Metformin is FDA-approved for type 2 diabetes. Any other use, including aging or weight loss, is off-label. FDA says providers "generally may prescribe the drug for an unapproved use when they judge that it is medically appropriate" ^[6].

Today, Glucophage itself is listed as discontinued ^[4]. Metformin lives on as a generic, like the tablets whose label is cited here ^[3]. Combination pills that pair metformin with other diabetes drugs exist too. They're different products, and they aren't part of the cost table below.

What to include in a metformin quote

Compare the generic pharmacy fill separately from a clinical program. A headline can include visits or membership, or cover only medication. Ask for the exact preparation, quantity, charge interval and mandatory fees. Program availability is not evidence that off-label longevity use will benefit you.

Use the billing calculator to separate calendar-month and fixed-day charges, and the provider checklist to organize the written quote. AgelessRx’s review explains one longevity program’s service and pricing limitations. A lower pharmacy price alone does not include clinical assessment or monitoring.

How to get metformin prescribed

Metformin needs a prescription. For diabetes or prediabetes, your regular doctor can prescribe it. For longevity, online programs run an intake and prescribe it off-label.

1. Labs first. A kidney test (eGFR) is the big one. Many programs also check A1C and a metabolic panel.
2. A clinician visit. Expect questions on kidney and liver health, alcohol, heart failure and upcoming scans or surgery.
3. A fill. Generic tablets from a retail or mail-order pharmacy, or from the program's pharmacy partner.
4. Follow-up. Repeat kidney tests, and a B12 check every few years, per the label.

Five questions to ask before you pay

A program worth your time answers these up front. If it can't, it isn't ready for you yet.

1. What evidence are you relying on for metformin and aging in someone like me, without diabetes?
2. Which labs do you check before and after I start, and are they in the price?
3. Immediate-release or extended-release, and from which manufacturer?
4. How should I think about metformin if I train hard or I'm trying to build muscle?
5. What does the monthly fee cover beyond the prescription?

Metformin next to the GLP-1 drugs: a different tool for metabolic goals

People weighing metformin for metabolic health often also ask about semaglutide and tirzepatide. They're different drugs with different evidence. Metformin is a daily pill that steadies blood sugar; the GLP-1 drugs are mostly weekly shots that cut appetite and weight.

Who metformin fits, and who should look elsewhere

• Strongest case if you have type 2 diabetes or prediabetes. That's where the trials are.
• Worth a careful talk if you're healthy and want it for aging. Clinical aging-marker work is limited and population-specific; lifespan extension remains unestablished.
• Discuss risks first if you have impaired kidney function, heavy alcohol use, liver disease, acute hypoxia or an exercise-focused goal. Severe renal impairment is a contraindication; stable heart failure or any kidney diagnosis is not automatically the same condition.
• If weight is the goal, the approved weight drugs are a different class. See semaglutide and tirzepatide.

Looking at other longevity compounds? NAD+ and NMN come up next to metformin a lot, with their own evidence gaps; does NMN work is the quick version.

Ready to compare programs? Start with the metabolic shortlist, and see how picks work. Not sure which goal fits? Take the quiz.