← scts1.com
Health Guide

Metformin for weight loss: what the trials actually show

Metformina para bajar de peso: lo que muestran los ensayos

A lit antique oil lantern beside a modern lamp on a wooden table: a 1960s medicine that is still in use

It's probably the most-prescribed medication in the world for type 2 diabetes. And for years, a lot of people have been asking for it for a different reason: weight loss.

It doesn't have the marketing budget of a GLP-1. It doesn't trend on social media. It costs a fraction of the price. And yet there are decades of real studies behind metformin — with a message that's less dramatic than the internet suggests, but more honest.

Here are the numbers, exactly as they appear in the trials and in the FDA's own label. Not overstating what it does, and not underselling what it doesn't.

The essentials in 30 seconds

Educational content. Not medical advice. Results vary.

What you'll read

What metformin is, and why it's used off-label

Metformin has been approved in the United States since 1995. It's a biguanide, and its main job is lowering glucose: it reduces how much the liver produces and improves how the body responds to insulin.

It was never designed or approved as a weight-loss drug. But from the earliest trials in people with diabetes, doctors noticed something: people taking it tended to lose some weight, or at least gain far less than they did on other diabetes medications.

That observation opened the door to decades of off-label use — for prediabetes, for polycystic ovary syndrome, for weight gain caused by certain antipsychotics, and for people with insulin resistance who want to lose weight without having diabetes.

"Off-label" doesn't mean illegal or poorly studied. It means the FDA never specifically reviewed or approved that particular use. Physicians can still prescribe it that way, using clinical judgment, but the official label doesn't back that indication.

The mechanism: not just about glucose

For years, the assumption was that if metformin helped with weight, it was simply because it improved glucose and insulin — less insulin circulating, less signal to store fat. That part is real, but it isn't the whole story.

More recent research identified a second, more direct pathway: metformin activates an enzyme called AMPK — an energy sensor present in nearly every cell — and that activation increases production of a hormone called GDF-15.

GDF-15 acts on specific receptors in the brainstem, in a region tied to nausea and appetite. When it rises, appetite drops. It's the same broad family of mechanisms — though not the same receptor — that's part of how GLP-1 medications work.

A placebo-controlled trial in children with obesity, treated with metformin for 24 months, confirmed this in real people: those with the biggest rises in blood GDF-15 were also the ones who lost the most weight and the most visceral fat.

AMPK → GDF-15
the pathway that links metformin to lower appetite, not just lower glucose

The trial that changed the conversation: the Diabetes Prevention Program

If there's one trial everyone cites about metformin, it's this one. It was called the Diabetes Prevention Program (DPP), published in 2002, and it's still a required reference more than twenty years later.

3,234 people without diabetes but with elevated glucose — meaning high risk — were randomly split into three groups: placebo, metformin, or a lifestyle-change program with the goal of losing at least 7% of body weight and doing 150 minutes of physical activity a week.

The result, after an average follow-up of 2.8 years:

31% vs. 58%
reduction in diabetes incidence with metformin versus the reduction with lifestyle change, both compared with placebo

Metformin worked. It cut new diabetes cases by 31% compared with placebo. But lifestyle change — diet and movement, no medication at all — was nearly twice as effective: 58%.

That's an inconvenient fact for any narrative that sells metformin as a shortcut. The trial's own conclusion is direct about it: the lifestyle intervention was significantly more effective than the drug.

How much weight actually comes off?

This is where strict clinical trials and real-world practice need to be kept separate, because the numbers don't always match.

In an outpatient study — not a randomized trial, but a real follow-up of 154 patients over 6 months, on doses up to 2,500 mg a day — average weight loss was 5.8 ± 7.0 kg, or 5.6%. The untreated comparison group gained 0.8 kg over the same period.

5.6%
average weight loss over 6 months in a real-world outpatient follow-up (not everyone responds the same)

That same study found something worth noting: people with more insulin resistance lost significantly more weight than people with normal insulin sensitivity. Metformin appears to work best in exactly the people who need it most for metabolic reasons.

A more recent trial, in 2025, compared metformin alone against metformin plus semaglutide in 100 overweight or obese women with polycystic ovary syndrome, over 16 weeks. Metformin alone produced a loss of 2.25 ± 4.27 kg. The combination with semaglutide lost 6.09 ± 3.34 kg — nearly three times as much.

That comparison sums up where metformin sits today: real, measurable, and modest next to the newer options.

The side effects, exactly as the FDA reports them

This is the part almost nobody mentions with actual numbers. Here it is, straight from the approved label.

53.2%
of people on immediate-release metformin had diarrhea, versus 11.7% on placebo

The FDA describes these symptoms as dose-dependent, transient, and generally resolving on their own with continued use. In practice, starting at a low dose and titrating up slowly — and using the extended-release version where possible — cuts the problem down substantially.

Vitamin B12: the quiet long-term effect

This is the effect almost never mentioned, and it only shows up with prolonged use.

A placebo-controlled randomized trial in 390 people with diabetes, treated for 4.3 years, found metformin lowered vitamin B12 concentration by an average of 19% compared with placebo.

+7.2 points
absolute risk increase for B12 deficiency with metformin versus placebo, after 4.3 years

In practical terms: for every 14 people treated for 4.3 years, one developed B12 deficiency they wouldn't have had on placebo. Untreated B12 deficiency can affect the nerves and raise homocysteine, a marker tied to cardiovascular risk.

It's preventable. Any responsible long-term metformin plan includes checking B12 levels periodically — this isn't an effect that should surprise anyone if it's monitored from the start.

Who shouldn't take it? Kidneys and lactic acidosis

The best-known warning about metformin is lactic acidosis — a dangerous buildup of lactic acid in the blood. It's real, but it's also far rarer than its reputation suggests.

A systematic review published in JAMA in 2014, covering 65 studies, found the actual incidence of lactic acidosis in people taking metformin is 3 to 10 cases per 100,000 person-years — essentially indistinguishable from the background rate in people with diabetes who aren't on metformin.

The risk rises in people with significantly reduced kidney function, because the body clears metformin through the kidneys. That's why kidney function is checked before starting and periodically afterward, and why the dose is adjusted or the drug avoided entirely when the filtration rate is too low.

Metformin versus GLP-1: why the gap is so large

The comparison is inevitable, so it's worth making with numbers instead of opinions.

Metformin, in the best-documented scenarios, moves the needle by 2 to 6 kilograms over several months. GLP-1 medications like semaglutide or liraglutide, in comparable trials, produce weight loss of 6% to 15% or more of body weight — several times larger.

That gap isn't a flaw in metformin. It's acting on a different, more limited metabolic pathway. What metformin does offer — and what GLP-1s don't match — is more than two decades of use, an extremely well-studied safety profile, low cost, and an oral pill instead of an injection.

That's why, in clinical practice, it's rarely an either-or decision. It's usually part of a broader plan, and in some cases it's used alongside a GLP-1 — not instead of one — to address both mechanisms at once.

What this evidence does NOT say

Who is this NOT for?

Frequently asked questions

Does metformin work for weight loss if I don't have diabetes?

It's used that way off-label, and there's evidence it can help — especially in people with insulin resistance — but no regulatory agency has approved it specifically for that purpose. The decision belongs to a physician with your full history.

How long does it take to see results?

In the outpatient study cited here, weight loss was measured at 6 months. It isn't a days-or-weeks effect; it's a gradual metabolic change.

Does the diarrhea always happen?

Not always, but it's the most common effect, especially with the immediate-release version and when starting out. The extended-release version, taken with food and titrated up gradually, cuts the problem down substantially for most people.

Is it dangerous for the kidneys?

In people with normal or mildly reduced kidney function, no. The real lactic acidosis risk shows up with significantly impaired kidney function, which is why it's checked before and during treatment.

Can I combine it with a GLP-1?

It's a combination used in clinical practice and one that's been studied — including the 2025 trial cited here, where the combination substantially outperformed metformin alone. That decision also depends on an individual medical evaluation.

How we handle this at SCTS1

Metformin is a real, inexpensive, well-studied tool — but a modest one if the goal is significant weight loss on its own. Our job is to help you understand where it fits, not to sell it as something it isn't.

Once you start your program, a licensed physician reviews your case — habits, stress, sleep and medical history — and orders labs only if they consider it necessary, including kidney function and B12 when appropriate.

We offer clinician-supervised hormone health and weight management programs, and GLP-1 programs with FDA-approved medications (Wegovy®, Foundayo™, Ozempic®) from $149 per month, by telemedicine, in Spanish and English.

Eligibility is determined by medical evaluation. Individual results vary and are not guaranteed.

Sources

Articles located through PubMed.

Related articles

Your trust is worth more than a sale.

More information →

This article is informational and is not medical advice. SCTS1 does not prescribe medications. All medical care, prescriptions and treatment plans are provided by licensed physicians through our partner platform. Eligibility is determined by a licensed provider. Metformin is FDA-approved only for type 2 diabetes; its use for weight loss is off-label everywhere and is mentioned here for informational purposes only. SCTS1 does not offer or prescribe it as a standalone weight-loss treatment. Individual results vary and are not guaranteed.