Does Metformin Cause Weight Loss? What the Trials Show
Key takeaways
- Metformin produces modest weight loss, averaging about 2.1 kilograms over 2.8 years in the Diabetes Prevention Program.
- Intensive lifestyle change outperformed it in that trial, at 5.6 kilograms, and reduced diabetes risk by 58 percent against metformin's 31 percent.
- It works mainly by reducing liver glucose production and improving insulin sensitivity rather than by suppressing appetite.
- Long-term follow-up suggests responders keep the weight off better, though that analysis was post hoc and weaker evidence.
- GLP-1 medications produce around 15 to 20 percent weight loss, so metformin is not a comparable alternative for that purpose.
Metformin occupies an odd position. It is not licensed as a weight-loss drug, it is not marketed as one, and yet it is one of the most commonly prescribed medications in the world and people reliably lose a little weight on it. That gap between what it is for and what it does has produced a lot of confusion, some unrealistic expectations and a steady stream of off-label prescribing. Here is what the trial evidence actually supports.
Does metformin cause weight loss: the short answer
Yes, but the amount is small. In the Diabetes Prevention Program, the largest and most-cited trial in this area, participants taking metformin lost an average of about 2.1 kilograms over roughly 2.8 years, compared with 5.6 kilograms in the intensive lifestyle group and 0.1 kilograms on placebo. After one year, around 28.5 percent of the metformin group had lost at least 5 percent of their body weight, against 62.6 percent in the lifestyle group.
Those numbers are worth sitting with, because they are far from the figures people have become used to hearing. Metformin produces a modest effect that helps rather than transforms. Its real achievement in that trial was different: it reduced the development of type 2 diabetes in high-risk adults by 31 percent, while the lifestyle programme reduced it by 58 percent. The weight loss was a secondary finding, not the headline.
That distinction gets lost constantly, including in advertising for telehealth services that offer metformin for weight. The drug earns its place in medicine by preventing and managing diabetes cheaply and safely over decades of use. Judging it as a weight-loss medication is judging it on the thing it does least well.
Why the weight comes off at all
Metformin was never designed to suppress appetite, and it does not work the way the newer injectable medications do. Its primary action is to reduce the amount of glucose the liver produces and to improve how sensitive your tissues are to insulin. Lower circulating insulin means less signal to store fat, which is part of the explanation.
There are secondary contributors. Many people eat somewhat less on metformin, partly because of mild nausea and reduced appetite in the early weeks. The drug also appears to influence gut hormones and the gut microbiome, and research has looked at effects on GDF-15, a hormone involved in appetite regulation. None of these effects is powerful individually, which is consistent with the modest results seen in trials. It is a gentle nudge to metabolism rather than an override of hunger.
The difference matters for expectations. Medications that act on appetite directly produce an obvious subjective change: people notice they are no longer hungry. Metformin generally does not feel like anything once the first weeks have passed, which is why people often assume it is doing nothing. The effect is real but quiet, showing up on the scale over months rather than in how you feel at the dinner table.
What the long-term data suggests
One genuinely interesting finding concerns durability rather than magnitude. In a follow-up analysis of Diabetes Prevention Program participants who had lost at least 5 percent of their weight in the first year, the average loss maintained between years 6 and 15 was 6.2 percent in the metformin group, compared with 3.7 percent in the lifestyle group and 2.8 percent on placebo.
In other words, the people who responded to metformin tended to keep what they lost, while the lifestyle group regained more over time. That is an appealing result, but it needs a caveat stated plainly: this was a post hoc analysis of non-randomised subsets, so it is weaker evidence than the trial's main findings. It supports the idea that metformin's small effect is durable, and it does not establish that metformin is a good weight-loss strategy for people without a metabolic reason to take it.
There is a plausible explanation for the pattern, which is simply that a tablet is easier to keep taking than a programme of diet and exercise is to sustain for fifteen years. If that is the mechanism, it says more about adherence than about the drug, and it is an argument for pairing medication with habits rather than choosing between them.
Who tends to respond, and who does not
Response varies considerably, and a few patterns show up consistently.
- People with insulin resistance or prediabetes tend to see more benefit than those without.
- People with polycystic ovary syndrome are commonly prescribed it, and it may help weight and other markers alongside its main effects.
- People with a higher starting weight often lose more in absolute terms.
- People on antipsychotic medication that causes weight gain have shown benefit in trials.
- People without insulin resistance typically see very little.
- Nobody should expect the sort of results associated with GLP-1 medications.
This is also why metformin is sometimes prescribed off-label for weight in people who are not diabetic, a practice that sits in a grey area. It is cheap and well understood, which makes it attractive, but the expected result is a few kilograms rather than a transformation, and that should be said out loud before anyone starts.
Cost is part of why the practice persists. Metformin is one of the cheapest prescription medications available, with decades of safety data behind it, so the risk-to-cost calculation looks favourable even for a small benefit. That is a reasonable argument as far as it goes, provided nobody is told to expect results the evidence does not support.
Metformin compared with the newer medications
The contrast is stark. Semaglutide and tirzepatide produce average weight loss in the region of 15 to 20 percent of body weight in trials, while metformin produces roughly 2 to 3 percent. They are not in the same category and it is misleading to present them as alternatives to each other for weight purposes.
They are, however, frequently used together, and that combination is standard practice in type 2 diabetes rather than an experiment. Metformin remains first-line because it is cheap, long-studied and safe, with a newer agent added when more is needed. If that is your situation, our guide to taking metformin and Ozempic together covers how the pairing works. For a cheaper comparison with the supplement often marketed as a natural equivalent, we looked at berberine and blood sugar separately.
When to see a doctor
Talk to a doctor before seeking metformin for weight alone. If you have prediabetes, polycystic ovary syndrome or clear insulin resistance, there may be a sensible case for it, and that case is about metabolic health with weight as a bonus. If you are looking primarily for weight loss and have none of those, it is worth discussing what the realistic options are rather than starting a medication with a small expected effect.
Contact your prescriber if you lose weight rapidly or unintentionally on metformin, since that is not the expected pattern and deserves investigation. Report persistent digestive symptoms, and ask about vitamin B12 testing if you have been taking it for several years. It is also worth asking what the realistic target is for you specifically, since a few kilograms maintained over years is a genuinely useful outcome even though it is not the one most people arrive hoping for, and naming that expectation early prevents an otherwise sensible treatment being abandoned as a disappointment within a couple of months. This is general information rather than medical advice, and prescribing decisions belong with a clinician who knows your history.
Frequently asked questions
How much weight can you lose on metformin?
Trial averages are modest, around 2 to 3 percent of body weight. In the Diabetes Prevention Program the mean loss was about 2.1 kilograms over 2.8 years.
How does metformin cause weight loss?
Mainly by reducing glucose production in the liver and improving insulin sensitivity, with smaller contributions from mild appetite reduction and effects on gut hormones.
Can you take metformin just for weight loss?
It is sometimes prescribed off-label, particularly with insulin resistance or polycystic ovary syndrome, but the expected effect is a few kilograms rather than a transformation.
Is metformin as good as Ozempic for weight loss?
No. GLP-1 medications produce around 15 to 20 percent weight loss in trials against roughly 2 to 3 percent for metformin. They are often used together rather than as alternatives.
Does metformin weight loss last?
Follow-up data suggests responders maintained their loss better than the lifestyle group over 15 years, though that analysis was post hoc and should be treated cautiously.
Read next
- Metformin Side Effects: What Is Common and What Is Serious
What to expect when you start it.
- How to Reverse Prediabetes With Diet: What the Evidence Shows
The lifestyle side that outperformed the drug.
- Berberine for Blood Sugar: What the Evidence Actually Shows
The supplement marketed as a natural version.
This article is for general informational purposes only and is not medical advice. Always consult a qualified healthcare professional about your individual situation, and never start or stop a medication without your doctor. See our Medical Disclaimer.