The evidence
GLP-1 weight-management medicines are among the most heavily trialled treatments in the field. Here is what the published studies found — and, just as plainly, what they didn't.
How it works
GLP-1 is a hormone your gut releases after you eat. It is one of the signals that tells your brain you've had enough. These medicines are engineered versions of it, designed to last days rather than minutes.
A once-weekly dose under the skin, started low and increased in steps over several weeks. The step-up (titration) exists to let your body adjust — it is the main reason a prescriber stays involved rather than handing over a script and walking away.
Where your own GLP-1 is broken down within minutes, these medicines persist for days. Food empties from the stomach more slowly and the appetite signalling stays switched on between meals.
Most people describe it as less hunger and far less mental noise about food — being able to leave half of something, or not think about the pantry at 9pm. Eating less follows from that, rather than from trying harder.
Published human trials
These are large randomised controlled trials, published in the New England Journal of Medicine. In every one, the medicine was given alongside diet and activity support — not instead of it. That detail matters, and it is why our membership exists.
A randomised, double-blind, placebo-controlled trial in 1,961 adults with overweight or obesity, all receiving lifestyle intervention. In this clinical study, mean body-weight change was −14.9% on semaglutide against −2.4% on placebo at week 68.
Honest note: these are group averages. The spread around them is wide — some participants lost a great deal more, and some lost little. An average is not a forecast for you.
Wilding JPH et al., New England Journal of Medicine 2021;384:989–1002 (STEP 1).
A randomised, double-blind, placebo-controlled trial in 2,539 adults, again with lifestyle support throughout. In this clinical study, mean body-weight change on the 15 mg dose was −20.9% against −3.1% on placebo at week 72. Lower doses produced smaller average reductions.
Honest note: the two trials above ran in different populations at different times, so the headline percentages are not a like-for-like race between the medicines. Which one suits you — if either does — is a prescriber's judgement, not a league table.
Jastreboff AM et al., New England Journal of Medicine 2022;387:205–216 (SURMOUNT-1).
The same programmes tested what happens when treatment comes off. In STEP 4, participants who continued treatment after week 20 lost further weight, while those switched to placebo regained weight over the following 48 weeks. In the STEP 1 extension, participants who came off treatment regained around two-thirds of the weight they had lost within a year.
Why we lead with this: it means a script on its own is a temporary result. What holds is the habit, the food, the tracking and the support built around it — which is the part of this that isn't a medicine, and the part a membership can actually do.
Rubino D et al., JAMA 2021;325:1414–1425 (STEP 4). Wilding JPH et al., Diabetes, Obesity and Metabolism 2022;24:1553–1564 (STEP 1 extension).
Read before you start
Both columns matter. We would rather you decide on what has actually been demonstrated than on what an advertisement implies.
| What the evidence shows | What it doesn't |
|---|---|
| Shown: large average reductions in body weight over 68–72 weeks in big randomised trials, against placebo. | Not shown: that you will get the average. Trial results are group means with a wide spread, and some participants respond very little. |
| Shown: the effect holds while treatment continues, and continued treatment produced further loss in STEP 4. | Not shown: that it lasts after stopping. The withdrawal studies point the other way — most of the loss came back. |
| Shown: every trial delivered the medicine with diet and activity support. That is the tested combination. | Not shown: that the medicine works as well on its own. No large trial tested it without lifestyle support. |
| Shown: a well-characterised side-effect profile — gastrointestinal effects such as nausea, vomiting, diarrhoea and constipation are the most common, usually mild to moderate, mostly early and dose-related. | Not shown: that it is suitable for everyone. There are real contraindications and cautions, and serious adverse effects occur. Only a prescriber can weigh them for you. |
| Shown: Medsafe has approved these medicines for chronic weight management in New Zealand. | Not the same thing: approved is not funded. Pharmac does not fund them for weight management, so they are paid for privately. |
When new trial evidence is published — including evidence that cuts against this — we'll report it here.⚠ Pre-launch: every figure on this page must be checked against the source papers by the medical director before launch, and the whole page TAPS/ASA pre-vetted to the Therapeutic & Health Advertising Code (new Code from 1 April 2026) before any paid promotion. Keep the trial name, duration, "mean/average" framing and the placebo comparator attached to every number. The side-effect row is a summary, not a substitute for the consumer medicine information.
A three-minute check, then a free first consult. A prescriber decides whether treatment is appropriate — never a salesperson, and never on the strength of an average.
Check your eligibility →First consult free · No charge unless treatment is prescribed