Who took part and what the trial tested
The study enrolled 1,961 adults and assigned them two to one to semaglutide or placebo for 68 weeks. To join, people needed obesity, or overweight with a weight-related condition. Both groups had a lifestyle programme. The semaglutide dose went up in steps. These details matter because the result reflects a structured treatment programme. Nobody started straight on the maintenance dose.
The headline finding
Mean weight change at 68 weeks was −14.9% with semaglutide and −2.4% with placebo under the treatment-policy analysis. The reported estimated treatment difference was −12.4 percentage points, based on the trial's statistical model. If subtracting the rounded headline numbers gives a slightly different figure, rounding or the model explains the gap. Quote the published estimate as reported.
Why you may see another percentage
A trial report can answer more than one statistical question. An analysis of outcomes if everyone took treatment as intended gives a different figure from one that allows for people stopping and other events. Quote neither without its context. When two websites give different numbers, check whether they use the same estimand, time point and population before deciding one source is wrong.
STEP 1 covers one semaglutide regimen
STEP 1 did not test an Ozempic diabetes regimen, a Wegovy tablet or a 7.2 mg injection. The trial also excluded people with diabetes. Keep those limits attached to the result whenever you reuse it. The shared active ingredient links related research, but differences in form, exposure, disease and treatment goals remain. An accurate summary names the product and setting instead of saying only "semaglutide".
Averages and your own experience
An average marks the middle of a spread of results. Some participants lost more, some lost less and some stopped treatment. Losing less than the average does not mean you used the medicine wrongly. Clinical decisions also weigh side effects, nutrition, daily function and changes in conditions such as blood pressure. The study points to a benefit across the population. Your own response still needs review.
Questions to ask after reading STEP 1
Ask whether the study population resembles you, which dose your prescriber proposes and how they will judge success. Discuss what happens if you cannot tolerate treatment or stop later. The published result supports an informed conversation. The trial gives no dose-escalation instruction and guarantees no target weight. Keep evidence on long-term maintenance apart from STEP 1's main 68-week endpoint.
Check who counts in a success rate
The share of participants who reached a weight-loss threshold can look impressive. What matters is who sits in the denominator and how the researchers handled missing measurements. Did the estimate include everyone assigned to treatment, or only people measured at the end? The methods section of the publication answers this. Asking the question does not undermine the trial. The answer tells you what the figure measures. Never mix a completers-only success rate with a whole-population average as if both described the same group.
Sources
Your own medicine leaflet and prescriber come first. A source from another country does not mean the same rule applies in the UK.
- STEP 1: original trial publication nejm.org
- Wegovy injection: UK prescribing information medicines.org.uk
Spotted something out of date? Report a mistake.