Tirzepatide vs semaglutide: what SURMOUNT-5 actually found

The study in one paragraph

SURMOUNT-5 randomized 751 adults without diabetes who had obesity, or overweight with a qualifying weight-related condition. This 72-week, open-label trial compared maximum tolerated once-weekly injectable tirzepatide with injectable semaglutide. The studied doses were 10 or 15 mg for tirzepatide and 1.7 or 2.4 mg for semaglutide; these describe the research, not instructions for treatment. Eli Lilly funded the trial. Open-label means the assigned treatment was known rather than concealed.

Source: Aronne et al.: SURMOUNT-5, New England Journal of Medicine, 2025

The result and its uncertainty

The reported estimated mean weight change was −20.2% with tirzepatide and −13.7% with semaglutide. The corresponding 95% confidence intervals were −21.4% to −19.1% and −14.9% to −12.6%. Gastrointestinal events were the most common adverse events in both groups, mostly mild or moderate. These results support greater average weight reduction with tirzepatide under the trial conditions; they do not promise the same outcome for each participant or future patient.

Source: Aronne et al.: SURMOUNT-5, New England Journal of Medicine, 2025

Percentage points are the clearer comparison

Subtracting the reported averages gives a 6.5-percentage-point difference. Dividing that difference by 13.7 produces approximately 47% greater relative weight reduction. That second number does not mean losing 47% of starting body weight. Keeping both the original group results and the absolute difference visible makes the headline easier to interpret. Neither calculation creates a personal forecast. A consumer comparison should show what the percentage measures and the time period, rather than repeating the largest-looking figure alone.

Why a direct comparison is useful

NIH explains that random assignment helps researchers assess cause and effect. Comparing two treatments inside one trial avoids some problems involved in placing unrelated studies side by side, such as different participants and follow-up periods. Still, the question answered depends on the trial design. This result should not be silently extended to every formulation, patient population or care setting. Bring the study to your clinician as a question about relevant evidence, not as an instruction to request a particular prescription.

Source: NIH: understanding clinical studies

Do not transfer the result to every offer

The branded injectable products studied are not interchangeable evidence for a compounded preparation or another formulation sold through a website. FDA states that compounded drugs are not FDA approved. When comparing a commercial offer, identify the exact product and ask what evidence supports it. A shared ingredient name is not enough to conclude that its outcomes match this trial. Likewise, no part of the result measures a telehealth brand’s billing transparency or quality of customer support.

Source: FDA: unapproved GLP-1 products

Turn research into better care questions

Ask your clinician how the study population compares with your circumstances, which benefits and risks matter to you, and what follow-up would be needed. Then compare providers on a separate set of questions: access to a clinician, pharmacy information, insurance support, total cost and renewal terms. The related reviews can help with those service questions. They do not select treatment or convert the trial average into a guarantee. Keep your research notes and commercial shortlist connected while preserving that distinction.

Explore relevant provider reviews

Use these reviews to compare the published services and terms. Inclusion is a research starting point, not a recommendation for treatment.

Continue your research

• How to read GLP-1 research without overreading a headline
• Weight regain after stopping GLP-1 treatment: what two studies show
• GLP-1 weight-loss plateaus: what the trial timelines mean