What Does “15% Weight Loss” Mean? How to Read Study Claims

First, translate a percentage into ordinary arithmetic

A 15% change means fifteen out of every hundred units of starting weight. In an illustrative example, 15% of 200 lb is 30 lb, while 15% of 240 lb is 36 lb. Those are calculations, not treatment predictions or recommended targets.

An average also does not describe every participant. For a simplified illustration, losses of 5%, 15%, and 25% average to 15%, even though only one of the three people has that result. This invented example demonstrates why the average alone cannot tell you the range of outcomes.

A real example: the STEP 1 trial

STEP 1 studied 1,961 adults with overweight or obesity who did not have diabetes. Participants received weekly injected semaglutide, targeting 2.4 mg, or placebo, alongside lifestyle intervention, over 68 weeks. The published mean weight change was −14.9% with semaglutide and −2.4% with placebo.

These are results for the study’s treatment and population at its specified follow-up point. They are not a first-month expectation, a guarantee, or a measurement of how well a telehealth provider performs. The trial also reported adverse events and treatment discontinuations, which belong in a benefit–risk discussion.

Sources: Wilding et al., NEJM: STEP 1 trial

Use this table when a website quotes a study

Keep a link to the actual study beside your notes. NIDDK recommends asking whether a weight-loss program has been studied and whether its findings have been published. A testimonial, a star rating, and a medication trial are different kinds of information.

Sources: NIDDK: Evaluating program evidence

Do not turn separate trials into a provider league table

Suppose one website cites a result from a 68-week study and another cites a different study with a different population and follow-up period. Placing the percentages side by side does not remove those differences. Before treating the larger number as evidence of superiority, ask whether the treatments were compared directly in an appropriate study.

A medication trial also does not test the service advertising it. Appointment access, pharmacy fulfillment, billing, and follow-up need their own evidence. A provider cannot demonstrate better clinical care simply by quoting a larger number from a drug study.

Check whether the advertised product matches the evidence

FDA warns against claims that compounded GLP-1 products are the same as approved drugs. The fact that a branded medicine has trial evidence does not establish the performance of a compounded formulation offered on the next page.

• Can the provider name and link the study supporting this claim?
• Does the study use the same product, formulation, and treatment schedule being discussed?
• Is the number an average, a subgroup result, or an ‘up to’ claim?
• What does the clinician think is relevant to my situation, including the risks and alternatives?

Sources: FDA: Unapproved GLP-1 products