SCALE Diabetes quantifies the smaller weight result in type 2 diabetes

Results from the SCALE Diabetes trial, published in JAMA, put a number on something many people with type 2 diabetes notice when they start a GLP-1 medication for weight: the scale tends to move less than it does for people without diabetes. In the 846-person randomized trial, adults taking liraglutide 3.0 mg lost 6.0% of their body weight over 56 weeks, compared with 2.0% on placebo ^[1].

The trial ran at 126 sites in nine countries between June 2011 and January 2013 ^[1]. Of 1,361 people screened, 846 were randomized in a 2:1:1 split to once-daily subcutaneous liraglutide 3.0 mg (n=423), liraglutide 1.8 mg (n=211), or placebo (n=212) ^[1]. Everyone, including the placebo group, was also asked to follow a 500-calorie-per-day dietary deficit and get at least 150 minutes of physical activity per week ^[1]. To enroll, participants needed a body mass index of 27.0 or higher, an A1C between 7.0% and 10.0%, and had to be taking zero to three oral diabetes drugs — metformin, a thiazolidinedione, or a sulfonylurea ^[1]. The study was double-blind and included a 12-week observational period after the drug was stopped ^[1].

The numbers

Average starting weight was about 105.7 kg in the 3.0 mg group, 105.8 kg in the 1.8 mg group, and 106.5 kg on placebo ^[1]. At week 56, weight loss was 6.0% (6.4 kg) with liraglutide 3.0 mg, 4.7% (5.0 kg) with liraglutide 1.8 mg, and 2.0% (2.2 kg) with placebo ^[1]. The estimated treatment difference for the 3.0 mg dose versus placebo was −4.00% (95% CI, −5.10% to −2.90%); for 1.8 mg versus placebo it was −2.71% (95% CI, −4.00% to −1.42%), with P < .001 for both ^[1].

The trial had three co-primary endpoints, including the share of people crossing common weight-loss thresholds ^[1]. Loss of 5% or more of starting weight occurred in 54.3% on liraglutide 3.0 mg and 40.4% on 1.8 mg, versus 21.4% on placebo ^[1]. Loss of more than 10% occurred in 25.2% and 15.9%, versus 6.7% on placebo ^[1]. The difference for the 3.0 mg dose at the 10% threshold was 18.5 percentage points (95% CI, 12.7 to 24.4; P < .001) ^[1].

On safety, the authors reported more gastrointestinal disorders with liraglutide 3.0 mg than with either liraglutide 1.8 mg or placebo, and no cases of pancreatitis ^[1]. The published paper later carried an erratum in JAMA on January 5, 2016, titled "Incorrect Data in Tables and eFigure" ^[1]. The trial is registered as NCT01272232 ^[1].

Why it matters for patients

A roughly 6% average weight loss is real but modest, and it sits at the low end of the 5% to 10% range that the authors describe as enough to improve type 2 diabetes and related conditions ^[1]. The spread also matters: about half the people on the higher dose hit 5%, and about a quarter hit more than 10% ^[1]. That means averages hide a wide range of individual results, with some people losing much more and others losing little.

It is worth being precise about what this trial does and does not show. SCALE Diabetes tested liraglutide, not semaglutide (Ozempic, Wegovy, Rybelsus) or tirzepatide (Mounjaro, Zepbound), and it did not enroll a comparison group of people without diabetes ^[1]. So while the 6.0% figure is often contrasted with larger results seen in people without diabetes, this trial alone does not quantify that gap — that comparison is not in the source. The trial also ran only 56 weeks on drug, and the authors concluded that "further studies are needed to evaluate longer-term efficacy and safety" ^[1].

One more design detail shapes how to read the results: both groups received diet and activity counseling, so the 4-percentage-point difference reflects what the drug added on top of lifestyle changes, not the drug alone ^[1].

Any decision about whether these numbers are meaningful for a specific person, including how they compare with newer drugs or with the cost and side-effect tradeoffs, is a conversation for a clinician.