ASMBS head-to-head study finds bariatric surgery far outperforms GLP-1s in the real world

Patients who had bariatric surgery lost about five times more weight over two years than patients prescribed weekly GLP-1 injections, according to a real-world study presented June 17, 2025, at the American Society for Metabolic and Bariatric Surgery (ASMBS) 2025 Annual Scientific Meeting in Washington, DC ^[1].

Researchers from NYU Langone Health and NYC Health + Hospitals reported that patients who had sleeve gastrectomy or Roux-en-Y gastric bypass lost an average of 58 pounds at two years, compared with 12 pounds for patients who had a GLP-1 prescription for at least six months — 24% total weight loss versus 4.7% ^[1]. Patients who stayed on continuous GLP-1 therapy for a full year did better, at 7% total weight loss, but still well below the surgery group ^[1].

What the study did

This was a retrospective comparative effectiveness study using electronic medical record data from NYU Langone Health and NYC Health + Hospitals ^[1]. It included patients with a body mass index (BMI) of at least 35 who either had sleeve gastrectomy or gastric bypass, or were prescribed injectable semaglutide or tirzepatide, between 2018 and 2024 ^[1]. Semaglutide is the active ingredient in Ozempic and Wegovy; tirzepatide is the active ingredient in Mounjaro and Zepbound ^[1].

After adjusting for age, BMI and co-morbidities using average treatment effect weighting, the researchers compared outcomes for 51,085 GLP-1 and surgical patients ^[1]. The work was supported by an NYU CTSA grant (KL2 TR001446) from the National Center for Advancing Translational Sciences at the National Institutes of Health ^[1].

The gap between these results and clinical trial results is the central point the authors raise. "Clinical trials show weight loss between 15% to 21% for GLP-1s, but this study suggests that weight loss in the real world is considerably lower even for patients who have active prescriptions for an entire year. We know as many as 70% of patients may discontinue treatment within one year," said lead author Avery Brown, MD, a surgical resident at NYU Langone Health ^[1]. "GLP-1 patients may need to adjust their expectations, adhere more closely to treatment or opt for metabolic and bariatric surgery to achieve desired results" ^[1].

The release cites a separate study finding that 53.6% of patients with overweight or obesity stopped GLP-1 therapy within one year, rising to 72.2% by two years ^[1]. About 12% of Americans say they have ever taken a GLP-1 drug, including 6% who say they are currently in treatment ^[1]. Meanwhile, more than 270,000 metabolic and bariatric procedures were performed in 2023 — roughly 1% of people who meet BMI-based eligibility requirements, according to ASMBS ^[1].

Why it matters for patients

The headline numbers are averages across a whole population of prescriptions, not a measure of what happens to someone who takes the medication consistently at a full dose. The study's own data show that distinction: 4.7% at two years for anyone with at least six months of prescriptions, versus 7% for those on continuous therapy for a year ^[1]. Stopping early — for side effects, cost, insurance changes or supply — pulls the average down.

That makes the finding less a verdict on how well the drugs can work and more a picture of how often real-world treatment gets interrupted. It is also worth knowing who presented it: the study came from bariatric surgeons and was released by the main US professional society for bariatric surgery ^[1]. ASMBS President Ann M. Rogers, MD, who was not involved in the study, said that "while both patient groups lose weight, metabolic and bariatric surgery is much more effective and durable," and that patients with insufficient weight loss or trouble staying on treatment "should consider bariatric surgery as an option or even in combination" ^[1].

Several things are not answered in the source material. The release does not break out results separately for semaglutide versus tirzepatide, does not report what doses patients reached, and does not describe complication rates or side effects in either group. Because this was an observational study rather than a randomized trial, statistical adjustment can reduce but not eliminate differences between people who choose surgery and people who are prescribed medication.

What happens next

The research team says future work will look at what providers can do to improve GLP-1 outcomes, which patients are better served by surgery versus medication, and what role out-of-pocket costs play in treatment success, according to senior author Karan R. Chhabra, MD, MSc, of NYU Grossman School of Medicine ^[1]. The findings were presented as a meeting abstract; the sources do not indicate whether or when a peer-reviewed publication will follow.