Bariatric surgery still beats GLP-1 drugs on weight loss, 44,000 patients show
Gastric bypass patients lost roughly 28% of body weight at three years, against 7-11% for tirzepatide or semaglutide — even among patients who stayed on the drugs continuously for at least a year.
| Group | Value (%) |
|---|---|
| Gastric bypass | 28.4 |
| Sleeve gastrectomy | 22 |
| Semaglutide | 7.4 |
Clinical trials have established how much weight GLP-1 drugs and bariatric surgery each produce separately, but direct, real-world comparisons between the two — as they're actually used, with real-world adherence patterns — have been limited. A study published this month in Obesity compares them directly, using electronic health records from two urban health systems [s1].
The design
This retrospective study included adults with BMI 35 or higher who received injectable semaglutide, injectable tirzepatide, sleeve gastrectomy, or gastric bypass at two urban health systems between 2018 and 2024 [s1]. Total weight loss was compared up to three years after treatment initiation, using inverse probability weighting and mixed linear models to help balance differences between groups [s1]. Two analytic approaches were used: an intention-to-treat analysis (counting anyone who received any GLP-1 receptor agonist order, regardless of how long they continued) and a per-protocol analysis (restricted to patients with at least one year of continuous GLP-1 drug orders, a stricter standard meant to better reflect people who actually stayed on treatment) [s1].
What it found
The study included 44,025 patients total [s1]. In the intention-to-treat analysis, total weight loss at one, two, and three years was: semaglutide (n=25,804), 5.4%, 6.5%, and 7.4%; tirzepatide (n=7,308), 9.1% and 10.8% (two available timepoints); sleeve gastrectomy (n=8,728), 24.4%, 22.4%, and 22.0%; and gastric bypass (n=2,185), 29.8%, 28.1%, and 28.4% [s1]. Bariatric surgery was associated with significantly greater weight loss than either GLP-1 drug at every timepoint measured [s1].
Restricting to patients with at least one year of continuous GLP-1 drug use — the per-protocol analysis, meant to isolate people who actually stayed on treatment as intended — narrowed but didn't close the gap: semaglutide total weight loss rose to 7.2%, 8.0%, and 8.8% at one, two, and three years; tirzepatide rose to 11.7% and 11.9% [s1]. Both remained well below the roughly 22–29% weight loss maintained by either surgical procedure across the same multi-year window [s1].
Why the surgery-versus-drug gap is larger here than headline trial numbers might suggest
Clinical trials of tirzepatide and semaglutide have reported considerably higher weight-loss percentages than what this real-world data shows — a gap that reflects a familiar and important distinction between trial efficacy and real-world effectiveness. Trial participants are closely monitored, dose-escalated on a strict protocol, and often more adherent than a general patient population; this study's own comparison between its intention-to-treat and per-protocol analyses illustrates the effect directly, since even restricting to patients continuously on the drug for a full year still produced considerably lower weight loss than trial-reported figures, let alone the intention-to-treat numbers that include people who stopped or never escalated to a full dose.
What this doesn't establish
This is a retrospective, two-center observational study, not a randomized trial — patients weren't randomly assigned to surgery or a GLP-1 drug, and the populations who choose or are selected for bariatric surgery versus a GLP-1 prescription likely differ in ways beyond what statistical weighting alone can fully correct for, including surgical eligibility, personal preference, and clinical risk factors. The study measures total weight loss specifically, not other outcomes that matter to patients and clinicians, such as cardiovascular events, diabetes remission, quality of life, or complication rates — all of which factor into real treatment decisions and none of which this analysis addresses. The drugs studied were injectable formulations only; oral GLP-1 options weren't part of this comparison, and this data predates or excludes some newer, higher-dose drug regimens.
What the authors conclude
The study's authors describe their finding plainly: among patients eligible for both options, bariatric surgery was associated with greater weight loss than GLP-1 receptor agonists [s1] — a real-world confirmation of a pattern that's been anticipated from trial data, now quantified with routine clinical adherence patterns factored in rather than trial-protocol adherence.
What to watch
Whether newer, higher-dose GLP-1 and dual-agonist drugs narrow this gap in future real-world comparisons, and whether outcome measures beyond weight loss alone — durability, complications, cardiovascular and metabolic outcomes — are compared with similar real-world rigor. This article is not medical advice; the choice between bariatric surgery and pharmacologic treatment involves considerations beyond weight loss alone and should be discussed with a clinician.
Sources
- Real-World Effectiveness of Semaglutide and Tirzepatide Compared With Bariatric Surgery — Obesity, 25 June 2026
Sources
- Real-World Effectiveness of Semaglutide and Tirzepatide Compared With Bariatric Surgery — Obesity , June 25, 2026
More on
A real-world comparison finds tirzepatide beating semaglutide and liraglutide
A single-clinic retrospective study in Istanbul followed patients on all three drugs for 36 weeks. The ranking matches what randomized trials have shown — with liraglutide the least tolerated of the three.
Weight loss modifies some diseases and is untested in most of the ones it is blamed for
A Lancet review triangulates the evidence linking adiposity to multisystem disease and finds robust trial support in a short list — while an expert panel argues the drugs must be lifelong.
In 160,000 nondiabetic adults, GLP-1 users had 41% lower obesity-linked cancer risk
The association held across most demographic subgroups, but not among Black patients, a gap the study doesn't explain. Two-year follow-up limits what it can say about longer-term risk.
Two large real-world studies find little separating tirzepatide and semaglutide
Head-to-head in US insurance data, the two drugs showed comparable cardiovascular outcomes and comparable rates of severe gastrointestinal events in type 2 diabetes. Neither study was a trial.