Zepbound vs Wegovy for the Heart: An Independent Study Found No Clear Winner

Share

Zepbound vs Wegovy is usually framed as a weight-loss contest, and on the scale tirzepatide tends to win. The harder question is the heart. Wegovy’s active ingredient, semaglutide, carries an FDA-approved indication for reducing cardiovascular risk in people with obesity or overweight and established cardiovascular disease. Tirzepatide, the active ingredient in Zepbound, does not have the same cardiovascular risk-reduction indication in that population.

A new independent analysis published November 10 in Diabetes, Obesity and Metabolism tried to determine whether that distinction shows up in real patients. Researchers followed insurance-claims records from 15,343 adults with established atherosclerotic cardiovascular disease who were starting one of the two medications. They found no statistically significant difference in the primary cardiovascular outcome.

That finding matters because a widely publicized study sponsored by Novo Nordisk, the maker of Wegovy, reported something different in 2025 using the same commercial data provider. The company highlighted a 57% lower risk of heart attack, stroke or death among semaglutide users who remained on treatment.

Two observational analyses, one data vendor and different results. The gap between them is a lesson in how much research design matters when comparing medicines outside a randomized trial.

What the Boston University Team Actually Did

The study, led by Gwen Aubrac of Boston University School of Public Health, with collaborators at Wake Forest University and McGill University, used a method epidemiologists call target trial emulation. Instead of assigning patients randomly to semaglutide or tirzepatide, the researchers used existing healthcare records to approximate a trial that has not yet been conducted in this population.

The analysis drew on national claims data from Komodo Health. Eligible participants had established atherosclerotic cardiovascular disease (ASCVD), a body mass index of at least 27 with a qualifying overweight or obesity diagnosis, and continuous insurance coverage before starting treatment. Patients with diabetes or previous use of GLP-1 drugs or other obesity medications were excluded.

The final study population included 6,817 new semaglutide users and 8,526 new tirzepatide users who began treatment between November 2023 and November 2025. Median follow-up was approximately 0.55 years, or about six and a half months.

The primary endpoint was revised five-point major adverse cardiovascular events, or MACE. It combined heart attack, stroke, hospitalization for heart failure or unstable angina, coronary revascularization procedures, and death from any cause. Secondary outcomes included a narrower three-point composite and the individual cardiovascular events.

Researchers used inverse probability of treatment weighting to balance measured differences between the two patient groups, then applied additional weighting to account for patients who discontinued or switched treatment. The aim was to reduce the bias that arises when people who receive one medication differ systematically from those receiving another.

The study was funded by the Robert Wood Johnson Foundation and the American Heart Association. No pharmaceutical-company funding was disclosed. One author reported consulting relationships with other companies outside the scope of the research.

Same data vendor, same broad cardiovascular question, different results. The research design is central to understanding why.

The Zepbound vs Wegovy Numbers, Read Carefully

There were 366 unadjusted primary-outcome events during follow-up. After statistical weighting, the hazard ratio comparing semaglutide with tirzepatide was 1.21, with a 95% confidence interval from 0.99 to 1.51.

The point estimate leaned against semaglutide. A hazard ratio above 1.0 means the estimated event hazard was higher in the semaglutide group than in the tirzepatide group. But the confidence interval included 1.0, so the comparison was not statistically significant at the conventional threshold.

That does not mean the medications are equally protective. The confidence interval remains compatible with no difference and with substantially greater cardiovascular risk among semaglutide users. Its lower boundary, 0.99, leaves little room for a semaglutide advantage under this particular analysis.

The narrower revised three-point MACE result was also inconclusive: hazard ratio 1.07, with a confidence interval of 0.75 to 1.55, based on 150 unadjusted events. Individual endpoints, including heart attack, stroke, hospitalization for heart failure or unstable angina, and coronary revascularization, were not statistically different between the two drugs.

The researchers reported too few deaths to estimate a separate mortality hazard ratio reliably. That is important because a headline about heart protection can sound broader than the event counts permit.

No statistically significant difference is not the same as proof that the two drugs protect the heart equally.

The authors’ conclusion was appropriately restrained: among adults with overweight or obesity and ASCVD without diabetes, the analysis found no significant difference in cardiovascular risk between semaglutide and tirzepatide.

Why Novo Nordisk’s STEER Study Told a Different Story

The contrast with STEER is what makes this new paper newsworthy. In August 2025, Novo Nordisk presented results from its own retrospective analysis of US Komodo data. STEER included 10,625 semaglutide users and 10,625 matched tirzepatide users, all at least 45 years old, with overweight or obesity, established cardiovascular disease and no diabetes.

In the analysis restricted to patients who remained on treatment without a gap exceeding 30 days, semaglutide was associated with a 57% lower risk of the revised three-point composite of heart attack, stroke or death. The result came from 15 events in the semaglutide group and 39 in the tirzepatide group.

In the broader analysis that included people regardless of treatment interruptions, the estimated reduction was 29%, with 56 events versus 83. For revised five-point MACE, STEER reported corresponding risk reductions of 43% in the treatment-persistence analysis and 22% in the broader comparison.

Novo presented these findings as evidence that semaglutide’s cardiovascular protection should not automatically be assumed for other GLP-1-based medications. The results were subsequently published in Diabetes, Obesity and Metabolism in January 2026.

But STEER was not randomized, and its most attention-grabbing result came from very few cardiovascular events. Percentages such as 57% sound definitive until the underlying counts come into view.

There is also a methodological problem with restricting analysis to patients who stay on treatment. Patients who tolerate a medication, continue filling prescriptions and remain under medical care may differ from those who discontinue. Those differences can influence cardiovascular outcomes independently of the drug itself.

That possibility does not invalidate STEER. It is one of the reasons observational findings require careful interpretation and, ideally, independent replication.

Fifteen events versus 39 can produce a striking percentage. It is still a small number of events on which to build a comparative claim.

Two Studies, One Database, Different Answers

The Boston University paper does not accuse Novo Nordisk or the STEER investigators of manipulating results. The authors instead identify methodological differences that could help explain the diverging estimates.

These include the study period, treatment definitions, length of the permitted medication gap, which confounding variables were included, how treatment groups were balanced, and how researchers adjusted for patients leaving the analysis.

The endpoints were not perfectly identical either. The independent study’s revised five-point MACE included hospitalization for unstable angina alongside heart failure, while STEER’s revised five-point definition did not include unstable angina.

Those details matter because a claims database is not a ready-made experiment. Researchers must decide which patients qualify, what counts as starting treatment, how long someone must be followed, how an event is recognized and what happens when a prescription is no longer filled.

In the new study, the researchers tested alternative assumptions about early cardiovascular events, off-label drug formulations, treatment discontinuation and outcome definitions. Point estimates generally remained in the same direction, but the analysis was sensitive to several of those choices.

One revealing difference concerned treatment persistence. The independent study recorded discontinuation rates of 31.0% among semaglutide users and 21.7% among tirzepatide users, with additional patients switching medications. These different patterns complicate any comparison that follows only patients remaining on their assigned drug.

The result is not a neat contest in which one study must be correct and the other false. Both are attempts to estimate treatment effects from imperfect real-world records. Their disagreement shows how strongly conclusions can depend on the assumptions used to turn those records into a comparison.

What This Study Cannot Tell You

The biggest limitation is that insurance claims are not randomization. A claims record can indicate that a prescription was filled and a cardiovascular diagnosis was billed. It cannot reliably establish whether a patient injected every dose, how much weight that patient lost, or what happened to smoking, diet, exercise and other major cardiovascular risk factors during treatment.

Inverse probability weighting can balance measured differences between groups. It cannot eliminate differences that were never recorded. The researchers acknowledge that they could not fully adjust for blood pressure, blood glucose, cholesterol measurements and behavioral risk factors.

The two treatment groups were also not identical before adjustment. Semaglutide users had a greater burden of some cardiovascular conditions, including hypertension, angina and heart failure. Tirzepatide users were more likely to have obstructive sleep apnea. Although the weighting process improved balance in recorded characteristics, unmeasured differences could remain.

Follow-up was short. Median observation lasted approximately six and a half months, while cardiovascular protection is often evaluated over several years. A longer study could accumulate more events, but it would also introduce additional complications involving treatment changes and adherence.

Drug dosing was another limitation. Only 26.1% of semaglutide users reached the 2.4 mg maximum weight-management dose, compared with 11.7% of tirzepatide users reaching 15 mg. The researchers could not establish from claims alone whether these differences reflected titration, tolerability or prescribing practices.

The results do not apply directly to patients with diabetes, who were excluded. Nor do they establish either medication’s absolute benefit over placebo. This was a comparison between two active treatments, not a test of taking medication versus taking nothing.

Grey Pepper has previously examined real-world cardiovascular comparisons of Ozempic and Mounjaro and cardiac MRI research involving semaglutide and heart failure. The new paper adds evidence specifically in people with overweight or obesity, existing atherosclerotic cardiovascular disease and no diabetes.

=== OPTIONAL SECOND INLINE IMAGE HERE — UPLOADED IMAGE 4: FEMALE PHYSICIAN DISCUSSING CLINICAL RESULTS WITH A MIDDLE-AGED WOMAN USING A TABLET IN A CONSULTATION ROOM. CAPTION: “Illustrative clinical consultation. Real-world cardiovascular comparisons must account for patient differences, treatment changes and incomplete medical records.” THIS IS NOT AN ACTUAL STUDY PARTICIPANT OR A DOCUMENTED CLINICAL VISIT FROM THE RESEARCH. ===

Zepbound vs Wegovy: The Evidence Still Missing

There is an important distinction between the evidence supporting semaglutide and the evidence supporting tirzepatide in this particular population.

Semaglutide’s cardiovascular indication rests on SELECT, a randomized, placebo-controlled outcomes trial that found a 20% relative reduction in major adverse cardiovascular events among adults with established cardiovascular disease and overweight or obesity who did not have diabetes. That evidence supported the FDA’s March 2024 label expansion.

Tirzepatide has demonstrated cardiovascular benefits in other settings, including obesity-related heart failure and studies involving patients with type 2 diabetes. But those results do not directly answer whether it prevents major atherosclerotic cardiovascular events in the same non-diabetic population studied in SELECT.

The cleanest comparative evidence would come from a randomized trial assigning otherwise similar patients to semaglutide or tirzepatide and following them long enough to assess cardiovascular outcomes. A dedicated tirzepatide outcomes trial against placebo in the relevant population could also establish a stronger independent evidence base for its cardiovascular effects.

Until that evidence is available, observational research will continue trying to answer the question with insurance records. Those studies can identify important signals, but they cannot fully remove the possibility that differences in patient selection, treatment patterns or outcome definitions influenced the result.

The new independent paper is valuable precisely because it does not reproduce the strong semaglutide advantage emphasized in Novo Nordisk’s earlier analysis. It provides another estimate from the same broad data environment and makes the methodological uncertainty more visible.

The honest headline is not that one drug wins. It is that the direct comparison remains unsettled.

For patients and clinicians, Wegovy’s randomized cardiovascular evidence still deserves greater weight than an observational comparison. For researchers and insurers, the new paper is a reminder not to confuse the absence of statistical significance with clinical equivalence, or a dramatic relative-risk percentage with proof of causation.

The Zepbound vs Wegovy cardiovascular question remains open. An independent team studied more than 15,000 real-world patients and found no statistically significant separation. That does not establish that the two medications are identical. It establishes that this analysis could not reliably determine a winner.

Semaglutide has randomized cardiovascular outcomes evidence and the corresponding FDA indication. Tirzepatide has compelling clinical results in other populations. Which better protects the heart in people with obesity and established atherosclerotic disease without diabetes remains an unanswered head-to-head question.

Sources

  1. Aubrac G, Raquib RV, Le MT, et al. Cardiovascular Outcomes in Semaglutide Versus Tirzepatide Users With Obesity and Atherosclerotic Disease: A Target Trial Emulation. Diabetes, Obesity and Metabolism, November 10, 2026. PMID: 43010688.
  2. Wilson L, Zhao Z, et al. Semaglutide and Tirzepatide Effects on Cardiovascular Outcomes in People With Overweight or Obesity in the Real World (STEER). Diabetes, Obesity and Metabolism, January 5, 2026. PMID: 41491349.
  3. Novo Nordisk. Wegovy Cuts Risk of Heart Attack, Stroke or Death by 57% Compared to Tirzepatide in Real-World Study. August 31, 2025.
  4. Drug Topics. Wegovy Outperforms Tirzepatide in Reducing Major Adverse Cardiovascular Events. September 2, 2025.
  5. Lincoff AM, Brown-Frandsen K, Colhoun HM, et al. Semaglutide and Cardiovascular Outcomes in Obesity Without Diabetes. New England Journal of Medicine, 2023.
  6. U.S. FDA. Zepbound Prescribing Information, updated August 2026.

Read more

Local News