Study finds similar heart risk with tirzepatide and semaglutide in US insurance data

A large study of US insurance records found that people with type 2 diabetes and elevated cardiovascular risk who started tirzepatide (Mounjaro, Zepbound) had about the same risk of heart attack, stroke or death as people who started semaglutide (Ozempic, Wegovy, Rybelsus). The head-to-head hazard ratio was 1.06 (95% confidence interval 0.95 to 1.18) for the combined endpoint of myocardial infarction, stroke or death from any cause ^[1].

The comparison included 297,842 people who began one of the two drugs, drawn from three national US claims databases covering 2018 to 2025. Researchers used propensity score matching to balance baseline characteristics between the groups, an attempt to make the two groups comparable on factors that could influence heart risk ^[1].

The result is notable because tirzepatide produces larger improvements in blood sugar control and weight loss than semaglutide, according to the paper ^[1]. Even so, the study did not detect a difference in cardiovascular outcomes between the two drugs.

The authors took several steps to test their method. They first emulated two randomized cardiovascular outcome trials using the same databases: SUSTAIN-6, which compared semaglutide with sitagliptin as a placebo proxy, and SURPASS-CVOT, which compared tirzepatide with dulaglutide. A total of 158,310 patients met eligibility criteria for the SUSTAIN-6 emulation and 44,671 for the SURPASS-CVOT emulation ^[1]. Benchmarking showed high agreement between the reference trials and their emulations for all individual endpoints except all-cause mortality in SUSTAIN-6, the paper says ^[1].

They then widened the eligibility criteria to reflect patients seen in routine care. In those expanded groups, semaglutide compared with sitagliptin had a hazard ratio of 0.82 (95% CI 0.74 to 0.91) for heart attack or stroke. Tirzepatide compared with dulaglutide had a hazard ratio of 0.87 (95% CI 0.75 to 1.01) for a composite that included death ^[1]. Those expanded analyses involved 453,201 people starting semaglutide or sitagliptin and 136,089 starting tirzepatide or dulaglutide ^[1].

The authors conclude that the findings support comparable cardiovascular benefit of tirzepatide and semaglutide in clinical practice, and argue that rigorously designed real-world evidence can complement randomized clinical trials ^[1]. The work was registered on ClinicalTrials.gov under NCT06659744, NCT07088718 and NCT07096063 ^[1].

Why it matters for patients

Semaglutide and tirzepatide have each been shown in randomized trials to reduce cardiovascular risk in certain groups, but no randomized study has directly compared the two, leaving treatment decisions to rely on indirect evidence from trial populations that differ from one another ^[1]. This analysis is an attempt to close that gap using data from ordinary practice rather than trial conditions.

For someone weighing the two drugs, the results suggest the difference in heart protection is small or absent over the period studied, at least among people with type 2 diabetes and elevated cardiovascular risk. That does not mean the drugs are identical. The confidence interval ran from 0.95 to 1.18, meaning the data leave open anything from a modest advantage for tirzepatide to an 18% higher risk, so a real difference cannot be ruled out.

Two caveats matter. First, this is observational data from insurance claims, not a randomized trial; researchers can balance measured factors but cannot rule out that unmeasured differences between the groups shaped the results. Second, claims data do not capture everything. How long each person was followed, what doses they took and how consistently they took them have not been reported.

What happens next

Randomized head-to-head data remain limited. The results of SURPASS-CVOT, which compared tirzepatide with dulaglutide rather than semaglutide, became available while this study was underway ^[1]. Whether a randomized trial directly comparing tirzepatide and semaglutide on cardiovascular outcomes will be conducted has not been disclosed in the paper. The three ClinicalTrials.gov registrations tied to this work are NCT06659744, NCT07088718 and NCT07096063; completion dates are not given in the abstract ^[1].