A new Iranian economic analysis favors injectable semaglutide over tirzepatide and liraglutide under its chosen prices and spending threshold. Yet the model estimates greater health benefits with tirzepatide. Those findings can coexist: a health system may judge an additional benefit too expensive without disputing that the benefit exists.
The study, published October 17, 2026, asks a payer’s question about value over 68 weeks. It does not establish which medicine is best for an individual, announce a coverage decision or report a new head-to-head clinical trial. Its usefulness lies in showing how local costs change the answer—and where the evidence remains difficult to audit.
What the model actually compares
Researchers constructed a decision tree for adults without diabetes who had a BMI of at least 30, or at least 27 with another weight-related complication. They drew treatment effects, adverse events and discontinuation from published literature, rather than enrolling a new Iranian patient cohort. The comparison covered subcutaneous tirzepatide, semaglutide and liraglutide from a payer perspective, according to the indexed abstract.
Costs included medicine acquisition and management of adverse events. Benefits were expressed as quality-adjusted life years, or QALYs: a measure combining time and health-related quality of life. Here, estimated benefits came from changes in BMI and deductions for adverse effects. A modeled QALY gain is therefore not a directly observed extension of survival.
For background on the medicine itself, our semaglutide evidence profile separates clinical research from regulatory status. An economic model answers a different question from either.
Greater benefit does not automatically mean better value
The authors report that tirzepatide produced more health benefit at greater cost than semaglutide, with an incremental cost-effectiveness ratio of US$34,423.89 per QALY. That ratio describes extra spending divided by extra modeled benefit. It is neither the price of a prescription nor an annual treatment bill.
The abstract describes its Iranian willingness-to-pay benchmark as US$15,558.5 per QALY on a purchasing-power-parity basis, tied to one GDP per capita in 2026. Against their chosen benchmark, the authors favor semaglutide. They also describe liraglutide as dominated by semaglutide: in the model, liraglutide costs more while delivering less benefit. These are the authors’ economic findings, not universal clinical rankings.
Currency treatment matters. The abstract explicitly labels the threshold as purchasing-power-parity based but presents the incremental ratio simply in US dollars. Without the full methods, we cannot independently verify the conversion conventions or turn these figures into a reliable discount target. Readers should not interpret them as current pharmacy prices.
Why a US analysis reached a different conclusion
A separate 2025 US economic study also used a 68-week decision tree for adults without diabetes. It favored tirzepatide under its assumptions and a willingness-to-pay threshold of $150,000 per QALY. It used 2024 US costs and included oral semaglutide as well as the injectable options; its authors found results sensitive to medication costs and effectiveness.
This is useful context, but not a controlled comparison between countries. The medicine options, input prices and thresholds differ. The US result cannot validate the Iranian model’s calculations, and the Iranian result cannot settle US insurance value. Together, the studies illustrate why a value conclusion needs its country, time horizon and assumptions attached.
The source limitations deserve attention
We reviewed the Iranian abstract, publisher disclosures and the accessible supplement’s baseline-trial table. The full main paper was restricted, so detailed cost inputs, dose assumptions, conversion methods and sensitivity-analysis calculations could not be checked. Although the abstract says sensitivity analyses supported the findings, we could not independently assess that robustness claim.
There is also a concrete discrepancy to resolve. The supplementary baseline table lists 478 participants in its SURMOUNT-1 row, while the original trial abstract reports 2,539 randomized adults. The available material does not explain that difference. The reason for the difference and any effect on the economic calculations are unknown. It is a reason to seek clarification before relying on the model for purchasing decisions.
The original SURMOUNT-1 trial followed participants for 72 weeks, whereas this economic model uses 68 weeks. Without the full modeling methods, we cannot verify how differing study durations and treatment-effect estimates were aligned. Our guides to reading weight-loss trials and trial estimands explain why those choices matter. We do not combine the studies’ weight-loss percentages into a ranking.
What readers can use from this study
Before treating a cost-effectiveness headline as an access promise, ask whose costs were counted, which alternatives were compared and how long benefits were modeled. A payer calculation does not answer what someone will pay out of pocket, whether supply is reliable or which medicine their clinician considers appropriate. Our guide to treatment access interruptions addresses the practical continuity questions.
The Iranian study offers a locally framed hypothesis about value, with a clear need for fuller methodological review. Its 68-week horizon cannot settle lifetime benefits or costs. No new approval or reimbursement decision follows from publication of the analysis, and its findings are not a reason to change treatment without an individual clinical discussion.


