Yes, in every US model published so far, although the size of the margin depends heavily on who built the model and what price it assumed. For adults with metabolic dysfunction-associated steatohepatitis (MASH) and moderate to advanced fibrosis, three models put semaglutide 2.4 mg at $19,911, $42,200 and $80,076 per quality-adjusted life year gained against standard care [1] [2] [3]. All three sit below the $100,000 to $150,000 thresholds those same models used to judge value.
Whether the drug works on the liver is a separate question, covered in what the liver trials found. Whether it is good value for weight loss is another, and there the obesity models reach a more mixed verdict. A cost per QALY is the figure a payer uses when it decides whether to cover a treatment, and a QALY is one year of life in full health.
Three models, three figures
The lowest ratio comes from a lifetime model taken from the perspective of a US third-party payer. Semaglutide produced 1.66 additional QALYs at an incremental cost of $33,031, for a ratio of $19,911 per QALY, and it was cost effective in 99.5% of simulations at a $150,000 threshold [1]. Two of its three authors are employees and shareholders of Novo Nordisk, which makes semaglutide, and the third works for a firm Novo Nordisk contracted to develop the model and write the manuscript.
An academic model from a Yale-based group ran shorter horizons for patients with F2 to F3 fibrosis and found semaglutide cost-effective at $42,200 per QALY over five years and $44,138 over ten [2]. When the model added semaglutide’s cardiovascular mortality benefit, the ratio improved to $38,324, and the authors declared no competing interests.
A third model, from Georgia Tech and Cleveland Clinic, used a lifetime horizon and a $100,000 threshold [3]. Semaglutide came in at $80,076 per QALY with an 89.8% probability of being cost-effective, and tirzepatide, which is not approved for MASH, came in lower at $42,705. The authors named drug price as the most influential parameter in the model, which is the point every one of these analyses eventually arrives at.
What the cash price does to the answer
The manufacturer’s model ran one scenario using the direct-to-consumer cash price of $499 a month instead of a payer-negotiated cost. At that price, semaglutide was less costly and more effective than standard care [1], which economists call dominant: the comparison needs no threshold because the drug wins on both counts.
Sellers tracked on this site publish a median of $189 a month for semaglutide by injection. Most of those figures are for compounded rather than branded product, which is a regulatory difference and not a discount. The spread is visible in what a month actually costs and in the price check, and none of these models was run on those prices.
That price sensitivity cuts both ways. A model is only as current as the price it was given, and the list price has been the subject of its own argument, laid out in what semaglutide costs to make.
Semaglutide against resmetirom
Resmetirom is the other drug approved for this stage of MASH, and it is not a GLP-1 drug at all. The models disagree about it more than they disagree about semaglutide. The manufacturer’s model put its two doses at $198,607 and $346,810 per QALY [1], and the Cleveland Clinic model at $273,445, above its threshold [3]. The Yale model put it at $95,981 over five years, below a $150,000 threshold [2].
Value and effect are different rankings. The same Yale model projected that treating 100,000 patients over ten years would prevent 1,040 liver-related deaths with semaglutide and 1,180 with resmetirom [2]. Semaglutide ranked better on cost per QALY while resmetirom prevented more liver events in that projection, and a reader choosing between them with a clinician is weighing both.
Outside the US
A generalized analysis across twelve countries, from Brazil to the United States, modeled people with type 2 diabetes from screening through treatment [4]. Screening followed by intensive lifestyle changes and semaglutide was cost-effective in 11 of the 12 countries, the exception being Tanzania, and the resmetirom pathway in 8. Whether screening itself is worth doing is a question with its own evidence, set out in whether people with diabetes should be screened.
None of these figures transfers directly to one person paying cash. A threshold is a judgment about what a health system should fund for a population, and an individual buyer has no discount rate or lifetime horizon. What the models establish is the ordering, and that ordering has held across three groups with different funding and different assumptions.