ACP review: which obesity drugs pay off?

A systematic review published on 16 June 2026 in the Annals of Internal Medicine assessed the cost-effectiveness of prescription weight-loss drugs in the United States^[1]. The review, commissioned by the American College of Physicians and led by Kevin Jenniskens at Cochrane Netherlands and University Medical Center Utrecht, compared five drugs against lifestyle modification alone and against each other. Two of the five, tirzepatide and phentermine-topiramate, were rated high-value. The rest were not.

What the review found

The authors used a three-tier value framework: high value (less than $100,000 per quality-adjusted life year), intermediate value ($100,000 to $200,000), and low value (above $200,000). Against lifestyle modification alone, tirzepatide and phentermine-topiramate both cleared the high-value threshold^[1]. Liraglutide landed in the low-value category against the same comparator.

Semaglutide produced mixed results depending on the comparator. Against liraglutide it looked cost-effective (high value). Against naltrexone-bupropion and phentermine-topiramate it did not (low value)^[1]. In practice, that means the answer to "is semaglutide worth the price?" depends entirely on what you measure it against.

The five drugs in the review span three decades of obesity pharmacology. Phentermine-topiramate (Qsymia) was FDA-approved in 2012. Liraglutide (Saxenda) followed in 2014 as the first GLP-1 agonist approved specifically for weight management. Naltrexone-bupropion (Contrave) came the same year. Semaglutide (Wegovy) arrived in 2021, and tirzepatide (Zepbound) in late 2023. The newer drugs produce larger weight reductions but carry higher list prices, and that price gap is exactly what cost-per-QALY modelling is designed to test.

Why the findings come with a large asterisk

Every study included in the review was model-based, not trial-based. That means researchers built economic simulations using clinical-trial efficacy data rather than tracking real patients' costs and outcomes over time. The review authors rated every included study as high risk of bias^[1]. They wrote plainly that "current evidence on cost-effectiveness of pharmacologic treatment is hampered by poor-quality studies," limiting how much weight any single conclusion can carry.

This is a common problem in health economics for newer drugs. Tirzepatide's Phase 3 obesity programme, led by the SURMOUNT trials, produced strong efficacy data. SURMOUNT-1 showed 20.9% mean body-weight reduction on the 15 mg dose at 72 weeks^[2]. But translating that into lifetime cost per QALY requires assumptions about drug pricing, treatment duration, weight regain after stopping, and downstream healthcare savings. Different modellers make different assumptions, and none of the existing models have been validated against long-term real-world spending data.

Consider one example. A model that assumes patients stay on tirzepatide for five years and captures the avoided cost of bariatric surgery, diabetes medications, and cardiovascular events will produce a very different ICER than one that assumes two years of treatment followed by partial weight regain. Both approaches are defensible. Neither has been tested against a decade of claims data, because the drug has not been on the market that long.

What this means for patients and prescribers

The review does not tell a patient whether their insurance should cover a specific drug. It tells clinicians and policymakers which drugs have the strongest economic case on paper. For tirzepatide, the signal is positive: it appears cost-effective against lifestyle modification using models built on SURMOUNT-1 trial data^[2]. For semaglutide, the picture is more complicated. And for liraglutide, the older GLP-1 agonist now facing generic competition, the economic case is weak.

The ACP will likely use this review as one input for a forthcoming clinical practice guideline on pharmacologic obesity treatment. When that guideline arrives, the cost-effectiveness tier assigned to each drug will shape formulary decisions at US health systems. That matters because obesity-drug coverage in the US remains patchy. Many employer plans and some state Medicaid programmes still exclude anti-obesity medications entirely, partly on the grounds that the economic case has not been made. A positive ACP guideline would weaken that argument for tirzepatide and phentermine-topiramate.

Patients considering these medications should discuss both clinical suitability and insurance coverage with a healthcare provider. The cost-effectiveness label does not guarantee access, and out-of-pocket pricing for GLP-1 class drugs remains high without coverage.

What we do not yet know

No study in this review modelled the cost-effectiveness of newer entrants such as retatrutide (Eli Lilly's triple-agonist, still in Phase 3) or oral semaglutide at higher doses. The review also excluded non-US settings, so it says nothing about NHS, EMA, or other national cost-effectiveness assessments. And because every included study was model-based, we are still waiting for real-world health-economic data from large claims databases or registry studies.