A 2024 Canadian cost-effectiveness model of weight-loss medicines

Why this study matters

Weight-management medicines can be expensive, and Canadian public and private drug plans ask whether the health gains justify the cost. Economists answer that question with cost-effectiveness models. This study built such a model for Canada, comparing four medicines with standard care. Canadian readers may see its numbers quoted in coverage debates, so it helps to understand what the model did and what it assumed.

Who was in the study

There were no real participants. The model simulated a Canadian cohort with a mean starting age of 50 years, a body mass index (BMI) of 37.5, and 27.6% living with type 2 diabetes. BMI is a number that compares weight with height. The model followed this simulated group over a lifetime.

What the researchers did

The model compared four medicines with the current standard of care, described as diet and exercise: orlistat (Xenical), naltrexone 32 mg/bupropion 360 mg (Contrave), liraglutide 3.0 mg (Saxenda, Victoza), and semaglutide 2.4 mg (Wegovy, Ozempic). It took treatment effects on BMI, blood sugar, blood pressure, and blood fats from the STEP trials of semaglutide and from a network meta-analysis (a statistical method that compares many trials at once). It then projected how often weight-related complications would occur, what they would cost, and how many quality-adjusted life years (QALYs) each strategy would produce. A QALY is one year of life in full health; a year in poorer health counts as less than one.

The main analysis took a societal perspective, meaning it counted costs to society as a whole, not only to drug plans. It used a willingness-to-pay threshold of CAD 50 000 per QALY, a common cut-off for calling a treatment cost-effective. The researchers also ran sensitivity analyses, which test how the results change when assumptions change.

What they found

In the main analysis, the model projected 18.32 QALYs with semaglutide 2.4 mg and 18.18 QALYs with standard care. The incremental cost-utility ratio (the extra cost per extra QALY) for semaglutide was CAD 29 014 versus standard care and CAD 31 243 versus orlistat, in 2021 Canadian dollars (R32-C136). Both are below the CAD 50 000 threshold. The summary we reviewed does not report confidence intervals for these figures.

The model reported that semaglutide "extendedly dominated" naltrexone/bupropion and liraglutide. In economics, that means those options were projected to cost more per QALY than a mix of other options, so they were not preferred in the model. The summary reports that semaglutide stayed below the threshold under public and private payer perspectives and across sensitivity analyses that varied weight regain after treatment, longer treatment, and real-world patient characteristics, with a 70% likelihood of being cost-effective at the CAD 50 000 threshold.

The key caveat came from one sensitivity analysis. When the modelled benefits of weight loss on cancer, mortality, cardiovascular disease, and osteoarthritis surgery were all removed at the same time, orlistat became the preferred option, at CAD 35 723 per QALY versus standard care. In other words, the result for semaglutide depended on assuming that weight loss leads to those longer-term health benefits.

Side effects and people who stopped

The summary we reviewed does not report side effects or treatment discontinuation. Models of this kind usually include assumptions about stopping treatment, but no figures were extracted.

What this study does not tell you

• It is a projection, not an observation. The QALY and cost figures are model outputs, not results measured in real Canadians over a lifetime.
• The result depended on assumed links between weight loss and cancer, mortality, cardiovascular disease, and osteoarthritis surgery. Removing those assumptions changed the conclusion.
• A cost-per-QALY comparison is not a ranking of how well each medicine works for a person. Effectiveness in the model came from trials with different designs and populations.
• Prices were 2021 Canadian dollars. Prices, generics, and coverage change, and the model does not reflect any change since then.
• A separate 2025 Canadian model used a different population, comparators, treatment-duration assumptions, and payer perspective and reached different conclusions. The two models are not directly comparable.
• The PubMed record lists authors affiliated with Novo Nordisk, the maker of semaglutide, and with a consulting firm. The summary we reviewed does not state the funding source or the authors' declared conflicts.

What it means in Canada

This model is one input that drug plans and health technology agencies may use when deciding what to cover. It is not a coverage decision, a price list, or a treatment recommendation. To learn how individual medicines are used in Canada, see semaglutide in Canada. For questions about what provincial plans pay for, see public drug coverage by province. To compare services, see compare services.

Related reading

Other studies in this library that bear on the same question.

Source