This desk looks at annual plans constantly, and the number that should sit next to every one of them is this: the median course runs 240 days [1]. Under eight months, in an insured population, in the largest dataset available. A twelve-month commitment priced against a twelve-month course is priced against something most people do not do — which is exactly the mismatch in why falling prices and annual plans do not mix.
Across 342,932 initiations, 60.8% were GLP-1 drugs, 35.7% SGLT2 inhibitors and 3.6% metabolic or bariatric surgery. Use of both drug classes grew enormously — GLP-1 initiations went from 1,988 to 44,852 a year — while surgery did not, accounting for 0.7% of 2024 treatments. By 2024, semaglutide was 62.2% of GLP-1 starts and tirzepatide 32.6%.
Everyone here was insured, in a private or Medicare Advantage plan, and everyone had both type 2 diabetes and obesity. That combination is the easiest case for staying on treatment: the drug is covered, the indication is established and the prescriptions renew. A self-pay buyer for weight alone faces more friction, not less, so 240 days and 53.0% are better than what this desk’s readers should assume for themselves.
One per-drug detail is reported without a number attached: tirzepatide had the highest adherence and persistence of the GLP-1 drugs. The published summary does not give the figure, so this desk will not supply one. Whether that reflects the drug, the people prescribed it, or when it entered the market is not separable here — the same problem as in the routine-care weight loss figures.
The practical arithmetic is simple. Multiply a monthly price by 8, not 12, when comparing what a course is likely to cost, and treat any plan that charges for 12 as charging a premium for months half of buyers will not use. Check the refund terms before the price — the reasoning behind why flat-rate pricing costs nothing to advertise.