Insurance and GLP-1s: Why Coverage Is a Bet on the Price Falling

Ozempic had a 2024 list price of $959 for a 30-day supply in the United States, and almost nobody with health insurance is supposed to pay that, which is sort of the point of insurance and sort of the problem with it. If you typed in the Ozempic price with insurance, the honest answer is that there is no single number. It depends on whether your employer or insurer chose to cover the drug, for which conditions, at what copay, and after how much paperwork. The sources I have do not give a typical insured copay, and I am not going to invent one.

What the sources do give is the shape of the decision. Without insurance, brand Ozempic or Wegovy runs $935 to $1,349 a month. Medicare’s negotiated price for Ozempic, Wegovy and Rybelsus is $274 per 30-day supply, effective January 1, 2027. Every health plan in America sits somewhere between those numbers, making a bet about how fast the price comes down. The price on your pharmacy receipt is the visible end of that bet.

Why a health plan would hesitate to cover a drug that works

Start with a small, reasonable version. (This scenario is invented.) An employer runs a health plan for 100 people. One of them gets prescribed Ozempic, and the plan pays something near list price, call it roughly $11,500 a year. It’s fine. That is what a health plan is for: lots of people chip in so that the one person with an expensive problem does not pay for it alone.

Now ten people are on it. Then the drug class picks up approvals for heart protection, kidney disease, fatty liver disease and sleep apnea (all of which happened before 2026), so the list of people with a good medical reason to take it gets longer. Now thirty people are on it. The line item is no longer a rounding error. It is a budget meeting.

And here is where the employer’s finance person has a thought. The price of this drug is going to fall someday, because patents expire and generics arrive. So every month the plan pays $959 is, in a sense, a month of paying full price for being early. The rational response is to be less early. First the plan requires a doctor to fill out a form before it pays. Then it requires the patient to try something cheaper first. Then it covers the drug for diabetes but not for weight. Then it covers it with a copay big enough to be a conversation.

Push it one more step and you get the logical endpoint: a plan that covers Ozempic in full, starting February 17, 2041, which is when U.S. Patent 11,318,191 (“GLP-1 compositions and uses thereof”) expires. That is a perfectly rational benefit. It is also perfectly useless. Nobody’s employer has written that exact rule, I assume, but the incentive points that way, and a lot of real coverage policy is somewhere along that road.

Prior authorization is a queue with a form at the front

Prior authorization is the mechanism by which a plan says “yes, but prove it.” The prescriber documents that the patient meets the plan’s criteria, the plan reviews it, and only then does the plan pay. The specific criteria vary by plan, and the facts here do not settle what any particular plan requires, so I will not pretend they do.

The cleanest way to think about it is a deli counter. Everyone is entitled to sliced turkey, but you have to take a number, and the number dispenser is in a different building, and it only accepts tickets signed by your doctor. Nobody is refused turkey. The line just moves slower than it would if turkey were cheap. When turkey gets cheap, the dispenser tends to disappear.

I picture the meeting where this gets designed going something like: “Hey, what if we covered it, but made it slightly annoying?” “How annoying?” “About $959 a month annoying.” (I have never attended this meeting. I write a newsletter.) To be fair, the people in the meeting are doing their jobs, and their job is to keep premiums from going up for the other 70 employees who are not on the drug.

The annoyance has spawned its own service industry. As of October 27, 2026, Walgreens added “virtual coverage navigation, Medicare GLP-1 Bridge support, and pharmacist follow-up calls for GLP-1 patients,” according to Pharmacy Times. “Pharmacist follow-up calls” is a very clinical phrase for what is, structurally, someone checking in after the first date to ask whether you’d like to keep seeing each other. When a pharmacy chain builds a department to help people get through their own insurance, the paperwork has become a product.

What generic entry did to the math in India

If you are a health plan making a bet on price declines, India in 2026 is the dream scenario. Over 40 generic semaglutide brands launched on March 21, 2026, the day after the patent expired there. Before that, Ozempic cost about ₹8,800 to ₹11,175 a month (roughly $100 to $130). The generics came in at ₹1,800 to ₹4,200 a month (roughly $20 to $50). And then, as the Generic Guide puts it, “Novo Nordisk responded by cutting Ozempic 36% and Wegovy 48% within 10 days.”

Ten days. That is less time than it takes most employers to schedule a benefits review. By October 8, 2026, the lowest stated monthly price for generic semaglutide in India was Rs 1,290 (roughly $15), for vials from Natco and Eris. That is a modest change.

The point is that this is what the bet looks like when it pays off. Patent expires, dozens of competitors show up the next morning, the brand cuts its own price almost immediately, and anyone who waited gets the drug at a fraction of what the early buyers paid. It is a bit like The Price Is Right, if the contestants were actuaries and the showcase were a five-year spending forecast.

Why the American version of that cliff is further away and shallower

The trouble with the bet in the United States is that the payoff date keeps receding. Semaglutide in America is covered by a stack of patents: three titled “Oral dosing of GLP-1 compounds” that run to May 2, 2034; three titled “GLP-1 compositions and uses thereof” that run to August 24, 2038; solid-composition patents running to February 1, 2039; and the one that runs to February 17, 2041. (The site has a whole article on the thicket, so I will leave the trees there.) Apotex received the first FDA tentative approval for a generic Ozempic on April 10, 2026. “Tentative” is doing real work in that sentence, and the facts here do not say when it becomes a launch.

Then there is the question of how far the price falls once a generic does arrive. America has a test case. Teva launched an authorized generic of Victoza, an older GLP-1, on June 24, 2024, priced at $469.60 for a two-pack against about $560 for the brand. DiaTribe called that “only a 16-17% price reduction.” Prost Biotech says peptide generics in the U.S. typically get 15 to 40% discounts, versus 80 to 90% for small-molecule generics, “due to the more expensive biosimilar approval pathway.” So an American health plan waiting for semaglutide to go generic may be waiting a decade for a discount smaller than the one India got in ten days.

And while it waits, the market moves. Eli Lilly overtook Novo Nordisk as the GLP-1 market leader by 2026, with tirzepatide protected by a U.S. compound patent until May 13, 2036. So the plan’s bet on semaglutide prices falling is partly a bet on a drug that may not be the one its members are prescribed. That is a little dumb, but it is the system’s dumb, not anyone’s in particular.

Medicare’s $274 is the closest thing Americans have to a generic

Semaglutide was included in the second round of Medicare price negotiations under the Inflation Reduction Act, and the result is $274 per 30-day supply starting January 1, 2027. Against the 2024 list price of $959, that is about 71% off. Nobody had to wait for a patent to expire. The government just asked, with leverage.

This matters for everyone else because it changes the reference point. A commercial insurer negotiating with a manufacturer now knows that a large buyer can get the same molecule for $274. Whether that number pulls commercial prices down, and by how much, is not settled by anything I have. I guess it would be strange if it pulled them up.

Look, the whole arrangement is coherent once you see it as a bet: plans restrict coverage now because the drug is expensive now, and they expect it to be cheaper later, through patents expiring, generics arriving, negotiation, or a brand cutting its own price to hold its ground, etc. The restrictions are the cost of waiting. Patients bear part of that cost in forms and copays. Pharmacies bear part of it in follow-up calls. Employers bear the rest in budget meetings.

Which means that, as of January 1, 2027, the cheapest coverage strategy for an American employer is to wait for its employees to turn 65.