Why Ozempic is so expensive

Two drugmakers now sit among the ten most valuable companies on earth, and they got there on obesity. Eli Lilly is worth $817 billion and Novo Nordisk $589 billion, against $844 billion for TSMC and $690 billion for Tesla. Neither firm was in this conversation a decade ago — in 2014 Lilly was worth $77 billion.

The reason is the size of the pool rather than the price of the drug. 890 million people worldwide are obese against 486 million with cancer, 262 million with asthma, 200 million with heart disease, 55 million with dementia and 45 million with HIV. Oncology is the biggest drug class in the world at a projected $370 billion; weight loss is already second at $133 billion.

Insulin is where that justification breaks. Eli Lilly and Novo Nordisk, two of the three manufacturers who make 90% of the world's insulin, charged $900 for a year's supply in 2007 — both of them. They then charged $1,200 in 2008, $1,440 in 2010, $1,500 in 2011, $1,680 in 2012 and $2,100 in 2013. The same figures, in the same years, from two rivals.

Then they accelerated. From 2014 the price roughly tripled again in five years: Lilly to $4,080 in 2015, $5,280 in 2017 and $5,864 in 2019, with Novo a few percent behind at every step. The only “cheap” insulin price was always last year's, and a Type 1 diabetic cannot opt out of insulin.

The patent was supposed to stop this. Humalog's expired in 2013 and generics should have flooded in — instead Lilly filed secondary patents on incremental improvements, a practice known as patent thicketing, and the monopoly ran another full decade. Challenging a thicket costs more in legal fees than a small manufacturer can raise, so the challenge never comes.

The motive was a patent cliff of its own. Lilly's 2000s were carried by five blockbusters — Zyprexa at $4.9 billion, insulin at $3.0 billion, Gemzar at $1.7 billion, Evista at $1.1 billion and later Cymbalta and Cialis. Zyprexa and Gemzar lost protection in 2011 and Cymbalta and Evista in 2014, and the drop was immediate.

So insulin was raised to cover the gap, almost line for line. Zyprexa went from $5.0 billion in 2010 to $1.7 billion in 2012 and $0.5 billion by 2018. Over exactly the same years Lilly's insulin sales climbed from $3.1 billion to $4.3 billion. Insulin was the one product where the patient pool was both large enough and desperate enough to absorb any price.

Lilly also invented this drug class and then lost it. Byetta, launched in 2005, was the first GLP-1 ever sold. It needed two injections a day, drew reports of kidney failure, pancreatitis and deaths, and took an FDA black-box warning. Sales peaked at $797 million in 2009 and were down to $278 million by 2012.

Novo Nordisk took the idea and executed it. Victoza launched in 2009 as a once-daily GLP-1, did $13 million in its first year, $1.4 billion by 2012 and $3.2 billion by 2017 — clearing a billion dollars in its fourth year, five years faster than Lilly's Trulicity managed a decade later.

The replacement is already bigger than the thing it replaced. Mounjaro went from $483 million in 2022 to $5.2 billion in 2023, against a peak of $4.3 billion for insulin in 2018. Zepbound, the same molecule branded for the obese rather than the diabetic, managed only $176 million — because insurers and Medicare still will not cover weight loss, so patients pay about $1,000 a month out of pocket or get the diabetic version instead.

Novo is further ahead. Ozempic went from $0.3 billion in 2018 to $1.7 billion, $3.1 billion, $5.0 billion, $8.8 billion and $14.2 billion in 2023, with Wegovy adding $4.6 billion on top. That single drug now grosses more in a year than the company ever made from insulin.

Specialization is why Novo out-earns a much larger rival. Novo ran 46% operating margins in 2015 and 44% in 2017 against 14% and 12% at Lilly, on revenue that grew from $7 billion in 2008 to $18 billion in 2019 with a fraction of the assets. Going all-in on one disease worked better than Lilly's decade of diversification.

None of the usual gatekeepers apply to this drug class. Manufacturers normally have to buy doctors — big pharma paid physicians $8.1 billion in cash and $13 billion in total in 2023 — and negotiate placement with pharmacy benefit managers who take a cut of every rebate: CVS Caremark at $187 billion of revenue, Express Scripts at $154 billion and Optum at $116 billion in 2023. With GLP-1s the patients arrive asking for the drug by name, which inverts every one of those negotiations.

Lilly has now removed the doctor from the transaction outright. LillyDirect is a company-run telehealth portal where an independent physician will write the prescription your own doctor declined and the drug ships to your door. Going direct-to-consumer is only possible with a product this viral — and the company that built it is the same one that spent a decade raising insulin prices on people who could not refuse.

The reclassification campaign is the actual product. Purdue built OxyContin into an epidemic by getting pain treated as a fifth vital sign; Lilly and Novo are now arguing that obesity is a disease rather than a condition. The medical case is arguable. The commercial case is not: disease status is what unlocks insurer and government coverage, and coverage is what turns 890 million people into a recurring, patented, permanent prescription.

And obesity is a symptom, not the root cause. American obesity is downstream of a food system that permits additives, dyes, and hormones banned in most of the developed world. The country already spends $13,493 per person on healthcare against $5,493 in the UK and still dies six years earlier than Japan. The proposal on the table is to bill taxpayers and patients for a $1,000-a-month injection rather than change what is in the food.