Why is insulin so expensive?

Why aren’t we seeing more companies making insulin? There are many reasons for this, but patent evergreening is a big one. Patents give a person or organization a monopoly on a particular invention for a specific period of time. In the USA, it is generally 20 years. Humalog, Lantus and other previous generation insulins are now off patent, as are even older animal based insulins. So what’s going on? Pharmaceutical companies take advantage of loopholes in the U.S. patent system to build thickets of patents around their drugs which will make them last much longer (evergreening). This prevents competition and can keep prices high for decades. Our friends at I-MAK recently showed that Sanofi, the maker of Lantus, is no exception. Sanofi has filed 74 patent applications on Lantus alone, that means Sanofi has created the potential for a competition-free monopoly for 37 years.

More here, and yes there are a multiple of reasons, not just that one.  Such as this:

… it is actually legal for one insulin producer to pay another one not to enter the market. A few years ago the company Merck announced plans to sell a biosimilar version of Sanofi’s Lantus. Sanofi sued, and eventually Merck announced that it was no longer pursuing it’s biosimilar, presumably due to payments from Sanofi to stay away.

Here is another relevant source.  And this:

…Sanofi has filed lawsuits against both Merck and Mylan to prevent them from going to market with a generic lantus insulin (the Sanofi blockbuster drug).

Here is Vox coverage.  Furthermore, fewer restrictions on foreign importation could solve much of the problem:

According to the Food and Drug Administration, “in most circumstances, it is illegal for individuals to import drugs into the United States for personal use.”

New bills by Peter Welch, Elijah Cummings, and Bernie Sanders would ease those restraints.  It seems easy enough to address this problem without having systematic government purchases of pharmaceuticals.  Insulin prices have risen as much as threefold over the last ten years, but that doesn’t have to be the case.


"People of the same trade seldom meet together, even for merriment and diversion, but the conversation ends in a conspiracy against the public“

Importation really is key. I have personally gone to Canada to buy Fiasp, and the price difference was 10x (in Canada OTC and ~$30USD/vial, in the US prescription and ~$300USD/vial).

"Sanofi has filed 74 patent applications on Lantus alone, that means Sanofi has created the potential for a competition-free monopoly for 37 years"

That makes no sense. Once a product is for sale, there can be at most 21 years of patent protection (typically less than 20). I notice there is o explanation whatsoever at the link for how they came to this conclusion.

My understanding is the problem is that the FDA requires full testing for generic drugs even though they may biochemically identical to an already approved drug. Given the lower profits on generics, there is no reason for a company to make this investment. Seems like something the FDA could easily solve if it wanted to.

I see the first patent was filed in 1994 just before the major patent law reform, so it is theoretically possible although unlikely that protection could extend for 37 years (for applications filed before 1995, their term is 17 years from grant, so if it took 20 years to get to grant that would be true- but 20 years to grant is a hell of a long time and extremely unlikely). This issue should no longer exist going forward unless these companies are somehow getting huge amounts of term extension due to PTO delay which again would not make sense.

Many thanks Melissa for pointing out why this narrative is preposterous or woefully incomplete. And for asking the right questions, viz. if Lantus's patent has run out, then what is stopping other companies from making generic versions? Your explanations are the only ones that make any sense.

Yes, thanks for the additional info. I too hoped the link would explain. So it seems Melissa rejects Ray's contention that the so-called "evergreening" represents eminently valuable innovations to the original product?

Curiouser and curiouser. If the higher prices of evergreened drugs truly reflected greater value for patients, then why do drug companies pay other drug companies to delay copies of the generic versions? The payments would be pointless because patients would always demand the evergreened version.

More questions.

For the FDA. Why are you (ex hypothesi) approving new patents on inventions that don't have any novelty and usefulness and that are then used to "protect" the original patent?

For everyone: If the new (evergreening) patents do not add genuine greater value, then why doesn't it make economic sense for a competitor to sell the "un-improved" insulin -- the excess of the margin of the new version surely presents a mouth-watering proft opportunity.

For the FTC: Why are brand name drug companies getting away with paying other drug companies to delay introducing generic versions of their off-patent drugs? Can any behavior better meet the description of an illegal restraint of trade?

If these agencies can't do their jobs, let's fire them. As Oliver Cromwell told the Rump Parliament: "You have sat too long here for any good you have been doing. Depart, I say, and let us have done with you. In the name of God, go!"

They get individual patents on new indications and formulations to increase patent life.

If they get a patent on a new formulation, it does not cover the original formulation. So that doesn't prevent someone from selling a generic of the original formulation.

You can get a patent for a new use if it's not obvious, but what is the other use for insulin?

Generics still have to go through a testing process, but it is less cumbersome than a new drug:


74 patents is proof that there is a lot of innovation, hard work, and good old fashion American ingenuity in healthcare that deserves to be richly rewarded by our free market system. If anything pharmaceutical companies are deeply underpaid with all the government red tape. Bernie Sanders and Donald Trump should both move to Venezuela.

Bwaahahahahaha! Hehehehehehe!

Woo boy!

This is awesome. This gets framed, right alongside my favorite rants about “they should pay ‘em in script!” and “if people didn’t want children’ pajamas to be highly flammable, a market would arise providing the public with such goods.”

You, my friend, are legend.

It's always interesting when someone takes seriously an obviously satirical post. Is that a failure of the ideological turing test or just someone so delighted to get an opportunity to rail on the "other side" that they don't want to look a gift horse in the mouth?

Another reason to open the border is so I can drive my car to the local farmacia and get my supplies on the cheap. F**** drug companies, they can eat a fat d*ck.

The GoodRX article that Tyler links makes clear that cheap classic insulin is available; it's just not as good as the hard-to-manufacture new stuff. Note too the anecdote in the article where an insurer dropped prices on alternative synthetics, and this was presented as a bad thing because it pressured people to switch away from Lantus.

Does any of this even matter when premiums continue to march up in price? I'm a healthy individual but I'm forced to pay these high rates because of all these little tweaks here and there that push up marginal cost for the insurer. This collective buying of medicine and healthcare is a private sector disaster. I will continue to vote for any and all populists who will do something about it. I don't care if they are racist, fascist, or socialist. I want to see more of my paycheck stay in my pocket not stolen by heathcare execs.

I'd understand that if the populists do something about it. We've seen how the current populist talked about healthcare, but there was never any plan behind any of those very good words: There are many possible conservative tweaks to healthcare, but the Republicans will not do any of them.

You can get inexpensive insulin that's structurally identical to human insulin. (Novolin is human insulin produced using genetically modified yeast and vials can be bought at Walmart inexpensively.) It needs to be refrigerated, you have to inject it about 30 minutes before meals, it only lasts for a few hours.

Other insulins are actually different molecules or different formulations which are less restrictive to use. Some are fast acting and can be taken just before eating. Others last for much longer, as long as all day. Some don't need to be refrigerated and can be stored at room temperature. There have been a few inhalable formulations.

Novolin, being identical to human insulin, must have been protected by patenting the specific means of production. Most drugs are protected both for the specific molecule (if possible) and the means to produce the drug. Generic manufacturers simply have to come up with different chemical routes to avoid still in effect patents, this is usually not a problem for small molecules, but biologics are more difficult case.

Interesting, thanks.

Dittoes! (Though I don't follow the last para).

Now, if this is the true story, then why are patients whining and yammering about steep price increases?

Is it so hard to refrigerate the insulin and to inject it about 30 minutes before meals? The only real issue seems to be that a does only lasts for a few hours. Does that mean waking up every few hours? Or more serious inconvenience?).

In either case, it seems we need a tax on whiners. Either the evergreened insulin is worth the higher price (so enough yammering) or it isn't (so buy Novolin at Walmart).

Wow, just ignorance, trying to beat down these crazy anti-patent ideas is like playing Whack-a-Mole (TM). "Patent Evergreening" he says. There's no such thing. In drug patents, the secondary patent inventions usually add more value, that's why people don't want to use the primary patent invention. Case in point: aspirin. You can scrape the bark of an elm tree to get salicylic acid (aspirin), which is not patented, but the 'value add' that Bayer has patented is the coatings on the generic, non-patentable active ingredient of aspirin to make it not affect your stomach lining. Same with insulin. Something in the 'improvement' or secondary patent is what the consumer wants, for free, and what AlexT calls "evergreening". It's not fair to insist the patient get it for free. If they just want animal insulin, at the lowest possible price, let them use the nasty, first generation, off-patent stuff.

Oops! Got me, it's TC not AlexT. He's been brainwashed by AlexT however...

from the "this" link: a Dr. Kevin Campbell makes this claim: "In the case of insulin (because even a generic is expensive to produce), it's estimated that a generic form will only reduce costs by 20 percent. Unfortunately, it's my belief that the large manufacturers of insulin – Sanofi, Eli Lilly and Novo Nordisk – are working to make the development of generic insulins more difficult. In fact, Sanofi has filed lawsuits against both Merck and Mylan to prevent them from going to market with a generic lantus insulin (the Sanofi blockbuster drug)." - So basically, taking the doctor's word for it, even generic, non-patented insulin is only 20% cheaper than the patented stuff. Hardly enough cost savings to crow about...

Is a generic insulin costly to produce? Then why do I read online estimates like one in BMJ Global Health that the cost of production for a vial of human insulin is between $2.28 and $3.42, while the production cost for a vial of most analog insulins is between $3.69 and $6.16, according to the study in BMJ Global Health.

Meanwhile the price appears to be in the $400-500 range.

Who is telling the truth and who isn't?

Even a non-chemist can see that acetylsalicylic acid is not the same as salicylic acid (SA). C9H8O4 vs. C7H6O3. Willow bark contains salicin which is metabolized to SA. SA is used as an antiseptic, wart remover, and acne treatment. Aspirin's pain & inflammation reducing ability is due to it's being rapidly metabolized to salicylic acid. The acetyl compound is responsible for blood thinning.

aspirin is more soluble in water than SA. hence it's 'better; to ingest.

The final, failing stages of crony capitalism. I'm sure all you market purists think that somehow this won't affect you.

It’s not much of a market to be honest. Imports are completely blocked, layered patent system, heavy handed involvement from the courts, all functioning within a health care system that itself is barely operating within a market.

No sorry this isn’t like buying socks at Walmart or Target. I think the market “purists” will sit this one out.

That's the result of crony capitalism - it's not much of a market.

Just looked it up and an Australian says he pays around $55 US a year for insulin, but he has additional expenses for monitoring his condition and injecting the drug. Of course, this comes with the intolerable burden of 28% of GDP being collected as tax revenue instead of the freedom inducing 26% rate in the US.

Putting Australia solidly where one has a health care system which is not exceptional, like America's.

'With an active, competitive biosimilar market, this study estimated a reasonable price for analog insulins to be between $78-130 USD per person per year (except insulin Detemir, which was higher). Regular and NPH was even lower — between $48-72 USD per year. Yearly costs are based on an average dosage of 40 units per day (a World Health Organization statistic).

The numbers proposed take into account not only manufacturing costs, but many of the other variables involved in production including the cost of active pharmaceutical ingredients, cost of other ingredients, cost of vials, cost of transportation, operating expenses and the added cost of bringing a new biosimilar to market. These numbers are “competitive but profitable” to manufacturers based on experts’ analysis.' https://beyondtype1.org/how-much-does-it-cost-to-produce-insulin/

I'm not sure I understand this, but I think Australia is paying under $300 US per person per year for insulin before subsidy, assuming an average of 40 units per day:


If what I just read online is correct, that's less than one tenth its cost in the US.

It's pretty complicated to tease all the factors apart, but a normal guide is that the U.S. health care system pays more than any other country, often in multiples or magnitudes, not simple percentages like 15% or 60%.

What makes it particularly hard is that insulin can be delivered so many ways at this point - and an insulin pump that senses insulin levels, and which can be easily adjusted in terms of what one plans to eat or drink, is certainly not capable of using the lowest cost insulin. There is a definite aspect where price plays a role in how insulin is handled as a medication/treatment and in terms of quality of life and mitigating the long term effects of diabetes. Which then turns into a fascinating cost/benefit analysis - how does amputation or eye repair/loss of vision compare as an expense to the best current technology delaying or preventing such costly diabetes side effects? A decision that is likely made in place like Australia or Germany in a very different fashion than in the profoundly profit driven American health care system.

Australia (like Canada) is too small a market to matter much in drug pricing. Drug manufacturers recover their development costs in the huge US market, and then they charge the rest of the world just for their variable costs. Basically, if US law mandated paying no more than what Australians do, then it's likely that neither Americans nor Australians would have new drugs anymore. So enjoy your subsidy.

Along these lines, the study prior cites expressly omits capital expenditures, quality control, and regulatory costs associated with manufacture, and it looks like it's silent on R&D and the FDA testing costs required for biosimilars. It does mention that costs in India (where there is no intellectual property protection) are 90% lower, but that Indian manufacturers can't get approved even for the European market, due to "challenges with impurities and within-product variability."

Germany is not so small a market, and though this article is 3 years old and in German, it still illustrates Crikey's basic point. Essentially, a German student studying in America was not able to bring a year's worth of insulin from Germany, and after checking American prices, thought she would pay around 250 dollars more than in Germany for a period of insulin supplies. She was shocked to get a bill for 1200 dollars, with the pharmacist saying that was with a rebate due to her being a student.

So, instead of continuing to pay that price for the next months, she and her family arranged insulin to be sent from Germany - including having her sister fly to the U.S. to bring enough for four months. http://www.spiegel.de/lebenundlernen/uni/mit-diabetes-in-die-usa-eine-studentin-erzaehlt-a-1119086.html

The American health care system is truly exceptional - and as Germans might say, nicht zu fassen.

'neither Americans nor Australians would have new drugs anymore'

Wrong - 'Just how important is our publicly funded research to Big Pharma and Biotech? According to a new study by a small, partly industry-funded think tank called the Center for Integration of Science and Industry (CISI), it is existentially important. No NIH funds, no new drugs, no patents, no profits, no industry.

The CISI study, underwritten by the National Biomedical Research Foundation, mapped the relationship between NIH-funded research and every new drug approved by the FDA between 2010 and 2016. The authors found that each of the 210 medicines approved for market came out of research supported by the NIH. Of the $100 billion it spent nationally during this period, more than half of it — $64 billion — ended up helping the development of 84 first-in-class drugs.' https://portside.org/2018-03-04/taxpayers-not-big-pharma-have-funded-research-behind-every-new-drug-2010

Of course the initial research is often public.

That's also irrelevant.

You're basically saying "DARPA invented the internet, therefore DARPA invented Google, Paypal, Tinder, and Amazon"

Initial research is great, it's also the easiest part of the drug creation process. $64 billion. Now look up what it costs to bring a drug to market from phase 1 through phase 4 clinical trials.

And then stop posting about markets you know literally nothing about apart from wikipedia.

Great DARPA line.

And you are right about the rest. There is a huge distance between and initial discovery and a patented drug.

Sampat & Lichtenberg recently reported that drugs with public-sector patents accounted for 2.5 percent of [the sales in their sample], while drugs whose applications cited federally funded research and development or government publications accounted for 27 percent."

Tom T. is right that without the higher prices American consumers pay "neither Americans nor Australians would have new drugs anymore."

Let's get tough on the free riders. An indirect approach has been suggested by Alex -- drug companies should be prohibited from selling drugs abroad (with exceptions for low-income countries) at a lower price than in the US. That might give them a backbone transplant in their bargaining with the Greeks and Italians -- and even Germans and Brits.

clockwork-prior's life may depend on it.

'while drugs whose applications cited federally funded research and development or government publications accounted for 27 percent'

Reformulations of already existing drugs are patentable. That is why the 210 figure involves actual new drugs, and not evergreened patents.

'Let's get tough on the free riders.'

You mean that the six of world's ten largest pharma companies in Europe need to what? Stop making a profit in Europe? The difference between profit and profiteering seems impossible for some Americans to recognize.

'drug companies should be prohibited from selling drugs abroad'

You really seem to believe that the pharma industry is exclusively American.

'That's also irrelevant'

Well, I did not actually quote the beginning of the article - 'Something odd happened when the Trump administration submitted the original version of its latest pro-corporate budget: Big Pharma didn’t like it.

The problem wasn’t a tax hike or new regulations: the problem was that the budget included deep cuts to the budget of the National Institutes of Health.

If those cuts had gone through, they would have exposed one of the biggest lies told about Big Pharma: that the current system of patents and price-gouging is just an unfortunate necessity to cover the cost of all their brave and noble R&D work.

Trump’s original spending proposal for fiscal year 2019, released last month, included major cuts to not just to the NIH, but the National Science Foundation as well. It is those two publicly funded entities — not Big Pharma — that support the bulk of the country’s basic research into diseases and pathways to new treatments.

That’s why the cuts were especially unwelcome in the executive suites of drug and biotech companies. Their business models depend on Washington subsidizing expensive, high-risk basic research, mostly through the vast laboratory network funded by the NIH.'

'"DARPA invented the internet, therefore DARPA invented Google, Paypal, Tinder, and Amazon"'

Not even close. But then, the difference between basic research into biological and chemical processes and claiming anyone 'invented' the complex system that is the 'Internet' seems just a bit too small for some to recognize.

'it's also the easiest part of the drug creation process'

Which explains the giant backlog of new antibiotics, doesn't it?

'Now look up what it costs to bring a drug to market from phase 1 through phase 4 clinical trials.'

Why not provide a link instead - and then one to what the drug companies spend on marketing? Does not even have to be to wikipedia - after all. none of my links for this post have been.

Many people here seem to live in a world where saying something makes it true - the point of citation is to provide at least an attempt at a factual basis. So, it seems about 50% of an approved drug's price is taxpayer provided basic research - 'However, various sources indicate that it can cost more than $1 billion to bring one product to the market, including approximately $50-840 million to bring treatments through the stages of Basic Research/Drug Development and Pre-Clinical/Translational Research, and approximately $50-970 million to complete the Clinical Trials (Phases 1, 2, and 3).' https://www.brightfocus.org/clinical-trials/how-clinical-trials-work/fda-approval-process And in the case of all approved new drugs between 2010 and 2016, the basic research was apparently not industry funded, thus saving them half the cost - as noted by the link.

Of course, you are welcome to provide your own link - it certainly improves the level of discourse to have more information available.

'...it is actually legal for one insulin producer to pay another one not to enter the market.' combined with '...presumably due to payments from Sanofi to stay away.' is still a higher standard than most comments here - at least 'presumably' gives an attentive reader (not to mention a foundation's lawyers) the opportunity to understand the author actually doesn't know what they are cliaming.

Over 30 million Americans have diabetes (roughly 10% of the population), and over 100 million have diabetes or prediabetes. Why? Eating and exercise habits. One might conclude that the makers of insulin and the producers of America's food are partners.

Well, they both tend to contribute to the same public policy institutes.

So malefactors of great wealth buy political power in America. I'm shocked.

This all might have been a plausible story to tell ten years ago, but some big facts are missing here when characterizing today’s US pharma landscape. Two in particular stand out:

1) There’s a difference between gross and net prices. Net prices realized by manufacturers of insulin have dropped sharply in the US over the past 5 years even as gross prices continued to increase. And yet no mention of supply chain in this piece.

2) Sanofi’s patent protection of Lantus is an interesting place to focus, given the existence of Basaglar - which is, for all practical purposes, a biosimilar.

Medicine is full of this charade. For instance hospital 'prices'. No one pays these artificially inflated numbers, as they are negotiating starting points. Same with drug prices. Hospitals in the end get 20-30% actual collections on their billing dollars. Some of my wife's meds are in the $100K/yr price range. Her insurance pays much less, although pinning that reduction down is difficult. Same with her massive hospital bills. Insurance pays 20-30%, and the hospitals are still operating!

This is exactly right. The US healthcare system is full of these largely opaque transactions where list prices are ludicrous & real prices paid are a lot lower.

There are drugs in other categories where net prices have grown quite a bit. But the insulin drugs are not those, as can be observed by the public disclosures of the major manufacturers (Eli Lilly; Sanofi, and Novo Nordisk)

This exact same Blog noted recently that:

"Médecins Sans Frontières are outraged by these restrictions but, as Tyler and I explain, the alternative is no sales in developing countries or one world-price and you can be sure that if there’s one world-price that price will be the US price and not the Egyptian price."


So your against reimportation there but for it here?

Net insulin prices after rebates haven't gone up in a long time. The main issue with insulin is that prices would go down drastically if rebates were applied at POS. Which is what the Trump administration is proposing right now. But people like Pelosi are in the way of that.

Allow re-importation from wealthy, industrialized countries only (EU, Canada, Japan, Australia).

Won't it be a similar concern for the manufacturer (If I give Canada a price break, its going to eat into my USA margins).

How about we just get rid of rebates and prices will fall to Canada levels.

If re-importation means that manufacturers won't be able to sell to Canada, Japan, Australia, and the EU at lower prices than are charged in the U.S., I am more than OK with that outcome. A single price across industrialized countries sounds like a distinct improvement from my U.S. perspective.

Exclusion of importation of patented drugs is part of the patent-owner's right right to exclude, which is a property right.

So importation would be infringement, and allowing importation of patented products is statutory patent infringement. Since patents are property, taking away a right in that bundle would likely constitute a taking, which the government would have to pay for anyway.


If it's rent seeking by drug companies - where are the excess profits? Why can't I invest in these money making machines? Hell if it is simple monopoly supply that is causing high prices, I am ready to invest in a competitor.

United States Constitution (Article I, Section 8, Clause 8):

This clause states that the United States Congress shall have power...
"To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries."

How does "Crony Capitalism" evolve from that?

Are government market-interventions the fault of the "capitalism" (market) system or the government system?

(and note that Congress has the option NOT to exercise this Constitutional power)

People of the same trade seldom meet together, even for merriment and diversion, but the conversation ends in a conspiracy against the public, or in some contrivance to raise prices.

Because Mick Jagger going to NYC for heart surgery. Maybe, if he showed up at a Brit NHS facility, they'd place him on a morphine drip. Then it's, "Nice to meet you hope you guess my name . . ."

If you don't like America, go to socialist paradise Venezuela.

I exercise (some) and I ain't 180 lbs. overweight. I don't need any effing insulin.

With 188,000 laws and regulations, you don't have free-markets. You don't live in a free country, either. Hell, I likely broke 16 laws this morning.

Plenty of people are insulin-dependent for reasons that have nothing to do with diet or lifestyle. Diabetes has many possible causes.

Type 1 diabetes is an autoimmune disease, not a overeating/sedate lifestyle disease. And I have been thinking of moving to a socialist hellhole in Europe as they seem to have much better policies towards their population than the US does.

Not necessarily a big fan of the NHS but I believe that you can go for private care in the UK which I am sure Jagger could afford. And of course the well known fact that British longevity is better than the US, not just at birth but at all ages.

Here is an easy way to reform healthcare: reduce the patent period for pharmaceuticals from 20 years to, say, 10 years.

That would probably make things much worse. Patent protection works best for pharma

>>According to the Food and Drug Administration, “in most circumstances, it is illegal for individuals to import drugs into the United States for personal use.”

This is technically true. But in practice unless you're importing something addictive it isn't enforced.

The solution is broad-based IP reform, not whatever hare-brained schemes proposed by the likes of Cummings or Sanders. Mickey Mouse is still under copyright! Where is the public benefit in that? Remember, we invented IP protections primarily to bring more to the public, protecting authors & inventors is just the means to that end, not the end in itself - or it was before Disney started greasing palms in Congress.

Boris, where is Natasha?

Not sure what copyright term extension has to do with term-limited patents, but you do you.

I'm from the government and I'm here to help.

You forgot the demand and focused only on supply. There is a worldwide Diabetes Epidemic brought on by cheap commodity agriculture starting with sugar, refined white flour, and then corn.

I'm surprised Tyler didn't find a bunch of view points that just support the status quo.

Pheeeuwww there is a lot of nonsense quoted above.

Humalog, Lantus and other previous generation insulins are now off patent, as are even older animal based insulins.

Humalog and Lantus are not "previous generation insulins." They are among the best insulins available on the market. The only newer insulin out there is stuff like Degludec, which is a once-every-three days insulin appropriate only for type 2 diabetics.

Sanofi has filed 74 patent applications on Lantus alone, that means Sanofi has created the potential for a competition-free monopoly for 37 years.

Poppycock. For one thing, there is already a generic version of Lantus available. For another thing, there are at least three other ultra-long-acting insulin brands that compete directly with Lantus.

Does the article mention that Sanofi literally gives away Lantus for free? Just go to the product website and sign up. Same with Apidra, their rapid-acting insulin analogue.

Comments for this post are closed