The Insulin Crisis: A Case Study in Pricing Failure

Insulin is not a new drug. It was discovered in 1921 by Frederick Banting and Charles Best at the University of Toronto. Banting sold the patent to the University of Toronto for one dollar, explicitly so that the drug would remain affordable and accessible. He believed that a medicine necessary for survival should not be a source of private profit. For decades, insulin was inexpensive. A vial cost a few dollars. People who needed it to live could afford it.

Between 1996 and 2019, the list price of insulin in the United States increased by approximately 1,200 percent. A vial of Humalog that cost $21 in 1996 cost $274 by 2019. A vial of Lantus that cost $35 in 2001 cost $290 by 2019. The insulin itself did not change. The patients who needed it did not change. The price changed — and the consequences for the people who could not keep up with it were, in some cases, fatal.

Who Depends on Insulin

Approximately 8.4 million Americans use insulin to manage diabetes. For people with Type 1 diabetes — an autoimmune condition in which the body produces no insulin at all — the drug is not optional. Without it, blood sugar rises to levels that damage organs, cause diabetic ketoacidosis, and produce death within days to weeks. Type 1 diabetics do not have the option of managing their condition through diet, exercise, or oral medication. They require insulin to survive.

Approximately 1.6 million Americans have Type 1 diabetes. An additional 6.8 million Americans with Type 2 diabetes require insulin because oral medications are no longer sufficient to manage their condition. For both groups, insulin is a maintenance medication taken multiple times daily for life. There is no course of treatment that ends. There is no generic substitution in most cases. There is no going without.

What Rationing Looks Like

When insulin prices rise faster than incomes and insurance coverage, patients ration. Rationing insulin means taking less than the prescribed dose to make a vial last longer — skipping doses, reducing units, spacing injections further apart than a doctor has prescribed.

The clinical consequences of insulin rationing are serious and well-documented. Underdosing insulin produces chronically elevated blood sugar. Over time, that elevation damages the kidneys, the eyes, the nerves, and the cardiovascular system. Diabetic retinopathy, nephropathy, neuropathy, and cardiovascular disease are the long-term consequences of inadequate blood sugar control. They are also expensive to treat — far more expensive than the insulin that would have prevented them.

In the short term, severe underdosing can trigger diabetic ketoacidosis, a life-threatening condition requiring emergency hospitalization. Several deaths from insulin rationing have been documented and reported in the United States, including the case of Alec Smith, a 26-year-old in Minnesota who died in 2017 after aging off his parents’ insurance and rationing insulin he could not afford at full price. His case drew national attention and became a reference point in legislative debates about insulin pricing.

A 2019 study published in JAMA Internal Medicine found that 25 percent of insulin-dependent diabetes patients in the United States reported rationing insulin because of cost. A subsequent survey by the Health Care Cost Institute found that out-of-pocket spending on insulin among insured patients doubled between 2012 and 2016 — even for patients with coverage.

The Pricing Mechanism

The price of insulin did not increase because insulin became more expensive to manufacture. The active pharmaceutical ingredient in modern insulin analogs is produced through recombinant DNA technology that has become more efficient, not less, over time. Manufacturing costs for insulin are estimated at a few dollars per vial. The gap between manufacturing cost and list price is not production cost — it is the product of the pricing chain described in the previous article, applied to a drug for which demand is inelastic.

Inelastic demand means that patients cannot reduce their consumption in response to price increases. A person who needs insulin to survive will find a way to pay for it, borrow money for it, ration it dangerously, or go without it at serious risk. That inability to walk away from the purchase gives manufacturers pricing power that does not exist for discretionary products.

Three manufacturers — Eli Lilly, Novo Nordisk, and Sanofi — produce the vast majority of insulin sold in the United States. The market is not competitive in the conventional sense. The three manufacturers do not compete primarily on price. They compete on formulary placement — on which of their products gets covered at the lowest tier by the largest insurers — and that competition runs through Pharmacy Benefit Managers who negotiate rebates in exchange for favorable placement.

The rebate structure creates a documented perverse incentive: manufacturers raise list prices in part to have more room to offer rebates to PBMs, which improves their formulary placement, which maintains or increases market share. Higher list prices fund larger rebates. The rebates flow to PBMs and insurers. Patients whose cost-sharing is based on list price — those who are uninsured, underinsured, or in the deductible phase of their plan — pay more as list prices rise, regardless of what rebates are flowing elsewhere in the system.

A 2019 Senate Finance Committee investigation found that the list price of Lantus increased 159 percent between 2012 and 2019, while Sanofi’s net price — what it actually received after rebates — increased only 8 percent over the same period. The list price increase was not primarily about what Sanofi received. It was about the rebate architecture.

What Has Changed

Legislative and regulatory pressure on insulin pricing has produced some changes. In 2023, Eli Lilly announced it would cap the out-of-pocket cost of its insulin products at $35 per month for patients with commercial insurance. Novo Nordisk and Sanofi followed with similar announcements. The Inflation Reduction Act of 2022 capped Medicare Part D out-of-pocket insulin costs at $35 per month beginning in 2023.

These changes are meaningful for the patients they reach. A $35 monthly cap is substantially less than what many patients were paying. But the caps apply differently depending on insurance status: the Medicare cap is statutory, the commercial insurance caps are manufacturer commitments that are not legally binding in the same way, and uninsured patients — those most exposed to list prices — are covered only through separate manufacturer assistance programs that require application and qualification.

The list prices themselves remain high. The underlying pricing mechanism — manufacturer pricing power, PBM rebate architecture, inelastic demand — has not changed structurally. The caps represent a partial mitigation of the most severe access failures. They do not represent a resolution of the pricing structure that produced those failures.

Why This Case Matters for the Hub

Insulin is not the only drug whose pricing has produced rationing and preventable harm. It is the most documented, the most politically visible, and the most structurally clear example of how the drug pricing system produces outcomes that are difficult to defend on any principled basis — a drug invented a century ago, sold at a manufacturing cost of a few dollars, priced at hundreds of dollars per vial because the patients who need it cannot go without it.

The structural features of the insulin market — three dominant manufacturers, PBM rebate architecture, inelastic demand, absence of meaningful price negotiation — appear in varying degrees across the drug pricing system. Understanding how they operate in the insulin case is a foundation for understanding how they operate elsewhere.

If you or someone you know has direct experience with insulin rationing, insulin costs, or the consequences of inadequate access to diabetes medication, the forum is where that testimony belongs. The Sentiment stage of this hub is designed to surface exactly that kind of grounded, specific experience.


This article was researched and drafted with AI assistance under human review. See our full AI and editorial practices.