Medicare insulin costs dropped from $51 to $22 per month after federal price caps took effect

Johns Hopkins Bloomberg School of Public Health / JAMA

Average out-of-pocket insulin costs for Medicare beneficiaries fell from $50.87 per month in 2019 to $21.98 in 2023. The share of patients paying $35 or less for a 30-day supply jumped from 48% to 75% over the same period. Every U.S. state saw declines.

Those numbers, published March 19 in JAMA, represent the first large-scale analysis of whether the Inflation Reduction Act's insulin price caps actually did what Congress intended. The short answer: largely yes, with a notable gap.

The $35 cap and its predecessor

The path to cheaper insulin for Medicare patients came in two stages. In 2021, the Centers for Medicare and Medicaid Services (CMS) introduced a voluntary program capping out-of-pocket costs at $35 for a 30-day supply. Then the Inflation Reduction Act of 2022 made that cap mandatory for all Medicare Part D beneficiaries, effective January 1, 2023. It was the first time the federal government had imposed insulin price caps across the entire Medicare population.

Researchers at the Johns Hopkins Bloomberg School of Public Health, led by Michael Fang, analyzed Medicare claims data covering nearly 3.8 million patients who had at least one insulin claim between 2019 and 2023. They excluded those receiving low-income subsidies, focusing on the population most directly affected by the new caps.

The trajectory was clear. Costs dropped, the number of patients paying above the cap shrank, and the decline held across all 50 states plus Washington, D.C.

The prorating gap

But roughly one in four Medicare beneficiaries still paid more than $35 for a 30-day insulin supply in 2023 — a finding the researchers themselves called unexpected.

The explanation lies in how CMS applies the cap. Official guidance pegs the $35 limit to full 30-day multiples. If a prescription covers an in-between quantity — say, 45 days — the plan can treat it as a 60-day supply and charge up to $70. A 90-day prescription might cost up to $105.

"If the prescription falls in between, the patient can be charged up to the next full multiple of a month," Fang explained.

This prorating mechanism creates state-by-state variation that has nothing to do with insulin prices themselves. In 2023, average 30-day costs ranged from $10.36 in Washington, D.C., to $31.09 in Minnesota — differences that may partly reflect how local Medicare plans handle the math on non-standard prescription lengths.

3.8 million Americans who depend on insulin

The stakes are not abstract. Approximately 3.8 million Medicare beneficiaries use insulin to manage type 1 or type 2 diabetes. For type 1 patients, the hormone is not optional — their bodies produce virtually none of it. Many type 2 patients also require insulin when other treatments cannot maintain adequate blood sugar control.

Before the federal caps, insulin costs had become a political flashpoint. Patients rationed doses. Some skipped them entirely. The list price of common insulin products had risen more than 1,000% over two decades, though the actual out-of-pocket burden depended heavily on insurance structure, deductible timing, and plan design.

The new data suggests the policy intervention worked at scale. A 57% reduction in average monthly out-of-pocket costs — from roughly $51 to $22 — is substantial by any measure. And the fact that it held across states with very different insurance market structures suggests the federal mandate, rather than local factors, was the primary driver.

What the numbers don't tell us

The study tracks out-of-pocket costs, not total insulin spending. The cap shifts costs within the system — from patients to plans, and potentially from plans to manufacturers through negotiated pricing. Whether total insulin spending in Medicare rose, fell, or simply redistributed is a separate question this analysis does not address.

It also cannot capture behavioral effects. Did lower costs change how patients used insulin? Did adherence improve? Did fewer people ration? Those outcomes, arguably more important than the cost data itself, would require clinical follow-up that this claims-based study was not designed to provide.

The researchers are now investigating the prorating issue more closely, examining how costs vary across individual plans and whether policy adjustments could close the gap for the 25% of beneficiaries still paying above $35.

For now, the headline finding is straightforward. Congress capped insulin prices. Prices came down. Not perfectly, not for everyone, and with a bureaucratic wrinkle that leaves some patients paying more than the law intended. But the direction is clear, and the magnitude is large enough to matter.