News|Articles|April 6, 2026

$35 Insulin Cap Cut Medicare Spending but Failed to Attract New Users

Fact checked by: Christina Mattina
Listen
0:00 / 0:00

Key Takeaways

  • A voluntary 2021 model and 2023 IRA expansion reduced mean quarterly insulin out-of-pocket spending by ~$106 overall, dropping the share paying >$35 per 30-day supply from 62.1% to 0.11%.
  • Insulin utilization increased primarily through higher daily units among prevalent users, while the proportion of beneficiaries with T2D filling any insulin prescription did not rise after universal implementation.
SHOW MORE

The cap cut quarterly spending by the hundreds and lowered A1c levels but failed to expand the pool of insulin users.

Capping Medicare beneficiaries’ monthly insulin out-of-pocket costs at $35 was associated with steep reductions in spending and modest but meaningful improvements in glycemic control—yet did little to draw new patients onto insulin therapy. The cap was initially introduced for some beneficiaries before expanding to all beneficiaries.

A large national cohort study published today in JAMA Internal Medicine reached these conclusions,1 with an accompanying editorial arguing that policy evaluation must look beyond costs to capture the full picture, including risks.2

Insulin affordability has been a persistent problem in the US health care system, with high costs driving some patients with diabetes to ration or forgo their medication entirely—leading to worse glycemic outcomes, the study authors explain. In response, CMS’ Innovation Center launched the Part D Senior Savings Model in January 2021, a voluntary program allowing participating plans to cap insulin out-of-pocket costs at $35 per month.3 By 2022, more than 17 million Medicare beneficiaries were enrolled in participating plans. Congress then extended the cap to all Part D beneficiaries in January 2023 as part of the Inflation Reduction Act (IRA). Despite the policy’s broad reach, however, independent evaluations linking the federal cap to clinical outcomes are lacking.

Researchers conducted a serial cross-sectional interrupted time-series analysis using data from the IQVIA Longitudinal Access and Adjudication Data database and linked ambulatory electronic health record (EHR) data, covering January 1, 2019, through December 31, 2023. The study cohort comprised 4.8 million Medicare Part D beneficiaries with type 2 diabetes (T2D), with a median quarterly cohort size of 1,393,402 patients (range, 1,261,976-1,475,460). Patients with commercial or Medicaid coverage and those receiving low-income subsidies were excluded.

The analysis was divided into 3 periods: before the monthly $35 cap (January 1, 2019-December 31, 2020), partial exposure via the voluntary Senior Savings Model (January 1, 2021-December 31, 2022), and universal exposure under the IRA (January 1, 2023-December 31, 2023). Outcomes included quarterly insulin out-of-pocket spending, daily insulin units, hemoglobin A1c (HbA1c) levels in a linked EHR subset of 207,197 patients, and incidence of severe hypoglycemia and acute hyperglycemic events.

The Impact on Out-of-Pocket Spending

Most patients in this study were 65 years or older and male. Diabetic eye disease, chronic kidney disease, and atherosclerotic cardiovascular disease were the top 3 comorbidities.

At baseline, 62.1% of insulin users paid more than $35 for a 30-day supply, but by the fourth quarter of 2023, that figure had dropped to just 0.11%. In addition, mean quarterly insulin out-of-pocket spending fell from baseline’s $192.66 (95% CI, $192.63-$192.68) by $47.90 (95% CI, −$48.95 to −$46.84; P < .001) following the 2021 initiative and by an additional $58.59 (95% CI, −$59.91 to −$57.27; P < .001) after the IRA took effect, reaching $95.42 (95% CI, $95.39-$95.45) by late 2023. Overall out-of-pocket spending across all prescriptions also declined, although out-of-pocket costs for noninsulin diabetes medications—including costly sodium-glucose cotransporter-2 inhibitors and glucagon-like peptide-1 receptor agonists—rose over the same period, partially offsetting insulin savings.

Insulin use increased modestly, but the gains were concentrated among existing users rather than new initiators. Among insulin users, mean daily units increased by 1.36 (95% CI, 1.15-1.58; P < .001) after 2021 and by another 1.16 (95% CI, 0.89-1.44; P < .001) after 2023. However, the proportion of patients with T2D who filled at least 1 insulin prescription did not increase after 2023. In the EHR-linked subset, mean HbA1c decreased by 0.06 percentage points following 2023 implementation, and the share of patients with poor glycemic control (HbA1c >9%) fell as well. Severe hypoglycemia incidence increased modestly, by 0.51 per 1000 patients after 2021 and 0.29 per 1000 patients after 2023, while rates of diabetic ketoacidosis and hyperosmolar hyperglycemic state were unchanged.

Looking Forward

In the accompanying editorial,2 the authors noted that although the cost reductions are a meaningful first step, the cap was more effective at improving adherence among existing insulin users than at spurring new initiations, pointing to the need for complementary strategies targeting patients, prescribers, and insurers. They also emphasized that the increase in hypoglycemic events underscores the importance of evaluating clinical risks alongside financial benefits, and called for longer-term study of cardiovascular and kidney outcomes to more fully assess the policy’s impact.

“Policy initiatives, such as the insulin co-pay cap, should be designed and evaluated in terms of their implications not only for costs of the drug but also for clinical outcomes in the short-term,” they emphasized, “such as diabetes control, adherence among current insulin users, and diabetes-related hospitalization or emergency department visits.”

The study authors noted that a separate IRA provision capping Medicare Part D annual out-of-pocket spending at $2000 starting in January 2025 may reduce reliance on the insulin-specific cap for high-cost medication users and that broader affordability policies may ultimately prove more effective than drug-specific measures for patients managing multiple chronic conditions, as “federal cost-sharing policies could also improve affordability and access to other essential medications.”

References

  1. Hong D, Cantos K, Morgenstern D, et al. Out-of-pocket spending for insulin by Medicare beneficiaries after monthly caps. JAMA Intern Med. Published online April 6, 2026. doi:10.1001/jamainternmed.2026.0255
  2. Hung A, Inouye SK, Durant RW. Insulin copay caps—impacts beyond costs. Published online April 6, 2026. doi:10.1001/jamainternmed.2026.0268
  3. McCormick B. Policy measures lower out-of-pocket insulin costs for Medicare Part D beneficiaries. AJMC®. March 23, 2026. Accessed April 5, 2026. https://www.ajmc.com/view/policy-measures-lower-out-of-pocket-insulin-costs-for-medicare-part-d-beneficiaries