Date of Award
January 2026
Document Type
Thesis
Degree Name
Medical Doctor (MD)
Department
Medicine
First Advisor
Kasia J. Lipska
Abstract
Insulin prices have skyrocketed in recent decades, imposing substantial financial burdens on Americans who rely on this medication to manage their diabetes. As a result, one in six Americans using insulin for diabetes rationed their insulin in 2021 because of its cost, a behavior linked to adverse health outcomes. In response, policymakers have passed legislation to limit out of pocket (OOP) spending on insulin. The landmark Inflation Reduction Act (IRA) capped OOP spending at $35 per monthly insulin prescription among Medicare Part D beneficiaries starting in 2023. However, the real-world consequences of high insulin costs and insulin rationing, and the early effects of policies that have capped OOP spending, remain understudied. To clarify these relationships, I worked with a multidisciplinary team to conduct two separate analyses using two nationally representative surveys.The first analysis used the 2021 National Health Interview Survey (NHIS) to examine the association between cost-related insulin rationing (skipping, delaying, or reducing doses to save money) and emergency healthcare utilization, a potentially costly consequence of insulin unaffordability. Age-stratified logistic regression and zero-inflated negative binomial regression models examined associations between insulin rationing and (1) emergency department (ED) visit or hospitalization or (2) number of urgent care visits. Among 982 respondents representing 7.5 million U.S. adults with diabetes on insulin (median age 61 years), 17% reported rationing insulin. Among adults 18–64 years old, rationing was not significantly associated with emergency healthcare utilization. However, among adults >65 years old, rationing was associated with more urgent care visits (relative risk 2.1, 95% CI 1.2–3.6). These findings raise the possibility that individuals who ration their insulin may also be rationing healthcare itself, thereby masking detectable associations. The second analysis used the 2022 and 2023 Medical Expenditure Panel Survey (MEPS) to compare OOP spending on insulin in the year before and after the implementation of the IRA. OOP spending on insulin exceeding $420 annually (equivalent to more than 12 fills of $35, “over the cap”) was modeled using logistic regression as a function of year (2022 vs. 2023), selected characteristics, and their interaction. Among 228 respondents in 2022 and 195 in 2023, representing 2.1 and 2.4 million US adults aged 65 and older with Medicare Part D coverage and at least one insulin prescription, the percentage of participants with OOP spending over the cap decreased from 27.1% in 2022 to 13.2% in 2023 (p=0.003). Adults without other public health insurance experienced greater spending reductions, while OOP spending reductions were absent among adults using multiple types of insulin. However, the prevalence of rationing any medication did not di]er significantly in 2022 (6.5%) vs. in 2023 (6.1%; p=0.86). These findings suggest that capping monthly insulin spending, while effective at lowering annual insulin spending, may not address medication rationing behavior. Taken together, these studies illustrate the power of national surveys like the NHIS and the MEPS to illuminate how cost pressures shape diabetes care. To date, the 2021 NHIS provides the only national estimate of cost-related insulin rationing, and our analysis of the MEPS provides the first nationally-representative, population-level evidence on the association of the IRA with OOP spending on insulin. As policymakers and commercial entities continue to redefine the landscape of insulin pricing, it becomes increasingly important to understand how costs influence access, behavior, and outcomes. It is essential to continue to examine the real-world consequences of these policies to ensure that they work as intended to secure healthcare access for all.
Recommended Citation
Borden, Caroline G., "Insulin Rationing In The United States: Scope, Consequences, And The Early Impact Of Policy Reform" (2026). Yale Medicine Thesis Digital Library. 4379.
https://elischolar.library.yale.edu/ymtdl/4379
Comments
This thesis is restricted to Yale network users only. It will be made publicly available on 06/01/2029