NR 506 Week 7 Health Policy Evaluation Write-Up Example

Evaluating the Medicare Part D Insulin Limit Against Three Criteria Set in Advance: More Fills Relative to 60- to 64-Year-Olds Without Part D, Savings Estimated but Not Yet Measured, and Premium Effects Nobody Has Isolated

Student Name

College of Nursing, Chamberlain University

NR 506: Health Care Policy

Instructor Name

Month Day, Year

Evaluating the Medicare Part D Insulin Limit Against Three Criteria Set in Advance: More Fills Relative to 60- to 64-Year-Olds Without Part D, Savings Estimated but Not Yet Measured, and Premium Effects Nobody Has Isolated

Criteria Before Evidence

This paper evaluates one policy already in force: the limit on cost-sharing for covered insulin products in Medicare Part D, in effect since January 1, 2023 (Inflation Reduction Act, 2022). The criteria are set here, before any evidence is examined, so that the evidence cannot choose them.

Criterion 1, access. The policy was meant to reduce cost-related rationing of insulin. It succeeds on this criterion if insulin use among Part D enrollees rose after the limit compared with what would have happened without it.

Criterion 2, financial protection. It succeeds if out-of-pocket spending on insulin fell for the enrollees who had been paying more than the limit.

Criterion 3, spillover cost. It is acceptable on this criterion if any increase in Part D premiums attributable to the limit is small relative to the savings to insulin users.

A fourth possible criterion, improvement in diabetes control and complications, is set aside for now, because two or three years is too short for such outcomes to appear in any data. It is named so that its absence is visible.

The Comparison, Named

A figure from after the policy, such as the number of insulin prescriptions filled by Medicare enrollees in 2023, proves nothing on its own; insulin use might have risen anyway. An evaluation needs a comparison. The strongest one available comes from Myerson et al. (2023), who used a difference-in-differences design. They compared monthly insulin fills from September 2021 through April 2023 among Medicare Part D enrollees aged 65 to 74, who became subject to the limit in January 2023, with fills in a slightly younger group, adults 60 to 64 who had no Part D coverage and were about to age into Medicare but were not affected by the limit. If the two groups were following similar trends before 2023, a divergence after January 2023 can reasonably be attributed to the policy.

Its Own Required Numbers

On access, the study found that after the limit took effect, total insulin fills among the Medicare group rose by an adjusted 21,719 fills per month relative to the comparison group (95% CI, 20,797 to 22,641), and fills with cost-sharing of $35 or less rose by 49,515 per month (95% CI, 48,480 to 50,549) (Myerson et al., 2023). The second figure shows that the policy changed what people paid at the counter; the first shows that it changed how much insulin people filled. On Criterion 1, the evidence supports success, at least in the first months.

On financial protection, there is an estimate but not yet a measurement. Ahead of the start date, an HHS issue brief applied the limit to 2020 claims and found $734 million in Part D savings spread across roughly 1.5 million insulin users (Sayed et al., 2023). That is a projection from past claims, not an observation of what happened after 2023. Criterion 2 is plausibly met, and the rise in fills at $35 or less is consistent with it, but it has not been directly measured in the sources available to me.

On spillover cost, I found no study that isolates the effect of the insulin limit on Part D premiums. The limit took effect at the same time as other parts of the Inflation Reduction Act's Part D redesign, which also affect premiums, so any premium change in 2023 through 2025 has many causes. Criterion 3 cannot yet be judged.

The Study and Its Sponsor

The main evidence comes from a short research letter using IQVIA's National Prescription Audit, a commercial prescription database. The authors disclosed their conflicts: one reported serving as a paid consultant to Public Citizen's Health Research Group, and another reported grants from a range of pharmaceutical companies, insurers, health systems and foundations. These disclosures point in different directions and do not suggest a particular bias in the finding, but a reader should know them, and should know that the data come from a commercial vendor rather than from Medicare claims.

The format matters too. A research letter reports a single main analysis in a few hundred words, with little room for checks such as testing whether the two groups' trends really were parallel before 2023. That does not make the finding wrong, but it means the evaluation rests on one short report. A longer study using Medicare's own claims, with the pre-policy trends shown, would make the access finding much firmer.

What This Cannot Settle

The authors named three limitations: the short period after the policy, the lack of demographic data, and the possibility that other changes happening at the same time affected insulin use (Myerson et al., 2023). The third deserves weight. From 2021 through 2023, the Part D Senior Savings Model already offered $35 insulin in participating plans (Centers for Medicare & Medicaid Services, n.d.). Some of the Medicare group was therefore already paying $35 before January 2023, which would make the measured effect smaller than the effect on people who were newly protected. And the younger comparison group may include many with employer coverage whose own insulin costs were changing because of state caps and manufacturer programs. If their fills also rose for those reasons, the true effect of the Medicare limit would be larger than measured; if their fills fell, it would be smaller.

None of this can settle whether the policy improved health. More insulin filled is a reasonable intermediate outcome, but it is not the same as better glucose control or fewer complications.

The Judgment, and Its Trigger

My judgment is that the Medicare Part D insulin limit is succeeding on its central aim, access, with moderately strong evidence from a well-chosen comparison; that it is probably succeeding on financial protection, with evidence that is projected rather than measured; and that its spillover cost is unknown. On balance, the policy should be kept, and nothing in the evidence argues for changing it.

The judgment comes with a trigger that would reopen it. If a study using Medicare claims through 2025 found that insulin fills among enrollees returned to their pre-2023 trend, or that the limit accounted for a meaningful share of Part D premium growth, the evaluation should be repeated. The 2026 change in the statutory formula, which lowers the limit below $35 for insulins whose negotiated price is low, is a natural point for that next evaluation.

References

Centers for Medicare & Medicaid Services. (n.d.). Part D senior savings model. https://www.cms.gov/priorities/innovation/innovation-models/part-d-savings-model

Inflation Reduction Act of 2022, Pub. L. No. 117-169, 136 Stat. 1818 (2022).

Myerson, R., Qato, D. M., Goldman, D. P., & Romley, J. A. (2023). Insulin fills by Medicare enrollees and out-of-pocket caps under the Inflation Reduction Act. JAMA, 330(7), 660-662. https://doi.org/10.1001/jama.2023.12951

Sayed, B. A., Finegold, K., Olsen, T. A., De Lew, N., Sheingold, S., Ashok, K., & Sommers, B. D. (2023). Insulin affordability and the Inflation Reduction Act: Medicare beneficiary savings by state and demographics (Issue Brief). Office of the Assistant Secretary for Planning and Evaluation. https://aspe.hhs.gov/reports/insulin-affordability-ira-data-point