Do Medicare Add-on Payments Improve Access to High-Cost Drugs in Hospitals Serving Poor Communities?

By Jermaine Piper, MPH, Inna Cintina, PhD, and Lane Koenig, PhD
November 12, 2025

High drug prices are a growing concern among patients and hospitals alike – with the median annual price for new drugs increasing by 35% to $300,000 between 2022 and 2023. The high costs for new drugs create challenges, particularly for those hospitals that rely on Medicare and other government funding sources, which tend to pay below hospital costs. Because Medicare payment rates do not adjust in the initial years after a new drug is introduced, hospitals that provide lifesaving, yet expensive, new drug therapies might incur large financial losses under the inpatient prospective payment system.

Residents of socially deprived communities, as indicated by the Social Deprivation Index (SDI), are more likely to rely on public health insurance options such as Medicare and Medicaid. Hospitals that serve a higher proportion of Medicare and Medicaid patients tend to have lower operating margins compared to those serving higher proportions of commercially-insured patients. As a result, hospitals providing care for lower-income communities may have difficulty providing access to new, high-cost drugs.

The Centers for Medicare & Medicaid Services’ (CMS) New Technology Add-on Payment (NTAP) program provides hospitals with an additional payment to recoup some of the financial losses for drugs (or devices) that are new, costly, and clinically effective. Typically, the NTAP reimburses hospitals either 65% of the cost of the drug or 65% of the money lost in providing care to a patient–whichever is less. The NTAP designation improves the ability of hospitals to provide new drugs to Medicare beneficiaries on an inpatient basis. However, increased reimbursements through NTAPs may not be sufficient to ensure access to new, high-cost drugs in hospitals serving poor communities.

In this blog, we assess the extent to which the adoption of NTAP-designated drugs is similar across communities of varying social deprivation. We report, by a hospital’s SDI, Medicare beneficiaries’ utilization of three NTAP-designated drugs: (1) CABLIVI® – a medication that treats a life-threatening blood disorder; (2) EluviaTM – a drug eluting stent used to treat peripheral arterial disease; and (3) XOSPATA® – a medication used to treat acute myeloid leukemia (AML). These drugs were selected, in part, due to their high prices: the maximum NTAP for these drugs (which represents 65% of their wholesale acquisition costs) ranged from $3,647 (EluviaTM) to $33,215 (CABLIVI®).  We identified each drug in inpatient fee-for-service Medicare claims based on assigned MS-DRG and ICD-10-PCS procedure codes. To calculate procedure-specific NTAP utilization rates, we identified inpatient admissions with these procedures and divided that number by the admissions potentially eligible to receive that treatment based on the drug’s therapeutic indication and diagnoses on a claim.

Areas with Higher Social Deprivation saw Higher Utilization of CABLIVI and XOSPATA®, but Lower Utilization for EluviaTM . We found inconsistent utilization rates of NTAP-designated drugs by SDI. XOSPATA’s utilization rate in the highest SDI quartile (i.e., patients treated by hospitals located in the most socially deprived areas) is nearly twice as large as in other quartiles. On the other hand, Eluvia™, the least expensive of the three therapies, has the highest utilization among providers practicing in lower SDI areas.

 

Utilization Rates of NTAP-Designated Drugs by SDI in FY 2020-22 (per 1,000 Admissions)

Despite the high cost of these new drugs, the observed used patterns do not suggest systematic underuse of selected drugs by hospitals located in areas with high levels of social deprivation. These findings may indicate that additional payments under the NTAP program help support access to novel treatments.

There are several important limitations to the analysis. First, our results may not generalize to other NTAP-designated drugs. Second, MedPAR claims data do not provide all the information needed to establish whether a beneficiary is a good candidate for the drug. Thus, the findings may reflect differences in patient needs across geographic areas. Third, the analysis may not capture patients who have the qualifying condition but have not sought out treatment. Finally, we define SDI based on hospital location and not on where a beneficiary lives. Teaching hospitals, for example, may be more likely to use these new treatments and tend to be located in large, urban areas, but draw patients from many locations.

Services : Blog, Payment Policy & Delivery System Innovation Expertise: , ,