High versus low-cost intravenous iron in Medicare: Utilization patterns of iron infusions in traditional fee-for-service Medicare and Medicare advantage for iron deficiency anemia Free

Background

Iron deficiency anemia (IDA) can be managed with intravenous (IV) iron when oral supplements are insufficient. There are several IV formulations available on the market that differ in price, infusion time and adverse effects. Given the differences between Medicare Advantage (MA) and Part B Traditional Fee-for-service Medicare (TM), research examining how medication cost and reimbursement impact drug utilization have significant ramifications for the growing Medicare budget. As of 2023, MA covers the majority (51%) of the 63 million beneficiaries enrolled in Medicare [Neuman, KFF Medicare Jan 2024]. MA offers a variety of plan types including healthcare maintenance organizations (HMO), preferred provider organizations (PPO), private fee-for-service (PFFS), point of service (POS) and special needs plans (SNPs) [Jacobson, Commonwealth Fund Oct 2021]. In contrast, TM only offers a classic fee-for-service model. Research on outpatient drug administration has found that MA plans promote greater use of lower cost alternatives compared to TM [Anderson, Health Services Research Nov 2021; Kozlowski, JAMA Health Forum Dec 2023]. Several IV iron formulations are available and include low-cost iron sucrose (200 mg), dextran (1000 mg) and ferric gluconate (125 mg) priced in Medicare at $46, $284 and $21, respectively per average clinic administered dose [Data.cms.gov, Medicare Part B Spending by Drug]. In comparison, ferric derisomaltose (1000 mg), carboxymaltose (750 mg) and ferumoxytol (510 mg) are higher cost pricing at $2,570, $765 and $459, respectively. Our study aims to compare utilization of IV iron formulations in TM to MA.

Methods

We performed a retrospective cohort analysis using 2021 claims linked to ICD-10 codes for IDA. Part B TM claims were drawn from the Medicare Beneficiary Summary File (MBSF) and MA claims from the Part C Medicare Encounter Data. Infusions were identified with HCPCS codes for ferric carboxymaltose, gluconate and derisomaltose; iron sucrose and dextran; and ferumoxytol. Beneficiaries were categorized by plan type as TM, MA-HMO, MA-PPO and MA Medicaid dual eligible plans. Hybrid HMO-POS plans were included in MA HMO. Encounters related to SNP, PFFS, Cost Plans, Medicare Savings Account (MSA) and at dialysis centers for end-stage renal disease (ESRD) were excluded. The 2021 average sales price (ASP) per unit was collected from Medicare Part B drug data at data.cms.gov.

RStudio 2024.12.1 was used to develop descriptive statistics and analysis via multinomial logistical regression using ferumoxytol as a reference for natural log odds ratio calculations due to its mid-range price point across formulations.

Results

In 2021, a total of 14,106,578 non-ESRD IV iron infusions were used for IDA across MA and TM. The most common infusions were iron sucrose (43%), ferric carboxymaltose (23%), ferumoxytol (18%), iron dextran (8%), ferric gluconate (6%) and derisomaltose (1%). The relative utilization of these varied greatly by plan type. For TM, utilization was comprised of both high and low-cost formulations. TM was 2.6 times more likely to use iron sucrose (OR 2.6, 95% CI [1.1-7.6*e^91]), 1.3 times more likely to use iron carboxymaltose (OR 1.3, 95% CI [1.1-5.9*e^18]) and 70% (OR 0.3, CI [7.1*e^-71,1.0]) less likely to use iron dextran. Conversely, MA-HMO favored low-cost formulations and were 3 times more likely to use iron dextran (OR 3.1, CI [1.1-3*e^55]) and 2 times more likely to use ferric gluconate (OR 2.0, CI [1.1-5.4*e^33]). MA-Medicaid selected strongly for low-cost formulations and were 2.8 times more likely to use iron dextran (OR 2.8, CI [1.1-1.1*e^5]), 6.5 times more likely to use ferric gluconate (OR 6.5, CI [1.1-1.8*e^11]) and 5 times more likely to use iron sucrose (OR 5, CI [1.1-1.1*e^12]). MA-PPO plans were 1.5 times more likely to use ferric derisomaltose (OR 1.5, CI [1.1-3.1*e^10]) and 1.6 times more likely to use iron dextran (OR 1.6, CI [1.1-1.08*e^5]).

Conclusions

TM and MA showed wide variation in IV iron usage in 2021. Our results show MA Medicaid and MA-HMO plans were 2-6 times more likely to use low-cost iron relative to other formulations. Conversely, TM and MA-PPO were similar in utilizing both high and low-cost formulations. This could be due to incentive differences offered by various plan structures in MA relative to TM. Future research should assess the downstream implications of these patterns overtime, on outcomes and on cost-effectiveness.