Medicaid Expenditures and Estimated Rebates for Epinephrine Autoinjectors, 2012 to 2016

The epinephrine autoinjector EpiPen received considerable scrutiny in the past year because Mylan, its manufacturer, increased its list price to $609 even though the only ingredient, epinephrine, was discovered over 100 years ago (the delivery system is patented). Mylan recently announced a $465 million settlement with the Department of Justice (DOJ) to resolve a fraud investigation,¹ alleging that Mylan misclassified EpiPen as a generic drug for the Medicaid Drug Rebate Program² to grant the government a lower rebate per unit sold. For brand-name drugs, the manufacturer pays a higher base rebate amount and an additional rebate connected to price increases above inflation. Since the economic impact of the misclassification of EpiPen by Mylan remains controversial,³ we used publicly available data to estimate the amount of money Medicaid lost due to Mylan’s classification of EpiPen as a generic product, to help policymakers evaluate the value of the proposed settlement.

We obtained volume and expenditure data for EpiPen, EpiPen Jr, EpiPen 2-Pak and EpiPen 2-Pak Jr since 2007, using the Medicaid Drug Utilization database.⁴ We calculated rebates under brand-name and generic scenarios for the latter 2 products by merging National Average Drug Acquisition Costs (NADAC) with Medicaid volume from the fourth quarter of 2012 to the first quarter of 2016. NADAC is derived from surveys of invoiced prices at retail community pharmacies and helps Medicaid directors evaluate pharmacy payments.⁵ We could not calculate rebates from 2007 onwards when Mylan acquired the product because the NADAC did not exist prior to 2012. We calculated the generic rebate amount per unit by multiplying the NADAC at each quarter by 0.13, the standard rebate for generic drugs. We multiplied the generic rebate per unit by the total volume per quarter to obtain the total generic rebate amount. The brand-name rebate calculation had 2 components: a base rebate amount (quarterly NADAC multiplied by 0.231)⁶ and an additional rebate amount to account for increases in price above inflation as measured by the Consumer Price Index for all urban consumers (CPI-U). We estimated this by subtracting the CPI-U inflation adjusted baseline NADAC (defined in October 2012) from the quarterly NADAC. We then added the base and additional rebate amounts per unit to obtain the total rebate amount per unit for the brand-name scenario. Finally, we multiplied the total rebate amount per unit by the total volume per quarter to obtain the expected brand-name rebate amount.

EpiPen volume and expenditures per unit grew substantially after 2007 (Figure 1A and B). We estimate that the total generic rebate amount paid by Mylan from the fourth quarter of 2012 to the first quarter of 2016 was $112.2 million (Figure 2). If Mylan had correctly classified EpiPen 2-Pak and EpiPen 2-Pak Jr as brand-name products, the total rebate amount would have been much greater—at least $538.3 million ($199.3 million from the base rebate amount of 23.1% multiplied by NADAC and an additional $339 million due to price increases above inflation). Thus, Mylan avoided paying Medicaid a minimum of $426.1 million in rebates by classifying its product as generic.

Our estimate of $426 million in excess Medicaid spending for epinephrine autoinjector sales covers only 2 formulations and the years 2012 through 2016. Therefore, we believe that the proposed $465 million settlement underestimates the actual cost of Mylan’s strategy, pointing to the limits of litigation as a way of recovering taxpayer funds. Other factors that may have affected the proposed settlement amount include supplemental rebate agreements, litigation costs, and uncertainty regarding the government’s responsibility. Despite these limitations, a better way to avert such profit-maximizing strategies would be to prevent manufacturers from classifying their own products as generic or branded for Medicaid rebate purposes.⁷

Corresponding Author: Jing Luo, MD, MPH, Program on Regulation, Therapeutics, and Law, Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital, 1620 Tremont St, Ste 3030, Boston, MA 02120 (jluo1@partners.org).

Correction: This article was corrected on May 1, 2017, to fix a typographical error in the Results section.

Published Online: March 27, 2017. doi:10.1001/jamainternmed.2017.0257

Author Contributions: Dr Luo had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Luo, Kesselheim.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Luo, Kesselheim.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Luo.

Obtained funding: Kesselheim.

Administrative, technical, or material support: Avorn.

Supervision: Avorn.

Conflict of Interest Disclosures: Dr. Kesselheim’s work is supported by the Laura and John Arnold Foundation, with additional support from the Engelberg Foundation, and Harvard Program in Therapeutic Science.

Additional Information: We thank Michael Sinha, MD, JD, MPH, and Sean Dickson, JD, MPH, for research assistance. They did not receive compensation for their contributions.