Who pays in Medicare Part D? – It’s time to give plans more skin in the game

Americans rank “taking action to lower prescription drug prices” as their top priority for Congress this year. [1] Policymakers seem to be listening. Proposals have emanated from Senate and House committees, the Trump administration, and several Democratic presidential candidates. While the specifics vary, many of them share a common theme: have the federal government take a more active role in determining drug prices. There is a sense the private market cannot wring enough savings out of the system. But before giving up on the current structure, it is worthwhile to assess the role that private plans play in the market today and whether incentives might be better aligned to achieve lower drug prices and other social objectives.

The Medicare Part D program – created by the Medicare Modernization Act of 2003 – is ground zero for policy consideration. Part D was a nationwide experiment in using private health plans to expand coverage with substantial public subsidies – and a precursor to the experiment with the Affordable Care Act. Today, Part D covers 45 million beneficiaries and costs the federal government $85 billion annually. [2] In contrast to other parts of the Medicare program, Part D is run and administered entirely by private plans. In order to gain legislative approval, the law explicitly prohibited the federal government from directly negotiating or setting drug prices in Part D, instead relying entirely on competition among private plans in designing formularies and negotiating with manufacturers to enhance choice and lower prices. However, Part D’s design also limited the incentives of these private firms to manage costs, particularly among high-cost beneficiaries.

Mostly, this lack of incentives reflects the concerns in Congress about whether the Part D market would even get off the ground. Prior to the 2003 legislation, a market for stand-alone prescription drug plans did not exist, and two policy questions were paramount. First, would private plans choose to participate in this new market? Second, would beneficiaries – especially healthier ones – choose to enroll in a voluntary program? Similar questions arose around the Affordable Care Act, although the architects of that law included a mandate to address adverse selection. A public-private partnership depends on voluntary participation by plans, and policymakers in the Part D debate took this concern very seriously. [3]

One result of these concerns was that the enacting law for Part D included several features to limit the financial risk to plans – especially due to high cost beneficiaries. Key among these is the federal reinsurance program. Once a beneficiary reaches a catastrophic limit – corresponding today to about $8,000 in total drug spending – the federal government steps in to subsidize 80 percent of remaining beneficiary spending for the year. The plan’s liability falls to just 15 percent, with patients responsible for the remaining 5 percent. The net effect is that the government – not the plan – bears most of the burden for right-tail spending in the program.

In the early years of Part D, government reinsurance seemed reasonable. Less than 20 percent of total drug spending occurred among beneficiaries above the catastrophic limit, and total federal reinsurance subsidies were less than $10 billion. But the Part D standard benefit design has not kept up with the marketplace, particularly the emergence and growth of specialty drugs. Today, nearly half of total Part D spending occurs among beneficiaries above the catastrophic limit, and reinsurance subsidies amount to around $43 billion – representing the largest component of federal spending on the program. As reinsurance spending grows disproportionately, plans bear less and less risk as a share of total spending. [2] The situation calls into question what is being gained through the use of private plans.

In addition to the limited risk in catastrophic coverage, plans also bear very little liability for spending in the so-called coverage gap or “doughnut hole” phase. The Affordable Care Act and the Bipartisan Budget Act of 2018 addressed the high and problematic cost-sharing for beneficiaries by “filling-in” this doughnut hole, so beneficiaries no long face full cost-sharing as they did in the early years of Part D. But these changes have primarily relied on contributions from drug manufacturers – rather than adding to Part D plans’ liability.

Other Part D program features also limit plans’ ability to effectively manage spending. For example, plans use formularies with varied levels of beneficiary cost-sharing as a tool to negotiate drug prices and incentivize beneficiaries to use lower-cost options. But Part D plans’ ability to do so is limited by federal requirements that they cover all or nearly all drugs in six protected drug classes. Moreover, these financial incentives are ineffective for low-income beneficiaries – who account for about one-third or enrollment and half of total drug spending – for whom the federal government pays nearly all of their cost-sharing. While these policies reflect important priorities related to access and affordability, they also undermine plans’ ability to effectively negotiate and manage spending.

The net result is that, today, Part D plans are only actually responsible for about 34 percent of total prescription drug spending (Figure). In contrast, according to data from the Health Care Cost Institute, health insurers in the commercial market paid an average of 85 percent of total drug spending in 2016. [4]

Figure: Distribution of Medicare Part D Spending by Benefit Phase and Payer.

Notes: Analysis of a 100 percent sample of 2016 Medicare Part D claims data, accessed through Center’s for Medicare and Medicaid Services (CMS) Virtual Research Data Center (VRDC). All spending for claims that “straddle” multiple phases is assigned to the final benefit phase of the claim. For beneficiaries who do not receive low-income subsidies (non-LIS), the distribution of payer liability in the coverage gap is adjusted to reflect 2019 parameters (i.e., for branded drugs, liability is adjusted to: 70 percent paid by manufacturers, 25 percent paid by beneficiaries, and 5 percent paid by plans; for generic drugs, liability is adjusted to 37 percent paid by beneficiaries and 63 percent paid by plans). The figure includes spending for both LIS and non-LIS beneficiaries. TDC=total drug costs. TrOOP = “True” out of pocket costs, which includes manufacturer-paid discounts.

Thus, despite an intention to rely on private plans to administer the Part D program and design formularies to constrain drug prices, the market has evolved such that the federal government – and not the plans – is directly liable for the majority of Part D spending.

As policymakers debate the pros and cons of various proposals to reform the prescription drug marketplace, at least one idea that should clearly emerge is the need to update the Part D benefit design. Plans now have more than a decade of experience operating in this market, and there is no good economic reason for government to shoulder such a large portion of the risk that plans face. Policymakers should consider reforms that would give plans more insurance risk – such as increasing plan liability in the reinsurance and coverage gap phases. But these should be coupled with important beneficiary protections – like a cap on out-of-pocket spending – and changes that enable plans greater flexibility to use formulary tools. [5] The federal government could re-direct its subsidies to help keep premiums low, but if we want private plans to deliver on value to beneficiaries and taxpayers, it’s time to give them more skin in the game.

Private plans have the potential to lead the way toward innovative contracting approaches that emphasize value and, in doing so, deliver on the intended visions of the Part D marketplace. But they can only be expected to do so with the right incentives in place. It took policymakers forty years to add an outpatient prescription drug benefit to the Medicare program. Let’s hope we don’t have to wait forty more to modernize Part D.