A Framework for Categorizing and Analyzing Prescription Drug Pricing Reform Options

Over the last several years, members of the public as well as policymakers have expressed growing concern over the problem of high prescription drug prices. Nearly one in four Americans reports difficulty affording their prescription medication, and system-wide costs for payers are rising as well. Several Congressional committees have investigated a range of particularly noteworthy cases, including Martin Shkreli’s actions increasing the price of Daraprim, Mylan’s price increases for the EpiPen over time, AbbVie’s efforts to keep competitors to Humira off the market, and the increasing list prices of insulin products. Along with these investigations, there is a renewed focus on prescription drug payment and coverage reform that would generate savings used to offset coverage expansions, infrastructure spending, or other priorities. But Congress has yet to coalesce around significant drug pricing reforms that would address important drivers of these problems.

In this post, we present a framework for categorizing and analyzing a wide range of proposed policy reforms for prescription drug pricing. We argue that policymakers should think about reform options as fitting within three sets of categories: (1) those focusing on patients’ out-of-pocket costs; (2) those targeting problematic incentives for various stakeholders within the system; and (3) those addressing the high prices of particular drugs.

Policymakers have many options to choose from within each of these categories, and given the fragmentation of the American health care system, multiple options within each category will be needed to ensure a broad group of patients benefit from the reform efforts. But it will also be important for policymakers to choose multiple options across categories. Choosing reform options primarily within a single category is unlikely to sufficiently address the many problems at play, and may even exacerbate other dimensions. For instance, reducing drug prices should help lower patient out-of-pocket costs, but solely reducing out-of-pocket costs can exacerbate problematic incentives and lead to higher prices. Throughout, this blog also discusses the trade-offs inherent in the myriad policy options, including between different types of innovation being encouraged or discouraged.

Reforms that would reduce patients’ out-of-pocket costs can be important to avoid some of the harmful health consequences patients may experience when they are unable to take their medications as prescribed and the financial consequences even if they are able to do so.

The most prominent efforts to address out-of-pocket costs at the federal level would take direct aim at Medicare Part D, which unlike private insurance, currently lacks an out-of-pocket spending limit for enrollees. Once patients hit the catastrophic phase of the Part D benefit structure, patients are responsible for 5% of the list price of their medications, without limit. Today, more than a million Part D beneficiaries have out-of-pocket spending above the catastrophic threshold in a given year, and, according to estimates from the Kaiser Family Foundation, 2.7 million people had spending above the threshold in at least one year over a five-year period (2015-2019).

Several Congressional proposals would impose annual caps on Part D out-of-pocket costs. The House Democratic proposal, H.R. 3, imposes a $2,000 cap on these costs. The Senate Finance Committee’s bipartisan approach from 2019 and the current House Republican proposal, H.R. 19, both cap patient spending at $3,100. Others have proposed additionally tying drug cost-sharing in Medicare Part D to net rather than list prices, which would directly benefit enrollees taking drugs with large rebates, such as insulin and Hepatitis C treatments. Although these proposals would be very impactful for certain Part D beneficiaries with high out-of-pocket costs, they would not be helpful for commercially-insured or uninsured patients with high medication costs.

Other proposals have targeted patients’ out-of-pocket costs for specific products, most notably insulin. At the federal level, the Trump Administration implemented a voluntary demonstration project in which beneficiaries in some enhanced Medicare Part D plans would be eligible to access certain insulin products for a flat fee of $35 per month, so long as both insulin manufacturers and the Part D plans elect to participate. States across the political spectrum have also passed laws to impose out-of-pocket limits on insulin, often applying to  uninsured patients as well. These programs typically do not target the underlying prices of insulin products, though, meaning that the costs of the drugs will be borne elsewhere in the system (such as in the insurance premiums paid by consumers, employers and governments).

Additionally, some of the broader coverage expansions under consideration would substantially reduce the out-of-pocket cost of prescription drugs for those newly gaining insurance coverage. Enhancements to cost-sharing subsidies for ACA Marketplace enrollees would similarly help lower out-of-pocket costs on prescription drugs, along with other services.

These proposals and others like them would matter greatly for patients needing high-priced drugs. However, these proposals do not directly address the high prices of those drugs themselves. Instead, they largely shift costs within our system, and may even lead pharmaceutical companies to raise drug prices if they know that higher prices will no longer impair patients’ ability to fill their prescriptions. Enacting reforms in this category alone, therefore, will generally increase prices, overall premiums, and spending.

The current legal and regulatory structures around prescription drug pricing and spending create incentives for different actors to drive drug prices and spending up over time, rather than down. Policymakers have identified many of these misaligned incentives and have proposed legislative and regulatory changes to address these issues. In this section, we detail some of the most prominent reform options available.

Currently, in the catastrophic phase of the Part D benefit, patients are responsible for 5% of costs, plans are responsible for 15%, and Medicare for 80% of the costs. Such limited liability for health plans in the catastrophic phase weakens their incentive to manage drug spending and in some cases undermines plans’ incentive to steer patients towards lower-priced drugs, likely enabling prices to remain high or even increase.

A number of proposed benefit redesigns, including the proposed Part D benefit redesigns in H.R. 3, H.R. 19, and the Senate Finance Committee package, would substantially reduce Medicare’s own responsibility in that phase, increase plans’ responsibility, and provide new financial obligations for manufacturers. For instance, H.R. 3 would reduce Medicare’s 80% responsibility to 20%, increase plans’ responsibility to 50%, and create a new 30% obligation for manufacturers.

Several proposals in this category would take aim at the issue of list-to-net spread. For certain drugs with therapeutic substitutes, both list prices and rebates have grown rapidly over time. Even where net prices have remained more stable, these increases in list prices create significant challenges principally but not only for patients whose cost-sharing is often tied to these ever-increasing list prices.

One potential response, included in a Congressional proposal put forward by Senator Ron Wyden (D-OR), would require Medicare Part D plan sponsors to tie patients’ cost-sharing to a drug’s net price (or an approximate thereof) rather than its list price. The Trump Administration proposed a similar approach in their FY2019 budget, but the Congressional Budget Office estimated that it would increase cumulative ten-year deficits by $43.4 billion, primarily because it would make the Part D benefit more generous. (Only a portion of the lower coinsurance costs would be recouped through higher premiums). Proposals like this could significantly reduce cost-sharing for patients needing medications with high list-to-net spreads, but would not meaningfully assist patients using high-cost drugs with small, if any, rebates.

Another proposal would aim to more directly discourage companies from raising the prices of their products over time more rapidly than inflation, by requiring them to pay rebates back to Medicare for doing so. This proposal, contained within both the Democratic-led H.R. 3 and the bipartisan Senate Finance Committee package, would bring into Medicare a policy tool that Medicaid already uses to manage increases in drug reimbursements. This proposal responds to unusual features of the prescription drug market that drive prices up rather than down over time, including exclusivity rights and patients’ challenges with changing drugs even where alternatives exist. In turn, this policy also goes toward addressing high underlying prices of certain drugs by encouraging them to increase more slowly than would otherwise have occurred. However, some expect these inflationary rebates to lead to somewhat higher drug launch prices going forward, modulating the rebates’ impact.

Other proposals would attempt to address the incentive in Medicare Part B to prescribe higher cost drugs. Currently, Medicare Part B reimburses providers for administering drugs or biologics based on its average sales price (ASP) – a price net of rebates and other price concessions – plus an additional six percent of the ASP to compensate them for acquiring and holding a drug in inventory. This payment structure creates an incentive to prescribe the more expensive drug when there are competing options, and evidence suggests that providers do respond to this incentive. Various proposals, including President Obama’s Fiscal Year 2017 Budget and CMMI efforts from both the Obama and Trump administrations, have sought to eliminate or ameliorate this perverse incentive by reducing the add-on payment and, in some cases, also converting it to a flat fee.

A number of separate proposals would target strategies used by pharmaceutical companies to delay generic entry, including pay-for-delay settlements (in which brand and generic or biosimilar firms reach settlements designed to extend the brand firm’s time without competition), product hopping (in which a manufacturer introduces a new form of their product in an effort to undermine the effect of an impending generic competitor), and citizen petition abuse (in which innovator firms file largely frivolous petitions with the Food & Drug Administration (FDA) in an effort to delay the agency’s approval of generic competitors). These proposals, which often have bipartisan co-sponsorship, are an attempt to more rigorously enforce the social contract embodied by our patent and exclusivity laws. That is, although companies receive exclusive rights from the federal government for their inventions, those rights are time-limited and policymakers have generally expected that, upon the expiration of patent or other exclusive rights, generic competitors will enter the market and bring down drug prices. But as companies are able to use the intellectual property system to secure their monopoly rights for twenty or even thirty years without competition, policymakers have begun to push back. These reactive proposals serve important roles in ameliorating the types of conduct they specifically enumerate. However, more will be needed, as these proposals are unlikely to prevent industry from developing additional, novel methods of extending their monopoly periods in the future.

A related set of criticisms has focused on the FDA’s accelerated approval program, which enables drugs for serious illnesses to come to market on the basis of improvements in surrogate endpoints (e.g. reduced cholesterol levels) that are “reasonably likely to predict” a clinical benefit, rather than true improvements in patient outcomes (e.g. a reduction in heart attacks). Although FDA requires manufacturers under this program to complete post-market studies to confirm that these products offer clinical benefits, many observers have argued that companies frequently fail to complete these trials within a reasonable period of time. In the meantime, there are no restrictions on companies’ pricing for their products, and many insurers are required to cover them on the same terms as they must cover products that have demonstrated clinical benefits. For instance, an Alzheimer’s treatment, Aduhelm, was recently granted accelerated approval despite the fact that no members of FDA’s own advisory committee thought the evidence was sufficient to support approval. Yet its manufacturer, Biogen, still set a $56,000 list price for the treatment that could add hundreds of billions of dollars to Medicare and Medicaid spending before any post-market review is complete; FDA has given Biogen nearly nine years to submit the results of its follow-on clinical trials.

Expressing concern about the impact of the accelerated approval pathway on state Medicaid programs, which must carefully manage their budgets, the Medicaid and CHIP Payment and Access Commission has proposed changing the terms on which state Medicaid programs provide payment for accelerated approval drugs, specifically by increasing the mandatory minimum Medicaid rebate either beginning at approval and lasting until clinical benefits have been demonstrated or beginning a certain number of years after approval, if confirmatory trials remain ongoing. Although outside the scope of existing proposals, similar minimum rebates also could be required in the Medicare program and to be offered to payers in the commercial market. Alternatively, if a drug price negotiation system is set up, reimbursement rates for these drugs could be set through that process, taking into account the lack of clinical evidence supporting the drug’s approval.

While the above ideas are important for the many misaligned incentives within our prescription drug pricing system, they largely fail to address the very high prices for many drugs. As lawmakers become more concerned about the problems of older drugs like Humira that still lack competition after decades on the market and highly-priced drugs that provide little value over existing treatments, they may consider reforms that would strengthen payers’ negotiating authority and enable payers to obtain better prices for these products, or going a step further to administratively set limits on what payers will reimburse for certain drug products.

Many such reforms offer the potential to reduce government, employer, and consumer spending on prescription drugs. Depending on the type and scale of the negotiation program, the ensuing lower expected profits for manufacturers of affected newly-developed drugs may also reduce the number of drugs that get developed, although the magnitude of any such effect is highly uncertain. Further, whether such innovation effects are concerning is highly dependent on the types of innovation being encouraged or discouraged – the value to patients will be different if the foregone products would have entered already crowded therapeutic classes than if they would have represented new treatment options.

In this section, we begin by discussing broad approaches to reducing drug pricing such as mandating minimum rebates beyond Medicaid, and close by discussing different more targeted price regulation options.

A number of proposals would require manufacturers to offer a minimum rebate for drugs sold to Medicare Part D patients (similar to what exists in the Medicaid program), with the minimum rebates sometimes restricted to enrollees in Part D’s low-income subsidy program who largely would have received drug coverage through Medicaid before Part D’s creation. Policymakers could also require minimum rebates be offered to payers in the commercial market.

H.R. 3 enables the Secretary of HHS to determine a maximum price that can be paid by Medicare to manufacturers of selected high-priced drugs and requires that manufacturers also offer this price to private payers in the commercial market. H.R. 3 establishes a ceiling (equal to 120 percent of the average international market price across six peer countries) and a floor (the lowest of the six peer country prices), with negotiation between HHS and the manufacturer determining the price within this range. HHS is directed to consider a range of other factors, including but not limited to information about the drug’s comparative effectiveness and whether the drug is a therapeutic advance over existing treatments. If the manufacturer and HHS fail to reach an agreement on a price, the manufacturer faces an escalating tax imposed on all U.S. sales of the drug, beginning at 65% and increasing to 95%.

Public discussion has often focused on the set of drugs to which a process of this type should apply. H.R. 3 limits this process to drugs that are either among the 125 highest-spend drugs in Medicare Part D, among the 125 highest-spend drugs in the United States across all markets, or are an insulin product. This focus likely will provide HHS with significant opportunities for achieving cost-savings. In practice, this limitation to drugs with high spending would also tend to focus the Secretary’s efforts on older products, which have had enough time on the market to garner a large enough patient population, but the bill is not formally restricted to older drugs. An implication is that newly introduced drugs would have many years before such restrictions in price would apply to them, helping limit any impacts on the number of new drugs coming to market.

Policymakers may further target their selection of drugs for analogous programs depending on their policy goals, including targeting drugs where existing market mechanisms are likely to be less effective at obtaining price discounts.

• Lower-Value Drugs
  For instance, negotiation could target lower-value drugs that provide patients with little clinical benefit, using cost- and comparative-effectiveness data to identify value-based prices for these products (potentially also incorporating drugs receiving accelerated approval until a clinical benefit is proven, as discussed in the previous section). Basing drug price negotiations on domestic analysis of cost- and comparative-effectiveness might focus price reductions on lower-value drugs and potentially even increase the incentives to develop high-value drugs.

• Older Drugs
  Negotiation might more explicitly target drugs that have been on the market for several years, either identifying a specified number of years or tying negotiation to the expiration of the original patents or FDA-administered exclusivity periods. Price reductions focused on older drugs that still do not have competition should have less impact on incentives for new drug development given both the greater uncertainty in revenue for a drug after many years on the market and the time value of money – such a focus can also help ameliorate the effects from manufacturer attempts to delay generic or biosimilar entry, as discussed in the previous section. Inflation rebates, discussed in the previous section, would also typically reduce net prices most for older drugs facing limited competition given their compounding nature over time.

• Orphan Drugs
  Some scholars have argued for negotiation or price regulation targeted at older orphan products, where the market size is likely to be insufficient to support the type of competition that would ordinarily drive down prices as patents and exclusivity periods expire.

• Physician-Administered Drugs
  Lastly, negotiation schemes might give a priority to  physician-administered drugs given the more limited tools payers have to negotiate price discounts for such drugs. For instance, physician-administered drugs could be given a preference for negotiation under a process like that in H.R. 3. Such a process could be used to determine the Medicare Part B reimbursement amounts (rather than using the ASP today) or to negotiate a price the must be offered to all payers (including providers), which would additionally allow employers, commercially-insured individuals, and the uninsured to benefit and avoid some administrative complexity.