This analysis is part of the USC-Brookings Schaeffer Initiative for Health Policy, which is a partnership between Economic Studies at Brookings and the University of Southern California Schaeffer Center for Health Policy & Economics. The Initiative aims to inform the national health care debate with rigorous, evidence-based analysis leading to practical recommendations using the collaborative strengths of USC and Brookings. The Commonwealth Fund has provided a grant to the Brookings Institution to support the work on which this post is based. This post originally appeared in Health Affairs on September 9, 2019.
Last fall, the Trump administration announced that it was exploring an international reference pricing model for drugs covered under Part B of Medicare (which are generally drugs administered by a physician, such as many chemotherapy drugs). Under this model, the amount Medicare would pay for these drugs would be set based on the average price paid in a selected group of other high-income countries.
The administration has argued for a focus on Part B drugs because of the limited role that competitive forces play in determining what Medicare pays for these drugs. Medicare Part D plans, Medicare Advantage (MA) plans, and commercial insurance plans all negotiate prices with manufacturers; they use a variety of tools—such as formularies, preferred tiers, step therapy, and prior authorization—to create leverage in those negotiations and to control costs directly. Medicare Part B, in contrast, does not use these tools directly.
The average sales price (ASP) that Medicare payment is based upon does incorporate competitively determined prices in the commercial market. (ASP reflects the average price, net of discounts and rebates, received by the manufacturer for virtually all sales in the US.) Medicare, however, represents a very large share of total revenue for many Part B drugs, weakening the competitive pressures on physician-administered drug prices compared to outpatient drugs that face these competitive pressures across more lines of business. The Trump administration highlights that prices for Part B drugs in the US appear to be roughly 80 percent higher than in other higher-income countries.
Notwithstanding the fact that Part D drugs face greater competitive pressures than Part B drugs, the US also appears to pay much higher prices for Part D drugs than other countries. While international drug price comparisons are challenging, two studies published in Health Affairs suggest that net drug prices in Part D and systemwide in the US are significantly higher than in other wealthy countries.
The existence of relatively high drug prices across US markets raises the question of whether it would be appropriate to extend the Trump administration’s international reference pricing approach beyond Medicare Part B to help control drug prices in Part D and outside of Medicare—especially for drugs without therapeutic alternatives, which face minimal competitive pressures (something recent reports suggest the administration is considering). On the other hand, using these tools in Part B or beyond to reduce drug prices would reduce the expected financial return to developing new treatments, which could reduce development of new drugs. While the magnitude of any reduction in new drug development is uncertain, if drug development exhibits sufficiently strong diminishing marginal returns as drug spending increases, then this reduction in innovation could be relatively small.
The rest of this post discusses considerations for expanding international reference pricing to Medicare Part D, the commercial market, and Medicaid; technical difficulties common to its application to any market; and the impact international reference pricing might have on access to drugs, international drug prices, and drug manufacturer revenue.
Expanding International Reference Pricing Beyond Medicare Part B
One important design decision in extending an international reference pricing model is determining which markets and which drugs should be included. In this section, we briefly discuss the considerations that would arise in extending such a model to all payers, to Medicare Part D, or to Medicaid, and then discuss what drugs might be encompassed in an expanded model.
A conceptually simple approach would use international reference pricing to set maximum prices across all markets for either all brand drugs or a subset of them. In this case, US drug list prices could simply be tied to the reference price as a condition of the manufacturer being allowed to sell any products in the country. Payments above the list price (in any form) would be prohibited, but payers could use all the tools they have to negotiate discounts below these new list prices, leaving market mechanisms intact under a price cap based on the reference price.
In addition to conceptual simplicity, the argument for this broadest approach is that US brand drug prices appear higher than international ones across all lines of business and for most classes of drugs. To the extent that US prices are lower for a certain drug, then that price would be unaffected by the reference pricing model. This approach would require legislative action, as the systemwide reach and specific requirements of the approach are beyond the scope of the demonstration authority of the Center for Medicare and Medicaid Innovation (the Innovation Center) and would constitute a marked change from current pricing.
Medicare Part D
As noted above, research suggests that drug prices in Part D, just like in Part B, are notably higher than those internationally. This can be viewed as an argument for the Innovation Center demonstration authority to extend the Trump administration’s proposed reform to self-administered outpatient drugs under Part D. Such an extension would have far reaching effects because many more drugs would be affected than under the administration’s proposal and spending on Part D drugs far exceeds that on Part B drugs. The Innovation Center authority can be viewed as similar in Part B and Part D, although there are significant limitations on this authority that would pose meaningful shortcomings compared to using legislation, as discussed further below.
One way to operationalize reference prices in Part D would be to require a minimum rebate for each affected drug. This amount would be automatically adjusted to any increase in a drug’s list price and recalculated each year, as a condition of the manufacturer being permitted to sell drugs in the Part D program. In conjunction, additional payments to manufacturers from Part D plans would be banned. As an alternative, payments (in any form) above the reference price to the manufacturer could be prohibited.
Drug payment reductions based on international reference pricing for Medicare or the commercial market may automatically flow through to Medicaid through the “best price” rule—the requirement that Medicaid receive the best price offered to any drug purchaser—depending on how reference price rules are implemented. To the extent that a policy goal is to have Medicaid also benefit from lower prices, it would probably be more straightforward to explicitly include the international reference price in the determination of the Medicaid best price mandated drug rebate amounts. The rebate required to match the reference price would bind for Medicaid when it is larger than the rebate that otherwise would have been mandated by Medicaid rebate provisions. Alternatively, international reference pricing could be used as a replacement for the current Medicaid best price rule.
The Innovation Center has the potential to pursue a demonstration along these lines within its authority. As noted, however, doing so through legislation would avoid many complications.
Should Reference Pricing Apply To All Drugs Or A Subset?
In addition to identifying the lines of business that should be subject to international reference pricing, a further question arises about whether reference pricing should be applied to every drug within a given market. To limit the scope of reference pricing while maximizing its effect, the model could apply only to selected classes of drugs that are deemed to face less competition or have especially high prices. Outside of Medicare Part B, insurers and pharmacy benefit managers have a series of tools at their disposal to help negotiate lower drug prices and steer use to lower-cost drugs, but these are far more effective where drugs face therapeutic alternatives that provide competing treatment options. Hence, the rationale for international reference pricing is in many ways stronger for drugs that face limited, if any, competition.
One way of pursuing this approach would limit the relevant international reference price to binding list prices solely for drugs without therapeutic alternatives. Reference pricing would be in effect until a competitor drug entered the market, at which point the constraint would be lifted or phased out, under the expectation that competition could begin and help determine a market price. In a variant, international reference pricing would apply broadly to Medicare Part B drugs (and potentially physician-administered drugs in the commercial market) but only to drugs in Part D and the commercial market that lack competitors. Medicaid would also benefit from these lower prices through either the effect of the “best price” rule or by more explicitly applying reference prices to Medicaid, as outlined earlier.
Key Questions Undergirding Any International Reference Pricing Approach
Regardless of how far or wide international reference pricing for prescription drugs is applied within the US, a few key questions arise.
Can An Accurate Reference Price Be Determined?
Implementing an international reference pricing model requires being able to accurately determine an average price paid by a reference group of foreign countries for a given drug, which could be quite challenging. There do not appear to be existing data sources that accurately capture final drug prices, net of all concessions, across other wealthy countries. Moreover, with US prices tied to international prices, both drug manufacturers and foreign countries (or payers within those countries) would have strong incentives to find creative ways to structure agreements with a superficially high price coupled with hidden discounts or other price concessions.
While this problem does not appear to have undermined existing international reference pricing structures in other countries, the incentives to mask or otherwise game prices would be much stronger if the US adopted this approach, since the United States is responsible for a high share of pharmaceutical company revenue and profits. A 2018 University of Southern California-Brookings Schaeffer Initiative paper estimates that the US accounts for between 64 percent and 78 percent of worldwide pharmaceutical profits. If gaming becomes prevalent, the international reference pricing model might have limited effect on prices in the United States (or in other countries).
The most direct method to accurately determine final drug prices internationally would be to compel data from every drug manufacturer that wishes to sell products in the US—similar to how ASPs are calculated. Some question whether the Innovation Center would have the authority to compel such disclosure, so it might be preferable to require disclosure through legislation. Even with strong legislative requirements, it might prove difficult to effectively limit gaming and establish accurate international reference prices, but a lack of authority could undermine the model and impair its impact.
New drugs or drugs that are poised to enter the market pose issues in determining an international reference price. It is not uncommon, for instance, for manufacturers to withhold selling a new drug in a country for a period such as a year as a condition of agreeing to price concessions within that country. During this window of time, drug prices would presumably continue to be set in the same manner they are today.
Would Manufacturers Continue To Sell Their Products In The United States?
Whether for a given line of business or across all payers in the US, establishing an international reference price relies upon the federal government being willing to walk away if a drug manufacturer refuses to sell in the US at that price. Theoretically, then, it is possible that US consumers might lose access to certain drugs. However, any reference price will still substantially exceed the marginal costs of drug production. Thus, it is unlikely that any manufacturer will refuse to sell in the United States unless it believes that doing so can influence the country’s political process to increase prices (a tactic that will almost certainly be tested).
How Would International Reference Pricing Affect Drug Prices Domestically And Abroad?
The Trump administration’s push for international reference pricing in Medicare Part B is premised on the theory that it would lower drug prices in the US while simultaneously increasing prices abroad, attenuating the revenue reduction incurred by drug manufacturers. The same rationale would apply to using an international reference pricing model beyond Part B. Assuming that an accurate reference price can be obtained—a big “if”—the model should reduce domestic prices, although multiple factors would influence the size of that reduction.
Simultaneously, the international reference pricing model would place upward pressure on prices in the reference countries for the affected drugs. This is because agreeing to a lower price in one of the reference countries would now also reduce the manufacturer’s revenue in the US. The relevant incentives would vary based on the share of a specific drug’s sales in the US versus the reference countries, as well as the current distribution of international prices. A similar phenomenon is seen today with the United States in contracts that include most-favored nation clauses across the health care system and in manufacturers’ responses to the Medicaid “best price” rule.
However, national governments are not typical market actors and the low marginal costs of production are well-known. For some nations where the government is effectively setting drug prices, prices may end up unaffected by the reference pricing model if the country can credibly threaten that it will not pay any more than the current price. On the other hand, political pressures might undermine this credibility. As a result, the impact on drug prices might vary by country, depending on the local politics and how insulated the drug price setting process is from political pressure.
The magnitude of the impact on domestic drug spending is also highly uncertain and depends in large part on how prices in the reference countries respond. Additionally, higher drug spending in the US stems both from higher prices for the same drugs and greater use of certain newer medications; many other countries restrict use of some new drugs, particularly those with less proven value over previous treatments. The international reference pricing approach would primarily affect prices but not use, which could limit its impact on US drug spending.
No other existing reference pricing approach is perfectly comparable to the ones contemplated in this post. However, in sum, the literature strongly suggests that, directionally, US drug prices would decline and prices in the reference group countries would increase, although the magnitude of these changes is highly uncertain.
Alternatives To International Reference Pricing
The attraction of an international reference pricing model is that it offers the potential to reduce drug prices in the US and may not commensurately reduce the expected financial returns to new drug development, to the extent it places upward pressure on drug prices in the reference countries. However, the magnitude of the model’s impact on reference country drug prices is highly uncertain, and similar reductions in US drug spending could also be achieved without putting upward pressure on prices in other countries. Some have also criticized the Trump administration’s international reference pricing proposal on the grounds that it would “outsource” a policy decision to policy makers in the countries whose drug prices go into the index used to determine reference prices.
Given these objections and the complications inherent in an international reference pricing approach, it might prove preferable to have the US government directly handle drug price regulation, potentially based on an assessment of the value of different treatments, rather than piggybacking on other countries and undermining their efforts to control drug costs. Such an approach might also allow the US to address drug spending stemming from our notably higher use of less effective drugs relative to other nations.
Organizations such as the Institute for Clinical and Economic Review conduct comprehensive clinical and cost-effectiveness analyses of new drugs that private insurers may use in their coverage decisions, analyses that parallel what governmental organizations in other countries conduct to make decisions on maximum drug prices. Regulators could draw on such analyses from organizations either inside or outside the federal government to determine reference prices for Medicare, Medicaid, and the commercial market. Depending on its structure, such an approach could save either more or less than relying on international reference pricing. US analyses plausibly might place a greater monetary value on health-related gains or be more (or less) susceptible to lobbying by industry than many governments abroad.
Steven Lieberman is the president of Lieberman Consulting, Inc, and routinely works with trade associations and private equity firms on general policy analysis on healthcare developments. The authors did not receive financial support from any firm or person for this article or from any firm or person with a financial or political interest in this article. Lieberman is currently a board member for Tuple Health, a healthcare design and technology company, and Primary Care Coalition of Montgomery County, a non-profit.