External Reference Pricing: The Drug-Pricing Reform America Needs?

Abstract

• Issue: To control prescription drug costs, some U.S. policymakers have proposed external reference pricing (ERP), which would tie prices for high-cost drugs to those in other countries.
• Goals: To examine ERP programs internationally, assess their impacts, and explore prospects for U.S. adoption of ERP.
• Methods: Review of published literature and technical reports.
• Key Findings: ERP features vary, including the drugs targeted, reference countries used, and pricing calculations. The impacts of ERP are difficult to isolate from those of concurrent pricing policies, and evidence on the durability of savings is mixed. Industry strategies to “game” ERP, such as delaying the launch of drugs in particular markets, are intended to keep prices high. While initial ERP savings in the United States could be substantial, pharmaceutical companies may employ similar tactics that could erode savings. Most countries proposed for inclusion in the U.S. reference price “basket” favor a value-based approach that ties the price of drugs to their benefits; using their prices as references would mean importing their valuations.
• Conclusion: While short-term savings from a U.S. ERP program may be substantial, they will be difficult to sustain. Over the longer term, U.S. pricing policy could develop its own value-based approach, and any role for ERP would be limited.

Introduction

External reference pricing (ERP), sometimes known as international reference pricing, refers to the practice of informing price negotiations in a given country by calculating a benchmark, or reference, price based on publicly available pricing data from one or more other countries.¹ The rationale for ERP is straightforward — to ensure that the maximum price paid for a drug is not excessive relative to its price in countries that are comparable in terms of economic indicators, structure of the health system, and other factors.²

While the technical aspects of ERP vary by country, the practice is widespread. For example, as of 2019, 23 of 27 countries in the European Union employed some form of ERP.³ ERP also is employed in settings as diverse as Australia, Brazil, Canada, Jordan, and South Africa.⁴ In addition, ERP has emerged as a key budgeting and procurement strategy in many middle-income countries as they seek to expand access to health care while managing costs.⁵

The growth of ERP has sparked interest among U.S. policymakers. In December 2019, the U.S. House of Representatives passed the Elijah E. Cummings Lower Drug Costs Now Act (H.R. 3), which included a provision to set ceiling prices for drugs that represent substantial expense to Medicare and face limited competition at 120 percent of the average list price across six countries (Australia, Canada, France, Germany, Japan, and the United Kingdom).⁶ The Congressional Budget Office (CBO) estimated this provision could save Medicare $450 billion over a 10-year period.⁷

Interest in controlling drug spending has continued under the Biden administration. Among its activities are: 1) consideration of whether to adopt or strike a Trump administration rule precluding the use of government “march-in” rights to eliminate patents for excessively priced drugs that were developed with government funding; 2) two new bills (the Prescription Drug Price Relief Act and Medicare Drug Price Negotiation Act) that use ERP to set the maximum price to retain market exclusivity and fallback prices for Medicare negotiations; and 3) efforts to allow drug importation and promote biosimilar education.⁸

Key questions remain about ERP’s effects on prices, the drug innovation pipeline, and access to existing therapies in the United States and abroad. For example, the CBO acknowledged significant uncertainty in their estimates for H.R. 3, depending on how the process is implemented and how the drug industry responds.⁹ This issue brief explores these questions and discusses lessons learned outside the U.S.

Approaches to ERP

While ERP is widely employed internationally, how it is implemented and maintained varies substantially. Several features of ERP bear mentioning as the U.S. considers its own program (Exhibit 1).¹⁰

Target Drugs

Some countries include all drugs in their ERP scope, while others focus on prescription-only products or those eligible for public financing. Other variations include programs that focus on:

• drugs administered in hospitals
• “innovative” medicines, such as new drug classes or specialized therapies
• drugs with an active patent (that is, excluding generics)
• drugs subject to “parallel import” provisions — medicines sold in a lower-priced country that are imported into a higher-priced setting, as the U.S. is currently discussing.¹¹

Prices Used

Most countries use the “ex-factory,” or list price, in ERP calculations. Because it is set before country-specific markups and concessions are applied, the list price facilitates international comparisons. Some countries rely on the price used by pharmacies to either buy or sell the drug. ERP prices do not, however, include confidential manufacturer discounts or rebates, a fact that has implications for any benefits the U.S. might realize, as discussed below.

Selection of Reference Countries

Geographic proximity, per capita income, comparable health system or health insurance infrastructure, and availability of pricing data all play a role in reference country selection. The number of referenced countries varies from one (as in Luxembourg, which uses the price in the country from which the drug is sourced) to as many as 38 (as in Egypt).¹²

Calculation of Reference Price

The most common method of calculating reference prices is to find the average price across the group, or basket, of countries selected for comparison. However, variations, such as the average of the three lowest prices, are increasingly used to control costs. Some nations require only that the price chosen not be higher than the highest price, or lower than the lowest price, in the basket. While most approaches account for exchange rate fluctuations in their calculations, only Germany adjusts for differences in market size and purchasing power across countries.¹³

Update Frequency

Most countries revise their reference price calculations — and, hence, manufacturer negotiations — on a periodic basis. Austria allows for updates every six months as protection against gaps in available pricing data elsewhere,¹⁴ while other countries update after longer intervals, such as annually or every two to five years. Belgium and Hungary do not revise their initial pricing at all. Germany and Estonia tie any revisions to the duration of the pricing agreement and/or the availability of new evidence that might affect negotiations.¹⁵

The Impact of ERP

Rather than serving as the primary negotiating tool, ERP is frequently employed to support other drug-pricing policies.¹⁶ The independent effects of ERP on pricing are therefore difficult to estimate. Other potential policies are shown in Exhibit 2. Examples include:

• parallel trade regulations
• “payback” provisions, in which pharmaceutical companies agree to refund any amounts in excess of prespecified budget or profitability thresholds¹⁷
• some type of health technology assessment, such as consideration of a drug’s clinical benefit relative to treatment alternatives, cost-effectiveness analysis, potential budget impact, and additional weighting for product innovation.¹⁸

These multipronged approaches may reflect the limitations of using list prices for ERP calculations. For example, a reference price may be used to determine a ceiling or maximum price, while the actual price that the country health system pays will be based on confidential discounts or rebates.

There are also technical and administrative challenges associated with implementing and maintaining ERP.¹⁹ For one, predictable calculations depend on the availability of other countries’ drug-pricing data, stipulations for a minimum dataset, or use of substitute data when a country’s data are unavailable. Countries also may purchase drugs in different package sizes, and prices may vary by dosage strength in some countries but not others. For reference baskets that include multiple currencies, the source of exchange-rate data, the rate used, and the frequency of monitoring currency fluctuations have major implications for ERP calculations. Finally, ERP is a time- and resource-intensive exercise, particularly if a comprehensive approach is taken with frequent revisions and a large dataset.²⁰

Evidence has been mixed on whether ERP results in substantial and durable price reductions. In many cases, savings have been realized in the short term only to deteriorate as prices converged among countries in a given reference basket.²¹ Savings from ERP implementation appear greatest in nations where price revisions are frequent, the number of countries referenced is high, the lowest-price countries are weighted more heavily in calculations, and exchange-rate fluctuations are monitored closely.²² No study appears to have examined ERP’s influence on health outcomes, but others have noted that ERP may negatively affect access to drugs as companies withdraw from markets or countries prioritize parallel imports.²³

There are also several potential consequences of ERP external to the country implementing it. For example, a price change in one setting may indirectly affect prices in another, if the two countries reference a common set of country comparators.²⁴ In addition, ERP has been described as a “gameable” policy, since drug manufacturers may delay or eliminate product launches in referenced countries to mitigate price effects in key markets.²⁵ A recent analysis of launch timing in Europe found that, while timing varied in higher-priced countries, drug launches were consistently delayed by three or more years in lower-priced eastern European countries.²⁶

Some countries have begun to limit ERP’s role or abandon it altogether. Italy replaced ERP as its main governmental price negotiation tool in 2001 because program savings had substantially declined. The country now uses a combination of health technology assessment, cost-effectiveness analysis, and safety evaluation, as well as separate considerations for a drug’s market size and budget impact, additional weighting for therapeutic innovation, and considerations of the prices charged elsewhere (as a supportive element only).²⁷ Sweden eliminated ERP in 2003 and replaced it with a system based on health technology assessment. Not only was drug spending increasing, but concerns were growing that simply linking prices to those charged elsewhere took no account of whether a given drug was of any benefit to Swedish society.²⁸

Implications for U.S. Adoption of ERP

Implementation of ERP in the United States would have both domestic and international implications. While ERP would allow U.S. policymakers to sidestep the thorny question of whether value-based pricing makes sense here, ERP as currently described amounts to importing value constructs and assessments from elsewhere.²⁹ Five of the six reference countries included in H.R. 3 (all but Germany), for example, use cost-effectiveness analyses to inform drug prices and have varying thresholds for what is considered cost-effective.

Lower revenues for the pharmaceutical industry also may reduce funds available for research. The CBO projects that H.R. 3 would result in eight fewer drugs being brought to market in the first 10-year period, and as many as 30 in the subsequent decade, a reduction of about 10 percent in the drug pipeline.³⁰

ERP has the potential to substantially lower U.S. brand-name drug prices, in no small part because drugs are so much more expensive here. But what about the possibility that drug companies would game this policy? Most new drugs are launched in the U.S. first. Drug companies could raise launch prices for new drugs and delay their launches in referenced countries to maximize revenues during the period when there are no other prices to reference.³¹

In addition, while there is little evidence that ERP has caused drug companies to leave international markets, pharmaceutical sales elsewhere are a small fraction of those in the U.S., where over 80 percent of profits from new drugs are realized.³² Drug companies could exit markets in referenced countries with the lowest prices to keep U.S. prices high. Finally, companies could increase list prices in countries the U.S. references while keeping net (and largely confidential) prices stable.

As an exploratory example, consider palbociclib, a single-source cancer therapy sold by Pfizer, Inc. (under the brand name Ibrance) that in 2019 ranked 11th in Medicare Part D spending.³³ In 2018, palbociclib was listed for $538.94 per pill in the U.S. Reducing this to 120 percent of the six-country average proposed under the House bill ($206.30) would produce more than $1 billion in annual savings.³⁴ However, savings would be eroded by price increases for palbociclib, whether across-the-board or targeted at lower-price countries. In what might be perceived as an extreme but nevertheless plausible scenario, market exit in the three lowest-priced countries and 25 percent price increases in the remaining three countries would reduce savings by more than $300 million (Exhibit 3). While savings would remain substantial in relation to current spending, prices would continue to increase in the referenced countries over time, further reducing the savings.

Conclusion

External reference pricing would more closely align drug prices in the United States with those in comparable countries, returning billions in savings to the health system. However, the international experience with ERP suggests that its immediate benefits would be difficult to sustain in the long term. It is a system ripe for manipulation, with the potential for deleterious effects both in the U.S. (launch price increases) and abroad (launch delays and market exits) as the drug industry seeks to preserve global profits.

Our review also suggests that ERP may be best utilized as a supportive pricing policy rather than the primary lever. The U.S. may benefit from following the lead of most of the countries in its proposed reference basket as well as the strategy advocated by the U.S. nonprofit Institute for Clinical and Economic Review: using value assessments to determine what society is willing to pay for a given drug in relation to its health benefits. While a value-based approach might inform the price for public and private payers, ERP could play a supplemental role by setting maximum allowable prices or serving as a check on prices determined through other means. ERP also could serve as a transitional policy prior to development of a formal value-based approach.

Other countries have used value assessments to help establish prices that make sense given societal values and constraints. Rather than import those norms, the United States might consider developing its own system to ensure that true innovation is rewarded, and that we do not overpay for products that deliver marginal health benefits.