As global health leaders gather in Berlin from May 19–20 for the first-ever G20 Health Ministers’ meeting, one of the main topics of discussion is expected to be how to best fight the threat of antimicrobial resistance (AMR). This reflects the growing recognition that AMR poses a significant threat to human health. An influential 2014 report by the Review on Antimicrobial Resistance, commonly referred to as the O’Neill Commission, estimated that antimicrobial-resistant infections currently claim 700,000 lives worldwide each year, a figure that could rise to as high as 10 million deaths per year by 2050. Estimates suggest that AMR could reduce world gross domestic product by 2–3 percent per year, imposing trillions of dollars in global economic burden.
Proposals to combat AMR are usually built around two pillars. The first is to extend the effective life of the existing stock of antimicrobials with disease prevention and measures to combat overuse. The second is to promote the development of new antimicrobial medications to fight drug-resistant infections. While both are critical to solving the problem of AMR, there are inherent structural barriers that limit the usefulness of traditional incentives for promoting the development and effective stewardship of new antimicrobials. While the O’Neill Commission and others have made specific recommendations about policies to overcome these barriers, each proposed solution comes with its own set of costs and benefits. Thus, questions remain about the most efficient long-term solution to tackling the problem of AMR.
The Rationale For Intervention
The US patent system incentivizes innovation by offering the creator of a new technology monopoly power over its use for a fixed period of time. Although monopolies are usually considered inefficient, the potential for profit makes it worthwhile for innovators to undertake the costs and risk associated with research and development. An attractive feature of this system is that, in theory, monopoly prices are based on consumer willingness to pay, which in turn reflects the value of the innovation. In the case of medications, we would thus expect market forces to give manufacturers incentives to invest in the development of new treatments that offer the most value to patients (Note 1).
However, while the development of new antimicrobials to combat AMR would seem to offer high value, our current system offers insufficient incentives to invest in them. This is because the profit from a new drug is mechanically related to the number of people who take it. In the case of antimicrobials, more use of a specific agent eventually leads to resistance, so the highest societal value from new antimicrobials comes when they are used as little as possible until needed. This significantly undermines the incentives for manufacturers to invest in the development of a new antimicrobial drug. Efficient incentives will come from policy interventions that dampen the tie between manufacturer profits and the volume of antimicrobial use.
Policy levers designed to increase innovation are typically divided into “push” and “pull” mechanisms. Push mechanisms are supply-side policy tools that make innovation less expensive, such as a direct research and development subsidy. Pull mechanisms, including patents, are demand-based tools designed to make the market more attractive for a successfully developed product.
In the case of AMR, the fact that the new drugs are intended to have limited initial use means that the current patent system provides insufficient pull incentives. This limited initial use also undermines the value of push mechanisms used in isolation; there is little point subsidizing research and development for a product that does not have a profitable market. Thus, while push incentives may be useful in the long run, we view reforms that generate pull incentives as more fundamental and necessary to solve the problem of AMR.
There are several policy proposals that have been offered by the O’Neill Commission and others that would increase pull incentives for new antimicrobials (for examples, see here, here, here, and here). Three of the most commonly proposed policies are market-based entry prizes for new molecules, exclusivity extensions tied to the successful development of a new medicine, and vouchers for priority Food and Drug Administration (FDA) review of new molecules. We review each in terms of the likely power of the incentive, the potential for administrative burden or unintended consequences, and the ease of implementation.
Market Entry Awards
The most common solution offered to generate pull incentives is to create some kind of market entry award. These are essentially structured payments made to companies that successfully develop and bring to market new antimicrobials. The key choice variables for policy makers are the size of the awards and the timing of the payout. Most observers agree that the awards should be tied to the value of the innovation, but exactly what is meant by “value” and how to calculate it may vary. Prior estimates of the payment necessary to recover a sufficient return net of research and development expenditures range from US$1 billion–$4 billion. This highlights a central challenge to designing market entry awards, and pull incentives in general, that it requires some agency or organization to estimate the value of new antimicrobials to set the size of the award. This poses practical and political challenges and increases the administrative burden associated with the policy.
It is often proposed to use the timing of the payments as a tool to promote appropriate use stewardship of new antimicrobials. For example, the award could involve an upfront payment coupled with future payments tied to benchmark measures of appropriate use. Such a structure has better incentives to avoid overuse than a single, lump sum payment in which the manufacturer maintains ownership of the new antimicrobial (Note 2). However, a delay in the timing of the award payment could make the award less attractive to the manufacturers, particularly if future payments were tied to the appropriations process. The optimal payment structure must strike the right balance between incentives for innovation and for appropriate use.
The funding for the entry awards would most likely come from taxes. The potential efficiency and fairness of this funding model depends on the particular nature of the tax. For example, the O’Neill Commission described a “pay or play” tax requiring manufacturers to either demonstrate that they are investing in research and development to develop new antimicrobials or pay a surcharge that would be used to fund the awards. This model has some intuitive appeal, since it targets manufacturers and reinforces incentives by giving them the ability to opt out through research and development.
However, this model also has some limitations. First, because it uses a narrow tax base that specifically targets the pharmaceutical industry, the taxes would likely be passed through to payers and consumers in the form of higher prices and market distortion. This could lead to a reduced quantity of medicine consumed, worse medication adherence, and potentially adverse health outcomes (Note 3). Furthermore, the pay-or-play model requires regulatory oversight to verify the validity of the research and development activities in question or else companies may have incentives to engage in unproductive activities simply to avoid taxation (Note 4).
In principal, a more efficient option would be to spread the cost of the tax across the widest possible base by funding the payments through general tax revenue. However, this would require either raising overall taxes or cutting spending somewhere else, which poses political challenges. There is also the question of who should pay, and how the tax burden should be divided. A means-tested global surcharge would spread the tax over the widest possible base, in principal the fairest and most efficient option, but this introduces even more political uncertainty and relies on voluntary commitments from many autonomous governments.
Transferable exclusivity extensions have emerged as an alternative type of market entry award aimed at generating incentives to innovate new antimicrobials. This policy would grant an extension to the successful innovator of a new antimicrobial that could be applied to some other, existing drug on a one-time basis. Importantly, the extensions would be tradeable between firms, so an innovator with no existing product could sell the extension to a manufacturer with a drug close to patent expiry.
The exclusivity extension has some attractive qualities. It is potentially of high value to innovators, the value is not tied to the sales of antimicrobials, administrative costs are comparatively low, and there would be no need for a separate funding appropriation. The optimal voucher would need to be designed such that its value—which is essentially determined by the length of the voucher and the annual sales of the product to which it would be applied—was linked to the value of fighting AMR.
This raises the problem discussed previously of how to estimate the value of a new antimicrobial. As a practical matter, a possible alternative is to implement the policy with a simple preset time designation, say a nine-month extension, and then tailor up or down depending on the success of the program in generating new research and development (Note 5). Alternatively, there could be different classes of extensions, say a six-month extension for medicines targeting a low-priority infection versus an 18-month extension for medicines targeting high-priority infections. But ultimately, some estimate of value would be needed to justify the length of the extension.
While the exclusivity extensions would not require new taxation, there would be a cost in the form of higher prices paid by users of the product subject to the extension. For example, if the extension were applied to a cancer drug, this means that patients who needed that drug would face higher prices during the extension (as would their insurers) and could have reduced access to care. This extends the efficiency costs of monopoly pricing, which would affect a comparatively small patient population (compared to, say, a tax on the general population).
Some have pointed to this issue to criticize exclusivity extensions as inefficient (Note 6). However, the reality in the United States is that the extent to which cancer patients bear the costs would depend in part on who pays for their care. If health care is financed through insurance, this spreads the risk over the broader insurance pool through premiums and reduces the burden on any specific disease class. This suggests that restricting the impact of the exclusivity to patient populations with high rates of insurance coverage would help “smooth” the cost over a larger base.
A final challenge with the exclusivity extension is that it does not include inherent incentives to avoid overuse of the new product. Upon market entry, a manufacturer would still have incentives to maximize profits from the new antimicrobial. Thus, using exclusivity extensions to combat AMR requires additional guardrails to promote good stewardship, such as provider awareness programs or some kind of commitment device on the part of the innovator.
Priority Review Vouchers
An alternative mechanism that has been proposed to promote innovation is a priority review voucher. In 2007, the FDA created a program that conferred vouchers to the innovators of neglected infectious and tropical diseases that grant them the right to obtain priority review for some other product. It has been suggested to extend this program to antimicrobials as a means to combat AMR. As with the exclusivity extensions, these vouchers are tradeable, so the innovator can sell the voucher to another manufacturer close to the development of a potential blockbuster.
The advantage of a priority review voucher is that it can provide incentives for innovation without necessarily generating significant new costs. Manufacturers benefit because their product is on the market longer and they get any first-mover advantages, and patients who need the new drug get it faster. There could be offsetting costs in the sense that, with fixed FDA resources, reviewing one medicine earlier probably means that some other medicine is reviewed later. But this is more of a transfer between patients and does not necessarily generate any new burden. This is an advantage over the exclusivity extension, which generates new costs from extending the monopoly period. Moreover, no new appropriations would be needed for priority review vouchers tied to new antimicrobials because they would build off of an existing program.
One potential drawback of priority review vouchers is that they may be less lucrative and thus offer less incentive for innovation than exclusivity extensions or market entry awards. Exactly how much less depends on the potential price of a voucher compared to the price of an exclusivity extension or market entry award. The highest price for a voucher to date from the existing FDA program is $350 million. By comparison, it seems reasonable that to gain an extra year of exclusivity on a drug with annual revenues of $5–$10 billion a company may be willing to pay much more. Most proposed market entry awards are $1 billion, while others estimate that the government could expect to earn $500–$750 million per year from selling exclusivity extensions.
Matching these amounts with vouchers would probably require issuing more vouchers or trying to differentiate them in some way (say by creating a “gold star” voucher that immediately put a medicine at the front of the queue). Ultimately, the optimal number of vouchers should be tied to the societal value of the innovation generated. As with the case of the exclusivity extension, this doesn’t avoid the need to understand the societal value, although the actual decision point (1 voucher, 2 vouchers, and so forth) is simpler than pinpointing an exact value.
Another potential issue with vouchers is that they are based on speeding the FDA approval process, but reforms made under 21st Century Cures and the Prescription Drug User Fee Act VI are already intended to modernize and enhance the efficiency of drug development and review. To the extent that the general FDA approval process is enhanced, it would lower the market value of a voucher and dilute the value of the incentive.
Finally, it is worth noting that the voucher is similar to the exclusivity extension in that it offers no inherent mechanisms to promote appropriate use, and additional policy guardrails would be needed.
A Combination Approach?
There are relative strengths and weaknesses for all three of the most commonly proposed policies to promote the development of new antimicrobials. Market entry awards are simple and direct, but they could have high administrative costs and require stand-alone appropriations. Exclusivity extensions and priority review vouchers offer alternative mechanisms that would provide incentives while being self-funded. Priority review vouchers have some efficiency advantage over exclusivity extensions because they shouldn’t increase price or restrict access, but exclusivity extensions probably offer stronger incentives.
A hybrid approach that combined a priority review voucher with a briefer exclusivity extension might be an effective compromise that lowered some of the cost, but these should be coupled with additional efforts to prevent overuse of any new antimicrobials developed.
This argument is weakened if market imperfections (for example, insurance coverage) distort the relationship between price and value, whereupon incentives for research and development may not be optimal. See Howard et al. (2015) and Besanko et al. (2016) for more detail.
An alternative approach might be to treat the market entry award as a buyout, giving the government the authority to directly regulate use of the new drug. This has similarities to the proposed approach of governments appropriating rights to distribute new pharmaceuticals through a form of eminent domain grounded in patent law, an approach that has proven controversial in part due to the potential chilling effect it could have on research and development.
One might also argue that the taxation could be inefficient because it distorts research and development investment decisions. However, that is true of any pull incentive designed to promote the development of antimicrobials, in the sense that it changes the relative value of investing in different disease areas. The extent to which this argument had merit would depend on the size of the tax compared to the value of new antimicrobials development.
The same is true of the market entry award, but choosing the time period is probably a simpler exercise than determining an exact value.
Outterson and McDonnell (2016) estimate that a 12-month exclusivity extension would cost $4.8 billion for every $800 million in research and development in a new product.
This piece is based on research conducted by Precision Health Economics (PHE), a health care consultancy, with financial support from Pfizer. In addition to their academic appointments, Drs. Seabury and Sood are consultants to PHE.