Chapter 3. Policy options to address current challenges

This chapter proposes a range of policy options to address the challenges described in Chapter 1. It begins by describing the principles and objectives that guided the development of these options: increasing the value of pharmaceutical spending, ensuring access in countries with different levels of development, supporting a rules-based system, fostering competition, and promoting better communication and dialogue among stakeholders. It goes on to describe sixteen policy options falling into five broad categories: reducing R&D costs and accelerating market access; increasing spending efficiency; determining willingness to pay for new treatments or health benefits; developing push and pull incentives to encourage innovation in areas of high unmet need; and strengthening the information base used to support policy decision-making.


Introduction and guiding principles

The two first chapters of this report described current trends in pharmaceutical markets and the many challenges currently facing policy makers. In preparation of this draft report, the OECD Secretariat engaged in extensive consultations with a broad range of stakeholders and experts. They contributed to the identification of a number of policy options with the potential to address some of the challenges highlighted in the first two chapters of this publication.

This chapter explores these policy options. It considers their strengths, weaknesses and feasibility. Since OECD countries have adopted very different approaches to pharmaceutical policy, certain options are not relevant or would be unlikely to work in some countries. Some would require international co-operation, while others would require changes only at the national level. In all cases, the policy options examined are not recommendations to governments, but rather, possible actions for consideration within each specific country’s national context and according to its priorities regarding the three objectives of providing access, ensuring spending efficiency, and fostering innovation.

The advantages and disadvantages of different options have been assessed using two broad criteria. First and foremost, the assessment is based on the available empirical evidence of effectiveness in achieving their stated objectives. In some cases, the evidence is substantial; for other options, however, there is little or no evidence of how effective they may be. In addition, when reviewing and assessing the potential of different policy options and the trade-offs among the three policy goals listed above, this report has been guided by five principles:

  1. 1. Increasing the value of spending on medicines. The overall objective is to ensure that maximum value is obtained from the expenditure made. This could lead to reduced (or curtailed) expenditure on low value items and/or increased expenditure on high value items; it may mean seeking to reduce prices (to ensure a desired level of cost effectiveness) or varying payment methods; or it may involve varying the ways in which certain products are deployed within the health care system. While payers may wish to reward innovation explicitly in order to encourage further, effective private investment in research and development (R&D), at the same time they may wish to send clear signals intended to guide investment toward the kinds of innovations that reflect their priorities.

  2. 2. Ensuring access in countries at different levels of development. The most effective way of ensuring that patients in countries at different levels of development can access innovative treatments is to apply differential (or tiered) pricing. Under this paradigm, wealthy countries pay more for medicines than poor countries and firms are able to earn sufficient profits in affluent countries to make further investments in R&D.

  3. 3. Supporting a rules-based system. The application by public payers of transparent criteria in determining willingness to pay for added health benefits would enable developers to know in advance what level of reward they may expect.

  4. 4. Fostering competition in both on-patent (between available treatments for the same indication) and off-patent markets (though the use of generic and biosimilar products) improves the efficiency of pharmaceutical spending. On-patent competition is not assured even where there are multiple therapies for the same indication, but can be fostered with appropriate procurement and payment policies.

  5. 5. Promoting better communication and dialogue to increase trust among stakeholders and improve alignment of industry R&D with societal priorities. Policy debates and decisions need to be informed by authoritative information on industry activities, R&D costs, and forthcoming products.

An overview of policy options

Policy options have been organised into five categories according to their main objective. These are shown in Figure 3.1.

Figure 3.1. An overview of policy options

Policy options considered in this report are described in detail in the third section of this chapter, together with their advantages, weaknesses and feasibility. Box 3.1 describes options that are not considered in the report, and explains the reasons why. Detailed descriptions of policy options also include definitions of technical terms where necessary, as well as bibliographic references supporting statements made in the paragraphs below. This section provides an overview of the policy options.

Box 3.1. Policy options which were not retained for consideration

Compulsory licensing. Compulsory licensing was rejected for inclusion in this report as a solution to balancing value for money and incentives to invest because its use reflects the failure of the pharmaceutical market to reach a price that all parties accept is appropriate, and this report focuses on how to ensure that this state of affairs does not occur in the future. A compulsory license granted to a generic manufacturer allows it to produce and sell a generic version of a patented medicine without the consent of the patent owner. International law allows for compulsory licenses in clearly defined circumstances, as long as certain procedures are followed and the patent owner is paid a royalty, leaving considerable leeway for governments in determining when to grant them. While compulsory licenses have been granted on a few specific occasions in OECD countries, wider use risks undermining the incentives that intellectual property protections create. Rather than debating whether the use of compulsory licensing might be justified, this report focuses on how an appropriate pricing policy framework might be agreed.

Rate of return regulation. Intellectual property rights give companies market power to maintain higher prices than they otherwise could, for a period of time, as an incentive to encourage greater than otherwise private investment in research and development. Consideration has been given to regulating the use of that market power, for example through regulation of the rate of return of firms in monopoly positions, or price capping. A system along these lines was put in place in the United Kingdom (the PPRS). This approach was, however, largely discredited in an assessment report by the UK Office of Fair Trading (OFT, 2007). Companies would, in effect, be paid for any costs they incurred, irrespective of how (in)efficient they had been. The OFT further noted that rate of return regulation amounts to paying companies for the amount of capital invested rather than for their output (i.e. benefits for patients) and may provide incentives for companies to over-invest in low-risk projects if the allowed rate of return is above the cost of capital. These reasons, along with practical problems in its implementation, led the OFT to recommend that other mechanisms be used to provide incentives for pharmaceutical innovation. It is therefore not considered further in this report.

Other policy options have been put forward by stakeholders, but have not been considered in depth in this report. For example, this is the case of the development of self-care and over-the-counter markets, which could relieve some pressure on public funding. This however, requires improvements in the health literacy of the population and is not considered practical in the short to medium term.

Involve stakeholders in joint efforts to reduce the costs of R&D and accelerate market access

Efforts could be made to reduce the overheads involved in complying with the evidentiary requirements of multiple regulators and HTA entities, for example through greater international co-ordination and harmonisation and even by way of joint or shared assessments. To achieve the objective of increased financial sustainability for payers, such measures would need to be accompanied by reductions in medicine prices and improvements in the value proposition.

Greater harmonisation of regulatory requirements for marketing approval is a means of reducing some development costs for companies and potentially accelerating market entry, benefiting patients needing treatment while at the same time generating earlier revenue flows for companies. Harmonisation can also reduce effort and costs for individual regulatory agencies. Many regulatory agencies are already collaborating through international networks to harmonise documentation and assessment standards, and some have bilateral agreements to share information on assessments of specific products. Other national agencies (e.g. in Mexico) go further, relying to some extent on market approval granted by another trusted agency, or establishing mutual recognition procedures (e.g. European agencies for medicines that are not approved centrally). See Option 1.

Accelerated marketing authorisation can provide faster access to treatments for unmet medical needs and can reduce the costs of evidence generation prior to market access. However, it also increase uncertainty about the safety and efficacy of new treatments and impose greater risks on patients. Ensuring its use is limited to situations where there is genuine and substantial unmet need as well as stringent rules for compliance with post-market collection of evidence and appropriate patient information are key for this option to work for the benefit of patients, payer and companies. See Option 2.

Increase spending efficiency

Countries need to seek efficiency gains and tackle waste in all parts of their health systems, including the pharmaceutical sector where policies could improve the value of both innovative and existing medicines. This could be achieved through co-operation in health technology assessment (HTA), international collaboration in price negotiation or drug procurement, more frequent assessment of the performance of medicines in clinical practice, policies that promote competition in both on-patent and off-patent markets, and measures that encourage appropriate prescribing and rational use.

Co-operation in health technology assessment (HTA) offers the potential to reduce duplication of effort and related costs for both companies and HTA agencies (see Option 3). Some national HTA agencies are already involved in regional and international HTA co-operation initiatives. These vary from loose networks that share information on methods or completed assessments, to full integration of activities. Greater integration of HTA (or joint evaluation) can be envisaged for clinical assessments, but is much more problematic for economic assessments. Greater co-operation requires, however, that participating countries share similar standards of care (to ensure the appropriate choice of comparator). It also means that national agencies need to agree on methods and parameters for HTA, which might be among the most challenging barriers to co-operation on HTA. See Option 3.

Greater international co-operation in price negotiations, contracting or procurement could increase bargaining power of buyers, competition among sellers, and impose greater discipline in negotiation and pricing processes through improving the information available to buyers. Co-operation could also potentially reduce transaction costs for industry and payers. International co-operation could take the form of joint price negotiations - i.e. exchange of information or joint drafting of technical specifications, without tendering together, or fully integrated international joint procurement. All these options have their own merits and weaknesses. Their appropriateness depends on market characteristics (e.g. the level of competitiveness in a therapeutic area); on political will to cede sovereignty and independence in order to benefit from co-operation; and on the level of harmonisation that can be achieved between national regulations and processes to evaluate medicines and make pricing or purchasing decisions.

Co-operation in price negotiation or contracting is compatible with value-based pricing and may even be compatible with tiered pricing, as long as participating countries have similar income levels and willingness-to-pay, or if the contract includes provisions allowing companies to charge different prices in different countries.

Pooled tendering or contracting can lead to the award of a significant proportion of, or the total market for a medicine, or class of medicines, to a single supplier. This can reduce the choice of products available to physicians and patients, and can lead to higher market concentration or an eventual supplier monopoly. This in turn can not only create the potential for price increases in the longer run but can also jeopardise the security of supply through increased dependence on a small number of, or even a single supplier. These risks can be mitigated by appropriate tendering and contracting mechanisms that include strong supply guarantees, or strategies that do not result in sole-supplier arrangements. See Option 4.

Countries could also use routinely collected data to assess the performance of medicines and adjust coverage conditions and prices after market entry to better reflect effectiveness and cost effectiveness of medicines in routine clinical practice. With the current capacity of health care systems to analyse routine data, this is unlikely to be practical for all products, and may be unnecessary for many products for which cost-effectiveness has been well established. However, it could be applied selectively to those treatments whose performance is uncertain and for which budget impact is high (see Option 5). Routinely collected data could also be used to measure performance in some MEAs and thus reduce their administrative costs. This is an area where advances in other parts of the health system – e.g. the use of ICT, and accessing data for health system governance – are necessary for improving effectiveness in the pharmaceutical system. See Option 5.

As highlighted in Chapter 1, countries could promote competition in on-patent markets. However, sometimes particular market features prevent competition from working effectively. In particular, in countries with national “positive lists” and regulated prices, payers are not always in a position to harness price competition. By contrast, formulary management1 and tendering are two options to foster price competition between products with similar indications and pharmacological profiles. Competitive tendering by indication could be used at the national level and follow a formal, transparent process to allocate the desired market shares to the lowest bidders, achieving an optimal trade-off between improving spending efficiency and maintaining a competitive market with more than one manufacturer in each market segment. See Option 6.

Bundled payments for episodes of care are another option to foster competition and increase the value of medical care. With appropriate quality standards, they could be used to incentivise providers to use the most cost-effective treatment for a given pathology and to negotiate procurement prices with companies. Such payments are being piloted in oncology in the United States, and preliminary evidence suggests that they offer an option for increasing the efficiency and quality of cancer care, as well as reducing the costs. With bundled payment mechanisms, monitoring quality and outcomes are important both to ensure patients do not receive sub-standard care and to link payments to value. See Option 7.

The first chapter of this report highlighted several challenges in off-patent markets: the fact that some countries do not fully exploit the potential of generics and biosimilars; the small number of competitors in some off-patent market segments; and as a consequence, sporadic price surges for a limited number of off-patent products in de facto monopoly positions. The risks of shortages are not addressed in this report, but these are obviously a concern when there is only one supplier for a given molecule.

Countries could consider a range of policy options to promote competition in off-patent markets. They could work toward getting generics approved more quickly, mobilise a range of incentives to encourage uptake of generics and biosimilars, and improve procurement mechanisms (see Option 8). These policy options are likely to minimise the risks of sporadic and sharp price increases. In addition, OECD member countries could implement a system to allow purchasers to report sharp price increases so that these could prompt investigation by competent competition or anti-trust authorities. See Option 8.

Determine willingness to pay for new treatments or health benefits transparently

Most OECD countries regulate or constrain the prices of publicly-funded on-patent drugs indirectly, through coverage conditions. Methods and processes for decision-making have evolved and over time have generally become more transparent. However, the number of medicines with high prices entering the market has been increasing, notably in oncology and for rare diseases. In some cases these medicines greatly improve patients’ health and quality of life, while in others the health benefits are more modest. Often these treatments do not meet pre-defined criteria for positive coverage and pricing decisions, which leads to case-by-case negotiations with unpredictable outcomes for all stakeholders. On a different note, the launch of new and very effective treatments for hepatitis C initially generated outcry and frustration because payers were not prepared for, and had not anticipated the launch of an extremely effective treatment for a disease of high prevalence. The combination of such a high price and high prevalence for a new product is rare. Nevertheless, this recent example demonstrated that existing models for determining societal willingness to pay, where they existed, were inadequate. The result was case-by-case negotiations.

Public payers could consider re-defining criteria for coverage and pricing decisions to anticipate more diverse scenarios, including circumstances in which effective treatments come to market for diseases with high prevalence. Decision rules may vary for a given medicine as a means of segmenting prices where cost effectiveness varies across different populations or indications, or where there is uncertainty about clinical and cost effectiveness in some indications. In some cases, the payer may initially commit the full price for the drug and require the manufacturer to refund all or part of the cost if the treatment does not work as anticipated, or the anticipated cost effectiveness is not achieved in practice. Information on how these agreements work is scant; as such, an important issue is that despite their widespread use the collective understanding of the performance of the medicines that are the subject of these agreements does not advance, as the results generally remain confidential. While the outcomes tracked via the implementation of such agreements cannot replace randomised control trials as a basis of effectiveness or value assessment, countries could optimise the use of managed entry agreements (MEAs) by addressing current limitations. See Option 10.

Develop new push and pull incentives to encourage innovation in areas with high unmet needs

Countries could consider the development of push incentives for targeted product development. The public sector already contributes substantially to pharmaceutical R&D but the allocation of funds could be improved. Current examples of public contributions include funding of public and academic health research, tax credits for private R&D spending or public-private partnerships at various levels of the R&D process. Public funding of R&D could seek to target more effectively those therapeutic areas of high unmet need that are not adequately addressed by private investment. High priority needs could be assessed both at national and international level, as they may differ, and some funds could be pooled to be distributed among these priorities (e.g. for the development of antibiotics). When public funds are targeted towards the development of specific products, governments could partner with industry to develop access plans, as is already done in the development of treatments for neglected diseases. See Option 11.

A number of alternative pull incentives have been proposed in the past, for example to spur the development of treatments for neglected diseases, and such approaches are envisaged to foster the development of new antibiotics to combat antimicrobial resistance. Alternative pull incentives could target priority product profiles that might be agreed internationally and made conditional on the affordability of the resulting medicines. To achieve this, countries could continue to explore alternative pull incentives to encourage R&D and obtain desired innovation for unmet medical needs, such as prizes and similar innovation rewards, or advance market commitments (AMCs). See Option 12.

Countries could review orphan drug policies to target more closely areas of unmet need. Orphan drug policies have been introduced in several OECD countries to encourage the development of treatments for rare diseases for which market incentives are too weak. The number of medicines and indications available to treat rare diseases has been increasing over time, and while this is good for patients with rare diseases, orphan designations and related advantages are sometimes granted for narrow indications of products with other registered indications that generate “blockbuster” revenues. The development of precision medicine implies that indications will increasingly target small populations, making them potentially eligible to receive advantages arising from orphan drug policies. These advantages often come at a cost to taxpayers, through reduced or absent evaluation fees, tax credits, and extended market exclusivity in some countries. Current trends suggest that these costs will increase, without necessarily spurring development of the types of medicines for which orphan drug policies were intended. It may be useful to assess whether existing policies are delivering the right outcomes, and to assess alternative options. See Option 13.

Strengthen the information base to better inform policy debates

Public debates reveal a lack of trust in available information on the activities and performance of the pharmaceutical industry, and perceptions that it generates excess profits from its activities. Publishing authoritative information on industry activities and performance, including R&D spending and profitability, could be a first step towards fostering trust among stakeholders and better informing policy debates. Stakeholders could agree on a set of relevant indicators and relevant sources (original data collection vs use of existing databases). The OECD could contribute to this effort by relying on its multi-sectoral capacity, that is, by involving several directorates (e.g. the Directorate for Employment, Labour and Social Affairs – responsible for health – together with the Directorate for Financial Affairs and the Directorate for Science, Technology and Innovation). Such an undertaking would require an agreement on relevant indicators and data sources, and would provide an opportunity for open dialogue. See Option 14.

OECD member countries could also increase price transparency in pharmaceutical markets. Levels of price opacity are high and increasing. The disconnect between list prices and prices actually paid by purchasers has a number of drawbacks: high list prices serve as an anchor in price negotiations; they blur international benchmarking; therapeutic choices cannot rely on comparisons of costs and benefits; patients sharing the costs of treatments may not benefit from confidential discounts or rebates; analyses of price trends have become partly irrelevant; and companies may be criticised for high list prices that are never applied. The effects of full price transparency are uncertain and full transparency may not be a realistic goal, especially if tiered pricing is to be applied in markets that are not segregated, and if countries continue to use international benchmarking. Co-operation among stakeholders could, however, improve price transparency. See Option 16.

OECD countries could also improve horizon scanning and promote co-operation at regional level to better prepare for the market launch and diffusion of new medicines. For example, some countries were taken by surprise when Sovaldi® entered the market in 2013, and had not made provisions for funding and delivering the new treatment. This made discussions over the price more difficult than necessary. Although horizon scanning is already undertaken in some countries, with more or less formal and institutionalised processes, health system preparedness and anticipation of the impact of new technologies and innovative medicines could be improved in many of them. Some parts of horizon scanning would remain country-specific, such as epidemiology or forecasting economic impact, but given the globalised nature of pharmaceutical markets, the identification of forthcoming medicines and their potential impact on health and health care systems would certainly benefit from co-operation. This could potentially be more comprehensive and less costly for individual countries. Pharmaceutical companies would in turn benefit from countries’ preparedness for more rapid adoption of the most effective technologies. See Option 16.

To conclude, a range of options is available to policy makers to create incentives for the industry to develop needed therapeutics, while providing patients with better access to products, improving the efficiency or value of pharmaceutical spending, and respecting the budgetary limitations of health systems. These can help to align the public interest more closely with that of the industry, and increase the aggregate value society obtains from spending on pharmaceuticals. Policy makers, however, need to find a balance between the objectives of ensuring patient access and financial sustainability, and allowing firms to earn sufficient return on their outlay to ensure continued investment in research and development of new products. This requires considering an overall policy package rather than picking and choosing among single measures to address only one policy goal or specific challenge.

A. Involve stakeholders in joint efforts to reduce the costs of R&D and accelerate market access

Option 1: Harmonise regulatory standards and promote mutual recognition


Co-operation among regulatory agencies could further accelerate patient access and potentially reduce costs of the later phases of R&D. Such co-operation could span the exchange of information on assessed products or harmonisation of evidence requirements, to reliance on assessments by other agencies to inform decision making.


Regulatory agencies responsible for the approval of medicines have varying capacity and apply different methods, evidentiary requirements and judgements. All these differences lead to differences in both the duration and outcome of the drug evaluation process. The approval time for new active substances by leading regulatory agencies (in United States, Japan, the European Union, Switzerland, Canada and Australia) has declined and converged over the past decade. However, while the average of median approval times was about one year in 2016, the difference between the fastest and slowest agencies was still 180 days (Bujar, McAuslane and Liberti, 2016). Several studies show that the FDA approval process – at least for oncology drugs – is shorter than in other countries, partly the result of its various accelerated approval pathways (Roberts, Allen and Sigal, 2011; Samueland Verma, 2016).

In terms of process, one study of all approvals by FDA and EMA between January 1999 and May 2014 showed that FDA was more likely than its European counterpart to approve drugs without randomised-controlled trials (RCTs) (Hatswell, Baio and Freemantle, 2016). A qualitative study on decision-making in oncology drugs at the FDA and the EMA outlined a number of differences. First, the study reported that the two agencies valued endpoints differently; while EMA respondents (to the survey) tended to identify progression-free survival (PFS) as a clinical benefit per se, FDA respondents considered that this surrogate endpoint had to be confirmed by a benefit in overall survival. FDA appeared more inclined to approve drugs based on activity data and phase II single-arm trials. Among factors not directly linked to the type and level of evidence but deemed to influence decision-making, the study mentioned that FDA had more interactions with the industry, and was also less risk-adverse and more willing to hasten access to medicines than its European counterpart (Tafuri et al., 2014).

Outcomes of the approval processes can also be different. Among 100 drug/indication pairs approved by the EMA between 1995 and 2008 for 42 anti-cancer drugs, 47 indications were assessed differently by the FDA. In 19 cases (indications), one agency had approved the indication but the other had not. In the remaining 28 cases, variations were observed in the framing of the indications, but only 10 were considered to be “clinically relevant” differences.

Although it appears plausible that seeking approval from agencies applying different methods and evidentiary requirements may impose costs in the last phases of clinical trials, there is no evidence from which to assess the magnitude of these additional costs.

International co-operation in regulation can take several forms, from information sharing between agencies to mutual recognition or fully integrated processes. Reliance is an intermediary step where assessments are shared, but the receiving authority remains responsible for the approval decision (Luigetti et al., 2016).

Several international networks already promote harmonisation of regulatory requirements. The International Conference on Harmonization of Technical Requirements for the Registration of Pharmaceuticals for Human Use (ICH), created in 1990, is the most influential initiative for medical products. It gathers representatives from regulatory authorities from Europe, Japan and the United States, and representatives of the pharmaceutical industry and from WHO, who work together to establish guidelines and standards for drug evaluation. Other initiatives include the Asia-Pacific Economic Co-operation Life Sciences Innovation Forum (APEC-LSIF); the Pan American Network for Drug Regulatory Harmonization (PANDRH), which supports regulatory convergence/harmonisation in the Americas; the International Pharmaceutical Regulators Programme (IPRP), which facilitates the implementation of internationally harmonised technical guidelines for pharmaceuticals; and a number of WHO programmes, such as the PAHO/WHO Collaborating Center for Biological Standardization, providing expertise and research in developing WHO written standards and guidelines (US International Trade Administration, 2016).

A number of initiatives promote information sharing between regulatory authorities. For instance, the International Generic Drug Regulators Programme (IGDRP) launched a pilot in 2014, which allows the EMA or EU countries’ national authorities sharing assessment reports with other members of the network, i.e. regulatory authorities from the European Union, Canada, Switzerland, and Australia (Luigetti et al., 2016). The WHO also launched in 2015 a pilot of collaborative registration for medicines approved by a stringent regulatory authority (SRA)2 where assessment reports from these authorities are shared with African countries (Luigetti et al., 2016).

Among OECD countries, some agencies are required to take into account overseas assessment reports where a product has been approved in a country with a “comparable control system” and the applicant requests such a review. This is the case for SwissMedic, for instance. According to Luigetti et al. (2016), the review time is reduced by 20% when EU assessment reports are used in the process. Health Canada is now permitted to use assessment reports produced by other agencies in its own evaluations. Mexico uses work produced by the EMA, the FDA, Health Canada, the Australian TGA and SwissMedic to enable approvals within 60 days (Luigetti et al., 2016).

The marketing authorisation of medicines in the European Union presents the highest level of regional integration, where medicines can be approved through centralised, decentralised or national procedures. Provisions for mutual recognition enable reliance on a decision of one member state by others within the European Union.


While the impact of greater harmonisation or co-operation on the reduction of R&D costs has not been evaluated, reliance on decisions of other agencies clearly reduces time to approval. Co-operation and coordination between agencies can generate transaction costs but reliance - or mutual recognition - where accepted by national authorities, holds the potential to reduce time and effort by both regulatory agencies and manufacturers. This can be particularly beneficial in countries with limited regulatory capacity, enabling them to draw on the expertise of stringent regulatory agencies.


Regulatory authorities operate within different cultural contexts and these are reflected in attitudes towards risks and potential benefits, which in turn influence regulatory decisions. Although scientific evidence is the cornerstone of regulatory approval, the appraisal of the evidence inevitably involves discretion and judgement. While regulatory authorities already cooperate to harmonise methods and standards, issues of sovereignty, and recognition of cultural, contextual and clinical differences among jurisdictions mean that few are willing to rely entirely on judgements made by other agencies to make major regulatory decisions.

Option 2: Accelerate market access for medicines with significant potential benefit


Accelerated and adaptive approval pathways can not only provide more rapid access to treatments for unmet medical needs but also have the potential to reduce the cost of producing evidence prior to marketing approval. However, in shifting some evidence requirements to after marketing approval, they also increase uncertainty about the safety and efficacy of new treatments and pose potential risks to patients who receive treatments early. Accelerated and adaptive approval may be most appropriate when used highly selectively, for those medicines promising the greatest potential benefit and in conjunction with stringent rules for compliance with post-market evidence requirements and patient information adequately conveying uncertainty as to the risks and benefits of treatment.


New drug approvals rely on well-established standards for evidence of safety and efficacy, usually based on randomised controlled trials (RCTs), which can take several years to conduct (see Chapter 2). The regulatory process itself takes several months, including review time by the competent authority and time used by the applicant to answer questions and update documentation. However, time to approval for New Active Substances (NAS) by stringent regulatory agencies has declined and converged over the past decade. The median approval time of six regulatory authorities3 declined from 565 days in 2006 to about one year in 2016, and the difference in the median approval time between the fastest and slowest agencies decreased from 530 to 180 days (Bujar, McAuslane and Liberti, 2016).

Since the end of the 1980s and following pressure from the HIV patient community to expedite access to new treatments, many regulatory agencies have implemented accelerated marketing authorisation pathways to approve promising treatments for high unmet medical needs earlier and more quickly (i.e. for severe diseases without any available treatments). Some regulatory agencies can approve treatments earlier in their development phase, with lower levels of evidence, usually conditional on provision of further evidence post-approval. In the United States, for example, the Food and Drug Administration (FDA) Accelerated Approval Program allows for early marketing authorisation of drugs for serious conditions / addressing unmet medical needs based on surrogate endpoints4, provided that manufacturers then submit results of phase 4 trials to confirm the expected clinical benefit (FDA, 2016). In the European Union, the EMA may grant conditional approval for medicines that address unmet medical needs – including medicines that treat life-threatening diseases, are intended for use in emergencies, or are orphan medicines – and where the benefit of early availability is considered to outweigh the risks of less comprehensive data than normally required (EMA, 2017b). Manufacturers are also required to provide comprehensive data after conditional approval to confirm that the benefit-risk balance is positive (EMA, 2017b). Accelerated assessment in the European Union, on the other hand, refers to a shortened period of review of an marketing approval application by the EMA when the application is for a product of major interest for public health and represents a therapeutic innovation (EMA, 2017a).

More recently, the Sakigake Review Designation System was launched by the Japanese Pharmaceuticals and Medical Devices Agency (PDMA) in 2015 to approve innovative products within six rather than 12 months of submission of phase 3 clinical trial data. Products are targeted for the Sakigake system based on criteria such as disease severity and evidence of substantial improvement over conventional therapies (Mori, 2017). In 2017 Japan also established the Conditional Early Approval System for early approval of medicines for severe and rare diseases without an existing effective treatment and for which conducting confirmatory clinical trials is difficult (ibid.). Early approval entails restrictions to ensure appropriate clinical use, and is conditional on post-market reconfirmation of safety and efficacy (ibid).

In 2014 the EMA launched a pilot project to explore the feasibility of “adaptive pathways” (Eichler et al., 2012, 2015). In its “adaptive licensing”5 pilot program the EMA takes a stepwise approach, with iterative phases of data gathering and regulatory evaluation; successive changes to marketing authorisation that take into account new evidence generated; and acknowledgement of remaining uncertainty at each stage (Eichler et al., 2012). Adaptive licensing can therefore enable, for example: the use of a new medicine in a restricted patient population with subsequent broadening of the marketing authorisation to a wider patient population, or a change to the indications for which a product is approved after confirming its benefit-risk balance, following a conditional approval based on surrogate endpoints. Adaptations could be based on a comprehensive development and licensing plan agreed in advance, not only between manufacturers and regulators, but also with payers that base decisions on the evidence generated (Eichler et al., 2012). Licensing frameworks based on the idea that knowledge and experience about a therapeutic product can be gained at every stage of its lifecycle have also been developed elsewhere, for example in Canada (ibid.).

While the trade-offs between the benefits and risks of earlier access and less certain treatment effects and toxicity need to remain at the forefront of decision making, accelerated marketing authorisation schemes may be useful in many situations where there is substantial potential benefit from early treatment. Beyond immediately life-threatening conditions, such as untreated HIV for which accelerated approval was initially devised, other examples include preventive therapies in situations where there is high risk of contracting the disease, or curative treatments where there is only a short period of time between diagnosis and the point at which treatment becomes impossible or futile (Eichler et al., 2015).


Accelerated marketing authorisation has the potential to provide patients in desperate need faster access to promising treatments. It could also provide the pharmaceutical industry with earlier returns on R&D investments. Because the cost of capital to finance R&D can represent a sizable share of total drug development costs, shortening the time between cash outlays for clinical trials and cash flow from product sales can reduce the overall cost of R&D.


Accelerated approval processes can pose greater risks for patients, particularly if uncertainty as to the risk/benefit balance of a treatment remains after approval. There is some evidence that drugs approved since the adoption of the Prescription Drug User Fee Act in the United States, which accelerated FDA approval processes, have been more likely to receive black-box safety warnings or be withdrawn, than drugs approved prior to it (Frank et al., 2014). On the other hand, a study of drugs approved by the EMA found that exceptional circumstances or conditional approval procedures were not associated with a higher probability of safety warnings than standard approval (Arnardottir et al., 2011).

It is not entirely clear whether prior efforts to accelerate marketing authorisation have been accompanied by sufficient efforts to reduce uncertainty after initial approval, or whether such accelerated procedures have increased safety issues. The EMA has reported that compliance with specific obligations related to conditional marketing authorisation was generally acceptable, while critics argue that evidence submitted subsequent to conditional authorisation has not provided sufficient certainty of the benefits and harms of new medicines (Banzi et al., 2017; EMA, 2017c).

Accelerated approval might also discourage investment in additional research on new agents. Manufacturers may take the view that by not generating further evidence on clinical effectiveness following accelerated approval they avoid the risk of unfavourable publicity and adverse regulatory action if later trials demonstrate poor clinical effectiveness (Naci et al., 2017). Thus, accelerated approval could create a disincentive to conduct additional research in those therapeutic areas in which accelerated approval has already been granted (ibid). A review of the available evidence base for novel therapeutic agents granted accelerated FDA approval between 2000 and 2013 found that a majority of RCTs of these products were not designed to evaluate their clinical benefits, but rather to incorporate them into standard treatment regimens, or to evaluate them for use in indications for which they were not approved (ibid).

Accelerated approval may also reduce the evidence available for subsequent health technology assessment (HTA) and coverage or pricing decisions by payers. This can make it more difficult to align prices with the relative effectiveness of treatments and to maximise the value of pharmaceutical expenditure.

Enabling conditions

There are a number of additional issues in deciding whether and how to implement accelerated marketing authorisation (OECD, 2017b). First, it is important to consider whether patients are adequately informed of the quasi-experimental status of products approved through such pathways, and that their safety and efficacy has not been established according to the criteria applied to other treatments. Second, countries could consider whether regulatory agencies have the means to ensure that companies comply with their commitments to produce further evidence post-approval within agreed timeframes. Stringent post-marketing evidence requirements, in terms of both safety and effectiveness, may be essential to manage uncertainty, as although incentives for firms to fund studies are reduced once a new product is approved for marketing, the need remains for robust clinical data to support regulatory, treatment, and pricing or coverage decisions. Third, countries could consider the option of withdrawal of authorisation in case of non-compliance with post-approval obligations, and whether to integrate products approved under accelerated pathways into standard treatment protocols prior to their safety and efficacy being established.

Reductions in the cost of R&D per se, however, do not automatically increase the efficiency of pharmaceutical spending or help achieve sustainable access to innovative medicines. Efficiency can only be increased if measures that reduce the costs of R&D are accompanied by mechanisms that can reduce medicine prices and improve value.

B. Increase spending efficiency

Option 3: Facilitate international co-operation in health technology assessment (HTA)

International co-operation in HTA could improve information available to payers; potentially avoid duplication of effort by HTA agencies; and reduce heterogeneity in evidentiary requirements among agencies, thereby reducing the burden of producing evidence for firms.

Implementation of co-operation initiatives may face significant challenges in gaining agreement on methods and parameters between countries, and in harmonising national processes. When considering such initiatives, it is therefore important that policy makers take into account the similarities and differences of national systems. Decisions about the breadth of co-operation, in terms of the countries involved, and the extent of co-operation, in terms of the level of integration of capacities and processes, or the type of assessments conducted jointly, may need to be adapted to the level of similarity of cooperating countries.

At the very least, however, HTA agencies could share clinical data on efficacy or comparative effectiveness that underlie assessments, or the results of those assessments, to reduce duplication and improve information available to payers. OECD countries could also attempt to define best practices in the conduct of HTA, or to standardise the health outcome measures to be used.


HTA is a mechanism for supporting coverage decisions and price negotiation processes. One option to improve the information available to payers is for countries to cooperate in assessing the comparative effectiveness of new medicines. This however, requires a certain degree of homogeneity in terms of existing standards of care. In contrast, assessments of cost-effectiveness and budget impact of new technologies must be country-specific, as they depend on the epidemiologic context, on costs of products and services and patterns of care in individual countries.

Population structures and burdens of disease, levels of income, healthcare system characteristics, standards of care and HTA infrastructure vary significantly among OECD countries. As a result, new products can have varying impacts on different healthcare systems. Co-operation in HTA can range from loose networks that share information on completed assessments to full integration of capacities and processes. Regardless of the level of integration, clinical data on efficacy or comparative effectiveness that underlie assessments could be shared across countries. Countries could also attempt to define best practices in conducting HTA, or to standardise the outcome measures to be used. Clinical studies often use a variety of outcome measures, which can make study synthesis difficult and can severely limit the generation of reliable research conclusions. A review of approximately 9 000 clinical trials in oncology, for example, found that more than 25 000 outcome measures were only used once or twice (Hirsch et al., 2013). A core set of standard outcome measures would also facilitate evidence generation by firms, as they would have a pre-specified set of outcome measures that would be accepted by HTA agencies.

Co-operation is already developing in regional and international HTA networks: examples include the Central and Eastern European Society of Technology Assessment in Health Care (CEESTAHC), Health Technology Assessment International (HTAi), HTAsiaLink, the International Network of Agencies for Health Technology Assessment (INAHTA), and the HTA Network of the Americas (Red de Evaluación de Tecnologías en Salud de las Américas, RedESTA). The United States Agency for Healthcare Research and Quality (AHRQ) created the Systematic Review Data Repository (SRDR) to make data underpinning systematic reviews and meta analyses and review reports freely available, and to reduce duplication among HTA agencies.

At the European Union level, the European Network for Health Technology Assessment (EUnetHTA), for example, has issued methodological guidelines to assess the comparative effectiveness of new technologies, and is piloting joint assessment for a small number of products.6 The European Commission made a legislative proposal for an initiative on EU-wide co-operation on HTA in January 2018, expected to be endorsed by the European Council and Parliament in 2019 (European Commission, 2018). Objectives of the proposal include improving the availability of innovative health technologies; reducing duplication of efforts by HTA agencies and industry to ensure the efficient use of resources; and strengthening the quality of HTA. The proposal foresees an amendment to the existing Directive on Cross-Border Healthcare (2011/24/EU), which provides for a network of HTA agencies of member states to facilitate co-operation and the exchange of information, to establish a joint clinical assessment process for medicines and medical devices. National assessments would be replaced by joint assessments for selected technologies and carried out by a coordination group of national authorities and bodies responsible for HTA. While non-clinical (e.g. health economic) assessments would remain national, the proposal also provides a mandate for the European Commission to support voluntary co-operation and the exchange of aspects of HTA that go beyond clinical assessment.

Some initiatives to standardise outcome measures and share information are also underway already. The development of core outcome sets (COS) has begun in recent years as a response to the heterogeneity in outcome measures, to provide standardised minimum sets of outcomes to be measured and reported in all clinical studies of effectiveness in a given therapeutic area (Gargon et al., 2014). For instance, the Core Outcome Measures in Effectiveness Trials (COMET) initiative, funded by the European Commission and the United Kingdom National Institute for Health Research (NIHR), promotes COS and published a handbook for COS development in 2017 (Williamson et al., 2017).

The BeNeLuxA initiative (involving Belgium, the Netherlands, Luxembourg, Austria, and Ireland) has included joint HTA among the potential activities listed in its Terms of Reference (BeNeLuxA, 2018). The focus is on high-cost and orphan drugs that are considered priorities in each of the participating countries, and for which assessment methods are deemed sufficiently similar to allow for such co-operation. An experimentation phase for this approach is currently underway for a small number of products.


Greater international co-operation in HTA has the clear potential to increase the level of information that underlies coverage decisions and price regulation or negotiations. Sharing of the same information can also increase harmonisation in decision processes between countries and facilitate greater co-operation in negotiations or procurement (see Option 4).

Co-operation reduces duplication of effort among countries and can generate administrative cost savings. While greater depth of co-operation may require a greater effort of harmonisation upfront, conducting joint assessments and the full integration of capacities, for example through establishing a centralised HTA function, could lead to greater cost savings. However, costs related to the conduct of HTA are modest in comparison to the costs of the technologies being assessed.

For manufacturers, greater alignment of submission requirements and methods between countries or joint assessment processes may also lower costs to comply with these requirements. Harmonisation of requirements can reduce the number of country-specific dossiers that need to be prepared and submitted in order to gain coverage by national payers and gain agreement on prices.


While the basic principles that underlie HTA are firmly established, substantial variation exists among countries in terms of the metrics used to assess technologies, assessment parameters, and other methodological choices. For example, a mapping of methods applied by 48 HTA agencies in 27 countries of the European Union and Norway found that approaches varied in terms of choice of comparators, or the use of network meta-analysis for indirect comparisons of treatments (European Commission, 2017). Co-operation may be particularly difficult where there are variations in burdens of disease, costs of health services, and treatment patterns. Domains that are context-dependent such as economic evaluation, or that require value judgments, will limit the scope for shared evaluation.

Enabling conditions

Because some domains of HTA are context-dependent or require value judgments, it is important that countries clearly define societal preferences and their parameters for HTA nationally (see Option 9), and assess the scope for co-operation accordingly. Continued and strong political commitment to co-operation will also be necessary to sustain initiatives. There may also be legal constraints on sharing of data between HTA agencies that need to be assessed.

Option 4: Encourage co-operation in negotiations, contracting or procurement


Countries could increase the efficiency of pharmaceutical procurement through international co-operation in negotiations, contracting, or procurement. This could increase their bargaining power, promote competition in pharmaceutical markets, and impose greater discipline in negotiation and pricing processes through improving the information available to buyers. Co-operation could also reduce administrative costs if duplication between countries is lessened.

International co-operation could span the exchange of information or joint drafting of technical specifications without joint tendering, to joint tendering, joint price negotiations, or fully integrated international joint procurement. All these options have their own merits and weaknesses. Their appropriateness depends on market characteristics (e.g. the level of competitiveness), the political will to cede sovereignty and independence in order to benefit from joint contracting opportunities, and the level of intercountry harmonisation achievable in evaluating medicines and making pricing or purchasing decisions. When substantial harmonisation effort is required, co-operation can also imply significant administrative costs. Depending on the nature and extent of co-operation, joint negotiations, contracting, or procurement may lead to greater price transparency.


International co-operation in medicines procurement can take different forms, with varying levels of integration among individual countries that agree to cooperate. As illustrated in Figure 3.2, in principle, co-operation initiatives can range from rather loose arrangements of sharing information to inform procurement, e.g. on prices, suppliers or methods and results of HTA, with countries continuing to regulate prices or purchase medicines individually, to more integrated forms, where participating countries may open joint tenders, enter into contracts jointly, or set up a central buying unit that enters into economic transactions with suppliers (World Health Organization, 2016).

Figure 3.2. Levels of international co-operation

Note: PAHO: Pan-American Health Organization

Source: World Health Organization, 2016, p. 8.

Examples of international co-operation in the procurement of medicines have emerged since the late 1970s. Initially, co-operation schemes developed mainly among smaller low- and middle-income countries, with a focus on vaccines or medicines deemed “essential”. As some schemes have matured, the product scope has expanded to cover medicines for a wide range of diseases, as well as other medical supplies. Examples that are considered successful, perhaps not least because of their longevity, include:

  • The Pan American Health Organization (PAHO) Revolving Fund for Vaccine Procurement, a highly integrated scheme, with tendering, contracting and payment performed centrally by PAHO was established in 1979 and now comprises 41 Latin American and Caribbean countries (DeRoeck, 2003; Huff-Rousselle, 2012; World Health Organization, 2016).

  • PAHO has also been operating the Strategic Fund for joint procurement of “strategic” public health supplies since 2000 ∗ comprising 25 countries in the PAHO region as at September 2015 ∗ to procure drugs for highly prevalent communicable and non-communicable diseases (PAHO, 2015).

  • The Gulf Co-operation Council Group Purchasing Program (GCC/GPP), a joint contracting scheme for vaccines, medicines and other medical supplies for seven countries in the gulf region, was also established in 1979 (Huff-Rousselle and Burnett, 1996; DeRoeck et al., 2006; World Health Organization, 2016).

  • The Eastern Caribbean Drug Service (ECDS) was established in 1986 (Huff-Rousselle and Burnett, 1996; Huff-Rousselle, 2012).

  • At a global level, the Global TB Drug Facility (GDF) established in 2001 has procured tuberculosis treatments for 90 countries, including most developing countries with high TB prevalence, providing drugs procured centrally with GDF funds as well as technical assistance to local procurement mechanisms (Stop TB Partnership, 2017).

  • World Health Organization Global Fund includes a pooled procurement mechanism for supplies to treat HIV, malaria and TB (Wafula, Agweyu and Macintyre, 2013; The Global Fund, 2015).

More recent examples can be found in Europe. In 2014 the European Commission established the EU joint procurement agreement for pandemic vaccines and other medical countermeasures for combating serious cross-border health threats (Azzopardi-Muscat, Schroder-Beck and Brand, 2016). By April 2016, 24 EU member states had signed the agreement (European Commission, 2016b). Formally, this mechanism represents a high degree of integration between participating member countries in that the European Commission takes central responsibility for the procurement process, opens tenders on behalf of countries, becomes the sole contractual counter party for suppliers, and allocates products to member countries (ibid.). However, there is no requirement for exclusivity, and member states can continue to contract individually with suppliers while participating in the scheme; they are also allowed to opt out of the scheme at any stage (Azzopardi-Muscat, Schroder-Beck and Brand, 2016). This EU joint procurement process has not been used thus far.

Belgium, the Netherlands and Luxembourg established a cooperative initiative in 2015, subsequently joined by Austria in 2016 and Ireland in 2018. The overall goal is to ensure affordable access to innovative medicines. The initiative comprises extensive co-operation in informing and preparing pricing and reimbursement decisions, including joint HTA, horizon scanning, and the exchange of information from national disease registries, as well as joint price negotiations with industry (BeNeLuxA, 2018; Department of Health, 2018). Internal transparency among the members of the cooperative is an essential principle. The focus is on high-cost and orphan drugs considered priorities in each of the participating countries, and for which assessment methods are deemed sufficiently similar to allow for such co-operation. The initiative is currently being piloted for a small number of products, in particular orphan drugs. Similar co-operation has been announced but not yet implemented by Bulgaria and Romania in the procurement of high-cost drugs (BMI Research, 2015; Petrovsky, 2015); by Poland, Hungary, Slovakia and Lithuania (Visegrad Group, 2017); and by ten Southern European countries that have signed the so-called Valletta Declaration (Infarmed, 2018).

There is little evidence available to assess the likely impact of this policy option in OECD countries. Joint procurement schemes in Europe are relatively new and the available literature mainly describes schemes in low- and middle-income country contexts, which may be of limited relevance to OECD countries (Azzopardi-Muscat, Schroder-Beck and Brand, 2016; Espín et al., 2016). Also, the publicly available studies were not always conducted independently, and did not always employ rigorous evaluation methods to show that prior co-operation or joint procurement schemes had been successful.


An advantage of looser arrangements to share information could be an increase in internal transparency, i.e. among, but not beyond, the cooperating parties, which could improve, for example, their understanding of commercial strategies of pharmaceutical firms involved in negotiations.

Greater international co-operation can impose greater discipline on negotiation and pricing processes through improving the information available to payers. Structured co-operation mechanisms can improve the information available to buyers and impose greater discipline on price negotiations, through the explicit definition among cooperating parties of criteria that underlie decision making, and therefore help align prices to the “value” of treatments as evaluated by health technology assessment (HTA). This can lead to higher prices for effective medicines that provide high value and, by the same argument, lower prices for medicines of lesser value, leading to greater alignment of prices with the relative effectiveness of treatments.

The primary objectives of past co-operation schemes have been to improve efficiency of spending, achieving overall cost reductions or offsetting savings with increased volumes to improve access (DeRoeck et al., 2006; Huff-Rousselle, 2012). The creation of buyer monopsony in entire market segments that span several countries and represent a large proportion of supplier revenue can help attract more competitive offers from suppliers (Management Sciences for Health, 2012). In low- and middle-income countries, co-operation has also aimed to improve the reliability of supply and avert shortages (DeRoeck et al., 2006). Other benefits can include (Huff-Rousselle and Burnett, 1996; DeRoeck et al., 2006; Huff-Rousselle, 2012; Azzopardi-Muscat, Schroder-Beck and Brand, 2016; Espín et al., 2016; World Health Organization, 2016):

  • More rational product choice through better-informed selection and standardisation.

  • Reduction in operating costs and administrative burden in the procurement process, for example through reducing duplication between countries and shared processes to gather information and monitor prices and supplier performance.

  • Improved quality assurance through restricted tenders and stricter up-front screening of suppliers.

  • Increased equity among member countries.

  • Reduction or elimination of the potential for corruption in procurement.

  • Transfer of technical capacity in procurement through the formal establishment of networks of procurement professionals, especially to resource-constrained environments.

  • Improvements in transparency and other aspects of supply chain management.

Aside from co-operation in operational processes for negotiation, contracting or procurement, countries could cooperate in skill development and capacity building within government agencies or payers involved in negotiation and procurement.

The limited evidence available suggests that some joint procurement schemes have been successful at achieving objectives related to efficiency, reliability of supply, and access. A review of the operating mechanisms and achievements of the GCC/GPP and the PAHO Revolving Fund found success in terms of the stated objectives of reducing cost and ensuring more reliable supply as a result of more efficient procurement and better forecasting of demand (DeRoeck et al., 2006). Both programmes were also found to have reduced prices versus local procurement and to have accelerated access to new medicines (DeRoeck et al., 2006; Organization, 2007). At a meeting on pooled procurement hosted by WHO in 2007, representatives of the Eastern Caribbean Drug Service (ECDS), the Gulf Co-operation Council Group Purchasing Program (GCC/GPP) and the Global TB Drug Facility (GDF) reported that joint procurement was successful at increasing bargaining power, achieving significant cost savings, enhancing quality control, and increasing access to medicines.


Gaining and sustaining political commitment to co-operation may be difficult. For example, key problems in procurement of H1N1 vaccines and antivirals by EU countries prior to establishment of the joint procurement mechanism were seen as the main triggers for the Council and Parliament to request establishment of the mechanism by the Commission. These were related to price, liability, confidentiality, and flexibility to adjust quantities ordered to actual needs (European Commission, 2016b). However, not all member states were supportive of the proposal at first, with disinterest especially among those who already had agreements in place with the industry (Azzopardi-Muscat, Schroder-Beck and Brand, 2016). The EU mandate on public health, and the strong legal basis in the Treaty on the Functioning of the European Union (TFEU) for combating serious cross-border threats to health, are considered instrumental in reaching an agreement. However, focus remained strictly on vaccines for serious cross-border threats to health, and the participation of countries remained relatively flexible, with opt-out clauses and no exclusivity requirements (ibid.).

Existing co-operation schemes in the procurement of medicines within the EU mainly involve small countries. However, it has been argued that larger countries may be reluctant to join, or may even perceive joint negotiation mechanisms as disadvantageous, by weakening their sovereign right to negotiate directly with suppliers (Azzopardi-Muscat, Schroder-Beck and Brand, 2016).

Depending on the depth of co-operation chosen, joint price negotiations or tendering can engender price transparency among cooperating countries. Although countries could agree to adopt tiered pricing between them, price transparency can make tiered pricing more difficult than where discounts are confidential, for example because the general public in a given country may be reluctant to accept higher prices than in other countries that are part of the cooperative. Joint negotiations or tendering can lead to price convergence among cooperating countries, and can disadvantage countries that were previously able to negotiate favourable terms with suppliers, at least in the short term.

Finally, given that co-operation schemes are costly to establish, purchasers may be tempted to award significant proportions of the total volume of a class of medicines to fewer or even a sole supplier in order to generate savings. This could reduce the number of different types of products in a class available to physicians and patients, and thus limit product choice, as well as leading to greater market concentration among suppliers, or eventual supplier monopolies. This may not only lead to price increases in the longer run but may also jeopardise the reliability of supply through increased dependence on few or only one supplier. Thus it is important to mitigate risk to competition by appropriate tendering and contracting mechanisms, and in particular, to avoid winner-takes-all tenders (see Option 8).

Enabling conditions

International co-operation in negotiations, contracting or procurement could be increased gradually, with initial steps including the building of capacity and co-operation in workforce training or exchange of information. These would require little effort to harmonise national processes. However, commercial information, such as prices, may be legally protected; rules that protect such information may need to be reviewed to enable information sharing.

Depending on progress and experience with initial steps, more demanding areas of co-operation that require more sustained political commitment and greater harmonisation might then be explored. Gaining and sustaining political commitment is important for deeper co-operation schemes that go beyond the sharing of information. The guide to drug procurement by Management Sciences for Health (2012), for example, concludes that continued political commitment is the cornerstone of joint procurement schemes. Noting that only a few multi-country procurement schemes have proven sustainable over time, such as the examples from PAHO, the Caribbean and the Gulf states, the guide also states that political commitment is a necessary but insufficient prerequisite to maintain buyer monopsony. Co-operation agreements could include mechanisms to enforce monopsony commitments, to avoid co-operation being weakened by the negotiations of individual members.

A significant challenge to co-operation in procurement can be that harmonisation in product registration and regulation are necessary, and that countries need to commit to aligning product preferences. Co-operation can lead to increases in overall costs if the administrative burden, for example from large harmonisation efforts, outweighs the savings generated from price reductions. This may require detailed technical feasibility assessments prior to launching co-operation schemes.

Option 5: Assess performance of medicines in routine clinical practice and adjust coverage and pricing accordingly


Instead of setting or negotiating definitive coverage conditions for medicines at market entry, adjusting them at subsequent points in time to better reflect performance in routine clinical practice could increase efficiency of spending, and could align incentives for innovation with the benefits medicines deliver to patients.

With the current capacity of health care systems to analyse routine data, this is unlikely to be practical for all products, and would be unnecessary for products for which effectiveness has been well established. However, it could be prioritised for treatments whose effectiveness is uncertain. The effective use of routinely-collected data would require countries to improve capacities and data governance to collect, link, process and analyse such data for the purpose of determining effectiveness.


Prices and reimbursement conditions for new medicines are usually defined at market launch, based on evidence of efficacy and safety generated in late-stage clinical development trials that take place prior to marketing authorisation. While studies of efficacy aim to establish the “extent to which an intervention does more good than harm under ideal circumstances” (James, 2017, p. 619), they do not necessarily demonstrate effectiveness, defined as the “extent to which an intervention does more good than harm under usual circumstances” (ibid.). Indeed, the use of medicines in routine clinical practice may often reveal a gap between the benefits and risks assessed in pre-market clinical trials and those observed in routine practice, not only due to factors unrelated to the medicine, such as adherence to treatment or other patient characteristics, patient-physician relationships, but also because of the possible discovery of previously unrecognised effects (e.g. rare adverse effects that were not detectable in pre-market clinical trials).

While this gap is widely recognised, the performance of medicines in routine practice is only rarely assessed, and usually not with the purpose of revising reimbursement conditions or prices. Several obstacles, including those that are described below and related to data availability and quality, analytical methods and patient privacy, may explain why this does not happen. Randomised controlled trials (RCTs), which are most often conducted by the pharmaceutical industry to establish efficacy prior to marketing authorisation, continue to be considered the “gold standard”, as they provide evidence under conditions that minimise bias and confounding. RCTs usually apply restrictive patient inclusion criteria to increase the probability of showing a treatment effect, and may exclude very young, very old and multi-morbid patients; use allocation concealment and blinding to avoid biases; and adhere to strict follow-up protocols that improve patient adherence. These methods increase the internal validity of studies and thus the confidence that an observed outcome can be attributed to the intervention (Revicki and Frank, 1999; Gartlehner et al., 2006). The hierarchies applied by most HTA agencies classify data from non-randomised studies as a lower level of evidence than RCT data.

Thus far, data from routinely collected sources have been used mainly to monitor safety and measure utilisation post marketing approval than for other purposes, largely because serious adverse events (death, hospitalisations) or other safety signals can be easier to discern. In HTA the use of routinely-collected data is more limited and policies of national HTA agencies vary (Makady et al., 2017). According to the most recent OECD surveys (OECD, 2013b, 2015a), nearly half the OECD member countries routinely collect data on pharmaceuticals prescribed or dispensed. However, linkage between these datasets and other health data is less common and limits their utility. In the European Union, several public-private initiatives have been launched to encourage co-operation between industry, regulators, HTA agencies, and other stakeholders in exploring tools and methods for the use of routine data (IMI, 2018).


Although deriving evidence from routinely-collected data still presents some challenges, in the long run this option is a promising route for providing credible long-term evidence of a product’s value to the health system. Pre-market RCTs have many advantages, such as pre-specified and well-defined endpoints and patient inclusion/exclusion criteria, randomisation of study subjects into intervention and control groups to control for bias and confounding, and blinding of patients and health care provider to provide unbiased measures of efficacy in the trial population. However, these features of RCT study designs that ensure high internal validity of their results can also limit their external validity and generalisability. Pre-market RCTs are carried out using selected populations under ideal conditions that but say little about which interventions work best when implemented in different routine settings and on different populations.

The advantage of studies of effectiveness in routine practice relative to RCTs is their greater external validity and generalisability. Such studies are thus useful for:

  • Examining clinical outcomes in a diverse study population that reflects the range and distribution of patients observed in routine clinical practice;

  • Comparing multiple alternative interventions (e.g. older vs. newer drugs) or clinical strategies to inform optimal therapy choices beyond placebo comparators;

  • Measuring the gaps between efficacy and effectiveness.

Studies that use routine data are significantly cheaper to conduct than RCTs. Through secondary use of data generated from the routine processes of healthcare delivery, such studies do not require the recruitment of healthcare professionals and patients to participate and can be conducted with smaller teams of researchers. With technological advances, the costs of data collection and storage can be expected to decrease further in the future.

Ultimately, using assessments of performance in routine clinical practice to adjust coverage and pricing could lead to greater alignment of financial incentives for manufacturers with the value delivered by the medicines they produce.


Data availability and quality remain problems in assessing performance in routine practice. Observational data in routine datasets are not collected for research purposes, which may imply that measures of health outcomes or confounding factors to assess performance may not be available or may be recorded with poor quality.

It is often methodologically difficult to disentangle effects of a medicine per se from confounding factors in non-randomised and retrospective studies based on routine data. Random allocation can balance confounding factors between intervention and control groups in RCTs if sample sizes are sufficiently large. Observational studies must instead rely on sophisticated statistical methods to control for confounding in the absence of randomisation. These methods, such as propensity score-matching, synthetic control groups, etc. are still an emerging field of research. While studies using routine data can enable greater generalisability of findings because they reflect the conditions of routine practice (in which patient populations are more heterogeneous and inappropriate prescribing or lack of patient adherence to treatment may be more common), they can have poorer internal validity and may fail to accurately isolate the effects of the treatment assessed (Revicki and Frank, 1999; Gartlehner et al., 2006). Examples of current initiatives that aim to develop methods for analysis of routine health data further, including the estimation of treatment effectiveness, include the Observational Health Data Sciences and Informatics (OHDS) initiative in the United States and the Electronic Health Records Research Group (EHRRG) in the United Kingdom.7

Enabling conditions

The effective use of routinely collected data would require countries to improve their capacities not only to collect and link data generated by health care providers, payers or other stakeholders in health care systems, but also to process and analyse the data for purpose of assessing effectiveness. The availability, governance and use of routinely collected data have been a recent focus of the OECD Expert Group on Health Care Quality Indicators. Based on a 2013/14 survey (OECD, 2015a), at least 15 OECD countries routinely collected data on prescription medicines. Most of these data come from reimbursement claims or billing data. In at least eight countries, prescription databases are linked with other databases such as hospital discharges and mortality data.

Assessing the performance of medicines post marketing authorisation might also entail expanding data collection and standardising the data to be collected, which can generate additional costs. For example, this could include defining minimum datasets or establishing additional registries for specific purposes. Countries would likely need to devote additional resources to expand their capacity for the collection, storage and analysis of routine data, and some countries may be unable to absorb the additional costs at current levels of health expenditure. Although studies that use routine data are cheaper than RCTs, assessments of the performance of medicines in routine practice would have to be done in addition to pre-market RCTs and other prospective studies that are currently required to support marketing authorisation or post-market surveillance. To produce generalizable results in each country, a given medicine would likely need to be assessed in each country using routine data from domestic sources, leading to some duplication.

While digitalisation has increased volumes of available data, computing power and infrastructure for analysis, not all data can be transformed into information, and not all information is relevant evidence. It is therefore important to examine the types of evidence that can be derived from different sources of routinely-collected data. There are differences in the potential value of different data sources according to their collection and storage methods. Claims data are the most commonly available source in many countries, but their purpose is to facilitate billing, so their scope is often limited to spending, utilisation and potentially appropriate use. Different databases sometimes utilise different coding schemes across different settings of care, such as physician records and hospital discharge data. Linking such sources to extract the information of interest may be difficult. The uptake of electronic health records (EHR) permits routine collection of additional information (e.g. diagnoses, test results, etc.).

Assessments of the performance of medicines from the perspectives of payers are inherently comparative. Adjustments to coverage and pricing of a given medicine may have consequences for treatments that are alternatives to the medicine assessed, and in particular, other medicines that serve as comparators. Countries may therefore need to consider more dynamic systems of iterative coverage and pricing as new evidence becomes available, as this could entail reassessments and adjustments for all products that may be compared to the product for which coverage or pricing is being changed. Such network effects may be complex to manage.

The use of routine data sources also requires robust data governance and security measures to preserve confidentiality and patient privacy. All countries that responded to the OECD surveys (OECD, 2013b, 2015a) have legislation on data privacy in general, which apply to health and health care data. Some also have laws that govern the use of health care data more specifically. A major problem, however, is that many legislative instruments governing data, privacy and security predate the digital era, and their interpretation in the context of secondary use of electronic health data is difficult, including requirements for obtaining informed consent. Although data that have been de-identified can generally be made accessible at lower risk to personal privacy, current data protection and privacy laws may be a barrier to greater use of routine data.

While analysis of data from routine practice cannot replace the safety and efficacy data generated by pre-market RCTs, evidence hierarchies and other methodological standards employed by some HTA agencies may need to be revisited. Post-market analyses can complement pre-market data by allowing actual effectiveness versus expected efficacy and safety to be evaluated in routine clinical settings. Analytical capacity would have to be expanded to apply rigorous observational methods that in the absence of randomisation can isolate effects of the treatment under evaluation from confounding factors. Many HTA agencies in OECD countries employ a formal hierarchy of evidence, attaching lower weight to observational studies, and are generally circumspect in considering evidence of effectiveness from routine data sources (Makady, Goettsch and Willemsen, 2015; Makady, 2017).

A further issue is the assignment of risks and costs between payers and industry while routine data are collected and analysed to reduce uncertainty around the effectiveness of a medicine. One option is to have government or HTA agencies collect and analyse data independently from the industry or payers, which could minimise conflicts of interest. If manufacturers are expected to collect and analyse data, policy needs to consider financial incentives for producing additional evidence. Payers may therefore need to be circumspect in initial coverage and pricing decisions where the effectiveness of a medicine is uncertain, and offer the prospect of coverage expansion or price increases if and when a medicine is shown to be effective using routine data. Initial coverage and pricing decisions also may also need to include binding conditions, enshrined in contractual or statutory law, to ensure that subsequent changes, such as delisting from coverage or price reductions, can be implemented.

Option 6: Promote competition in on-patent markets


To improve efficiency in pharmaceutical markets one approach is to promote competition among on-patent medicines with similar indications and pharmacological profiles, as seen recently with hepatitis C medicines. However, certain market features can prevent competition from working effectively. For example, in countries with national “positive lists” and regulated prices, negotiators are not always well placed to harness competition. Since positive lists usually determine both market access for companies and patient access to alternative treatment options, national “purchasers” may not always be in a position to negotiate favourable contract terms in exchange for listing.

There are several options for promoting on-patent competition; two in particular are tendering by indication (or requests for proposals, RFPs), and formulary management. While tenders for sole supplier arrangements can lead to the largest discounts, multiple supplier contracts retain therapeutic choice, and enhance security of supply. Formulary management can encourage competition among therapeutic alternatives by giving the most cost effective options preferential placement in a tiered formulary and in formulary-compliant treatment guidelines.


Tendering is well recognised as an effective mechanism for generating competition. While sole supplier tenders arguably create the strongest incentives for discounting, they are not without attendant risks. Multiple supplier arrangements can preserve therapeutic choice, support the retention of multiple suppliers in the market (thus avoiding risks of future monopoly pricing), and enhance the security of supply.

Competitive tenders could be offered at national level and be followed by a formal and transparent process to allocate market shares to the best bidders. This can provide an optimal trade-off between competitive prices and product availability sufficient to ensure that all patients have access to treatments based on need, particularly where products are not clinically interchangeable.

Norway recently shifted a large share of medicines from the outpatient market to hospital markets,8 where they are procured through a centralised process. The Norwegian Hospital Procurement Trust (LIS) is in charge of procurement of all of these products and uses European-wide requests for tender (RFTs). LIS drafts RFTs by indication,9 and in co-operation with physicians, to ensure that patients will be offered a choice between alternative treatments where needed, and does not adopt a winner-takes-all approach. Alternative products are ranked within treatment guidelines according to the attractiveness of the bids submitted by manufacturers, and prescribers are required to use the medicines according to their order of priority unless a different choice is justified for clinical reasons - but all products remain available. The impact on price of the Norwegian tenders by indication is not known, except for cases where biosimilars are available (see Option 8). Since 2016, LIS may no longer disclose the prices that result from its tendering process.

Formulary management can also be used to encourage competition among therapeutic alternatives, by giving the cheapest or most cost effective options preferential placement in a tiered formulary and in formulary-compliant treatment guidelines. This form of formulary management is used mainly by insurers and PBMs in the United States (see Box 3.2).

Box 3.2. Formulary management, definition

“Formulary management” is a term typically used in the United States to refer to the practice of constructing a reimbursement formulary that steers prescription and consumption towards certain categories of drugs, usually those offering the most favourable terms. Constructing the formulary includes defining a list of covered products, setting the associated co-payments for patients, and determining coverage conditions (e.g. stepped therapy, prior authorization).

The list of covered drugs (and associated co-payments) usually falls into one or more of three broad categories:

  • Open formulary: Where the insurer pays a proportion of the cost of all or virtually all marketed drugs, although some OTC of life-style drugs may be excluded.

  • Closed formulary: Where the insurer only covers drugs listed on the formulary.

  • Tiered formulary: The insurer offers differential co-payments or other financial incentives to encourage the use of preferred formulary drugs, but still pays a portion of the costs of the non-preferred drugs. Over time, the number of levels in tiered formularies has tended to increase. According to a 2016 survey of employer-sponsored plans, which cover about half of the non-elderly population, one third of covered workers face 4-tiered co-payments, and about half 3-tiered co-payments. Others have only one- or two-tiered co-payments or no co-payments after an initial deductible amount (The Kaiser Family Foundation and Health research and Educational trust, 2016).

A tiered formulary thus creates an opportunity to promote price competition in return for “preferred formulary status”, which provides an advantage to a manufacturer in terms of market share, particularly where products are considered to represent comparable degrees of clinical benefit. Importantly, a tiered formulary can also create opportunities to increase the value of pharmaceutical spending, by steering prescription and consumption towards medicines with higher value. This requires the selection of products, tier placements and coverage conditions to be informed by pharmacoeconomic analyses.

Many payers in OECD Member countries already use pharmacoeconomic assessments to inform not only formulary listing and coverage decisions, but also clinical treatment guidelines. While the use of pharmacoeconomics was uncommon in the United States in the 1990s, more recently there has been renewed interest in its application, and increasing recognition of its potential to enhance value through drug benefit design (Grabowski and Mullins, 1997; Neumann, 2004; Lising et al., 2017).

Source: See for example

To date, assessments of the impact of formulary management strategies by PBMs have generally focused on use, patient adherence, and health care costs, rather than effects on competition (see, for example, Huskamp et al., 2005). One study examined the impact of the implementation of a close national formulary in the Veterans Health Administration drug benefit in 1997, and found that it significantly impacted market shares and reduced the price per pill in the therapeutic classes affected by the change in the two years following implementation (Huskamp, Epstein and Blumenthal, 2003). Clearly, a significant obstacle to assessing the impact of formulary management on prices is that actual prices paid generally remain confidential. Recent analyses show that both list prices and rebates have increased substantially in the United States in recent years (see Chapter 1), but confidentiality prevents any analysis of the impact of price negotiations on prices for health insurance plans and consumers.


Tendering by indication has the potential to foster competition between medicines that are still patented, providing a lever to increase spending efficiency not commonly applied in countries with price regulation, particularly where the use of reference prices (maximum reimbursement amounts) discourages discounting to the payer.

A tiered formulary may be thought of as a form of value-based benefit design that can be used to promote the preferential use of high value products, and to discourage, through higher co-payments or narrower coverage conditions, the use of lower value products. Tier placement has the potential to send unambiguous signals to prescribers and patients alike regarding care choices that are of greater or lesser value.


Tendering processes that reduce prices very aggressively risk creating market segments that may become unattractive to manufacturers. This can lead to future problems of supply, and with market concentration, potentially, monopoly pricing. Clinical guidelines or formulary placement driven by the results of tendering processes may be seen by some patients and clinicians as limiting their therapeutic choices. Guidelines, however, could be developed by physicians’ societies or HTA agencies, and routinely incorporate cost effectiveness considerations in prioritising treatment options.

Where tiered formularies are determined by price rather than cost effectiveness, priority may not be given to the highest value or most effective products.

Enabling conditions

Tendering by indication is most effective in reducing prices when the outcome is “winner takes all”, but as noted above, this carries attendant risks to ongoing supplier participation in the market and security of supply. By contrast, not limiting the number of potentially successful suppliers creates minimal incentive for deep discounts.

One possible approach is to limit the number of bids accepted and set the reimbursement amount at the level of the lowest unsuccessful bid. This ensures a profit margin for successful tenderers by paying them a price higher than their tendered price (thus encouraging them to remain in the market), but retains incentives for suppliers to offer their best prices in order to be among the successful bidders.

Tiered formulary management is widely used by PBMs and health insurance plans in the United States. Its use in countries with national positive lists is not known, though it is not incompatible with value-based pricing as applied in 4th hurdle systems.10

Option 7: Explore bundled payments for episodes of care in oncology


Payment per episode of care rather than for individual services and products is expected to encourage health care providers to the most efficiency mix of inputs and engage in strategic purchasing to get better prices. In oncology, current pilots suggest that episode-based payments can generate savings, mainly stemming from more efficient use of resources. The ability of such payments to contain medicine costs might be more limited, especially when there is no competition for a given indication. Countries could, however, explore the feasibility of bundled payments in oncology.


In oncology, many new treatments are used in combination with existing ones and/or in step therapy protocols, leading to escalating treatment costs. Yet the prices of individual products are often negotiated with individual companies sequentially, and in some cases, with multi-year commitments.11 In parallel, provider payments for oncology treatments sometimes create incentives to use the most expensive treatment when cheaper alternatives exist. For example in France, in order to guarantee patient access, the most expensive oncology medicines are paid for over and above DRG tariffs, creating an incentive for hospitals to use these drugs instead of cheaper medicines included in the tariffs. In the United States, providers administering chemotherapies to patients covered by Medicare Part B are paid through a “buy and bill” system in which physicians receive a payment equal to the average sale price of the drug in private markets, plus a margin of 6%, thus incentivising the use of the most expensive medicines (Newcomer, 2012).

A bundled payment is a fee intended to cover the costs of a suite of services, delivered by one or several providers, as an alternative to individual payments for each service or product used in the process of care. The term is very generic and covers a number of different concepts. Payments per case (DRG-like), used in hospitals in many countries are bundled payments. Bundled payments per episode of care pay for a suite of services provided around an acute “episode of care”. They typically cover inpatient activities (e.g. elective surgery), and pre and post intervention visits for a set period of time (OECD, 2016).

The main objective of bundled payments is to encourage providers to be more efficient when treating patients. Increased efficiency can be derived from changes in the mix of inputs or from changes in the prices of individual inputs. In oncology, physicians typically make decisions concerning the mix of inputs. These decisions can be influenced by clinical guidelines or even constraints imposed by national or local formularies, but a bundled payment per episode could also encourage physicians to use the most cost-effective mix of inputs. The ability of individual providers to influence oncology drug prices may, however, be limited. Active formulary management or tendering may offer some opportunities to foster competition (see Option 6), but this requires the existence of several therapeutic options perceived as alternatives.

Bundled payments in oncology are currently being tested in the United States (Deloitte, 2016; Social and Scientific Systems, 2016; Spinks et al., 2017). Identified payment models falling into this category are presented in Table 3.1:

  • The 21st Century Oncology’s radiation bundle model, introduced in 2012, provides an episode-based payment which covers external beam radiation therapy services for 13 common cancer diagnoses, from the first consultation to 90 days after treatment (Falit, Chernew and Mantz, 2014).

  • The UnitedHealthcare Oncology Episode Pilot Program, tested between 2009 and 2012 in five sites and about 1000 patients, proposed bundled payments for 33 episodes of care related to breast, colon and lung cancer. The bundle included physician hospital care, hospice management care, and case management, while other services remained paid on a fee-for-service basis. Oncologists received a single fee for chemotherapy medications, set at the average sales prices and without a margin.

  • The Head and Neck Bundled Payment Pilot provided a one-year episode payment for patients with head and neck cancer. Providers were paid through four risk-adjusted bundled payments, with stop loss provisions and the possibility of adjusting payments in case of need (Spinks et al., 2017).

Table 3.1. Oncology bundled payments tested in the United States


Services included in the bundled payment

Contracting parties and payment arrangements

21st Century Oncology’s Radiation Oncology Bundled Payments

(From 2012)

The bundle includes professional and technical services delivered in its facilities for external beam radiation therapy services for 13 common cancer diagnoses, from the first consultation to 90 days after treatment.

Payment does not cover indirect treatment expenses such as medications, laboratory tests, and diagnostic imaging, and are not risk adjusted.

21st Century Oncology Center (largest provider of radiation therapy, with 180 facilities) and Humana

UnitedHealthcare Oncology Episode Pilot Program (2009-2012)

Bundled payment covered physician hospital care, hospice management and case management.

All other services (i.e. physician office visit, chemotherapy administration, diagnostic radiology, laboratory) were paid as per the pre-existing FFS arrangement.

Drugs were paid at average sales price.

UnitedHealthcare partnered with 5 US oncology practices to pilot an episode payment model for treating nineteen discrete “episodes” in breast, colon, and lung cancer with evidence-based treatment regimens.

Head and Neck Bundled Payment Pilot

Four bundles defined for different treatment plans. Bundles included multidisciplinary cancer care for one year.

UnitedHealthcare and University of Texas

MD Anderson Cancer Center (provider)

Note: FFS: fee-for-service; P4P: pay-for-performance.

Source: Social and Scientific Systems, 2016; Spinks et al., 2017.

Bundled payments for a number of chronic diseases treated in hospitals have been used in Portugal since 2007: HIV infection, multiple sclerosis, pulmonary hypertension, different lysosomal storage diseases, familial amyloid polyneuropathy and selected cancers (i.e. breast cancer, cervical cancer, and colorectal cancer). Hospitals receive an annual payment for two years, and outcomes are being monitored (survival rates at 6, 12, 18 and 24 months). Payment rates were initially defined by applying average costs of individual services to standards of care (Lourenço, 2016). Further information was not available at the time of writing.


Evidence on the impact of bundled payments for episodes of care in oncology or other expensive treatment settings is emerging. In the United States, for example, the costs of erythropoiesis-stimulating agents were included in Medicare payments for end-stage renal disease (ESRD) in 2011, to replace payments for the medicines on top of the treatment. Erythropoiesis-stimulating agents were one of the largest cost line-items for the Medicare program. The new bundled payment has been successful in containing spending growth (Levinson, 2014). In oncology, an assessment of the United Healthcare Oncology pilot program showed a 34% cost reduction for the entire episode of care, in spite of a 179% increase in chemotherapy costs (Newcomer et al. 2014). Interestingly, studies show that other alternative payment models, such as “clinical pathways” or add-on payments to providers for enhanced care delivery (e.g. Oncology Care Models tested in the United States12), also have the potential to improve cancer care and reduce costs, even though they may not affect pharmaceutical costs (McClellan and Thoumi, 2015).


If a payer’s objective is to contain medicine prices and costs, bundled payments may not be the solution. Even though providers might be incentivised to negotiate prices or tender to get better prices for the medicines they use, they will only be able to obtain price advantages if several treatments compete within a given indication. Opportunities for intra-class competition in oncology would need to be assessed. For example, approval of multiple PD-1 and PD-L1 inhibitors might trigger price competition (Seiden, 2016). However, competition may remain limited among medicines recently approved for a previously unmet need. The advent of precision medicine, with treatments increasingly tailored to individual patient characteristics, may also reduce opportunities for competition. Finally, evidence from one pilot of bundled payments showed an increase in pharmaceutical costs – not specifically attributed to a price increase - despite a decrease in total costs.

If poorly designed or executed, systems that provide bundled payments may potentially put providers or patients at risk. These problems, and possible mechanisms to avoid them, have been widely documented in the literature on DRGs (Busse et al., 2011). Because bundled payments shift some financial risk from payers to providers, the latter may respond to financial incentives inherent in the system by providing sub-standard care, engaging in risk-selection among patients, denying high-risk patients access to innovative treatments, or choosing inferior products. This may happen, for example, if payment tariffs are too low or not updated frequently enough, or when the quality of care is not rigorously monitored. Bundled payments are therefore only suitable where treatments are homogeneous enough to establish meaningful average costs across providers, appropriate payment tariffs can be set and updated regularly, and sufficient data are available to monitor the quality of care and take remedial action if sub-standard care is observed.

Enabling conditions

Bundled payments for episodes of care have the potential to increase efficiency in cancer care delivery through more efficient use of inputs. Given the financial incentives that bundling can create for providers, appropriate quality and outcome indicators would be needed to ensure quality and protect patients from risks of substandard care (Mcpherson, Hedden and Regier, no date; McClellan and Thoumi, 2015). In addition, if bundled payments were to include pharmaceutical costs – which has not always been the case in pilot projects – tariffs defined for episodes of care would require regular revision to stay abreast of rapid treatment evolution in oncology. Otherwise, patient access to new treatments could be compromised by the emergence of large gaps between actual costs and payments. It would also be necessary to take into account the common practice of off-label prescribing in oncology.13 Payers may need to decide whether to encourage or discourage this practice, which would in turn influence approaches to tariff setting.

The ability of bundled payments to influence medicine purchase prices would depend on the existence of intra-class competition. If bundled payments were introduced with this objective, they would need to target conditions, indications, and stages of treatment in which several drugs can be reasonably considered as suitable alternatives.

Option 8: Promote competition in off-patent markets


To improve efficiency of spending through savings derived from generics and biosimilars it is important to promote and maintain competition in off-patent product markets. Increasing the penetration and uptake of generics and biosimilars can deliver savings without loss of benefit for patients, by moving prices closer to marginal costs of production.

A number of policies can support the uptake of generics and biosimilars, such as encouraging early entry of new suppliers upon loss of exclusivity (LoE) of originator or reference medicines, encouraging physician prescribing by INN, strengthening the role of pharmacists, and incentivising and educating patients. In addition, price competition can be fostered by appropriate procurement mechanisms, provided that several manufacturers are active in each market segment.

Mechanisms to influence generic and biosimilar prices include competitive tendering, albeit avoiding sole-supplier or “winner-takes-all” arrangements, as these can lead to suppliers exiting the market, creating risks not only to the security of supply but also of market concentration that can increase prices in the long run.


The potential of generic markets is not yet fully realised in all OECD countries. For example, while the market share of generics by volume exceeds 80% in countries such as Germany and the United States, it is only approximately 52% across all OECD countries and below 30% in countries such as Greece, Italy and Switzerland.14 The penetration of biosimilars also varies significantly between countries, depending on the therapeutic class (Rémuzat et al., 2017). The savings potential of increased generic and biosimilar uptake therefore remains substantial (QuintilesIMS Institute, 2016).

In the United States, where the overall market share of generics is near 90% of the total volume of medicines dispensed by community pharmacies (OECD, 2017a), a recent study found that the market is highly concentrated: 40% of generic molecule-dosage-forms are supplied by a single manufacturer, 20% by two or three manufacturers, and the remaining 40% by more than four manufacturers. In addition, the share of molecule-dosage-forms supplied by one or two manufacturers increased between 2004 and 2016, and price increases were positively correlated with supplier concentration (ibid.). Although increases in the prices of off-patent drugs have so far had limited impact on overall spending (see Chapter 1), there is a risk that in the future prices of single-source generics will increase further due to the lack of competition.

Countries can improve policies to increase the efficiency of spending through greater use of generics and biosimilars, and can implement mechanisms to ensure that markets remain competitive. The 2017 OECD report on “Tackling Wasteful Spending in Health” suggested a number of policies in two main categories: 1) policies that aim to increase the availability of generics on the market, including entry-level legislation and pricing methods; and 2) policies that steer behaviour of main stakeholders – physicians, pharmacists, and patients – towards the use of generics already available on the market (OECD, 2017d).

Market entry of generic products after loss of market exclusivity (LoE) of originators could be facilitated through several mechanisms. For example, allowing generics manufacturers to complete regulatory requirements prior to LoE allows for more rapid entry to market. Readily accessible and easily searchable databases could be created to compile patents and exclusivity status of originator medicines. This is already the case in the United States, for example, where all approved small molecule pharmaceutical products, information on associated key patents and their expiry dates, and dates of LoE are available in a single database, known colloquially as the “Orange Book”15. Regulators could also identify areas where generic competition is limited or lacking, to prioritise generic approval in these areas. In June 2017, the US Food and Drug Administration (FDA) began publishing a list of branded medicines that had lost exclusivity but for which there were no approved generic competitors. It also expedites the review of generic approval applications for products with fewer than three approved generic suppliers (FDA, 2017).

Competitive processes that aim to balance short- and long-term efficiency and sustain strong competition could also be used to moderate generic and biosimilar prices. In general, evidence suggests that systems that use direct price regulation of generics, for example by imposing fixed discounts relative to originator products or using therapeutic reference pricing, are less effective in reducing prices than systems in which prices are established through competitive mechanisms such as tendering or negotiations (OECD, 2017d).

Countries could consider increasing the use of tendering for multi-source products. The three cornerstones of efficient tendering are: binding bids; a mechanism for setting the tender price that incentivises manufacturers to reveal their real costs of production; and appropriate selection of the number of suppliers (OECD, 2013a). Tenders can be designed to take into account criteria other than price, such as product quality or reliability of supply, and to avoid awarding the entire market volume in a segment to a single manufacturer. Reducing prices too aggressively can be counter-productive in the longer term, as individual market segments may become insufficiently attractive for manufacturers to remain active (Kanavos et al., 2012). Sole-supplier contracts can be particularly counter-productive in terms of preserving long-term competitiveness and avoiding monopolies, and can lead to higher prices in the longer run.

In countries where retail price competition in off-patent markets is impacted by the use of internal reference pricing,16 competition for market shares may take the form of discounting in the supply chain, and discounts may not be passed on to either the payer or the patient. To benefit from this type of discounting, in 2007 Australia introduced a programme of “price disclosure” (see Box 3.3).

Box 3.3. Use of price disclosure in Australia

Under Australia’s price disclosure policy, suppliers of multi-source medicines to the Pharmaceutical Benefits Scheme (PBS) are periodically required to submit data on sales revenue, sales volume, and the value of incentives (such as discounts or bonus stock) offered to community pharmacies to preferentially dispense their products.

Based on this information, the reimbursement price paid by the government is adjusted to reflect more closely the true average market price at which the medicines are supplied.

First introduced in 2007, price disclosure arrangements now apply to more than 400 products. The policy has resulted in very significant price reductions and the consequent savings to the PBS are expected to reach approximately AUD 20 billion by 2019-20.

As part of its price disclosure policy, Australia has also implemented moderating policies which aim to ensure that more than one supplier of each off-patent medicine remains in the market. This is intended to minimise the risk of exposure to large price increases that could arise if competition were to reduce prices to a level at which suppliers dropped out of the market leaving a single monopoly supplier of the remaining brand of a medicine.

Source: /industry/pricing/eapd/faq-for-pharmacies-and-wholesalers.

Policies that encourage generic or biosimilar penetration through steering the behaviour of main stakeholders, include (OECD, 2017d; Rémuzat et al., 2017):

  • Mechanisms that encourage or incentivise physicians to prescribe generics or biosimilars, such as guidelines; prescribing software that highlights price differences or selects the cheapest alternative by default; financial incentives for pharmacists to dispense, and for patients to accept generic/biosimilar products; and mandating prescribing by international non-proprietary name (INN).

  • Allowing or requiring (unless explicitly precluded by the prescriber) pharmacies to substitute originators with generics (and reference products with biosimilars where appropriate) and avoiding pharmacy remuneration schemes that encourage the dispensing of more expensive drugs (e.g. use of progressive mark-ups).

  • Educating professionals and patients through information campaigns and incentivising patients to be sensitive to product prices, for example through larger co-payments for more expensive products.

Finally, countries could also implement a system to monitor market dynamics and allow purchasers to report sharp price increases if procurement mechanisms are unsuccessful in avoiding market concentration. When the number of competitors becomes low (for example, less than three) payers could begin negotiating with generics manufacturers to re-enter the market. This might entail a commitment to a minimum price for a limited period of time in order to increase the reliability of supply. Enforcement of competition and antitrust laws could also provide recourse for countries.


Creating conditions that promote competition in the off-patent market using tendering or price disclosure can dramatically reduce generics prices.

Encouraging generic and biosimilar uptake is a straightforward means of increasing spending efficiency without reducing benefits to patients. Fostering competition in off-patent markets, through facilitating market entry of competitors and appropriate procurement mechanisms that avoid the formation of monopolies, can help moderate prices while avoiding the need for direct price regulation.

Making information on patent content is an integral part of the social contract that underlies patent protection, which aims to make innovations widely available to society. It could also be argued that access to readily available information on the status of patents and of other forms of market exclusivity is in the public interest. Providing such information in an easily accessible format is a simple way of enhancing transparency and facilitating competition. In the longer run, legal action against firms that abuse dominant market positions can act as a deterrent.


Overly aggressive discounting by manufacturers in order to gain market share can drive competitors out of the market and reduce competition in the longer term. Policy approaches may therefore need to include mechanisms that increase the likelihood that several suppliers remain active in each market segment, such as tenders that allocate specified market shares to competing manufacturers. Competitive mechanisms that lead to market concentration, such as sole supplier arrangements are best avoided. Policies may also need to include mechanisms for monitoring the competitiveness of market segments and provide for remedial action when segments become overly concentrated.

Enabling conditions

Mechanisms for making patent information readily available are inherently valuable and should not be associated with the imposition of barriers to generic market entry, in particular patent linkage. Patent linkage refers to linking marketing approval of a generic medicine to the patent status of the originator. This requires the regulatory authority to delay marketing authorisation until either the expiry of patents covering the originator (or reference product in the case of a biologic) or a decision that the relevant patents are not infringed, or are invalid or unenforceable (Roox et al., 2008; Bhardwaj, Raju and Padmavati, 2013). While this policy is intended to protect originators from lost sales caused by infringing products, it can lead to delays in market entry or unwarranted delay in authorisation if, for example, a patent is subsequently challenged successfully. Allowing generics manufacturers to receive marketing authorisation while the originator is still under patent can accelerate the entry of generic products upon LoE.

C. Determine willingness to pay for new treatments and health benefits

Option 9: Define explicit and firm criteria for coverage and pricing


Aligning coverage and pricing of new medicines with societal preferences and willingness-to-pay for healthcare may enhance both efficiency of spending and patient access. National authorities responsible for pricing and coverage could establish clear criteria for coverage and pricing decisions that reflect societal preferences, through a transparent and procedurally fair process. This could improve dynamic efficiency by making coverage and payment decisions more predictable to industry, rewarding effective treatments and providing consistent signals to guide investment decisions in the development of innovative treatments.

However, to achieve these goals, countries would need to adhere firmly to their stated criteria and not deviate from established rules. Governments or payers may nevertheless want to retain some flexibility in decision making to take into consideration constraints or circumstances specific to a decision that were not foreseen in pre-specified criteria. Willingness to pay for a given drug may legitimately vary across countries and decision rules will therefore generally be defined at the national or sub-national level, where health care budget decisions are made.

A rules-based process to make coverage and pricing decisions could also provide for resolution mechanisms that are invoked when the pre-specified rules preclude coverage of a new medicine at the price offered by the manufacturer, and negotiations to reach an agreement within the rules are unsuccessful. Such mechanisms may need to take into account budget constraints and afford significant authority and flexibility to governments, payers or independent arbiters to incentivise manufacturers to operate within the pre-specified rules, and avoid imposing decisions on payers that are not affordable.


Most OECD countries regulate the prices of publicly funded on-patent drugs, with the objective of striking a balance among access to medicines, efficiency of public spending, and encouragement for companies to invest in R&D. While some countries directly regulate prices,17 others impose coverage conditions that affect prices indirectly. Both types of regulation require that authorities make decisions centrally on what the right price of a medicine ought to be. Decision makers use a mix of methods or criteria to make such decisions, and the methods and criteria are not uniform across countries. These methods include, for example, international benchmarking; internal reference pricing (reference to the price of existing competitors); and economic evaluation (OECD, 2008; Panteli et al., 2016).

In OECD countries that use economic evaluation, assessments of value are often based on comparative cost-effectiveness analysis (CEA)18 to determine whether the price of a treatment is “acceptable”, or to express its value to the healthcare system in terms of its opportunity cost. CEA19 is usually preferred over cost-benefit analysis for two reasons. First, most experts consider that health budgets should only be used to maximise health improvements and not for other purposes (such as increasing labour productivity). Second, using cost-benefit analysis requires the monetisation of health gains, which raises many theoretical questions and practical issues. In principle the use of CEA requires the definition of a cost-effectiveness threshold, beyond which new medicines would not be covered (Culyer, 2016).

Cost-effectiveness is, however, generally not the sole criterion and new treatments may be accepted with very different levels of cost-effectiveness due to a number of factors (see Figure 3.3). The market for anti-hypertensives, for instance, is crowded with multiple products, many of which are relatively inexpensive off-patent drugs. This makes it challenging for manufacturers to develop new treatments with sufficient incremental benefits and prices that would allow them to approach a pre-determined cost-effectiveness threshold. On the other hand, some treatments may be accepted with incremental cost-effectiveness ratios (ICERs) higher than the implicit or explicit pre-defined threshold (see Chapter 1 and OECD, 2017d). This is currently the case in particular for oncology and orphan medicines, which are often funded through “special access” paths that circumvent rules applied in other areas. This is partly explained by the fact that the assessment of the “value” of a medicine may include factors other than cost-effectiveness (see Box 3.4).

Figure 3.3. Incremental cost-effectiveness ratios and burden of disease associated with some treatments

Note: X-axis represents the burden of disease expressed as “proportional shortfalls”.

Source: Adapted from Stolk et al., 2004) and Zwaap et al. (2015).

Box 3.4. Value-based pricing and societal willingness to pay

Value-based pricing commonly refers to one of the strategies firms can use to set the prices of their products. “Value-based pricing is the setting of a product or service’s price based on the benefits it provides to consumers. […] it relies on the consumers’ perception of the value” (Investopedia, 2017). More precisely, “Value-based pricing is the method of setting a price by which a company calculates and tries to earn the differentiated worth of its product for a particular customer segment when compared to its competitor.” Setting such a price requires referring to a specific market segment; identifying attributes that differentiate this product from existing competitors; and understanding how consumers value this differential or how much they are willing to pay for it (Dholakia, 2016). Market research, such as consumer surveys, can help to determine this amount. A company can for instance determine how much a consumer is willing to pay for a TV with a larger screen -all other things being equal- or how much a consumer values a specific brand in fashion products (ibid.).

Pharmaceutical companies may use value-based pricing to determine their pricing strategies. If a new product has no comparative advantage over existing competitors, it might be expected to enter the market at a lower price to capture market share. If the product offers some benefits over existing treatments, it is likely to command a premium over the price of competitors. The US Senate report on Gilead’s pricing strategy for Sovaldi® describes an example of how companies may think when setting the price of their products according to such a rationale (Committee on Finance of the US Senate, 2015, pages 29-30).

In many countries, however, governments or compulsory health insurance schemes are the main payers for “covered” pharmaceuticals. Their willingness to pay (on behalf of tax payers or the insured population) is therefore what companies need to understand in order to propose a price that will be considered acceptable. Many dimensions may be taken into account when assessing the value of new products: improvement in length and quality of life, obviously, but also convenience or cost-savings in other parts of the health system; gains in labour productivity for patients and carers, etc. Previous studies of “value-based pricing” have shown that countries most often adopt a health system perspective, rather than a wider societal perspective when making coverage or pricing decisions, but also that added therapeutic benefits (QALYs in the most formal and quantified approaches) are not the only attributes public payers value when they determine their willingness to pay. The rarity and the severity of the disease treated, as well as a high unmet medical need, often lead payers to accept higher prices for products for some disease categories (Paris and Belloni, 2013).

Many public payers have been reluctant to publish an explicit cost-effectiveness threshold beyond which they would decline to cover products. Most countries using cost-effectiveness as a criterion for coverage and/or pricing, such as Australia and the United Kingdom, use an implicit or explicit threshold to support decisions, but remain flexible in taking into account other factors that may be relevant to the decision. Signals about willingness to pay for different factors, such as rarity or severity of the disease or degree of unmet need, are thus not always clear and consistent.

While value-based pricing remains an appealing mechanism in principle, a consistently workable way to apply it in practice has yet to be found. Defining the concept of value and corresponding assessment methods may require additional debate among stakeholders (including government, payers, patients, the industry and the general public) to reach agreement. This would help payers to behave consistently in their decision making so that signals to industry reflect this concept of value. For example, formal Multi-Criteria Decision Analysis (MCDA) methods could be developed to incorporate factors other than health gains into a definition of value (Angelis and Kanavos, 2016). Such methods allow an explicit consideration of stakeholder and public preferences in trade-offs.

One option to account for severity while retaining a central role for the cost-effectiveness criterion is to adopt different ICER thresholds for different levels of disease severity. In 2015, the Dutch Health Care Institute proposed ICER reference values that varied with the level of severity of the disease treated, measured in terms of proportional shortfalls20 (see Zwaap et al., 2015 and Figure 3.3). In Norway, rules and criteria for priority setting in health care were recently updated after several commissions and a wide consultation on priority setting (see Box 3.5).

Box 3.5. Priority setting in Norway

In the past 30 years, five government commissions have been appointed to evaluate principles for priority setting in the health care sector, in 1987, 1997 (2), 2014 and 2015. In 1997, the Lønning II Commission recommended that priority setting in the Norwegian health care services should be based on three criteria: severity, expected benefit and cost-effectiveness. These criteria have been further defined and recommendations submitted to a consultation. Updated criteria for priority setting are the following (Norwegian Ministry of Health and Care Services, 2017):

  • The benefits of a new intervention are assessed qualitatively and quantitatively. Clinical benefits include direct health improvements, as well as indirect benefits such as improvement in family members’ health. The economic value of increased productivity at work is not included. In HTA, quality-adjusted life years (QALYs) are used as a quantifiable measure of healthy life years.

  • An intervention is assessed against its opportunity cost, which is the benefit to other patients that could have been realised with the same resources. A cost-effectiveness ratio is calculated to assess opportunity cost. Any ICER below or at NOK 275 000 (~ USD 32 000) per QALY is acceptable.

  • This ICER threshold is adapted to take into account the severity of the disease treated. Disease severity is assessed using the criteria of absolute shortfall. Absolute shortfall expresses the number1 of healthy life years lost by a patient group as a result of a disease as compared with the average expected healthy life years for the population of the same age. Six severity classes are proposed, with different ICER thresholds and the threshold for the highest level of severity is set at NOK 825 000 (three times the lowest level).

  • Additional discretionary assessments must be used, along with HTA to make decisions, which are: the quality and uncertainty of the documents (greater uncertainty justifies a lower priority) and the overall budget impact of an intervention. Interventions with a high overall budget impact will have higher opportunity costs, which may justify phased adoption.

  • The rarity of a condition does not in and of itself justify a different set of criteria for priority setting. However, less stringent requirements for documentation may be acceptable when assessing interventions targeted towards small patient groups with a severe condition where it is difficult to perform controlled outcome studies. In addition, higher resource use may be accepted for interventions targeted towards very small patient groups with an extremely severe condition, such as children with congenital genetic diseases, where there is often a lack of good documentation of the benefit of an intervention.

  • There are no discrete criteria for “end-of-life” therapies.

The transparency of the process facilitates equitable health care services and ensures the democratic legitimacy of decisions. However, it means that buyers, i.e. the hospitals and the authorities, could miss out on substantial discounts. The Norwegian government acknowledges that, ideally, discounts should be public, but this may require greater coordination at the European level to safeguard the principle of tiered pricing.

1. By contrast, proportional shortfall (suggested to assess severity by the Dutch Health Care Institute) expresses the proportion of anticipated life expectancy and quality of life lost by a patient group as compared with the average anticipated life expectancy and quality of life for the population of the same age. In the opinion of the Norwegian working group, it is more serious to lose 20 of 40 remaining healthy life years than to lose one of two remaining healthy life years.

Some OECD countries give explicit consideration to budget impact when determining their willingness to pay for particular products, such as those that treat large populations. Given that production costs of medicines are often relatively low, high volumes (e.g. due to high prevalence and/or life-time treatment) may still be attractive for investors even if the unit price of the product is lower than it would be if the cost-effectiveness of the product were considered as the sole criterion for pricing. For example, many payers have already entered into agreements with manufacturers that maximise the number of patients treated with the available budget or that include volume-price discounts. This approach was eventually taken in several countries for direct-acting antivirals (DAA) for hepatitis C. Rather than ending up with this outcome after some negotiations, the conditions which would trigger this could also be defined in advance. Even though the occurrence of such events is likely to be infrequent, it is nevertheless important to give innovators an indication of what they can expect to receive for a breakthrough treatment for a high-prevalence disease, with the obvious example being a treatment for Alzheimer’s disease.

In the Netherlands, for example, horizon scanning is used to anticipate the market entry of new drugs that, based on list prices in other countries, are expected to have a high budget impact. Medicines with a budget impact estimated to exceed EUR 40 million per year, or those exceeding EUR 10 million per year and a cost per patient of EUR 50 000, are temporarily suspended from reimbursement and subject to HTA and price negotiation by the Ministry of Health, Welfare and Sport. In England, the National Institute for Health and Care Excellence (NICE) introduced a “budget impact test” in 2017 whereby NHS England may enter into price negotiations for health technologies with an anticipated NHS budget impact exceeding GBP 20 million in any of the first three years of marketing (NICE, 2017).

A rules-based system could also provide for a resolution mechanism, to ensure that effective medicines are available to patients even if national authorities or payers are not able to agree prices with manufacturers within a reasonable timeframe after marketing authorisation. Such a system exists in Germany, for example, where an arbitration board (composed by an impartial chairman, two other impartial members, in addition to two members appointed by both parties to the negotiations) is empowered to determine a price for the German market within three months of a failed negotiation (Wenzl and Paris, 2018). Mechanisms like this may need to take into account budget constraints and confer significant authority and flexibility on governments, payers or an independent arbiter in order to avoid imposing decisions on payers that are unaffordable and to incentivise manufacturers to operate within the pre-specified rules. In Germany, for example, prices set by arbitration have been found to be significantly lower than those offered by the industry during preceding negotiations (Ludwig and Dintsios, 2016). Other countries, such as Australia, only allow for judicial review of procedural issues in coverage decisions, with the authority to invalidate decisions for procedural reasons, but not to overturn or invalidate decisions on their merits. Such a mechanism avoids introducing parallel routes to coverage decision making, which could produce decisions that are unaffordable for the public payer.


Transparent consensus criteria would make coverage and pricing decision more acceptable to societies and relieve decision-makers of some of the costs of making decisions on a case-by-case basis. As there is no purely technical solution, the legitimacy of any decision-making criteria would be enhanced were they to be widely debated and agreed.

A rules-based process for making decisions on coverage and pricing that includes firm indications of the willingness-to-pay of national authorities or payers, would increase their ability to align their expenditure better with public preferences. Enshrining pre-defined resolution mechanisms to stalled negotiations in legislation could ensure that public authorities are able to enforce these rules. Efficient resolution mechanisms, such as arbitration, could also help ensure that decision outcomes are aligned with pre-defined criteria while ensuring each party due process. Such mechanisms could be designed to give manufacturers an incentive to operate within the rules and avoid situations where rules may be circumvented through parallel paths to coverage.

At the same time, a more rules-based process would provide clearer signals and predictability for pharmaceutical companies, and align their incentives for R&D with public preferences. It would provide manufacturers with larger rewards for treatments to which society attaches high value, and smaller rewards for those valued less highly.


There is no straightforward means of setting an optimal spending cap for highly prevalent diseases for which effective medicines become available. What constitutes an adequate but not excessive budget reflects social value judgments, and would need to be determined through a political process and by authorities responsible for insurance coverage and pricing decisions.

An important factor to take into consideration when defining pricing criteria is that explicit thresholds or reservation prices can have powerful anchoring effects for subsequent negotiations, as well as for other mechanisms used for setting transaction-specific prices. Anchoring effects have been documented widely in the psychology and behavioural economics literature (see Box 3.6).

Conducting economic evaluations, such as cost-effectiveness analysis, requires resources and skills that may not be readily available in all OECD countries. More co-operation between countries has the potential to create efficiencies, especially in the assessment of clinical benefits. However, due to differences in the patterns of care and costs of health care inputs, including prices of comparators, economic evaluation can generally only be performed at the national level.

Box 3.6. Behavioural economics and anchoring effects

Anchors are cognitive heuristics that allow people to process numerical information but lead to biases because, when required to value an option under uncertainty, any relevant information presented subsequent to exposure to the anchor induces people to adjust their valuation from the anchor rather than valuing it based only on the relevant information (Tversky and Kahneman, 1974). Research has shown that these adjustments are usually insufficient given the information people receive so that an initial anchor has a significant effect on the final outcome of the valuation (ibid.). This can be applied directly to price negotiations and it has been shown consistently, across a wide variety of settings and population groups, that exposing people to a number prior to a price negotiation biases subsequent offers and final negotiated prices significantly (Furnham and Boo, 2011). People who have significant knowledge of the field in which prices are negotiated, such as professional purchasers, are also prone to this bias, although anchoring effects have been found to operate differently in experts (ibid).

Buyers can improve negotiation outcomes from their point of view by defining their optimal prices ahead of negotiations and making optimal prices the anchor. The literature suggests that there are risks in only defining a reservation price. Indeed, evidence shows that when buyers anchor their negotiation in a reservation price, subsequent offers and the final prices negotiated are significantly higher than when buyers make their first offer based on their optimal price (Galinsky and Mussweiler, 2001). Subsequent offers and final negotiated prices are also higher when sellers make the first offer, for instance by offering a discount from a high list price (ibid.). The latter scenario currently occurs when pharmaceutical manufacturers negotiate confidential discounts from a list price with individual purchasers.

Enabling conditions

Eliciting public preferences coherently is methodologically difficult, in part people need to be presented with decisions that involve complex trade-offs, and extensive data collection is also required. All of the proposals described above require value judgments. This calls for a more structured and transparent approach to eliciting such preferences. Possible approaches include broad public consultations, population surveys, or a political process that debates these concepts and aims to enshrine preferences in explicit rules or legislation.

To ensure predictability for stakeholders, it is important that clear criteria are established in each country as to when resolution mechanisms may be invoked as a response to stalled negotiations. The criteria could, for example, include the absence of therapeutic alternatives, limits to delays between marketing authorisation and reimbursement and pricing decisions, minimum thresholds of effectiveness, or budget impact.

Option 10: Optimise the use of Managed Entry Agreements


Performance-based managed entry agreements (MEAs) could be used in a targeted manner for new medicines whose effectiveness or cost-effectiveness is particularly uncertain, in order to align coverage and pricing with clinical outcomes. Similar to using routinely collected data, information generated from data collected under MEAs can be used to adjust prices and coverage conditions to reflect “real world” performance of new medicines, increasing efficiency of spending and aligning incentives for innovation with the benefits that medicines deliver to patients.

The following policies may increase benefits of performance-based agreements to health systems: 1) making evidence of clinical benefits and effectiveness generated under such agreements available to the scientific community; 2) designing agreements to incentivise firms to demonstrate performance of the product; 3) disclosing the existence of agreements in order to reduce opacity in international markets; and 4) clearly defining performance measurements. At the same time, MEAs can impose additional administrative burden on providers, payers and patients, which may offset some of the potential benefits.


In some OECD countries, MEAs (sometimes also referred to as risk-sharing agreements, special pricing arrangements, performance-based agreements, outcome-based agreements or patient access schemes) have increasingly been adopted between payers or regulators and pharmaceutical companies to accelerate access to innovative medicines or manage their financial impact. However, recent reviews find that little is known as of yet about the actual effects of MEAs and that evaluations are often limited to perceived advantages and disadvantages (Gerkens et al., 2017; Kanavos et al., 2017). In many countries confidentiality about the existence of MEAs and their content also makes it difficult to provide a complete and accurate overview of the agreements currently in place.

The main goals of MEAs are to reduce uncertainty about the safety, effectiveness or cost effectiveness of treatments (often present at the point of marketing authorisation or reimbursement decision), or to manage the financial impact of new medicines on healthcare budgets, for example, through confidential price discounts or rebates (Kanavos et al., 2017). Based on their goals, MEAs can be grouped into two broad types. These can then be broken down into further categories as illustrated in Figure 3.4, which shows the taxonomy of MEAs developed by the Belgian Healthcare Knowledge Centre (KCE).

Financial agreements often consist of confidential discounts or rebates, which may or may not be conditional on factors such as volume or expenditure thresholds, but are not linked to the performance of treatments and do not require the collection of data on health outcomes. Such agreements may reinforce opacity in the pricing of medicines, allowing net prices to remain below list prices that are referenced by other payers or regulators, or to serve as anchors in price negotiations elsewhere.

Performance-based agreements entail the collection of data on health outcomes, and make insurance coverage, payment or rebates contingent on the clinical performance of the product in practice. These agreements contain a financial provisions but are mainly prompted by the need to reduce uncertainty about the clinical or cost effective of a medicine through additional data collection. They also occasionally serve as a mechanism through which to pay different prices for different indications of a single medicine.

Figure 3.4. A taxonomy of managed entry agreements (MEA)

Source: Gerkens et al., 2017.

Recent studies of MEAs in Australia, Canada and European countries, as well as a number of non-OECD member countries, have reported a proliferation in MEAs in the recent past, with the majority of agreements being financial (Ferrario et al., 2017; Kanavos et al., 2017; Pauwels et al., 2017; Rotar et al., 2018). A large number of MEAs are now found in Australia, Belgium, Italy, the United Kingdom and in a number of Central and Eastern European countries, such as Bulgaria, Estonia, Hungary and Slovenia. Significant activity related to MEAs has also been reported, for example, in France and Sweden.

  • In Australia, 98 active agreements were reported in 2015, most of which were financial arrangements, although some performance-based agreements were also implemented (Lu et al., 2015). By March 2018 the number of agreements had increased to more than 150 (Australian Department of Health, personal communication) but information on their nature and content was not public.

  • In France, neither the existence nor the content of MEAs is made public. A total of EUR 1 billion in refunds related to MEAs was reported for 2016, of which more than EUR 400 million was derived from price-volume agreements and only some EUR 120 million was attributed to performance-based MEAs (CEPS, 2017). This does not include ex post rebates on direct-acting antivirals (DAA) used for hepatitis C, which are subject to a specific budget cap.

  • In Italy, as at March 2018, there were 88 active MEAs using registries to monitor patient-level data. Of these 88, 46 were performance-based between payers and manufacturers, including discounts for non-respondent patients (locally referred to as “risk sharing”) and full refunds for manufacturers for all non-responding patients (referred to as “payment by results”); 42 were financial MEAs, through upfront discounts or volume caps beyond which manufacturers are liable for costs (AIFA, 2016, 2018). This number increased from 29 MEAs as of October 2012 (11 financial and 18 performance-based, related to 25 different products) (Gerkens et al., 2017). In 2016, EUR 693 million in refunds related to all active agreements were reported by AIFA (OsMed, 2017).

  • In Sweden, in 2017 the national Dental and Pharmaceutical Benefits Agency (TLV) reported 22 active MEAs at both the national level and the level of individual county councils. Refunds under all active agreements were estimated to amount to SEK 940 million in 2017 (TLV, 2017). Agreements were implemented to manage uncertainties regarding use and effect in clinical practice, cost-effectiveness (for example of cancer treatments) or high budget impact and displacement of alternative therapies (for example of HCV treatments) (ibid.); the exact nature of the agreements was not reported.

A survey of the experience in 16 Central and Eastern European countries21 identified 678 MEAs across these countries, of which 668 were financial agreements (Ferrario et al., 2017). The largest number of agreements took the form of confidential discounts (495, 73%).

According to the KCE review (Gerkens et al., 2017), information on the impact of performance-based MEAs on filling knowledge gaps or decisions related to coverage were either not publicly available or reported to be relatively minor. Reasons for the latter related to, among other factors, concerns over study quality (e.g. an inability to adjust for confounding factors in analysis based on observational study designs) or the accumulation of evidence outside the MEA which increased the redundancy of the agreements. In the Netherlands, for example, while evidence of insufficient cost-effectiveness was found for three orphan drugs, for none of them was coverage withdrawn, however (ibid.).


MEAs can have a number of advantages in theory, which include (Gerkens et al., 2017; Kanavos et al., 2017):

  • Accelerating access to innovative and potentially effective treatments while uncertainty remains, and avoiding rejection of insurance coverage as a result of uncertainty.

  • Improving the alignment of prices with comparative effectiveness.

  • Greater flexibility in funding new and expensive treatments, for which coverage may not be possible with more rigid payment methods.

  • Expanding the time horizon for data collection, thus enabling the capture of information on safety, effectiveness and cost-effectiveness in routine clinical practice (rather than in a controlled clinical research context) over longer follow-up periods than in pre-market RCTs.

Performance-based agreements can help focus treatment on patient groups likely to benefit most from an innovation, and can make use of existing sources of routinely collected data (see Option 5) to increase scientific knowledge (Gerkens et al., 2017).


General weaknesses of MEAs and related challenges include (Gerkens et al., 2017; Kanavos et al., 2017):

  • Significant administrative burdens and costs for providers, manufacturers and payers involved in executing the agreement and collecting data. Pure financial agreements are easier to administer than performance-based agreements and their proliferation suggests that payers consider that their benefits outweigh their costs.

  • Payers may face difficulties in reducing prices, recouping payments already made to manufacturers, or de-listing treatments from coverage if the data collected under MEAs show that the treatment is less effective (or less cost effective) than expected.

  • Regulators or payers may face challenges in generalising results to settings of care or countries other than those in which the data were collected, in particular when agreements take the form of coverage with evidence development (CED).

  • Confidentiality clauses can prevent the dissemination of new knowledge on the performance of medicines that is generated from performance-based agreements.

  • Uncertainty for manufacturers as to the financial returns from additional research produced, and the potential impact that the new evidence could have on future prices or revenues, can create dis-incentives for additional data collection once an agreement has been put in place.

In Italy, where experience with performance-based MEA has accumulated since 2006, the absence of transparency, high administrative burdens for payers and providers, and difficulties in obtaining refunds from manufacturers have been identified as issues (Garattini and Casadei, 2011; Garattini, Curto and van de Vooren, 2015). In 2016, the Italian Medicines Agency (AIFA) reported that registries captured 1.2 million treatments for 1.1 million patients as well as EUR 693 million in related refunds (OsMed, 2017). Recent evaluations of the effectiveness of MEAs in terms of accelerating access or reducing uncertainty, administrative costs, or the proportion of contractual refunds recovered by payers are not available. However, in 2012, of EUR 46 million in MEA-related refunds reported by AIFA, only EUR 31 million could be collected, representing 5% of the total expenditure for the products subject to the agreements (Garattini, Curto and van de Vooren, 2015). Disputes with manufacturers and delayed requests by hospitals were reported as the main causes of difficulties in collection.

Enabling conditions

A number of practical difficulties need to be overcome for MEAs to be implemented effectively. Difficulties have been reported thus far in establishing clear governance structures to determine when MEAs are appropriate, which data should be collected, how success and failure should be defined, what conclusions should be drawn from failures, or when the execution of an agreement should be discontinued (Gerkens et al., 2017). The increased use of MEAs and post-market data collection may undermine the role of RCTs before marketing approval and increase uncertainty regarding the safety and efficacy of treatments at the time of initial introduction into clinical practice (ibid.).

Capacity for data collection and analysis might need to be expanded in some countries. Performance-based MEAs have thus far been difficult to implement in many countries, among other reasons because data infrastructure and governance is not adequate to support the determination of outcomes. For example, few countries have adequate electronic health or medical records that can capture all the data necessary to determine outcomes.

D. Develop new push and pull incentives to encourage innovations in areas with high unmet needs

Option 11: Develop push incentives targeting product development


This policy option focuses on push incentives designed to encourage the development of specific products in therapeutic areas with unmet medical needs and suggests models to ensure wide accessibility of treatments subsidised by the public sector.


Public private partnerships (PPPs) can take multiple forms and their number has been increasing steadily in recent decades. Precompetitive PPPs “aim to generate novel scientific concepts (e.g. disease targets and research models) and infrastructures (e.g. databases) through effective collaboration between multiple public and private entities based on mutual trust, pooling of complementary expertise and knowledge, and sharing of rewards” (de Vrueh and Crommelin, 2017). They can involve a wide range of stakeholders (universities, industry, charities, patient organisations, regulatory agencies). The Innovative Medicines Initiative (IMI) for instance, which was set up to enhance the competitiveness of the pharmaceutical sector in Europe in 2008, is the largest life sciences R&D PPP. It is jointly and equally supported by monetary contributions from the European Commission and in-kind contributions from company members of the European Federation of Pharmaceutical Industries and Associations (EFPIA). The performance of precompetitive PPPs can be assessed against their capacity to generate immediate tangible scientific knowledge, products and/or services (ibid). They play an important and growing role in pharmaceutical R&D processes but do not generally aim to deliver new medicines.

Product development partnerships (PDPs) are a type of PPP that focuses on the development of a specific product. Examples include the Medicines for Malaria Venture; the Drugs for Neglected Diseases initiative (DNDi), the International AIDS Vaccine Initiative, and the Pediatric Praziquantel Consortium. PDPs were originally developed to address the gap between R&D investments and the unmet need for new treatments for diseases with a high burden in the developing world. For these diseases, market-based incentives are not sufficient for private sector organisations to invest in the discovery, development, and deployment of new interventions.

PDPs enable the development and introduction to market of products at affordable prices. In the DNDi model for example, R&D costs are funded up-front through public and philanthropic institutions. DNDi has built close collaborations with several pharmaceutical companies. Since 2007, it has delivered one new treatment per year and met its 2007 target of six novel treatments for malaria, Human African trypanosomiasis (“sleeping sickness”), visceral leishmaniasis and Chagas disease by 2014. All of these were able to be developed rapidly and with a high probability of success as they involved either re-purposing existing drugs or developing combination therapies. Development costs are low22 by comparison with available estimates for the industry, partly due to the unique features of the project (Maxmen, 2016).

Other options exist to encourage the development of specific products. For example, governments could fund development through long term contracts awarded on a competitive basis, with government retaining IP rights to the developed product to enable generic manufacture. Using the current grant-funded medical research model in most countries, a health research agency could offer grants to develop drugs vital for public health that are of no commercial interest to for-profit companies. An alternative could be that the firm that undertakes the R&D retains IP rights, but commits to share them through a patent pool (see Box 3.7).

Box 3.7. The Medicines Patent Pool

The Medicines Patent Pool (MPP) was created in 2010 and is funded by Unitaid. Its goal is to provide access to affordable, quality assured HIV, hepatitis C and tuberculosis medicines in low and middle-income countries (LMICs). In May 2018 the MPP announced expansion of its mandate to include licensing of other patented essential medicines, such as those included in the WHO Model list of Essential Medicines.

The MPP signs licence agreements with patent holders to allow generic manufacturers to develop and supply licensed medicines in a large number of LMICs.

As of now, the MPP has signed agreements with 9 patent holders for 17 products. The MPP partners with multiple generic manufacturers to promote competition and drive further price reductions. These partners have already delivered 17 million patient-years of treatment in approximately 130 countries. Each licence agreement specifies in which countries generic medicines may be manufactured and/or distributed. The access-oriented voluntary license agreements signed through the MPP are negotiated from a public health perspective and are publicly available on the MPP website.

Importantly, the MPP licences enable generic manufacturers to develop new formulations (e.g. paediatric formulations or fixed-dose combinations) that may be particularly important for treatment scale-up in LMICs. High-demand for some of these formulations in LMICs (e.g. first line HIV combination products, pan-genotypic hepatitis C regimens) provide a suitable incentive for the development of these new formulations. A good example is the new fixed-dosed combination of tenofovir/lamivudine/dolutegravir developed by MPP licensees and first approved by the FDA in August 2017.

The quality of the generic medicines manufactured and distributed by the MPP licensees is assured through the WHO Prequalification programme or approval by a stringent regulatory authority, which to date has generally been the FDA though its “tentative approval” programme. A tentative approval from FDA signifies that although the product cannot be sold in the United States (due to existing patent protection or marketing exclusivity), the product meets all safety, efficacy and manufacturing quality standards for marketing in the United States.

Originator companies can receive royalties from MPP licensees. Sliding-scale royalties, linked to the income level of purchasing countries have also been included in some licences. In other cases, however, there are no royalties (e.g. licences for paediatric formulations).

Delays in access to new medicines in developing countries have been reduced significantly. The time it takes from first FDA approval of a new drug and “scale-up” of generic production/procurement for low-income countries is decreasing – from 11 years for tenofovir (standard HIV treatment) to 4-5 years for more recent products such as dolutegravir.

The licences concluded through the MPP are wide in geographical scope and generally cover all low-income countries, all countries in Sub-Saharan Africa and all lower middle-income countries. A number of licences also include additional upper middle-income countries in other regions of the world, which are not considered by companies as “commercial markets”.

The recent expansion of the MPP model could facilitate affordable access to essential medicines in other therapeutic areas over the coming years, including non-communicable diseases. It could also help facilitate access and stewardship for new antibiotics able to address antimicrobial resistance.

Source: and personal communication.

Governments could also encourage the development of publicly-funded clinical trials to generate evidence about existing or new treatments that could benefit patients and health systems. The Belgian Health Care Knowledge Centre (KCE) identified cases where publicly-funded research would be of clear benefit: 1) comparative effectiveness trials of medicinal products, 2) trials of medicinal products in children and in rare diseases, 3) non-commercial trials to counteract possible publication bias, 4) trials of medical devices, 5) trials of diagnostics and screening, and 6) trials in medical areas not addressed by private companies (Neyt et al., 2015). For example, the Australian Government is providing AUD 69 million through the Medical Research Future Fund (MRFF) for clinical trials over two financial years (2017-18 and 2018-19), to support researchers in their efforts to find cures for rare cancers, rare diseases, and other areas of unmet need; to support research into cancers with low survival rates, such as brain cancers; and improve treatment outcomes and quality of life. This work will sponsor new drugs, devices and treatments to prevent, better diagnose and cure illness and disease available to Australians and global markets.


PDPs allow the targeting of push incentives to the development of products responding to a high unmet health need that are insufficiently addressed by the private sector. Experience shows they can be very powerful vehicles to promote the development of such products. Mossialos et al. (2010), for instance, provide examples of medicines developed through PDPs for malaria, leishmaniasis, and tuberculosis. In a systematic review and assessment of incentives for the development of new antibiotics, Renwick et al. listed a number of advantages of PDPs. They allow sponsors to set the target product profile and guide development. They have the potential to attract both large firms that would not find the project attractive enough without public participation, as well as SMEs that struggle to raise the capital to overcome early-stage development barriers (Renwick, Brogan and Mossialos, 2016). Non-profit PDPs also remove the incentive to pursue high sales volumes, which is useful in the case of antibiotics for example.

Funding trials upfront and making drugs available without patent protection, or with a commitment to pool patents and agree to voluntary licensing, has the potential to enhance access and thus health impact.


On the downside of PDPs, Renwick et al. noted that the sponsor bears the risk of failed projects and that sponsors/governments may not be the best placed to evaluate the viability of a project (Renwick, Brogan and Mossialos, 2016). Firms who perform the R&D have better information about the quality of the R&D project and, for example, its chances of success, so that there is informational asymmetry between the funder and the funding recipient.

The Belgian KCE review identified several key hurdles to the option of conducting publicly-funded RCTs, among which were insufficient funding; difficulties in recruiting patients and persuading and incentivising clinicians to participate; lack of research infrastructure; and access to relevant industry-owned comparator products (Neyt et al., 2015).

Enabling conditions

Clearly defined research priorities are crucial to use push incentives effectively. Priorities that are global in nature would need to be defined and based on burden of disease and high unmet needs. Pooling funds from different countries would be desirable, as is the case in the development of antibiotics.

Public-private partnerships require long-term commitment from governments, which may not align with electoral cycles and related changes in government priorities. Another significant challenge for partnerships is the management of intellectual property rights given the multiple and sometimes divergent interests of consortium partners.

Option 12: Explore alternative pull incentives to encourage R&D for unmet medical needs


One option to spur drug development in areas of unmet need is the use of alternative pull incentives that are publicly-funded, such as market entry or innovation rewards. Alternative pull incentives are particularly appropriate in OECD countries for neglected areas of research, and for diseases where the current framework of incentives, including patent protection and market exclusivity, has not delivered needed innovation, such as in antibiotics. Orphan diseases may be another area where innovation rewards could be beneficial.

Alternative pull incentives have the potential to align pharmaceutical industry incentives more closely with public health needs. They could further improve population access to the treatments developed as a result of such incentives by requiring, for example, that specified population groups be treated, or that firms may not inhibit competition or set prices that pose barriers to access.


The current framework of incentives for innovation, together with manufacturer remuneration based on price and volume, encourages investment in R&D of treatments with large market potential (through high volumes and/or high prices) in populations with high ability to pay. This can lead to the neglect of disease areas where markets are small. Examples of areas with small market volumes include antibiotics, which should only be prescribed where absolutely necessary to prevent further development of resistance, or rare diseases that only affect small populations. Medicines for neglected tropical diseases prevalent in low-income countries are examples of areas with limited revenue potential due to inability to pay.

The idea of providing innovation inducement prizes dates back to the 19th century and prizes have been used in other fields (Williams, 2012). More generally, various types of alternative pull incentives to reward firms for innovation have been proposed in the past. Most of these mechanisms follow a similar economic rationale, though they differ in terms of the suggested timing of disbursements and the conditions attached to them. These characteristics, however, can have a significant impact on “if” and “to what extent” they can harness competition among innovators, as well as on their ultimate effects on encouraging R&D activity and ensuring population access to treatments. Examples of alternative pull incentives that have been put forward so far include:

  • A substantial one-time payment for the treatment of a specified disease shortly following marketing approval, based on development costs incurred and the therapeutic value of the novel medicine. Such payments have also been referred to as “innovation prizes” (Stiglitz and Jayadev, 2010; UNITAID, 2016).

  • A variation of innovation prizes could be a staged approach, in which a base payment could be made to compensate manufacturers for the costs of development and distribution, followed by subsequent periodic payments. This may be especially appropriate where there is high uncertainty around the therapeutic value of the new medicine so that payments can be increased on verification of clinical effectiveness (Hollis and Pogge, 2008).

  • Advance market commitments (AMCs) or advance purchase commitments, under which sponsors make a firm commitment before a new treatment is available, to purchase a specified volume of the new treatment at a specified price, as long as it meets predetermined requirements. Such mechanisms have already been piloted for vaccines (Kremer, Towse and Williams, 2005; Levine, Kremer and Albright, 2005). AMCs do not decouple incentives from price and volume, however. Rather, they set a fixed price and volume upfront in return for a commitment by the manufacturer to supply units beyond the committed volume at a lower price.

A pilot advance market commitment (AMC), with a total value of USD 1.5 billion funded by the governments of Canada, Italy, Norway, Russia, and the United Kingdom as well as the Gates Foundation, was set up in 2007 for a vaccine against pneumococcal disease (Cernuschi et al., 2011). The AMC guaranteed a price per treatment that was deemed affordable, supplementing this price up to a specified number of treatments, and has so far been considered successful in accelerating access to a life-saving product (GAVI Alliance, 2013). Because AMCs have so far been designed to supplement the existing market-based system and provide incremental sources of revenue for industry, they have received substantial political support (Hollis and Pogge, 2008).

Depending on the need for incremental innovation following a breakthrough in an area, alternative pull incentives might be structured in a way to foster competition and encourage improved follow-on products (Farlow et al., 2005; Kremer, Towse and Williams, 2005). One-off prizes that lead to “winner-takes-all” competition may therefore not be suitable in all cases. Rewards staggered in several tiers could favour follow-on innovation, by providing substantial payments to the first manufacturer that meets specifications, but also by reserving funds for subsequent competitors that offer improved treatments in the same class or for certain target populations or conditions (Berndt and Hurvitz, 2005; Levine, Kremer and Albright, 2005). This argument may also favour solutions that do not entirely decouple incentives from price and volume, but aim to make price and volume commitments up-front, such as AMCs. Mechanisms that allow follow-on products to take market share and thus portions of the rewards from earlier entrants could also provide additional incentives for first movers to develop highly effective treatments and bring products to market more quickly (Kremer, Towse and Williams, 2005; Levine, Kremer and Albright, 2005).

If alternative pull incentives are to harness competition among private firms and expect private capital to bear the risk of failure, they have to meet two fundamental conditions (Hollis and Pogge, 2008). First, the conditions for receiving the reward must be specified in a sufficiently clear and detailed manner, so that innovators understand the ultimate goal and deliver the innovations society values. Second, the magnitude of the reward must considerably exceed the amount of investment each competitor expects to make in delivering the innovation. This is because firms face two main sources of risk: they can fail because they are unsuccessful in developing an effective new treatment, or because a competitor is able to develop an effective treatment more quickly. Although staggered rewards discussed above may reserve some funds for follow-on products and reduce the risk arising from being slower than a competitor, rewards must compensate private capital for the risk of failure (ibid.).

Innovation rewards might be coupled with conditions that firms cannot use exclusionary rights to inhibit generic competition once the reward has been disbursed, and extract economic rents (Stiglitz and Jayadev, 2010). At the same time, because a part of the R&D process may need to continue to rely on publicly funded push incentives, the magnitude of the rewards might also be reduced to account for the net public contribution across the entire development process, for example through “clawback” provisions in the reward.


While there is limited empirical evidence thus far, there are some compelling arguments in favour of alternative pull incentives for the development of medicines. For example, a key strength of mechanisms that provide a reward for a pre-specified treatment is that, similar to current price- and volume-based incentives, they encourage competition among manufacturers. They provide an incentive to work quickly and efficiently towards approval of a successful treatment. Also, the risk of failed research is borne by private investors. Such pull incentives can avoid problems of informational asymmetries associated with push incentives by encouraging self-selection of projects that are pursued and abandoned based on information in the possession of the innovating firms (Kremer, Towse and Williams, 2005; Hollis and Pogge, 2008). If the conditions for receiving a reward are sufficiently general, for example by specifying only the disease and the level of efficacy that has to be achieved, such mechanisms can leave substantial leeway for the private sector in selecting the best avenues of research to achieve the desired outcomes.

Another argument in favour of alternative pull incentives is that they can define explicit conditions for the disbursement of reward payments that improve both access to the treatments and efficiency of spending. This could, for instance, include a requirement as a condition of payment that specified populations receive the treatment, or that generic competition is permitted as soon as the reward has been disbursed, to encourage lower prices. However, areas where market volumes are small may not be attractive for generic competition.

Other potential advantages of alternative pull incentives include (Kremer, 1998; Hollis and Pogge, 2008; Williams, 2013; UNITAID, 2016):

  • Accelerating development of follow-on products using the newly developed technology because patent protection would not be necessary to ensure return on investment;

  • Improving alignment of incentives for private R&D with public health priorities and a reduction of socially wasteful spending on promotional activities;

  • Enhancing efficiency in healthcare spending by precluding the extraction of profits beyond what public policy determines to be the “appropriate” reward for innovation.


There is also little empirical evidence of potential disadvantages of the use of alternative pull incentives and only very limited experience in applying these to medical technology or medicines. Prior experience is limited to vaccines for diseases of high burden in developing countries (see, for example, Levine, Kremer and Albright, 2005). Arguments against such mechanisms are largely theoretical or based on hypothetical problems in their implementation.

Designing effective alternative pull incentives may encounter significant practical obstacles in, for example, defining technical specifications, setting eligibility criteria for disbursing rewards, or determining the optimal amounts to be allocated to individual innovations. Such incentives are likely to be less effective in incentivising R&D in areas where there is significant uncertainty surrounding the costs of development and the value of the desired treatment, and which may make it impossible to specify the goals of research upfront (Williams, 2012). These areas could continue to rely on push incentives (see Option 11).

One argument against alternative pull incentives that involve reward payments is the difficulty of specifying rewards (Stiglitz and Jayadev, 2010; Williams, 2012). This applies both to the criteria to be fulfilled for receiving the reward and the amount to be awarded. The challenge of determining an amount sufficient to spur innovation while avoiding overpaying relative to the societal value of the product exists with all types of pull mechanisms, and indeed, is arguably similar to setting the optimal duration of patent protection or market exclusivity. A number of approaches have been suggested for determining the magnitude of innovation rewards, such as using the net present value of comparable compounds already developed (Levine, Kremer and Albright, 2005; Fisher and Syed, 2010) or auctions (Kremer, 1998). However, all have major shortcomings and no straightforward solution is currently available.

If alternative pull mechanisms condition the payment of a reward on development milestones that are unrelated to the diffusion of the medicine to patients, their effectiveness in achieving patient access may be reduced. This may be a particular problem if, for example, public entities acquire drug development projects early in the development process as a result of disbursing a reward and subsequently provide insufficient incentives for trials to achieve marketing approval (Kieff, 2001).

Alternative pull mechanisms that condition payments on entry of a new medicine to market may also provide insufficient incentives for continued investment post marketing authorisation. Post-marketing studies may be required to demonstrate long-term and relative effectiveness, and these might not be undertaken if there are no commercial incentives (Hollis and Pogge, 2008). For example, the trade association Pharmaceutical Research and Manufacturers of America reports that approximately 17% of its members’ R&D expenditure in 2014 was for post-approval research (PhRMA, 2016).

Enabling conditions

Depending on the treatments targeted, alternative pull incentives may require substantial amounts of funding and long-term commitments to be effective. If these treatments address a global need, they may require broad international co-operation to raise the necessary funds and ensure that countries contribute to development of novel drugs.

To provide effective motivation for private investment, alternative pull incentives would ideally be based on credible commitments made long before disbursements, which must be honoured once conditions for disbursement are met. The highest level of certainty would be provided by rewards established in law, as these would provide some protection from political interference and changes in leadership. Drug development cycles are typically longer than electoral cycles and critics have argued that investors and industry executives are unlikely to trust successive governments to honour commitments on entry rewards over the long periods of time required to bring new medicines to market (Farlow et al., 2005). If the institutions charged with assessing the value of novel products are able to undervalue the technology and soften commitments to releasing funds, incentives may be ineffective (Berndt and Hurvitz, 2005; Kremer, Towse and Williams, 2005).

AMCs for vaccines have been criticised for requiring detailed technical specifications, that can be difficult to develop (Farlow et al., 2005). These specifications may also discourage the pursuit of unexpected but beneficial options in the development process. In addition, this implies that AMCs can only be issued once product characteristics are reasonably well known, and thus restrict their utility to incentivising late-stage development (Hollis and Pogge, 2008). The pneumococcal vaccine pilot, in particular, was criticised for allocating a large sum to late-stage development of a vaccine that might well have been commercialised without the AMC (ibid). Other commentators have argued that the vaccine was adopted much more quickly in developing countries than vaccines developed without AMCs (Williams, 2012).

Option 13: Amend orphan drug policies to target more closely areas of unmet need

Objective: Encourage the development of affordable medicines to treat rare diseases


Although a more precise assessment of the benefits and shortfalls of different orphan drug policy frameworks would be needed, a number of policy options are proposed to improve the targeting of incentives, such as limiting financial inducements, restricting eligibility conditions, and developing other push and pull incentives.


A number of policy options have been suggested to repurpose orphan drug policies (especially in the United States) to ensure that they provide appropriate incentives for the development of medicines that would not be developed otherwise, and that patients are able to access them (Drummond and Towse, 2014; Daniel et al., 2016).

The first option is to limit financial incentives to ensure that societies do not pay more than is necessary to encourage the development of drugs for rare diseases. This could be achieved through a reduction in the market exclusivity period (where it exceeds the exclusivity afforded to non-orphan drugs), or through a claw-back on sales beyond a certain threshold. This already exists in some jurisdictions. In Europe, for example, the period of marketing exclusivity can be reduced from ten years to six, if, at the end of the fifth year criteria for orphan designation are no longer met, and thus that the extended market exclusivity is no longer justified. The process must be initiated by an EU Member State, but it is worth noting that this has not been used to date. Japan introduced a clawback system to recover a share of grant monies awarded for the development of orphan indications between orphan drug designation and marketing approval.23 Companies whose orphan-designated drugs are later approved and generate over JPY 100 million in sales are required to pay back 1% of the amount over the threshold to the National Institute for Biomedical Innovation (which disburses grants). This requirement is effective during the first ten years of the drug’s approval or until the amount of the grant funding provided to the company is paid back in full (Kelley, 2016).

Box 3.8. Review of the period of market exclusivity of orphan medicinal products

In Europe, according to Articles 3 and 5 of Regulation (EC) No 141/2000, a medicinal product shall be designated an orphan medicinal product if its sponsor can establish:

(a) that it is intended for the diagnosis, prevention or treatment of a life-threatening or chronically debilitating condition affecting no more than five in 10 thousand persons in the Community when the application is made (so-called “prevalence” criterion),


that is intended for the diagnosis, prevention or treatment of a life-threatening, seriously debilitating or serious and chronic condition in the Community and that without incentives it is unlikely that the marketing of the medicinal product in the Community would generate sufficient return to justify the necessary investment;


(b) that there exists no satisfactory method of diagnosis, prevention or treatment of the condition in question that has been authorised in the Community or, if such method exists, that the medicinal product will be of significant benefit to those affected.

The prevalence criterion is much more frequently used by companies to obtain an orphan designation. Between 2000 and 2015, only one orphan designation was granted on the basis of insufficient return on investment.

Article 8(2) of Regulation (EC) No 141/2000 establishes that the period of market exclusivity may be reduced from ten to six years if at least one of the designation criteria on the basis of which market exclusivity was granted is not met. This can only happen at the request of one member state, which triggers and assessment by the Committee on Orphan Medicinal Products (COMP) or the MEA. COMP will first review the criteria on which orphan designation was granted. If these criteria are still met, nothing will happen. If they are not, COMP will assess whether other criteria for orphan designation are met. If it is the case, COMP will recommend to not reduce the period of exclusivity. If none of the criteria for orphan designation are met, COMP may recommend a reduction of the period of exclusivity.

Source: European Commission, 2008, 2016a; Kelley, 2016.

Another option is to restrict eligibility for orphan designation to situations in which market-based incentives are insufficient. New products used in oncology, for example, may have mechanisms of action that make them potentially suitable therapies for several indications with small population targets (QuintilesIMS Institute, 2017). Companies may already have sufficient incentives to develop as many indications as possible for a given product, without benefitting from advantages granted to orphan medicines.

Other options could be considered, such as pull and push incentives developed in policy options 11 and 12. Not-for-profit development of indications for rare diseases could also be envisaged. Models such as the DNDi’s have shown that repurposing existing drugs for treatments without viable markets can sometimes be done at low cost. Since in many countries orphan medicines are fully paid for by the government or compulsory insurance schemes financed through social contributions, public payers might have an interest in subsidising product development for these indications and making products available at affordable prices for patients with rare diseases.


Limiting financial returns for orphan medicines that achieve high levels of sales, through truncated market exclusivity or partial refunds beyond a revenue cap, would enable public payers to save money. This could improve access, if prices fall as a result of increased competition after the loss of market exclusivity. The option of targeting incentives to those developments that would have not occurred in their absence, if possible (see below), would result in better value from public spending.

Not-for-profit development and/or publicly-funded trials could also potentially support the development of orphan medicines at lower costs than at present, especially when repurposing or developing a new indication for an existing product. This type of redevelopment is usually quicker and cheaper than developing an entirely new product, and the medicines could be made accessible at affordable prices to patients.


Reducing market exclusivity, or using a ‘clawback’ for successful products, may not solve problems related to patient access and unaffordable prices. In addition, there is no straightforward way to define a “cap” beyond which sales would be considered “high enough” to trigger limitations on orphan drug status.

Targeting incentives to limit eligibility to those medicines or indications that would not have been developed without specific inducements is not straightforward. The extent to which specific incentives (such as tax rebates in the United States, extended exclusivity, regulatory fee waivers) and the possibility of charging an “orphan price” play a role in the development of each indication is only known to the companies themselves.

As discussed under Option 11, publicly-funded trials raise a number of difficulties that would need to be overcome, among them the lack of research infrastructure and access to relevant industry-owned products (Neyt et al., 2015).

Enabling conditions

Orphan drug policies have spurred the development of treatments for rare diseases and any potential changes should be undertaken cautiously to ensure that R&D continues. Some basic research aimed at gaining an understanding of the impact of particular government-provided incentives would be needed in order to assess the potential impact of policy changes.

E. Strengthen the information base to better inform policy debates

Option 14: Publish authoritative information on industry activities and the risks, costs and returns from R&D, to better inform policy decisions.


An authoritative dataset assessing industry activity and performance would allow for a more fact-based debate of pharmaceutical policies, and would benefit all stakeholders, including the general public, policy makers and the pharmaceutical industry. It would also help restore trust among stakeholders.

OECD countries accounted for about 70% of global drug sales in 2016 (QuintilesIMS Institute, 2016). About 70% of global business expenditures in pharmaceutical R&D were incurred in OECD countries in 2015 (OECD, 2017a)24 and all but two of the top 50 pharmaceutical companies have their headquarters in OECD countries (Christel, 2017), thus reporting of industry activity across the OECD would capture a large share of global activity.

This reporting could be done in a number of different ways. For example, it could draw on periodic data collection of industry activity and performance. The data would be most effective if they were collected in accordance with an agreed set of indicators and used consistent reporting standards. Data could be submitted to and published by an intermediary, such as the OECD Secretariat. Indicators would have to be agreed internationally because many pharmaceutical firms are multinational. They could span subjects such as key financial metrics and R&D activity.


Public debates on pharmaceutical policy are often marked by a lack of authoritative and commonly accepted information supporting the arguments of the stakeholders involved, and this has arguably compromised trust among them. The absence of commonly accepted information to underpin debate may also undermine the ability of policy makers to make informed and balanced decisions. This is a particular issue in the pharmaceutical market, as demand is largely subsidised by public funds, the industry often benefits from publicly funded R&D, and policy makers use a range of policy instruments in an attempt to strike a balance between providing the necessary incentives for the industry and ensuring patient access to treatments.

While the private sector plays an essential role in health and pharmaceutical R&D, only firms possess rich information about their own activities. Information in the public domain is often unreliable, inaccurate or incomplete. For example, estimates of the average R&D costs of a successfully launched newly developed medicine range widely, from several hundred million to USD 2.6 billion (Morgan et al., 2011; Mestre-Ferrandiz, Sussex and Towse, 2012; DiMasi, Grabowski and Hansen, 2016; Prasad and Mailankody, 2017). The data used to develop these estimates are often not disclosed and the methods have been the subject of controversy and debate (Reuters, 2013; Avorn, 2015).

Publicly listed firms are already required to publish certain financial information to ensure functioning of capital markets and taxation systems, but the information is not structured in a way that facilitates pharmaceutical policy analysis. Furthermore, accounting methods and standards vary between countries, allowing some leeway for individual firms in defining their accounting and reporting policies. This can limit comparability and make data aggregation difficult. Smaller firms that engage in early-stage R&D are often not publicly listed and are financed through private equity and venture capital, with the result that no data on their activities are publicly available. When estimating R&D costs, for example, relying only on reports by publicly listed firms introduces a risk of bias because project failure rates are high at early stages of R&D. Industry costs from unsuccessful R&D might therefore be underestimated when considering only data from firms traded on the stock market.

A number of commercial databases provide access to firm-level financial reporting data, such as the ORBIS database, and the Bloomberg database used in this report (see Chapter 2). Some information on the R&D pipeline and on existing technologies may also be available from commercial databases, such as Adis Insight or GlobalData. The representativeness, accuracy and suitability of existing databases would need to be assessed in detail for the purposes described in this policy option.

Additional data collection and reporting of industry activity and performance could encompass topics such as key financial metrics and R&D activity. Financial metrics could include, for example, total revenue, operating and capital expenditures, profits, assets and liabilities and cash flow, a breakdown of expenditures and cash outflows by purpose, such as R&D, employee and executive compensation, sales and marketing, and/or returns to shareholders through dividends or share buy-backs. Breakdowns of R&D activity could include the number of active and discontinued product development projects or R&D expenditure by disease area, project or development phase. Data collection could be established or executed in co-operation with other OECD directorates and working parties, in particular the Directorate for Financial and Enterprise Affairs, the Directorate for Science, Technology and Innovation and the OECD Working Party of National Experts on Science and Technology Indicators, to ensure consistency with existing reporting standards.


Greater transparency and the availability of a common dataset for assessing industry activity and performance would provide at least three broad advantages for the general public, policy makers and the pharmaceutical industry. First, it would increase the legitimacy and accountability of pharmaceutical policy and allow the general public to scrutinise more objectively whether public funds are spent responsibly to serve public interests. Second, it would inform policy making and reduce the susceptibility of decision-making to be capture by stakeholder groups promoting their own rather than the broader public interest. Third, it would help restore trust between the industry and other stakeholders.


Primary data collection, validation, anonymization, and analysis via a survey of companies could be costly. Using existing data sources, such as the databases of financial reports or R&D pipeline information, may be a cheaper alternative, albeit with some potential limitations in the scope of possible analyses. However, available databases are subscription-based, so access would also involve financial costs.

The release of data on pharmaceutical R&D and industry performance may carry the risk of disclosure of some strategic information. However, the risk could be mitigated through anonymising data submissions by firms and publishing indicators only in aggregate to prevent re-identification (e.g. for all companies or for sub-groups of companies).

In addition, greater transparency of R&D costs may encourage countries to introduce cost-plus criteria into pricing and reimbursement decision frameworks. However, cost-plus pricing could create undesirable incentives for the industry, such acting inefficiently in R&D resource allocation or inflating R&D costs, and is at odds with tiered pricing based on societal willingness-to-pay.

Enabling conditions

Collecting and publishing such information would require significant effort to achieve consensus among countries and stakeholders on the identification of relevant metrics, detailed reporting standards, and collection methods that would ensure that data are accurate and comparable.

Implementation of this option would also require substantial effort in at least two other areas. First, preparatory work would be necessary to define the relevant indicators more specifically and to review in detail the availability of data from which to compute them. Second, a neutral third party would have to be identified to oversee data collection and ensure the protection of data privacy where necessary.

Option 15: Increase price transparency in pharmaceutical markets


Improving the information available to health care systems has the potential to enhance the efficiency of spending, as well as the transparency, legitimacy and accountability of decision making by public payers. The increasing divergence of list prices from net prices actually paid by purchasers, in part a result of the proliferation of financial managed entry agreements, not only raises concerns regarding the accountability and legitimacy of decision-making by public authorities but also has a number of practical drawbacks. While full transparency could compromise tiered pricing and have uncertain effects on the results of price negotiations, payers and other stakeholders in health care systems would benefit from increased transparency.


Price opacity is increasing in pharmaceutical markets both, within and between countries, due to the development of confidential agreements between the industry and private and public payers (see Chapter 1). Price opacity allows for tiered pricing, or price discrimination, between purchasers. This benefits firms because they can price products according to the ability and willingness to pay of each purchaser, thereby increasing revenues. Tiered pricing can also benefit payers and increase patient access, especially in settings where ability to pay is low and firms are willing to offer prices that are below the average.

Yet, price opacity also has a number of practical shortcomings. First, in coverage schemes where patient out-of-pocket payments are proportional to the price of the medicine, patients may not benefit from confidential discounts or rebates negotiated further upstream in the supply chain. Second, confidential prices may compromise comparisons of the costs and benefits of medicines competing for a given indication, undermining the capacity of decision-makers and providers to take comparative cost effectiveness into account when making coverage or treatment decisions. Third, list prices serve as an anchor in price negotiations between payers and manufacturers, and thus may lead to poor negotiation outcomes if anchors are far from the true price of the product (see Box 3.6 on anchoring effects). At the international level, price opacity blurs international price benchmarking, which is used by many OECD countries to regulate the prices of medicines, and makes price comparisons between countries partially irrelevant. Finally, while confidential agreements are often promoted as a way to adapt to countries’ ability to pay, there is no guarantee that low-income countries get lower prices.25 Indeed, anecdotal evidence suggests that prices are not necessarily lower in low-income countries (Iyengar et al., 2016; Vogler, Vitry and Babar, 2016).

In addition, confidential prices prevent the general public from scrutinising coverage and pricing decisions made by public authorities. This poses a problem for both the accountability and legitimacy of public policy, because the public cannot discern whether tax or social health insurance contributions are spent efficiently. Opacity also impedes analyses of net prices and price trends, and leads the public to overestimate pharmaceutical prices. This is not conducive to informed debate between stakeholders and the pharmaceutical industry, and may expose the latter to undue criticism of its pricing strategies.

To balance these concerns, national payers could cooperate to increase price transparency. A study on enhanced cross-country coordination in pharmaceutical pricing in EU countries, suggests making transparent the existence of a discount for a given product, so that the use of its list price as a reference in international benchmarking can be subject to appropriate caveats (Vogler et al., 2015).


Price transparency would enhance the capacity to integrate economic considerations into decision making and increase efficiency of spending. Price transparency improves the utility of economic evaluation in HTA and, crucially, enables the results of such evaluations to be shared with other stakeholders and the general public. Economic considerations could be included in establishing treatment guidelines and can help prescribers and patients make more informed treatment choices based on cost-effectiveness or other measures of value.

Increased price transparency would also increase the accountability and legitimacy of coverage and pricing decisions, allowing the general public to scrutinise more objectively whether public funds are spent efficiently. It would also allow for more accurate analyses by researchers of prices and price trends that inform policy debates. This would help restore trust between the industry and other stakeholders in pharmaceutical policy.


While policymakers and other stakeholders frequently call for price transparency to inform negotiations with manufacturers or to address legitimate concerns about the accountability of decisions made by public authorities, it is not clear how increased price transparency would affect prices across different countries. For example, increased transparency could lead to price convergence and potentially result in higher prices or lower patient access in countries with low ability to pay. Many OECD countries use international benchmarking to regulate pharmaceutical prices and some of them reference a wide range of countries with varying income levels. If prices were transparent, firms might be less willing to agree to reduced prices in low-income countries if these could influence prices in higher-income countries. In addition, the pressure from public opinion in high-income countries to reduce prices to match those obtained elsewhere could become intense if prices were made fully transparent.

In regions such as the European Union with free movement of goods between countries, the combination of price transparency and tiered pricing encourages parallel export26 from lower-income countries (e.g. Greece and Spain) to higher-income countries (e.g. Denmark and Germany). Parallel exports may undermine patient access in exporting countries and create shortages. In addition, parallel trade represents a loss of revenue for companies, which means that they may be less likely to consent to discounted prices if these are in the public domain and arbitrage cannot be effectively prevented.

Enabling conditions

Partial price transparency, such as revealing the existence – but not the magnitude – of confidential discounts can mitigate the risks associated with full transparency. For example, Australia discloses the existence of confidential discounts or rebates in a note linked to the name of the corresponding medicine in the national reimbursement formulary and in associated “therapeutic relativity sheets”.27 Partial transparency does not address, however, the problems price opacity poses for international price benchmarking, the financial burden for patients charged co-insurance based on list price, or the lack of information about value to inform treatment decisions by doctors and patients.

Another option that could be considered to reduce the incentive for parallel trade within the EU would be to allow companies to charge lower prices in countries with lower incomes by agreeing transparent, ex-post rebates. Such an approach, however, would require commitments from countries with higher incomes not to use these prices as benchmarks.

Box 3.9. A game theory approach to assessing the effects of price transparency

A game theory model of sequential price negotiations between a single pharmaceutical firm in a monopoly position and several national buyers suggests that, on average, prices would be lower under price transparency than under price opacity. This is because subsequent buyers learn about the “impatience” of manufacturers to close a deal by observing prices negotiated between the firm and prior buyers, and can better anticipate the minimum price the firm is willing to accept in a negotiation. The patience or impatience of a firm depends on the cost of revenues forgone while a negotiation is ongoing and negotiated prices crucially depend on the patience of firms to close a deal. Transparency decreases the information asymmetry of this parameter between the firm and buyers.

The model also suggests that transparency would increase overall welfare because, as a result of improved buyer information about the firm’s negotiation parameters, deals would be closed quicker, allowing firms to earn higher profits and buyers to treat patients earlier.

However, these results are contingent on a number of assumptions that, at best, may only approximate the structure of actual pharmaceutical markets:

First, the model assumes that the firm is a monopolist. If more than one firm competes in an oligopoly and prices are easily observable to all firms, this can risk tacit collusion between firms. This contrasts with a scenario of price opacity, where firms are uncertain about prices offered by competitors and try to undercut each other.

Second, the model assumes that firms are not forward-looking and do not anticipate the effect of closing a deal on subsequent negotiations. If firms are forward-looking, they have an incentive to signal patience by rejecting low prices in earlier negotiations, and this can lead to higher prices for earlier buyers and lower prices for subsequent buyers. The overall effect depends on whether the signalling effect that leads to higher prices for earlier buyers is stronger or weaker than the learning effect for subsequent buyers.

Third, the model assumes that all buyers commit to transparency. If prices of only some buyers are disclosed, then forward-looking firms have an incentive to take a particularly tough negotiation stance with those buyers that disclose prices. This phenomenon can also explain why buyers currently accept opacity. As long as some buyers believe that they can negotiate better deals under opacity and do not disclose prices, all other buyers face the high cost of unilaterally disclosing their own prices.

Source: Ettinger and Sidartha (2016), “Would Transparency in Negotiations Between Countries And Pharmaceutical Firms Lead To Lower Drug Prices?”, Université Paris Dauphine, unpublished.

Option 16: Improve horizon scanning activities and encourage co-operation at regional level


Improving the capacity to anticipate the opportunities and challenges represented by new medicines, emerging technologies, and disruptive innovations could speed patient access, enhance uptake by professionals, and improve budget provision. A number of OECD countries have been using horizon scanning (HS) to better prepare for market launches and adoption of new technologies. Given the globalised nature of health innovation, the process of identifying upcoming technologies and assessing their likely impact on health systems could benefit from international exchange, and regional co-operation could reduce duplication of effort. Some elements of HS would need to remain country-specific, such as epidemiology or forecasting of economic impacts, but international and multi-stakeholder co-operation has the potential to reduce duplication and be more comprehensive and less costly for individual countries.

Efforts have been launched at regional level, the most recent being the BeNeLuxA Initiative, which includes HS among the set of cooperative activities undertaken by the four countries involved (BeNeLuxA, 2018). The OECD would be well positioned to convene interested stakeholders to share information on the status of technologies and discuss methods of ensuring appropriate access (OECD, 2015b). Whether this occurs in the form of a data-sharing platform or additional data gathering on the part of the OECD Secretariat in collaboration with national and international partners, the information obtained would help member health systems prepare more effectively for the potential impact of innovative therapies on patient care, outcomes, costs, and society.


Drawing on public and private sector horizon scanning initiatives, many countries are already thinking proactively about medical technologies not yet on the market. Most OECD countries have some form of publicly-supported health horizon scanning, either undertaken within the government itself or contracted to non-profit organisations (Slawomirski, Colbert and Paris, 2017). These systems create a repository of information on emerging technologies with high potential impact on health care systems. They include information on patient needs, implementation barriers, and the benefits of new technologies in comparison with current alternatives. The functionality of horizon scanning activities varies widely across OECD countries. For example, the Italian Horizon Scanning Project (IHSP) is developing a forecasting model to predict impact of emerging medicines on the health system, including economic analyses, while the ECRI-AHRQ platform in the United States does not include predictions of the future costs of any health care technology (ECRI Institute, 2015). HS systems recently introduced in Nordic countries such as Sweden or Norway, aim to promote timely diffusion of cost-effective technologies in health systems (WHO Regional Office for Europe, 2017). The Dutch HS system established in 2012 aims to identify the financial risk associated with new inpatient or outpatient drugs, in terms of cost per patient and total budget impact and to determine drugs eligible for price negotiations28 (Lepage-Nefkens et al., 2017).

When budgeting for health, however, countries often adopt “siloed” budgets with spending caps based on historical allocation of resources rather than on prioritising and disinvesting where needed to free up resources for the most (cost-) effective health care interventions and technologies. Costs can be particularly difficult to project before a new therapy comes to market, however failure to do so can also have a detrimental effect on patient access through overburdening budgets unprepared for the new technology. This was illustrated in the health care systems of many OECD countries with the launch of highly effective direct-acting antivirals (DAAs) for hepatitis C in 2013 and 2014. An effective HS system would have alerted decision makers and budget holders not only that a new, very promising medicine was about to be approved, but also that it would be followed very quickly by potential competitors likely to help drive prices down.

Given the globalised nature of drug development, co-operation in HS, at least at the regional level, could potentially avoid duplication of effort and promote information sharing. The role for OECD could be defined in relation to existing national and international initiatives or projects. It could potentially range from brokering knowledge produced by others to the production of new information in partnership with various stakeholders, including regulatory agencies and industry.


Sharing information could reduce the information asymmetry between manufacturers and payers, which is currently pervasive in pharmaceutical markets. While manufacturers hold confidential information on their product development pipelines and the likelihood and timing of market launches, payers are currently reactive to launch strategies adopted by manufacturers. Manufacturers would nonetheless benefit from better preparedness by health care systems and more rapid adoption and diffusion of the most effective technologies.

Improved capacity and collaboration in horizon scanning would also allow health systems to assess whether emerging therapies adequately address unmet health needs in their populations. This could prompt alternative mechanisms to propel the development of needed therapies.

Horizon scanning is a resource-intensive endeavour, and it is arguably inefficient for countries to conduct horizon scanning activities in parallel. Pipelines and pharmaceutical R&D are global in nature, which means that the identification of products likely to have a significant impact on health systems could take place at a global level, even though entry to national markets is often sequential. While publication of horizon scanning reports already enables other countries to benefit, achieving a greater degree of cooperation and participation would allow the information gathering to address the specific needs of contributing countries.

Areas where international collaboration could be quite straightforward include capacity-building, in terms of people, data (epidemiologic, consumption, patent expiries, and upcoming treatments) and modelling. EuroScan has published a methods toolkit outlining a three-step process to identify new technologies, prioritise those which must be assessed, and then assess their potential impact (Simpson and EuroScan International Network, 2014).


Prior experience with horizon scanning has shown a number of limitations (OECD, 2017c). First, the impact of horizon scanning on technology diffusion is not straightforward. A study assessing its impact on the diffusion of six technologies in ten EU countries from 1995 to 2004 revealed that initial recommendations (positive or negative) did not have a significant impact (Packer, Simpson and Stevens, 2006). One reason for this might be that the results of horizon scanning do not always reach their target audience, i.e. governments and/or health providers (Martino et al., 2012).

A key element in assessing the feasibility of an international horizon scanning collaboration is determining the objective and scope. Such an initiative could focus on analysing public health priorities and unmet needs proactively, or could be more “reactive” in gathering information on forthcoming technologies.

Enabling conditions

In order to provide timely information to policy makers, the choice of the time horizon is important and may vary according to the technology. Drug development has a well-defined development and regulatory route, and horizon scanning usually aims to identify drugs in phase 3, or at least phase 2. However, orphan drugs may be eligible for expedited development, making the time horizon more difficult to define.

Collaboration on assessment of dimensions such as health impact, expected utilisation, time to adoption, and process impact could help health systems improve their forecasting activities. The following dimensions could be examined in an international horizon scanning exercise: potential health impact at population level (impact on morbidity, mortality); at patient level (data on effectiveness derived from clinical trials); expected utilisation of the technology (ease of acquisition, manner of administration, concomitant resource utilisation; ease of compliance; degree of invasiveness; physical and mental capacity required for use; anticipated side effects, risks, and adverse events). Time to adoption and budget impact are likely to be more country-specific, as well as effect on the healthcare system at an organisational level (e.g. structural changes, and staff training).

Several questions remain to be answered in assessing the feasibility of an integrated horizon scanning platform. Aligning group participation, objectives, and methods to ensure consistency would be a key challenge, as would be division of labour among relevant topics to ensure adequate coverage. Ideally, such a platform would also include a feedback loop on accuracy of predictions to further refine future scanning. Existing initiatives such as EuroScan, EUnetHTA and regional horizon scanning collaborations could be used as a potential model or input to any broader initiative under consideration.


AIFA (2018), Registri e PT attivi (aggiornamento 01/03/2018), Agenzia Italiana del Farmaco, default/files/Registri_PT_Attivi_20.06.2017.ods (accessed 27 March 2018).

AIFA (2016), Legenda e definizioni, Agenzia Italiana del Farmaco, definizioni_10022016.pdf (accessed 7 July 2017).

Angelis, A. and Kanavos, P. (2016), “Value-Based Assessment of New Medical Technologies: Towards a Robust Methodological Framework for the Application of Multiple Criteria Decision Analysis in the Context of Health Technology Assessment”, PharmacoEconomics, 34,

Arnardottir, A. H. et al. (2011), “Additional safety risk to exceptionally approved drugs in Europe?”, British Journal of Clinical Pharmacology, 72(3), pp. 490–499,

Avorn, J. (2015), “The $2.6 Billion Pill — Methodologic and Policy Considerations”, New England Journal of Medicine, 372(20), pp. 1877–1879,

Azzopardi-Muscat, N., Schroder-Beck, P. and Brand, H. (2016), “The European Union Joint Procurement Agreement for cross-border health threats: what is the potential for this new mechanism of health system collaboration?”, Health Economics, Policy and Law, pp. 1-17,

Banzi, R. et al. (2017), “Conditional approval of medicines by the EMA”, BMJ, p. j2062,

BeNeLuxA (2018), The BeNeLuxA Initiative -Collaboration on pharmaceutical policy - Terms of Reference,

Berndt, E. R., Conti, R. M. and Murphy, S. J. (2017), “The landscape of US generic prescription drug markets, 2004-2016”, National Bureau of Economic Research, NBER Working Paper, No. 23640.

Berndt, E. R. and Hurvitz, J. A. (2005), “Vaccine Advance-Purchase Agreements For Low-Income Countries: Practical Issues”, Health Affairs, 24(3), pp. 653–665,

Bhardwaj, R., Raju, K. D. and Padmavati, M. (2013), “The impact of patent linkage on marketing of generic drugs”, Journal of Intellectual Property Rights, 18(4), pp. 316–322.

BMI Research (2015), Joint Procurement To Increase Price Pressure On Drugmakers, (accessed 15 November 2016).

Bujar, M., McAuslane, N. and Liberti, L. (2016), “The impact of the evolving regulatory environment on the approval of new medicines across six major authorities 2006-2015”, Centre for Innovation in Regulatory Science (CIRS), R&D Briefing, No. 59.

Busse, R. et al. (2011), Diagnosis-Related Groups in Europe - Moving towards transparency, efficiency and quality in hospitals, Open University Press, Maidenhead, UK.

CEPS (2017), Rapport d’Activité 2016, Comité Economique des Produits de santé, Paris.

Cernuschi, T. et al. (2011), “Advance market commitment for pneumococcal vaccines: putting theory into practice”, Bull World Health Organ, 89, pp. 913–918,

Christel, M. (2017), “Pharm Exec’s Top 50 Companies 2017”, Pharmaceutical Executive , 37(6).

Committee on Finance of the US Senate (2015), The pricing of Sovaldi and Its Impact on the U.S . Health Care System, Committee on Finance United States Senate.

Culyer, A. J. (2016), “Cost-effectiveness thresholds in health care: A bookshelf guide to their meaning and use”, Health Economics, Policy and Law, Cambridge University Press, 11, pp. 415–432,

Daniel, M. G. et al. (2016), “The Orphan Drug Act”, American Journal of Clinical Oncology, 39(2), pp. 210–213,

Deloitte (2016), The evolution of oncology payment models: What can we learn from early experiments? Deloitte Center for Health Solutions,

Department of Health, I. (2018), Ireland to open negotiations with Belgium, the Netherlands, Luxembourg and Austria on drug pricing and supply – Minister Harris | Department of Health, Press release 13/02/2018, (accessed 18 May 2018).

DeRoeck, D. (2003), Group Procurement of Vaccines for Central / Eastern Europe and Newly Independent States: Feasibility, Issues and Options. Final Report.

DeRoeck, D. et al. (2006), “Regional group purchasing of vaccines: Review of the Pan American Health Organization EPI revolving fund and the Gulf Co-operation Council group purchasing program”, International Journal of Health Planning and Management, 21(1), pp. 23–43,

Dholakia, U. M. (2016), “A Quick Guide to Value-Based Pricing”, Harvard Business Review, (accessed 7 November 2017).

DiMasi, J. A. J., Grabowski, H. H. G. and Hansen, R. W. R. (2016), “Innovation in the pharmaceutical industry: New estimates of R&D costs”, Journal of Health Economics, 47, pp. 20–33,

Drummond, M. and Towse, A. (2014), “Orphan drugs policies: A suitable case for treatment”, European Journal of Health Economics, 15(4), pp. 335–340,

ECRI Institute (2015), AHRQ Healthcare Horizon Scanning System – Status Update 2015, Rockville, MD, United States.

Eichler, H. G. et al. (2015), “From adaptive licensing to adaptive pathways: delivering a flexible life-span approach to bring new drugs to patients”, Clinical pharmacology and therapeutics, 97(3), pp. 234–246,

Eichler, H. G. et al. (2012), “Adaptive licensing: Taking the next step in the evolution of drug approval”, Clinical Pharmacology and Therapeutics, 91(3), pp. 426–437,

EMA (2017a), Accelerated assessment, European Medicines Agency, curl=pages/regulation/general/general_content_000955.jsp&mid=WC0b01ac05809f843a (accessed 18 August 2017).

EMA (2017b), Conditional marketing authorisation, European Medicines Agency, index.jsp?curl=pages/regulation/general/general_content_000925.jsp (accessed 18 August 2017).

EMA (2017c), Conditional marketing authorisation: Report on ten years of experience at the European Medicines Agency, European Medicines Agency, London.

Espín, J. et al. (2016), “How can voluntary cross-border collaboration in public procurement improve access to health technologies in Europe”, Policy Brief, No. 21, World Health Organization 2016 (acting as the host organization for, and secretariat of, the European Observatory on Health Systems and Policies), Copenhagen.

European Commission (2018), “Proposal for a regulation of the European Parliament and of the Council on health technology assessment and amending Directive 2011/24/EU”, European Commission, Brussels.

European Commission (2017), “Mapping of HTA methodologies in EU and Norway”, European Commission, Brussels.

European Commission (2016a), Inventory of Union and Member State incentives to support research into, and the development and availability of, orphan mecidinal products — state of play 2015, European Commission, Brussels.

European Commission (2016b), Joint procurement of medical countermeasures, European Commission, Brussels,

European Commission (2008), Guidelines on aspects of the application of Article 8(2) of Regulation (EC) No. 141/2000: Review of the period of market exclusivity of orphan medicinal products, European Commission, Brussels.

Ettinger and Sidartha (2016), “Would Transparency in Negotiations Between Countries And Pharmaceutical Firms Lead To Lower Drug Prices?”, Paper Commissioned by the OECD, Université Paris Dauphine, unpublished.

Falit, B. P., Chernew, M. E. and Mantz, C. A. (2014), “Design and implementation of bundled payment systems for cancer care and radiation therapy”, International Journal of Radiation Oncology Biology Physics, Elsevier Inc., 89(5), pp. 950–953,

Farlow, A. et al. (2005), “Concerns regarding the Center for Global Development Report “Making Markets for Vaccines”, Submission to the Intellectual Property Rights, Innovation and Public Health WHO, WHO, Geneva.

FDA (2017), FDA Tackles Drug Competition to Improve Patient Access, US Food and Drug Administration, (accessed 5 October 2017).

FDA (2016), Accelerated Approval Program, US Food and Drug Administration, healthprofessionals/ucm313768.htm (accessed 18 August 2017).

Ferrario, A. et al. (2017), “The Implementation of Managed Entry Agreements in Central and Eastern Europe: Findings and Implications”, PharmacoEconomics,

Fisher, W. W. and Syed, T. (2010), “A prize system as a partial solution to the health crisis in the developing world”, in Pogge, T., Rimmer, M., and Rubenstein, K. (eds.), Cambridge University Press, Cambridge, United Kingdom.

Frank, C. et al. (2014), “Era Of Faster FDA Drug Approval Has Also Seen Increased Black-Box Warnings And Market Withdrawals”, Health Affairs, 33(8), pp. 1453–1459,

Furnham, A. and Boo, H. C. (2011), “A literature review of the anchoring effect”, Journal of Socio-Economics, Elsevier Inc., 40(1), pp. 35–42,

G20 Health Ministers (2017), Berlin Declaration of the G20 Health Ministers: Together Today for a Healthy Tomorrow, Berlin.

Galinsky, A. D. and Mussweiler, T. (2001), “First offers as anchors: The role of perspective-taking and negotiator focus”, Journal of Personality and Social Psychology, 81(4), pp. 657–669,

Garattini, L. and Casadei, G. (2011), “Risk sharing agreements: What lessons from Italy?”, International Journal of Technology Assessment in Health Care, 27(2), pp. 169–172,

Garattini, L., Curto, A. and van de Vooren, K. (2015), “Italian risk-sharing agreements on drugs: are they worthwhile?”, The European Journal of Health Economics, 16(1), pp. 1-3,

Gargon, E. et al. (2014), “Choosing Important Health Outcomes for Comparative Effectiveness Research: A Systematic Review”, PLoS ONE. Edited by R. W. Scherer, 9(6), p. e99111,

Gartlehner, G. et al. (2006), Criteria for Distinguishing Effectiveness From Efficacy Trials in Systematic Reviews. Technical Review 12 (Prepared by the RTI-International-University of North Carolina Evidence-based Practice Center under Contract No. 290-02-0016.), Agency for Healthcare Research and Quality, Rockville, MD, United States.

GAVI Alliance (2013), Advance Market Commitments ‘promising solutions’ to global health challenges, (accessed 15 April 2017).

Gerkens, S. et al. (2017), How to improve the Belgian process for Managed Entry Agreements? An analysis of the Belgian and international experience, KCE Reports 288, Health Services Research (HSR) Brussels: Belgian Health Care Knowledge Centre (KCE). .

Grabowski, H. and Mullins, D. (1997), “Pharmacy benefit management, cost-effectiveness analysis and drug formulary decisions”, Soc. Sci. Med, 45(4), pp. 535–544.

Hatswell, A. J., Baio, G. and Freemantle, N. (2016), “Regulatory approval of pharmaceuticals without a randomised controlled study: Analysis of EMA and FDA approvals”, BMJ Open, 6,

Hirsch, B. R. et al. (2013), “Characteristics of Oncology Clinical Trials”, JAMA Internal Medicine, 173(11), p. 972,

Hollis, A. and Pogge, T. (2008), The health impact fund: Making new medicines accessible also for all, Incentives for Global Health, New Haven, CT, United States.

Huff-Rousselle, M. (2012), “The logical underpinnings and benefits of pooled pharmaceutical procurement: A pragmatic role for our public institutions?”, Social Science and Medicine, 75(9), pp. 1572–1580,

Huff-Rousselle, M. and Burnett, F. (1996), “Cost containment through pharmaceutical procurement: A Caribbean case study.”, The International Journal of Health Planning and Management, 11(2), pp. 135–157,<135::AID-HPM422>3.0.CO;2-1.

Huskamp, H. A. et al. (2005), “The impact of a three-tier formulary on demand response for prescription drugs”, Journal of Economics and Management Strategy, 14(3), pp. 729–753,

Huskamp, H. A., Epstein, A. M. and Blumenthal, D. (2003), “The Impact Of A National Prescription Drug Formulary On Prices, Market Share, And Spending: Lessons For Medicare?”, Health Affairs, 22(3), pp. 149–158.

IMI (2018), Project Factsheets, (accessed 30 April 2018).

Infarmed (2018), La Valletta Declaration Committee met in Lisbon, (accessed 18 May 2018).

Investopedia (2017), “Value-Based Pricing”, Investopedia, (accessed 7 November 2017).

Iyengar, S. et al. (2016), “Prices, Costs, and Affordability of New Medicines for Hepatitis C in 30 Countries: An Economic Analysis”, PLoS Medicine, 13(5), pp. 1–22,

James, J. E. (2017), “Reviving Cochrane’s contribution to evidence-based medicine: bridging the gap between evidence of efficacy and evidence of effectiveness and cost-effectiveness”, European Journal of Clinical Investigation, 47(9), pp. 617–621,

Kanavos, P. et al. (2017), “Managing Risk and Uncertainty in Health Technology Introduction: The Role of Managed Entry Agreements”, Global Policy, 8(March), pp. 84-92,

Kanavos, P. et al. (2012), Tender systems for outpatient pharmaceuticals in the European Union: Evidence from the Netherlands , Germany and Belgium, LSE Health, Report for the European Commission - DG Enterprise.

Kelley, A. (2016), “Orphan drug legislation in four countries/regions”, Paper commissioned by the OECD, unpublished.

Kieff, F. S. (2001), “Property Rights and Property Rules for Commercializing Inventions”, Minnesota Law Review, 85, p. 697.

Kremer, M. R. (1998), “Patent buyouts: A mechanism for encouraging innovation”, The Quarterly Journal of Economics, 113(4), pp. 1137–1167,

Kremer, M., Towse, A. and Williams, H. (2005), Briefing Note on Advance Purchase Commitments, London.

Lepage-Nefkens, I. et al. (2017), Horizon scanning for pharmaceuticals: proposal for the BeNeLuxA collaboration, KCE Reports No. 283, Health Services Research (HSR) , Belgian Health Care Knowledge Centre (KCE), Brussels.

Levine, R., Kremer, M. and Albright, A. (2005), Making Markets for Vaccines: Ideas to action, Report of the Center for Global Development Advance Market Commitment Working Group, Washington, DC.

Levinson, D. R. (2014), Update: Medicare payments for end stage renal disease drugs, Department of Health and Human Services, Office of Inspector General, Report No. OEI-03-12-00550.Lising, A. et al. (2017), “Payer’s use of independant reports in decision making - Will there be an ICER effect?”, Value & Outcomes Spotlight, March/Apri, pp. 1–10.

Lourenço, A. (2016), Bundled Payment in Portugal - Country Background Note: Portugal, (accessed on 12 November 2017).

Lu, C. Y. et al. (2015), “Patient access schemes in Asia-pacific markets:Current experience and future potential”, Journal of Pharmaceutical Policy and Practice, 8(1), pp. 1–12,

Ludwig, S. and Dintsios, C.-M. (2016), “Arbitration Board Setting Reimbursement Amounts for Pharmaceutical Innovations in Germany When Price Negations between Payers and Manufacturers Fail: An Empirical Analysis of 5 Years’ Experience”, Value in Health, 19(8), pp. 1016–1025,

Luigetti, R. et al. (2016), “Regulatory collaboration - Collaboration, not competition: developping new reliance models”, WHO Drug Information, 30(4).

Makady, A. (2017), “Implementation of CRS in HTA practice: experiences from the Netherlands. Presentation to the NVTAG Symposium”, Zorginstituut Nederland, Utrecht, Netherlands.

Makady, A. et al. (2017), “Policies for use of Real-World Data in Health Technology Assessment: A comparative study of 6 HTA agencies”, Value in Health, forthcoming.

Makady, A., Goettsch, W. and Willemsen, A. (2015), Review of Policies and Perspectives on Real-World Data – Draft Report 30.01.2015.

Management Sciences for Health (2012), “MDS-3: Managing Access to Medicines and Health Technologies”, Management Sciences for Health, Arlington, VA, United States.

Martino, O. I. et al. (2012), “Innovation and the Burden of Disease: Retrospective Observational Study of New and Emerging Health Technologies Reported by the EuroScan Network from 2000 to 2009”, Value in Health, 15(2), pp. 376–380,

Maxmen, A. (2016), “Big Pharma’s cost cutting challenger”, Nature, 536, pp. 388–390,

McClellan, M. B. and Thoumi, A. I. (2015), “Oncology payment reform to achieve real health care reform.”, Journal of oncology practice, American Society of Clinical Oncology Alexandria, VA, 11(3), pp. 223–30,

Mcpherson, E., Hedden, L. and Regier, D. A. (n.d.), “Impact of oncologist payment method on health care outcomes, costs, quality: a rapid review”,

Medicines for Europe (2017), Market review – biosimilar medicines markets, Brussels.

Mestre-Ferrandiz, J., Sussex, J. and Towse, A. (2012), The R&D Cost of a New Medicine, Office of Health Economics, London, 86 pages.

Morgan, S. et al. (2011), “The cost of drug development: A systematic review”, Health Policy, Elsevier Ireland Ltd, 100(1), pp. 4–17,

Mori, K. (2017), “Global Co-operation by Regulatory Authorities and Japan’s Activities”, Ministry of Health, Labour and Welfare, Kyoto.

Mossialos, E. et al. (2010), Policies and incentives for promoting innovation in antibiotic research, European Observatory on Health Systems and Policies, 195 pages.

Naci, H. et al. (2017), “Timing and characteristics of cumulative evidence available on novel therapeutic agents receiving FDA accelerated approval”, The Milbank Quarterly, Vol. 95/2, pp. 261-290,

Neumann, P. J. (2004), “Evidence-based and value-based formulary guidelines”, Health Affairs, 23(1), pp. 124–134,

Newcomer, L. N. (2012), “Changing physician incentives for cancer care to reward better patient outcomes instead of use of more costly drugs”, Health Affairs, 31(4), pp. 780-785,

Newcomer L.N. et al. (2014), Changing Physician Incentives for Affordable, Quality Cancer Care: Results of an Episode Payment Model, Journal of Oncology Practice, 10(5), pp. 322-326.

Neyt, M. et al. (2015), Publicly funded practice-oriented clinical trials, KCE Reports 246, Health Services Research (HSR), Belgian Health Care Knowledge Centre (KCE), Brussels.

NICE (2017), Budget impact test, (accessed 30 April 2018).

Norwegian Ministry of Health and Care Services (2017), Principles for priority setting in health care: Summary of a white paper on priority setting in the Norwegian health care sector, Oslo.

OECD (2017a), Health at a Glance 2017: OECD Indicators, OECD Publishing, Paris,

OECD (2017b), New Health Technologies: Managing Access, Value and Sustainability, OECD Publishing, Paris,

OECD (2017c), New Health Technologies, OECD Publishing, Paris,

OECD (2017d), Tackling Wasteful Spending on Health, OECD Publishing, Paris,

OECD (2016), Better Ways to Pay for Health Care, OECD Health Policy Studies, OECD Publishing, Paris,

OECD (2015a), Health Data Governance, OECD Health Policy Studies, OECD Publishing, Paris,

OECD (2015b), “Scientific Advice for Policy Making: The Role and Responsibility of Expert Bodies and Individual Scientists”, OECD Science, Technology and Industry Policy Papers, No. 21, OECD Publishing, Paris,

OECD (2013a), Implementing the OECD Principles for Integrity in Public Procurement, OECD Public Governance Reviews, OECD Publishing, Paris,

OECD (2013b), Strengthening Health Information Infrastructure for Health Care Quality Governance, OECD Health Policy Studies, OECD Publishing, Paris,

OECD (2008), Pharmaceutical Pricing Policies in a Global Market, OECD Health Policy Studies, OECD Publishing, Paris,

OFT (2007), The Pharmaceutical Price Regulation Scheme An OFT market study - An OFT market study, Office of Fair Trading, London.

OsMed (2017), L’uso dei farmaci in Italia. Rapporto Nazionale 2016, Agenzia Italiana del Farmaco (AIFA), Rome.

Packer, C., Simpson, S. and Stevens, A. (2006), “International diffusion of new health technologies: A ten-country analysis of six health technologies”, International Journal of Technology Assessment in Health Care, 22(4), pp. 419–428,

PAHO (2015), PAHO Strategic Fund : A Regional Platform to Improve Access to Public Health Supplies, The Pan American Health Organization, Washington, DC.

Panteli, D. et al. (2016), Pharmaceutical regulation in 15 European countries, WHO Regional Office for Europe, Copenhagen.

Paris, V. and Belloni, A. (2013), “Value in Pharmaceutical Pricing”, OECD Health Working Papers, No. 63, OECD Publishing, Paris,

Pauwels, K. et al. (2017), “Managed Entry Agreements for Oncology Drugs: Lessons from the European Experience to Inform the Future”, Frontiers in Pharmacology, 8,

Petrovsky, B. Y. K. (2015), The Joint EU Member States Procurement of Rare Disease Medicines Initiative, WINTER/6._Joint_EU_Member_States_Procurement_of_Rare_Disease_Medicines_Initiative.pdf.

PhRMA (2016), “2016 PhRMA Annual Membership Survey”, Pharmaceutical Research and Manufacturers of America, Washington, DC.

Pinto Ribeiro, S., S. Menghinello and K. De Backer (2010), “The OECD ORBIS Database: Responding to the Need for Firm-Level Micro-Data in the OECD”, OECD Statistics Working Papers, No. 2010/01, OECD Publishing, Paris,

Polite, B. et al. (2016), “A Pathway Through the Bundle Jungle”, Journal of Oncology Practice, 12(6), pp. 504–509,

Prasad, V. and Mailankody, S. (2017), “Research and development spending to bring a single cancer drug to market and revenues after approval”, JAMA Internal Medicine, 10065,

QuintilesIMS Institute (2017), Global Oncology Trends 2017 - Advances, Complexity and Cost, QuintilesIMS Institute, Parsippany

QuintilesIMS Institute (2016), Outlook for Global Medicines through 2021- Balancing costs and value, QuintilesIMS Institute, Parsippany.

Rémuzat, C. et al. (2017), “Key drivers for market penetration of biosimilars in Europe”, Journal of Market Access & Health Policy, 5(1), p. 1272308,

Renwick, M. J., Brogan, D. M. and Mossialos, E. (2016), “A systematic review and critical assessment of incentive strategies for discovery and development of novel antibiotics”, The Journal of Antibiotics, Nature Publishing Group, 69(2), pp. 73–88,

Reuters (2013), GlaxoSmithKline boss says new drugs can be cheaper,

Revicki, D. A. and Frank, L. (1999), “Pharmacoeconomic evaluation in the real world. Effectiveness versus efficacy studies.”, PharmacoEconomics, 15(5), pp. 423–34.

Roberts, S. A., Allen, J. D. and Sigal, E. V (2011), “Despite criticism of the FDA review process, new cancer drugs reach patients sooner in the United states than in Europe”, Health Affairs, 30(7), pp. 1375–1381,

Roox, K. et al. (2008), Patent-related Barriers to Market Entry for Generic Medicines in the European Union, European Generic Medicines Association (EGA), Brussels.

Rotar, A. et al. (2018), “Rationalizing the introduction and use of pharmaceutical products: the role of managed entry agreements in Central and Eastern European countries (submitted)”, Health policy, Elsevier Ireland Ltd, pp. 1–7,

Samuel Phd, N. and Verma, S. (2016), “Cross-comparison of cancer drug approvals at three international regulatory agencies”, Current Oncology, 23(5),

Seiden, M. V. (2016), Immuno-Oncology 2016 and Beyond: The Opportunities, Challenges, and Risks, The American Journal of Managed care, evidence-based-oncology/2016/february-2016/immuno-oncology-2016-and-beyond-the-opportunities-challenges-and-risks?p=1 (accessed 25 June 2018).

Simpson, S. and EuroScan International Network (2014), A toolkit for the identification and assessment of new and emerging health technologies, University of Birmingham, Birmingham.

Slawomirski, L., Colbert, A. and Paris, V. (2017), “The past and potential future impact of new health technology”, in New Health Technologies: Managing Access, Value and Sustainability, OECD Publishing, Paris,

Social and Scientific Systems (2016), Examples of Health Care Payment Models Being Used in the Public and Private Sectors, Silver Spring.

Spinks, T. et al. (2017), “Development and Feasibility of Bundled Payments for the Multidisciplinary Treatment of Head and Neck Cancer: A Pilot Program”,

Stiglitz, J. and Jayadev, A. (2010), “Medicine for tomorrow: Some alternative proposals to promote socially beneficial research and development in pharmaceuticals”, Journal of Generic Medicines, Nature Publishing Group, 7(3), pp. 217–226,

Stolk, E. A. et al. (2004), “Reconciliation of economic concerns and health policy: Illustration of an equity adjustment procedure using proportional shortfall”, PharmacoEconomics, 22(17), pp. 1097–1107,

Stop TB Partnership (2017), What is the GDF?, (accessed 17 May 2017).

Tafuri, G. et al. (2014), “How do the EMA and FDA decide which anticancer drugs make it to the market? A comparative qualitative study on decision makers’ views”, Annals of Oncology, 25, pp. 265–269,

The Global Fund (2015), “Pooled Procurement Mechanism Operational Policy Note”, The Global Fund to Fight AIDS, Tuberculosis and Malaria, Geneva.

The Kaiser Family Foundation and Health research and Educational trust (2016), Employer Health Benefits - 2016 Annual survey.

TLV (2017), The development of pharmaceutical expenditure in Sweden, The Dental and Pharmaceutical Benefits Agency, Stockholm.

Tversky, A. and Kahneman, D. (1974), “Judgment under uncertainty: Heuristics and biases”, Science, 185(4157), pp. 1124–1131.

UNITAID (2016), An economic perspective on delinking the cost of R&D from the price of medicines, Discussion Paper, UNITAID, p 59.

US International Trade Administration (2016), 2016 Top markets report - Pharmaceuticals - A market assessment tool for US exporters.

Visegrad Group (2017), Visegrad Bulletin 5, (accessed 18 May 2018).

Vogler, S. et al. (2015), Study on enhanced cross-country coordination in the area of pharmaceutical product pricing - Final Report,

Vogler, S., Vitry, A. and Babar, Z. U. D. (2016), “Cancer drugs in 16 European countries, Australia, and New Zealand: A cross-country price comparison study”, The Lancet Oncology, Elsevier Ltd, 17(1), pp. 39–47,

de Vrueh, R. L. A. and Crommelin, D. J. A. (2017), “Reflections on the Future of Pharmaceutical Public-Private Partnerships: From Input to Impact”, Pharmaceutical Research, pp. 1985–1999,

Wafula, F., Agweyu, A. and Macintyre, K. (2013), “Regional and temporal trends in malaria commodity costs: An analysis of Global Fund data for 79 countries.”, Malaria Journal, 12, p. 466,

Wenzl, M. and Paris, V. (2018), Pharmaceutical Reimbursement and Pricing in Germany, OECD, Paris.

WHO Regional Office for Europe (2017), Report from technical working meeting on access to new medicines -joint horizon scanning and strategic negotiations opportunities, Copenhagen.

Williams, H. L. (2013), “Intellectual Property Rights and Innovation: Evidence from the Human Genome”, Journal of Political Economy, 121(1), pp. 1–27,

Williams, H. (2012), “Innovation Inducement Prizes: Connecting Research to Policy”, Journal of Policy Analysis and Management, 31(3), pp. 752–776,

Williamson, P. R. et al. (2017), “The COMET Handbook: Version 1.0”, Trials, 18(S3), p. 280,

WIPO, WHO and WTO (2012), Promoting Access to Medical Technologies and Innovation Intersections between public health, intellectual property and trade, Geneva.

World Health Organization (2016), Challenges and opportunities in improving access to medicines through efficient public procurement in the WHO European Region, Copenhagen.

World Health Organization (2007), Multi-country Regional Pooled Procurement of Medicines, Meeting Report 15-16 January 2007, World Health Organisation, Geneva. Geneva.

Zwaap, J. et al. (2015), Cost-effectiveness in practice, Amsterdam.


← 1. See Option 6 for a definition of “formulary management”.

← 2. SRAs are defined as regulatory authorities in countries that are members or observers of the International Conference on Harmonisation (ICH), or regulatory authorities associated with an ICH country through a legally binding mutual recognition agreement.

← 3. The European Medicines Agency (EMA),the United States Food and Drug Administration (FDA), the Japanese Pharmaceuticals and Medical Devices Agency (PMDA), Health Canada, Swissmedic and the Australian Therapeutic Goods Administration (TGA).

← 4. A surrogate endpoint is a proxy measure, such as a laboratory marker or physical sign or other measure that is expected to predict clinical benefit, but is not itself a measure of clinical benefit. A valid surrogate endpoint must be a correlate of, and fully capture the net effect of treatment on the true clinical endpoint.

← 5. In this context, the EMA uses the term licensing to refer to the granting of marketing authorisation.

← 6. See

← 7. See and

← 8. “H-prescription medicines” are medicines used in the outpatient sector but prescribed by specialists and financed by Regional Health Agencies (RHA) rather than National Insurance. In the past decade, a number of therapeutic classes have been transferred from National Insurance to RHA funding, such as TNF-inhibitors (2006), multiple sclerosis (2008), some oncology indications (2014), hepatitis C, growth hormone, colony stimulating factors, coagulation factors and treatments for anaemia (2016), pulmonary arterial hypertension and the majority of oncology indications (2017).

← 9. In 2016, LIS launched a call for tender for pharmaceutical treatments used in rheumatoid arthritis, ankylosing spondylitis / axial spondylitis, juvenile idiopathic arthritis, systemic lupus erythematosus, Crohn’s disease, ulcerative colitis, psoriasis and psoriatic arthritis. In December 2016, LIS launched a call for tender for pharmaceutical treatments used in the treatment of hepatitis C.

← 10. The term “4th hurdle” refers to the requirement imposed by many OECD Member countries that pharmaceutical manufacturers demonstrate that their products represent good value for money as well as being of good quality, effective and safe in order to be included in a national formulary and eligible for public subsidy.

← 11. In the United States, prices for oncology drugs are most often negotiated between contracting parties. By contrast, the prices of oncology medicines administered by physicians and covered by Medicare Part B are regulated. The reimbursement basis y are based on the average sales price (ASP) - the reimbursement basis for Medicare Part B drugs - on the prices negotiated in the private sector.

← 12. See

← 13. Off-label use is estimated to account for 20-30% of annual spending on oncology medicines in the United States (Polite et al., 2016).

← 14. Based on volumes, in the market of reimbursed medicines in Germany and Greece and the community pharmacy market in the United States, see Chapter 1.

← 15. Approved Drug Products with Therapeutic Equivalence Evaluations. See However the Orange Book does not list all small molecule patents, nor does it include patents on biological medicines.

← 16. Where internal or therapeutic reference pricing is used to set a single reimbursement price for all versions of a product or for all products within a therapeutic class (“reference prices” or fixed reimbursement amount), there is little incentive for suppliers to offer discounts to the payer.

← 17. Authorities often set price caps, leaving the firm free to price products below the cap for individual purchasers (e.g. wholesalers or insurers). For the sake of simplification, this price cap is often referred to as ‘the regulated price’. Many countries only regulate the prices of patented medicines dispensed to outpatients and publicly covered. Two OECD countries, Canada and Mexico, regulate the (maximum) prices of all patented drugs, covered or not, to prevent abuse of monopoly power and protect consumers.

← 18. Cost-effectiveness analysis (CEA), which is one form of economic evaluation, entails computing the incremental benefit of a new medicine relative to standard treatment and its incremental cost per unit of incremental benefit. The ICER is then benchmarked against a pre-defined ICER threshold, beyond which the treatment is usually not funded. Defining such a threshold is in principle necessary to support decision-making (Culyer, 2016).

← 19. Strictly speaking, HTA bodies most often use cost-utility analysis, weighting health gains with utility derived from different health states to define benefits in terms of quality-adjusted life years (QALY) gained. However, stakeholders often refer to this analysis in a generic sense as cost-effectiveness analysis.

← 20. Proportional shortfalls refer to the proportion of anticipated life expectancy and quality of life lost by a patient group as compared with the average anticipated life expectancy and quality of life for the population of the same age.

← 21. Albania, Bosnia-Herzegovina, Bulgaria, Croatia, Czech Republic, Estonia, Hungary, Kosovo, Latvia, Lithuania, Poland, Romania, Russia, Serbia, Slovakia and Slovenia.

← 22. The Drugs for Neglected Diseases initiative (DNDi) has produced several drugs in the past decade for a fraction of what pharmaceutical companies are said to spend. Factoring in the cost of failed candidates, the DNDi estimates that it can develop combination therapies for between USD 10 million and USD 45 million, and make a completely new drug from scratch for USD 110 million to USD 170 million.

← 23. These grants cover 50% of direct costs incurred up to three years.

← 24. The OECD database does not include information on BERD in all OECD and non-OECD countries in 2015. For non-OECD countries, data is only available for four countries, including China and Russia. Therefore, this figure is likely to be over-estimated.

← 25. Economists could argue that what counts for price discrimination is willingness to pay (WTP) and not ability to pay, but ability to pay obviously acts as a constraint on WTP.

← 26. “Parallel trade” refers to cross-country imports/exports operated by intermediaries –usually wholesalers- without the consent of the manufacturer. It is likely to occur when the price difference is high enough to make this trade profitable.

← 27. See and

← 28. If budget impact is expected to exceed EUR 10 million and annual cost per patient is likely to exceed EUR 50 000 or if budget impact is expected to exceed EUR 40 million, the government will seek a price agreement with the company.

End of the section – Back to iLibrary publication page