# Chapter 1. Medicines in health systems and society

This chapter begins by highlighting the benefits medicines bring to patients and health systems. It then describes recent trends in pharmaceutical expenditure and financing, and the different approaches to coverage and pricing used in OECD countries. Finally, it discusses key challenges currently faced by policy-makers – the growing launch prices of new medicines; assessing the value for money of spending in certain therapeutic areas; anticipating the arrival of high cost effective medicines for highly prevalent diseases; sharp price increases in off-patent products; and the potential misalignment of current incentives for development of treatments for rare diseases.

The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.

## Introduction

This chapter reviews the role of medicines in health systems and society, and sets pharmaceutical spending levels and trends in a broader context. It describes the role of governments and other payers in financing these expenditures, as well as pharmaceutical policies related to coverage and pricing of medicines. Lastly, it highlights some of the current challenges in pricing and affordability of medicines. The key findings may be summarised as follows:

• Innovations in medicines have not only improved survival rates and the quality of life for many patients, but have also changed the nature of diseases like HIV and cancer. Pharmaceutical spending often represents good value for money in health systems. Many medicines prevent disease complications and the use of costly health services. Non-adherence to pharmaceutical treatment is estimated to cost EUR 125 billion in European countries and USD 105 billion in the United States.

• In 2016, retail pharmaceutical expenditure accounted for 16.5% of current health expenditure, on average, in OECD countries. This represented 1.4% of GDP on average, exceeding 2% of GDP in only four countries (Greece, Japan, Hungary, and the United States). Total pharmaceutical spending is estimated at between 9% and 30% higher, taking into account drugs dispensed in hospitals or administered in physician settings.

• Since the onset of the global financial and economic crisis, per capita retail pharmaceutical expenditure has been declining on average in OECD countries by 0.5% per year in real terms, contrasting with 2.3% growth seen during the period 2003-2009. This is largely due to patent expiries of “blockbusters” and cost-containment policies implemented by governments. Conversely, over this period a majority of countries saw continued real expenditure growth in outpatient, inpatient and long-term care, albeit at slower rate than previously.

• Governments or largely publicly-funded health insurance schemes finance nearly 55% of retail pharmaceutical spending in OECD countries and an even higher proportion of expenditure in hospitals. Most OECD countries regulate the price of pharmaceuticals, at least for some market segments, either directly or through defining the conditions for coverage by public payers or social insurance schemes. They use a mix of several instruments, the most common being international benchmarking (basing the price a country pays for a medicine on what other countries are paying) and the use of health technology assessment (HTA) (which may include consideration of economic aspects, such as the cost-effectiveness of the new treatment relative to existing therapies).

• Finally, this chapter highlights a number of challenges countries face regarding the affordability of medicines. First, policy makers, patients and clinicians are increasingly concerned about drug pricing trends; these raise concerns about the future sustainability of health spending and the value of some high-cost new treatments; payers may be ill-prepared for the arrival of new treatments for diseases or conditions of high prevalence; high prices can compromise access; and incentives for the development of orphan medicines have had varying effects.

## Pharmaceutical treatments bring significant benefits to patients and society

Medicines play a significant role in the health care system and in society. Approximately half of all adults in OECD countries take prescribed medicines regularly (see for example Chaplin, 2015; National Center for Health Statistics, 2017). Vaccines and antimicrobials that prevent or treat communicable diseases have had a large and measurable effect on the life expectancy of the population as a whole. Many medicines improve the quality of life of patients with chronic diseases – alleviating or preventing pain, disfigurement, functional decline, disability, and premature death. They can have a substantial effect on productivity and prevent absenteeism due to ill-health, and in many cases, can reduce downstream healthcare costs by preventing avoidable complications or delaying disease progression.

Antimicrobials and vaccines have been pivotal in extending life expectancy and reducing the burden of communicable and vaccine preventable diseases. In 2003, Ehreth estimated that globally, vaccination prevented almost 6 million deaths annually, and higher vaccination rates could save another 3 million children’s lives (Ehreth, 2003). Antibiotics have not only saved patients’ lives, but have also played a role in achieving major advances in medicine and surgery. They have successfully prevented or treated infections in patients undergoing chemotherapy; in those suffering from chronic diseases such as diabetes; in patients who are immunosuppressed (e.g. being treated for autoimmune disorders such as rheumatoid arthritis); and in those undergoing complex procedures such as organ transplant, joint replacement, or cardiac surgery (Ventola, 2015).

Pharmaceutical innovations have changed the management of HIV and more recently, have transformed the treatment of hepatitis C. Life-expectancy post-HIV infection has improved dramatically since the early 1990 due to anti-retrovirals. The life expectancy of treated HIV-positive individuals is now close to that of the general population in the United States and Canada (Lacey et al., 2014). Combined with increased screening and prevention efforts, anti-retroviral treatments have led to a decline in the number of deaths from HIV/AIDS in Australia, Canada, France, Spain and the United States, from a peak in the mid-1990s, by 84% to 90% through 2013-14 (OECD, 2018).

Box 1.1. What is an innovative medicine?

“Innovative” and “innovation” are widely used terms but are rarely defined explicitly. For the purposes of this report a medicine may be described as innovative if it:

• meets a previously unmet or inadequately met, substantive (i.e. non-trivial) health need

• offers enhanced effectiveness (e.g. greater efficacy, reduced toxicity or both) or other incremental benefit (e.g. a substantive improvement in patient convenience) relative to existing therapeutic alternatives.

Conversely, a product that is new or novel, but does not offer additional benefit over existing therapies would not per se be considered innovative (Morgan, Lopert and Greyson, 2008; Bruen et al., 2016).

As recently as five years ago, the available treatment options for hepatitis C were associated with debilitating side effects, and achieved sustained viral responses (SVR) in only half of the patients over a course of treatment lasting up to 48 weeks (Shepherd et al., 2004). Today, a number of treatment options are available offering cure rates of more than 90%, with minimal side effects, in as little as 8 weeks (Asselah, Marcellin and Schinazi, 2018). A study estimated that the implementation of a one-time birth-cohort screening and the use of the new treatments could avert 78 800 cases of liver cancer, 9 900 liver transplants and 126 500 liver-related deaths in the United States by 2050 (Kabiri et al., 2014).

Prescription medicines are a key component of chronic disease management guidelines to reduce the mortality and morbidity burden of many non-communicable diseases. Outcomes for patients with diabetes and cardiovascular disease have improved substantially over the last few decades (OECD, 2015a, 2017a). Across the industrialized world, age-adjusted mortality rates from ischemic heart disease and stroke had declined to about one third of their 1960s baseline by the year 2000, catalysed by advances in both prevention and treatment, including significant declines in cigarette smoking, better hypertension control, widespread use of statins to reduce cholesterol levels, and the development and use of thrombolytics and stents in acute coronary syndromes to limit or prevent infarction (Mensah et al., 2017).

In the United States the age-adjusted annual heart disease mortality rate fell by more than 50% from 1950 to 1996, and by a further 22% from 1990 to 2013 (Mensah et al., 2017). Similar declines were recorded in nearly all regions of the world, particularly in North America and Western Europe, as well as in Japan, Australia, and New Zealand. Coronary heart disease mortality rates in England and Wales decreased by more than 50% between 1981 and 2000. Medical and surgical treatments contributed approximately 21% of total life-years gained during the period, of which 32% were attributable to secondary prevention in patients post myocardial infarction or revascularisation; 13% to heart failure treatments; and 9% to anti-hypertensive therapies (Unal, Critchley and Capewell, 2004; Unal et al., 2005). Cutler, McClellan and Newhouse (1998) estimated that nearly 30% of the decline in 30-day mortality due to heart attacks between 1975 and 1995 could be attributed to pharmaceuticals. A large US study of myocardial infarction in patients aged 65 and over found that mortality had declined by 3% per year between 1995 to 2004, and attributed the observed improvement primarily to increased use of cardiovascular medications (e.g. statins, beta blockers, angiotensin converting enzyme inhibitors and angiotensin receptor blockers, antiplatelet drugs) post discharge (Setoguchi et al., 2008). Finally, the United Kingdom Prospective Diabetes Study (UKPDS) established that retinopathy, nephropathy, and possibly neuropathy can be reduced in people with type 2 diabetes by lowering blood glucose levels through intensive pharmacological therapy. The impact of better glucose control has resulted in life expectancy improving substantially in people with type 1 and type 2 diabetes over recent decades (Miller et al., 2012; Home et al., 2014).

The momentum of biomedical research in rare diseases has resulted in new treatments that have dramatically improved outcomes for patients for whom treatment options had been limited. Ivacaftor, a drug used to treat cystic fibrosis in people with certain mutations present in 4–5% of cases has shown broad benefits of treatment across most outcomes assessed, including lung function and patient reported outcome measures, such as physical and social functioning, health perceptions and vitality (Whiting et al., 2014; Quittner et al., 2015). A decade ago, patients with pulmonary arterial hypertension (PAH) had limited treatment options and many required a lung transplant as their condition worsened. The last decade has seen large gains for these patients with the development of endothelin receptor antagonists. Macitentan, approved in 2013, has been shown to slow disease progression and reduce morbidity from PAH (Pulido et al., 2013). Fifteen years ago, the treatment of chronic myeloid leukaemia (CML) was dramatically transformed by a new class of medicines known as tyrosine kinase inhibitors (TKIs). Imatinib was the first of several TKIs, and was approved by the FDA in 2001. In patients with CML, the estimated 5-year survival rate has improved from a historical level of less than 20%, to >90% in the TKI era (Woessner, Lim and Deininger, 2011; Kantarjian et al., 2012).

Cancer mortality rates have been declining and five-year survival rates for most cancers have improved over recent decades, reflecting improvements not only in prevention and early detection, but also in treatment. Sun et al. (2010) estimated that overall survival increased by 3.9 years for all cancer combined in the United States between 1988 and 2000, with variations across cancer sites. Survival gains ranged from 0.5 years for pancreatic cancer to 3.5 years for non-Hodgkin’s lymphoma and 3.6 years for breast cancer.1 Improvements in treatment explained about 81% of these survival gains on average. More recently Dubois and Kyle (2016), using data from 11 countries over the period 1990-2011, showed that every new cancer medicine in a country was associated with a decline in cancer mortality of 8% for men and 9% in women. New cancer medicines such as novel immunotherapies offer better survival rates and provide patients with options that in many cases may be more tolerable, thus improving their quality of life. Immune checkpoint inhibitors (nivolumab, pembrolizumab, and atezolizumab) have achieved better survival rates than chemotherapy for patients with advanced non-small-cell lung cancer (Kim, Kim and Kim, 2017). Additionally, many emerging therapeutic options can be taken orally, which can help reduce patients’ time in clinics or hospitals.

Prescription medicines not only improve health outcomes for individuals but can also reduce costs in other parts of the health system, by reducing utilisation of health care services, especially hospitals. Khan and Socha-Dietrich (2018) reported that non-adherence may cost European governments as much as 125 billion euros annually in excess health care services and in the United States, USD 105 billion per year of avoidable hospitalisations alone (New England Health Care Institute, 2009; IMS Institute for HealthCare Informatics, 2013; Iuga and McGuire, 2014). Patients with chronic diseases who are non-adherent to their medicines use more secondary care services, such as outpatient care, emergency department visits and hospitalisations, than adherent patients. A large observational study of patients with diabetes, hypertension, high cholesterol and congestive heart failure found that, for all four conditions hospitalisation rates were significantly higher for patients with low medication adherence. Higher levels of adherence were found to be associated with significantly fewer annual inpatient hospital days (Roebuck et al., 2011). Combining the increases in pharmaceutical spending with the decreases in medical spending, average cost-benefit ratios from adherence for the four conditions examined were 1:13 for hypertension, l:9 for congestive heart failure, 1:9 for diabetes, and 1:3 for hyperlipidaemia. Thus just one extra USD spent on medicines for adherent patients with congestive heart failure, high blood pressure, diabetes and high cholesterol can generate between 3 to 13 USD in savings on emergency department visits and inpatient hospitalisations (Roebuck et al., 2011 quoted in Khan and Socha-Dietrich, 2018).

## Over the past decade, retail pharmaceutical spending as a share of GDP has been relatively stable in most OECD countries

Retail pharmaceutical spending2 accounted for 1.4% of GDP across OECD countries in 2016 and has on average remained stable over the past decade. The share of spending in GDP decreased significantly in Mexico, Portugal, Hungary, the Slovak Republic, and Poland, and increased significantly in Japan and Latvia (Figure 1.1). Retail pharmaceutical expenditure, however, does not provide a full picture of spending for pharmaceuticals. Total expenditure on pharmaceuticals is higher when expenditures for pharmaceuticals used in inpatient care are taken into account (see Figure 1.2). These expenditures account for less than 10% of total pharmaceutical expenditures in some countries (e.g. Korea or Germany) but for 30% or more in countries like Israel, Portugal and Denmark.

Retail pharmaceutical spending was 16.5% of current health spending in OECD countries in 2016, a decline from 19.2% in 2006. In real terms and on average in OECD countries, retail pharmaceutical spending growth has been declining almost every year from a high 8% growth in 2001 to negative growth rates after 2009, before a rebound to growth in 2014 (see Figure 1.3). Over this period, real expenditures in other parts of the health system, such as outpatient and inpatient care, continued to grow.

In individual countries, spending trends vary. Since 2008, real spending in retail pharmaceuticals has declined in one third of OECD countries, while it has continued to grow, sometimes moderately, in other countries (Figure 1.4). These trends are explained by both cost-containment measures adopted in the aftermath of the economic crisis and by patent expiry of top-selling products (Belloni, Morgan and Paris, 2016).

### “Specialty medicines” account for one-third of pharmaceutical spending

“Specialty medicines” account for an increasing share of global pharmaceutical sales3 (see Box 1.2). According to QuintilesIMS, the share of specialty medicines increased from 20 to 35% of pharmaceutical sales in the United States between 2007 and 2016; from 15-25% to 35-45% in the five largest European markets; and from 12-16% to 18-35% in South Korea, Japan, Canada and Australia. The proportion of specialty medicines is expected to further increase (QuintilesIMS Institute, 2016).

The growing share of expenditure attributable to “specialty medicines” arguably reflects the combined effects of therapeutic progress and market dynamics (e.g. loss of market exclusivity and generic competition). For example, Figure 1.5 shows changes in market shares by drug class from 1995 to 2015 in France. In the 1990s, anti-hypertensives and other drugs for cardiovascular diseases accounted for about 20% of pharmaceutical spending. Today, despite a continuous increase in consumption, they represent only a small proportion of spending (about 5%). In contrast, expenditures in oncology, autoimmune disorders and hepatitis C were much smaller in the early nineties with fewer, and less expensive treatments, and have grown steadily in the last ten years.

Although the increasing share of “specialty medicines” in pharmaceutical spending has raised concerns about the sustainability and efficiency of pharmaceutical spending, it should not be seen as a problem per se. As long as new products bring benefits to patients and represent value for health systems, paying for them may be a good investment.

Box 1.2. Specialty medicines – definition

“Specialty medicine” is a term mainly used in the United States. However, there is no standard definition and the scope of drugs included in this group varies across stakeholders (ASPE, 2016). That said, the term usually refers to injectable and non-injectable drugs that are typically used to treat chronic, complex conditions and may have one or more of the following characteristics: a requirement for frequent dose adjustment or intensive clinical monitoring; intensive patient training and compliance assistance; limited distribution; and specialised handling or administration. Specialty medicines are mainly used in cancer, rheumatoid arthritis, haemophilia, HIV/AIDS, psoriasis, inflammatory bowel disease and multiple sclerosis.

Several definitions also include price as a key criterion. For Quintiles IMS, for whom medicines with list prices in excess of USD 6 000 per patient per year are likely to be classified as specialty medicines, the share of pharmaceutical sales4 (IMS off-invoice data) accounted for by specialty medicines in the United States in 2016 was about 40%, albeit representing a volume of only 1-3% of prescriptions. The proportion was higher in the non-retail market (58%) than in the retail and mail-order market (32.5%) (QuintilesIMS, 2017).

Using the threshold of USD 600 per month (actual price) defined by CMS for inclusion of medicines in the specialty-tier of Medicare formularies, the share of retail expenditure represented by so-called “high-cost drugs” was 22% in 2015 (ASPE, 2016).

### Competition has helped contain pharmaceutical costs and improve access

In pharmaceutical markets, products can compete on quality or prices. Competition in off-patent markets usually reduces prices while both the existence and the effect of competition between patented drugs varies across settings and therapeutic areas.

#### Competition in off-patent markets drives prices down

Generic entry at the end of patent term offers opportunities to reduce costs without affecting the quality of care. In the United States, for instance, where the generic market is very dynamic, generic drugs generate significant savings, estimated at USD 1.68 trillion in the last decade (Uhl, 2017). Yet, in a number of OECD countries, there remains untapped potential for greater generic competition (see Figure 1.6), due in part to regulation and stakeholder reluctance (see OECD, 2017c).

Biologic medicines represent a growing proportion of pharmaceutical sales (prescription medicines and OTC) and this share is predicted to increase from 24% in 2015 to 29% in 2022 (EvaluatePharma, 2016). Several biologics have just begun to go off-patent in the past decade and the current size of the biologics market losing patent exclusivity between 2015 and 2020 is significant (IMS Institute for Healthcare Informatics, 2016b). The introduction of biosimilars is at various stages in OECD countries, and it has already and should continue to yield savings for health systems (Belloni, Morgan and Paris, 2016). However, this savings potential is not as high as with small molecule generics, due to longer and costly development and production processes for biologics and biosimilars. Price discounts of biosimilars are thus far in the range of 15-40% versus up to 85% for generics (Paris and Belloni, 2014; FDA, 2015; Harris, 2016; Simon Kucher, 2016; OECD, 2017d).

The uptake of biosimilars has been slow due a number of challenges, such as the complexity of the manufacturing process, which relies on reverse engineering, uncertainty around the timing of market entry (for example due to extensions of market exclusivity granted to originators as incentives for the development of paediatric indications), and both clinical and regulatory constraints on interchangeability and substitution in many countries (Moorkens et al., 2016). Payers, regulators, physicians, and patients have all expressed concern regarding biosimilar substitutability, particularly in light of the issues inherent in the biologics production process itself, and the severity of many conditions treated by biologics (ACS CAN, 2013). Such concerns are beginning to wane, as evidence of early experiences with substitution (and even switching) of biosimilar products has raised no cause for clinical concern (Jørgensen et al., 2017). Biosimilar penetration varies widely across OECD countries, for example from 4% for infliximab in France to 98% for filgrastim in the United Kingdom in 2015 (Simon Kucher, 2016). In 2015, biosimilars had 100% of the epoetin market share in Finland, Hungary, Poland, the Slovak Republic and the Czech Republic, but only 2% in Belgium and 6% in the United Kingdom. For TNF inhibitors, biosimilars had 90% and 82% of the market share in Denmark and Norway respectively, but only 2% in Switzerland and 5% in Belgium and Ireland (OECD, 2017a).

The opportunity for cost savings from increased generic and biosimilar uptake remains substantial. A wide array of policies can increase uptake, such as encouraging early entry, competitive pricing, encouraging physician prescribing, increasing the role of the pharmacists, and incentivising and educating patients (see Chapter 3). According to QuintilesIMS (2016), expected savings from loss of market exclusivity in the United States alone could amount to as much as USD 143.5 billion in the period 2017-2021, substantially more than the USD 91.1 billion achieved between 2012 and 2016.

#### The impact of competition in on-patent markets varies across settings and therapeutic areas

The effects of competition in on-patent markets on prices and volumes vary across therapeutic classes and across countries. Medicines with similar mechanisms of action and indications are not usually interchangeable but often constitute reasonable treatment alternatives for a given condition. Thus in competitive markets, the existence of multiple products in a homogeneous therapeutic class might be expected not only to drive prices down but also to increase use through expansion of the eligible population and increased promotional efforts. Evidence of the effects of such oligopolistic competition in OECD countries, however, remains unclear. One study of the peptic ulcer and gastroesophageal reflux disease market (mainly H2-antagonists and proton pump inhibitors) in the United States between 1991 and 2001 suggested that “me-too” drugs competed on price, but also increased use and spending (Arcidiacono et al., 2013). One Canadian study found that manufacturers of brand-name drugs launched between 1994 and 2003 did not compete on price with other products in the same class until there were 4 or more competitors (Lexchin, 2006). Similarly, a study of 458 new molecules launched in Germany between 1993 to 2008 found that the initial two entrants in a new class expanded the market and competed on quality but not on price, while price competition set in with the third competitor. The authors concluded that this pattern was related to prescriber behaviour (Mueller and Frenzel, 2015).5 In the United States, PBMs (specialists in managing pharmaceutical spending) and health insurers are able to negotiate rebates in some therapeutic classes in exchange for formulary listing, which results in a form of competition. The impact of this competition, however, remains unclear, as list prices of some on-patent medicines are increasing at a rapid rate and are not directly related to proportional increases in rebates (see below and Visante, 2017).

Recently, competition has reduced the prices of direct antiviral agents (DAAs) for the treatment of hepatitis C. The first entrant, Gilead’s sofosbuvir, entered the US market in 2013 at a list price of USD 84 000, with lower prices in other markets as a result of a tiered-pricing strategy defined by Gilead, linking the price of the medicine to the income level and prevalence of the disease in different countries. Between August 2013 and August 2017, five new single-component DAAs and six new fixed-dosed combinations obtained regulatory approval (Unitaid and WHO, 2017). Competition helped lower prices and the last competitor entered the US market in 2017 with a list price of USD 26 400 per treatment course (before discounts), and a monthly price 30% to 60% lower than those of its competitors (Sagonowsky, 2017). The most recent negotiations in OECD countries with regulated pricing also achieved lower prices thus enabling expanded population coverage (see Box 1.6 later in this chapter).

## Governments or compulsory health insurance schemes finance nearly 55% of retail pharmaceutical spending in OECD countries on average

In OECD countries, some coverage for prescription medicines is usually included in health coverage schemes. Most OECD countries define positive lists of reimbursed medicines at the central level and assess new medicines before inclusion in these lists. Canada and the United States, where public and private insurers define their own formularies, are exceptions. The United Kingdom and Germany do not use positive lists and medicines that are not excluded from reimbursement by law are paid for as soon as they are authorised for use in the country, although assessment bodies in parts of the United Kingdom may subsequently recommend not to continue funding.

As a result, governments or compulsory schemes finance nearly 55% of retail pharmaceutical spending in OECD countries, albeit with wide variations (see Figure 1.7), and most cover almost all costs of medicines dispensed to inpatients. Private voluntary health insurance schemes finance about one third of retail pharmaceutical spending in Canada, where they cover medicines used in outpatient care for a large share of the population. VHI also finances nearly one quarter of pharmaceutical spending in Slovenia, where it covers co-payment gaps left by basic health insurance.

Households financed 40% of retail pharmaceutical on average in OECD countries in 2016, and more than half the spending in five countries: Australia, Denmark, Iceland, Latvia, and Poland.6 Out-of-pocket payments for pharmaceuticals include both cost-sharing requirements for medicines that are not financed by third-party payers as well as for the costs of medicines (prescribed or not), not included in the range of benefits covered. The respective shares of these two components of household payments vary widely across countries, depending on the market share of non-covered medicines and on levels of co-payments for covered medicines.

The coverage of prescribed medicines is on average better than the coverage of total retail pharmaceutical spending (which includes OTC medicines and medical non-durables). The Australian government, for example, finances 76% of the expenditure for prescribed medicines, while social health insurance in Poland covers almost 70%. Hungary, Latvia and Norway are the countries with the highest share of household financing of prescribed medicines, above 45% (see Annex Table 1.A.1).

## Most OECD countries regulate pharmaceutical prices directly or indirectly through coverage determinations

Most OECD countries also regulate the prices of medicines directly or indirectly (through coverage determinations), at least in some market segments (OECD, 2008; Panteli et al., 2016). They generally use a mix of instruments, among which international benchmarking and health technology assessment (HTA) are commonly used.

### Almost all OECD countries use international benchmarking, at least for some market segments

With the exception of Australia, Sweden, the United Kingdom, and the United States, all OECD countries use international benchmarking as one tool to regulate prices, at least in some market segments. For example, Canada references prices paid in seven OECD countries to regulate the price of patented medicines; France refers to prices paid in four European countries to regulate the prices of novel drugs reimbursed in outpatient care. International benchmarking can be used as supportive information in price negotiations (e.g. Germany, Spain) or as the main criterion (e.g. Canada, France). It can be used for new products only or for price revisions over time. The basket of countries considered varies across countries, from 3 for New Zealand to 28 for some EU countries. The method to compute the benchmark price also varies (see Table 1.1 and Vogler et al., 2015; Rémuzat et al., 2017).

Table 1.1. Use of international benchmarking in OECD countries

Note: 1 – Australia may check the published prices in other countries fora small number of medicines, but countries benchmarked are not predefined. 2 – Countries referring to other EU countries. 2. Luxembourg refers to the country of origin of the product. 4. Mexico refers to prices in the six countries where the medicine has the highest sales.

Source: Vogler et al. (2015) and authors’ compilation for other OECD countries.

International benchmarking is perceived as a policy that is relatively easy to implement, especially for countries with low capacity for health technology assessment. However, it also has a number of drawbacks, as it does not necessary reflect country-specific value of new products; it is increasingly blurred by confidential agreements; and it is a barrier to tiered pricing7 (Vogler et al., 2015).

### OECD countries increasingly use health technology assessment to inform decisions on coverage and prices

Many countries use health technology assessment to inform coverage or pricing decisions for medicines. Some of them mainly rely on clinical evaluation while others consider economic aspects to make coverage decisions. Some countries assess all new patented products (e.g. Germany8) or all products applying for coverage (e.g. Australia, France), while others only assess medicines with high costs and/or uncertain effectiveness (e.g. England) (Auraaen et al., 2016). Differences in methods and criteria notwithstanding, two simplified archetypes of evaluation for the purpose of reimbursement or pricing have been identified (Paris and Belloni, 2013; Panteli et al., 2016):

• Economic evaluation at the price set by the manufacturer to determine whether the medicine will be reimbursed at this price; or

• Clinical evaluation, where the assessment of the incremental benefit of the medicine in question is the basis on which to negotiate a reimbursement price.

The first approach applies, for example, in Sweden, Norway, the Netherlands, Australia, Canada and Korea. The second is used in France, Italy, and Germany.

The idea behind these processes is that improved health benefits over existing competitors deserve a price premium or, said differently, that public payers are ready to pay additional costs for additional benefits. In principle, the use of the cost-effectiveness criteria requires the definition of threshold for the incremental cost-effectiveness ratio (ICER) beyond which new products would not be covered (Culyer, 2016). However, payers or evaluation bodies have been reluctant to establish ICER thresholds (Auraaen et al., 2016), notably because defining such a threshold is not straightforward.9 Where such thresholds have been determined, they have often been altered or not applied consistently, for instance for severe or terminal illnesses and rare diseases (Polton, 2015; Schuller, Hollak and Biegstraaten, 2015). In addition, even in countries using economic evaluation to inform coverage decisions, cost-effectiveness is often not the only criterion taken into account in decision-making (Auraaen et al., 2016).

### Pressure to accept higher price per unit of health gain has been growing

Over time, public payers have been pushed to accept higher prices for a given level of health benefit (e.g. for a QALY). In England for instance, cost-effectiveness thresholds initially recommended by the National Institute for Health and Care Excellence (NICE) were increased for treatments for patients with short life expectancy when there is sufficient evidence that the treatment offers an extension to life (as, for example, for terminal cancer patients). Treatments for rare diseases were also accepted with incremental cost-effectiveness ratios higher than pre-defined thresholds (Timmins, Rawlins and Appleby, 2016). The government also established the NHS Cancer Drug Fund (CDF) in 2010, with an initial annual budget of approximately GBP 50 million to give cancer patients access to drugs not routinely covered by the National Health Service (NHS) (NHS England, 2016). This scheme funds cancer drugs not recommended by NICE on grounds of insufficient cost-effectiveness as well as drugs not yet evaluated by NICE to accelerate access (Palnoch, 2016). A study estimated that the existence of the CDF resulted in considerable opportunity costs within NHS England, which can be measured in terms of forgone population health. Making reference to the NICE cost-effectiveness threshold of GBP 20 000 per QALY, the study estimated a net loss of 8 808 QALYs per annum (Leigh and Granby, 2016). Annual expenditure of the CDF had increased to approximately GBP 400 million by 2015, which led to the delisting of some drugs covered by the fund, the establishment of new criteria for funding and a mandate for NICE to appraise all new cancer drugs as well as the introduction of managed entry agreements with manufacturers (NHS England, 2016; Palnoch, 2016).

Similarly, the Australian government maintains the Life Saving Drugs Program (LSDP) through which patients can obtain funding from an annual budget for expensive drugs for very rare and life-threatening conditions that do not meet the cost-effectiveness criterion for reimbursement under the public Pharmaceutical Benefits Scheme (PBS) (Paris and Belloni, 2014). In September 2016, eight indications and corresponding medicines were listed in the LSDP (Australian Government Department of Health, 2016). Although overall expenditure for LSDP remains relatively small, it grew at an average annual rate of approximately 13% between 2010 and 2014 (Harvey and de Boer, 2015).

Research on stakeholders’ role in the decision-making process and anecdotal evidence from approval of high-cost drugs suggest that pressure from the general public, industry and special interest groups representing patients or physicians exert influence on decisions (Vuorenkoski, Toiviainen and Hemminki, 2003; Ozierański, McKee and King, 2012; Aggarwal, Ginsburg and Fojo, 2014; Campillo-Artero, Garcia-Armesto and Bernal-Delgado, 2016). While stakeholders’ participation in assessment and decision-making processes is being implemented in several countries, studies show that patients sometimes over-estimate the potential of new treatments and that mass-media may not contribute to a well-informed public debate (Weeks et al., 2012; Robertson et al., 2013; Mack et al., 2015).

Some evidence indicates that, when asked in surveys, the general public may give priority to treatment for rare diseases or to treatments for which no alternatives are available, but results from different studies are contradictory (Mentzakis, Stefanowska and Hurley, 2011; Aggarwal, Ginsburg and Fojo, 2014; Drummond and Towse, 2014). To encourage the development of treatments for small patient populations that might otherwise be neglected, legislation in many countries grants special status to orphan10 drugs, including an acceptance of lower levels of evidence, and higher prices for benefits comparable to those delivered by treatments for more prevalent conditions.

Even when orphan drugs do not meet country-specific criteria for reimbursement, it is politically difficult to deny funding. In the Netherlands, for instance, pharmaceuticals to treat Pompe and Fabry disease were initially covered for a limited period in 2009, with conditions for further evidence development (CED). Treatment with these medicines cost EUR 200 000 to 700 000 per patient and year. The Dutch Healthcare Insurance Board subsequently suggested discontinuation of coverage on grounds of poor cost-effectiveness, but reimbursement was maintained following publication of the proposal in the media (van den Brink, 2014). Several countries exempt orphan drugs from regular assessments (OECD, 2017c). In Korea for instance, products are exempted from economic evaluation in the following situations: 1) the condition is severe and life-threatening, such as certain cancers and rare diseases, and no alternative intervention exists; 2) it is difficult to generate evidence because of a paucity of patients; 3) the drug is listed in at least three out of seven countries (France, Germany, Italy, Japan, Switzerland, the United Kingdom, and the United States).

All these exceptions may well be justified by societal preferences, but they create some frustration among decision-makers, price negotiators and budget holders who, as a result, do not always have well-established rules to set a limit on prices claimed by companies, nor to determine which benefits should be displaced in other parts of the health system to fund these new treatments under budget constraints. Finally, in some circumstances, patients get access to medicines though judicial procedures invoking their right to health (see Box 1.3).

In some countries, public payers have been required to grant access to medicines not covered by statutory benefit package following court decisions in favour of patients claiming access on the basis of their legal rights to health. Although rights to health are enshrined in constitutional legislation in many Latin American countries, Colombia and Brazil are often cited as examples where such rights have led to poor allocation of public funds (Vargas-Peláez et al., 2014). In Colombia, for example, the 1991 Constitution allowed for enforcement of the right to health in judicial claims by individual patients for reimbursement of specific treatments, including pharmaceuticals. In a 1995 decision that set a precedent for further litigation and was subsequently adopted as a general rule by the National Social Security Health Board, the Constitutional Court ruled that medications not covered by the statutory benefits basket must be provided using public funds to gravely ill or terminal patients even if the condition is not curable (Lamprea, 2014). The number of claims increased dramatically in the early 2000s. Favourable court decisions and increasing prices following price deregulation in 2006 led to an increase in expenditure of the national FOSYGA (Fondo de Solidaridad y Garantía) fund (which is required to honour court decisions for reimbursement of non-covered treatments to health insurers) from USD 160 million to more than USD 1 billion between 2006 and 2010, 87% of which was spent on reimbursement of excluded pharmaceuticals (Lamprea, 2014; Gaviria, 2016). Price regulation was reintroduced for some drugs in 2013 and subsequently expanded, in an attempt by a new government to regain control over costs.

### US pharmaceutical policy relies on competition to achieve efficiency

In the United States, the population is covered for health care costs through several schemes, which all contribute to the financing of pharmaceutical expenditure. In 2016, more than two thirds (67.5%11) of US residents were covered by private health insurance; 16.7% by Medicare, the government programme for the elderly and disabled (see Box 1.4); 19.4% by Medicaid, the joint Federal-State programme for low-income individuals and families; and 4.6% by programs dedicated to military forces. About 8.8% of the population was uninsured (Barnett and Berchick, 2017). In 2016, 43.4% of expenditure for prescription drugs sold in retail outlets was financed private health insurance; 29% by Medicare; 10.2% by Medicaid; 3.1% by other health insurance programs; and 13.7% by households themselves12 (CMS, 2018).

The US pharmaceutical market is complex (National Academies of Sciences, Engineering and Medicine, 2018), more than those of many OECD countries. A number of private companies serving as managed care organisations, pharmacy benefit managers (PBMs), drug distributors, and pharmacies, play a role in benefit design (which drug is covered and under which conditions) and in the purchasing and distribution of pharmaceuticals. PBMs serve as third-party administrators of pharmacy benefits for private and public plans.13 They administer benefits such as formulary design and electronic prescribing as a commercial service. After a pharmacy and therapeutics (P&T) committee review regarding clinical utility and cost-effectiveness, PBMs can negotiate with manufacturers, on behalf of insurance plans, based on the status of utilisation management in their formulary. The three largest benefit managers increased their share of the total commercial prescription volume from 42% in 2005 to 66-68% in 2015 (Aitken et al., 2016; Sood et al., 2017).

PBMs aggregate purchasing power to negotiate lower prices for prescription drugs from manufacturers and discounts from pharmacies, with many also providing direct mail-order pharmacy services to further lower costs. With these tools, payers can obtain significant discounts or rebates14 from manufacturers in therapeutic areas where competition exists. Recent estimates suggest that rebates can amount up to 60% when brand competition is high and that average off-invoice discounts and rebates amount to nearly 30% in the United States (see Box 1.4).

The system, however, shows a number a limitations (National Academies of Sciences, Engineering and Medicine, 2018). Most PBMs do not disclose the prices they pay to retail pharmacies or their rebates from a drug manufacturer, although contracts may guarantee a health plan a specific percentage of that rebate. While purchasers of health insurance can benefit from these confidential rebates through premium reductions, patients purchasing the medicines do not directly benefit from them and may face high-cost-sharing requirements. This has led to a controversial practice whereby pharmaceutical companies help patients get access to costly medicines through co-pay assistance or coupons, which offset some of their costs. This practice can distort incentives used in benefit design. A 2013 analysis noted that 62% of coupons were for drugs with lower-cost therapeutic alternatives available (Ross and Kesselheim, 2013) and thus encouraged the use of high-cost medicines. The association of Pharmacy Benefit Managers (PBMs) estimates that couponing practices could raise 10-year prescription drug costs by USD 32 billion (Visante, 2011). More recently, there has been controversy with some PBM providers regarding fiduciary, transparency, and disclosure practices (National Academies of Sciences Engineering and Medicine, 2018). Particular features of Medicare drug coverage also limit the opportunities to exploit competition (see Box 1.4). For instance, the obligation to cover all drugs in six protected classes in Medicare Part D plans prevents insurers and PBMs to negotiate rebates in exchange for formulary listing.

In order to have their covered outpatient drugs paid for by Medicaid, manufacturers must agree to provide rebates to State and Federal Medicaid programs of an amount such that Medicaid does not pay more for brand name drugs than the lowest or “best price” that the manufacturer negotiates with others within the United States during a rebate period.15 Providing these rebates to Medicaid is also a requirement for receiving payments under Medicare Part B.

Box 1.4. Coverage of prescription drugs by Medicare in the United States

## Medicare covers pharmaceuticals through several programs, including Medicare Part B and Part D

Medicare Part B covers specialty drugs that are administered by a provider in a physician office or hospital outpatient setting. Medicines covered by this programme are overseen either by CMS nationally or by 1 of 12 local contractors. CMS cannot generally use cost information in benefit basket design, instead relying solely on the criteria “reasonable and necessary”. There have been attempts to control drug costs, notably by reimbursing certain drugs at the level of the “least costly alternative” (LCA) from 1995 to 2010, but CMS discontinued the use of the LCA after a court ruled that it was contrary to the law that requires reimbursement at 106 percent of a drug’s average sales price (ASP) (US Court of Appeals- District of Columbia Circuit, 2009). This removed the ability to incentivise use of the most cost-effective alternative, and further efforts have sought legislative authority for policies such as LCA (Levinson, 2012).

Medicare Part D was introduced in 2006 to provide coverage of outpatient medicines for Medicare beneficiaries. It is a voluntary programme, administered by private insurance providers. In 2016, about 43.2 million people were enrolled in Medicare Part D (13.8% of the population). Providers use the same tools to manage their Medicare Part D and private plans, but for Part D they carry some additional obligations. The authorising legislation for Medicare Part D specified a standard benefit design for plans to follow, but most offer alternative benefit designs including multi-tier formularies. All Part D plans, regardless of their benefit design, must cover “all or substantially all” products in six “protected classes”: immunosuppressants, anti-depressants, anti-psychotics, anti-convulsants, anti-retrovirals, and anti-neoplastic drugs (Bach, 2009; Frank, 2012; CMS, 2016). Part D plans may still place products from protected classes on high cost-sharing tiers and employ utilisation management tools, but the constraints imposed by law on the selection of drugs limits opportunities to promote competition.

Insurers can also offer Medicare Advantage Prescription Drug (MA-PD) plans that offer both medical and pharmacy drug coverage as part of managed care, but the majority of Medicare beneficiaries receive medical and drug coverage from different sources, resulting in potential access issues in therapeutic areas with both self- and physician-administered approved products.

## Policy makers face a number of challenges in pharmaceutical markets

Against the background presented in the previous sections, a number of challenges have emerged in pharmaceutical markets which are dominating policy debate. These challenges are linked to trends taking place in pharmaceutical markets:

• List prices of new medicines have been increasing at a high pace over time. This is true for launch prices, especially in some therapeutic areas, but also for the prices of existing on-patent drugs, at least in the United States.

• Increasing prices of new medicines particularly in oncology and orphan drugs are not always justified by commensurate increases in benefits for patients. All these trends have raised concern about the value and the sustainability of pharmaceutical spending.

• The high prices of treatment for highly prevalent diseases have raised concerns about budget impact.

• The prices of some off-patent medicines, in a de facto monopoly position, have gone up rapidly.

• Pharmaceutical treatments with high costs are not always available to patients who need them, because of those costs, but also other reasons.

### The prices of new medicines have been increasing at a high pace

Pharmaceutical prices have recently received a lot of attention from policy makers, media, the general public and clinicians (Abboud et al., 2013; Tefferi et al., 2015; Council of the European Union, 2016). Concerns over increasing prices are not new. An analysis of media over the past 30 years identified several crises in the past (Leopold, Chambers and Wagner, 2016). In the late 1980s, media coverage on high drug prices focused on the novel AIDS treatment zidovudine (AZT), which cost USD 10 000 (USD 21 322) per patient per year in 2014. In the 2000s, media coverage spiked following the launch of new cancer medicines such as bevacizumab (Avastin®) for metastatic colon cancer and trastuzumab (Herceptin®) for breast cancer, with price tags of USD 100 000 (USD 126 450 in 2014 dollars) per treatment course (Leopold, Chambers and Wagner, 2016). Early in 2014 policy debates were dominated by the launch of sofosbuvir, a new and transformative treatment for hepatitis C, with an initial list price per treatment course of USD 84 000.

Authoritative information on prices and prices trends is scarce. The growing disconnection between list prices and actual prices that are paid by payers due to the use of confidential rebates makes analyses of price trends or price comparisons difficult. Furthermore, price indexes are only available in a few countries (see Box 1.5). However, a number of trends can be identified, described below.

Box 1.5. Price comparisons and analyses of price trends are increasingly difficult

## The disconnect between list prices and actual prices paid has been increasing

Public reports and research studies most often rely on “list prices”, which are the only prices available to researchers and citizens. Actual prices paid by health insurers, providers or governments may be lower thanks to off-invoice discounts and rebates.

In high income countries, discounts and rebates vary widely across products and across countries, from below 10% to more than 60% (Crédit Suisse, 2016; QuintilesIMS Institute, 2016a; Sawhney, Gordian and Behnke, 2016; Morgan, Vogler and Wagner, 2017). In 11 high-income countries/payers1 surveyed in 2016, discounts and rebates were most frequently in a 20-30% range (Morgan, Vogler and Wagner, 2017). An analysis of 25 companies operating in the United States in 2015 estimated that rebates accounted for 37% of gross sales and that four companies had rebates higher than 49% (Crédit Suisse, 2016). In 2016, Quintiles IMS estimates that average off-invoice discounts and rebates amount to nearly 30% in the United States, 17% in Europe and are lower in the rest of world2 (QuintilesIMS Institute, 2016).

Evidence suggests that confidential agreements on prices are becoming more frequent (Morgan, Vogler and Wagner, 2017) and that list prices are increasingly disconnected from actual prices paid. In the Unites States, for instance, discounts and rebates increased from about 18% to 28% of total spending on brand-name drugs between 2010 and 2014 (Aitken et al., 2016). As a percentage of gross Medicaid drug expenditures, rebates increased dramatically from 17.6% in 2003 to 47.6% in 2013. Rebates in the Medicare programme have grown from 8.6% of total costs of brand-name and generic drugs in 2006 to 12.9 in 2013. As a result, while the compounded annual real growth rate of invoice prices in the period 2005–14 was 6.4%, the rate for net prices was only 5.4%—with most of the difference occurring after 2009 (ibid.). Off-invoice discounts and rebates in the United States typically apply to medicines competing in a therapeutic area where they can be negotiated in exchange of a “preferred” status in health plans’ formularies.

In other OECD countries, health care payers (mainly governments and compulsory health insurance schemes) are also increasingly negotiating rebates on list prices through managed entry agreements (see Chapter 3). Managed entry agreements are more likely to affect newly launched medicines, with or without competitors for a short period after market entry.

Pharmaceutical spending as reported in the System of Health Accounts and OECD statistics is net of discounts and rebates.

## International price comparisons and analyses of price trends are difficult

Confidential agreements essentially allow manufacturers to charge different prices to purchasers with different price sensitivity (price discrimination), within national markets (typically in the United States) or between countries with national payers or negotiators. Tiered pricing is generally considered as desirable in a global market (Danzon and Towse, 2003; OECD, 2008). The disconnection between list prices and actual prices, however, has a number of drawbacks. One of them is that analyses of price trends or price comparisons have become very difficult.

Price indices reported in national statistics reflect the evolution of prices of existing treatments. These changes are measured by comparing prices in year t and t-1 for an identical basket of medicines present in both years, holding volumes constant. Only a handful of countries regularly publish specific pharmaceutical price indices. A comparison of price trends in Finland, France and the United States using such indexes showed that between 2002 and 2012, prices have been declining in both European countries every single year, while prices in the United States have been increasing by 3% on average (Belloni, Morgan and Paris, 2016).

1. Australia, Austria, Canada, England, Germany, New Zealand, Norway, Scotland, Sweden, the Netherlands, and the United States (Department of Veterans Affairs).

2. Estimates from Morgan et al. are based on a survey of public payers while estimates from Quintiles IMS are based on IMS data on volumes and transaction (“off-invoice”) prices, adjusted using sales reported in SEC filing for specific products to establish “net prices”.

#### New medicines, especially in some therapeutic areas, are launched at high prices

Many medicines now have annual costs per patient exceeding USD 100 000. In the United States, the association of health insurers published a list of 150 drugs with annual average wholesale price higher than USD 10 000 in 2016. For about half of the indications treated with these drugs, annual costs exceeds USD 100 000 per year. Several treatments for rare genetic disorders cost USD 500 000 to USD 800 000 annually (AHIP, 2016). Although these prices are list prices and only relate to the US market – where prices of patented medicines are typically higher than in other markets -- they give an indication that the prices of new treatments are getting higher.

In the United States, the monthly treatment costs of oncology medicines at launch have been increasing steadily since 2000, with an increase in the median price per patient per month from some USD 5 000 to more than USD 10 000 in 2015, adjusted for inflation. New treatments for chronic myeloid leukaemia (CML) were recently launched at prices ranging from USD 118 000 to USD 138 000 per year (Abboud et al., 2013). In 2017, a new cell-therapy was introduced for young adult and paediatric acute lymphoblastic leukaemia, at a list price of USD 475 000 per treatment, with an outcome-based agreement by which the company will charge for the drug when it achieves clinical response within 30 days (Kaltenboeck and Bach, 2017). Many of these high cost medicines target small populations, which moderates their impact on total pharmaceutical expenditures.

Increases in launch prices are also observed in treatments for multiple sclerosis. In the United States again, a study on nine disease modifying therapies approved between 1993 and 2013 showed that the annual cost of therapy at launch time increased from USD 8 300 for the cheapest drug introduced before 1996 to USD 58 000 for the drug approved in 2013 (a 7-fold increase in list prices at launch). Price increases in other countries might be more limited: the analysis of 2013 list prices observed for the same products in two countries showed that annual costs of the latest introduced therapy was twice that of the first introduced in Australia (USD 22 000 against 11 000) and in the United Kingdom (USD 29 711 against 12 018) (Hartung et al., 2015). Figure 1.8 presents an update of this study up to 2017 for the United States.

Orphan drugs are sometimes launched at high prices. A study of 74 medicines approved by the EMA between 2002 and 2014 for 63 indications, showed that 18% of them had annual treatment costs higher than GBP 100 000 in the United Kingdom; 58% had costs between GBP 10 000 and GBP 100 000 per year; and the remaining 24% had annual costs below GBP 10 000 (Onakpoya et al., 2015). Another study looked at 335 drugs16 with at least one orphan indication17 available in the United States in 2016. It showed that annual treatment costs ranged from less than USD 6 000 to more than USD 500 000. Among them, 8.4% had annual costs exceeding USD 200 000 per patient, accounting for 5.7% of sales for orphan indications; 15.8% had an annual cost between USD 100 000 and 200 000 and represented 27.6% of orphan drug/indications sales (QuintilesIMS Institute, 2017b).

#### In the United States, list prices of existing on-patent medicines are increasing rapidly in some therapeutic areas

The list prices of existing on-patent medicines in certain therapeutic areas are increasing, especially in the United States. Such trends are generally not observed in other countries where regulation prohibits or limits price increases for existing medicines.

The Canadian Patented Medicine Prices Review Board recently compared the price increases from 2011 to 2015 of biological disease-modifying anti-rheumatic drugs (DMARDs) in several countries (see Figure 1.9). Over this period, the list prices18 of biologic DMARDs increased by 65% in the United States and by 5% in Canada. By contrast, several countries, including Switzerland, France and Sweden, had marked price reductions over the same time period (PMPRB, 2016). Hartung (2017) showed that the list prices of disease modifying therapy drugs (DMT) for multiple sclerosis have increased over time after market launch in the United States, with an annual average price increase of 35% for two of the first drugs launched (see Figure 1.8). Part of the increase is offset by subsequent rebates for some purchasers, but these are unknown. Sharp increases in the list prices of insulin have also been reported in the United States. For seven common insulins, prices rose by 93% to 325% between 2010 and 2015 (The Alliance of Community Health Plans, 2015). From data submitted to the OECD in the course of this project and other sources, such price increases have not been observed in Australia or the Netherlands, for instance.

In current debates in the United States, PBMs are often blamed for high and increasing list prices, due to the rebates they negotiate (Walker, 2016). In response to these allegations, the association representing PBMs commissioned an analysis of trends in list and net prices for the 200 top selling brand self-administered drugs19 between 2011 and 2016. This study showed no correlation between increases in list prices and the percentage rebates consented over this period (Visante, 2017).

### Increasing prices of new medicines particularly in oncology and orphan drugs have raised concerns about the sustainability and value of pharmaceutical spending

Projections of future trends do not permit the drawing of firm conclusions about future impact on cost at aggregate level and therefore on financial sustainability. Nevertheless, concerns have grown about the value of pharmaceutical spending in certain therapeutic categories.

#### Pharmaceutical spending is expected to increase in the next five years but financial sustainability may not be the main issue

Since the economic crisis, pharmaceutical spending has grown at a slower pace than spending in other parts of the health systems in many OECD countries. According to IMS estimates, global pharmaceutical spending is expected to grow by 4 to 7% per year between 2016 and 2021 (reduced to 3 to 6% when confidential rebates and discounts are taken into account), a slow-down by comparison to growth in 2014 and 2015. Growth will be driven by treatments in oncology (+9 to 12%), for diabetes (+8 to 11%) and for autoimmune diseases (11 to 14%) but on the other side, curbed by loss of market exclusivity of a number of products. Growth is expected to be higher in the United States, Korea and Canada than in many large European markets (QuintilesIMS Institute, 2016).

Trends in spending in oncology and treatments for rare diseases (some of which are rare cancers) have been identified as particularly worrying, due to the combination of high treatment costs; the multiplication of new medicines in these areas, offering new treatment opportunities; and a rich pipeline of potential new drugs. However, the impact this will have on future cost dynamics remain difficult to predict:

• Oncology medicines account for a significant share of pharmaceutical spending, ranging from 10.2% to 15.9% of pharmaceutical spending in G7 countries in 2015 (IMS Health, 2017), and their share in health expenditure is likely to increase over time. The share of health spending allocated to all types of oncology treatments might also be increasing though evidence is scarce and not always consistent. For two countries, data are available over a decade. In Germany, the share of cancer care related expenditure in total spending for health increased from 7% to 8.4% between 2004 and 2013, while in the Netherlands, it increased from 5% to 7.7% between 2003 and 2013 (Eurostat, 2016). That said, a report from the Office of Health Economics and the Swedish Institute for Health Economics published in 2016 noted that such a level of expenditure was relatively low given that cancer accounts for a large share of the burden of disease (Jönsson et al., 2016b). Future growth in global spending for oncology medicines until 2021 was estimated by IMS in two reports published in 2016 and 2017. While the first report anticipated an average annual growth during this period of 9-12%, the second report anticipated an average annual growth at 6-9% until 2021 (QuintilesIMS Institute, 2016, 2017a). These figures show that uncertainties about approvals and take-up of new medicines and about their prices make predictions quite difficult.

• Orphan medicines account for a smaller share of total spending. In the United States, pharmaceutical sales for orphan indications20 account for 7.9% of total drug spending (QuintilesIMS Institute, 2017b). In European countries, this share is a little lower but increasing. In the Netherlands, between 2006 and 2012, the number of orphan drugs increased from 11 to 43, the number of patients treated from about 2 200 to about 9 800 and associated spending from EUR 61.2 to 260.2 million. The proportion in total pharmaceutical spending quadrupled in 6 years, from 1.1% to 4.2% (Kanters, Steenhoek and Hakkaart, 2014). Another study assessed that spending for orphan medicines accounted for 2.1% of pharmaceutical spending in Sweden and 3.1% in France in 2012 and predicted that these shares could increase to 4.1% and 4.9% respectively in 2020 (Hutchings et al., 2014). A sensitivity analysis however, with different assumptions on the number of medicines approved, trends in launch prices and the impact of loss of market exclusivity, led to estimates of 9% and 11% respectively in 2020 (ibid.).

Although future spending dynamics remain difficult to predict, the most important question will be to assess whether increased spending in these two areas comes with commensurate increases in value for patients and societies.

Box 1.6. Sustainability of health spending

As on average three-quarters of health spending is financed publicly in OECD countries, sustainability is often viewed through the lens of fiscal sustainability, which is defined as “the ability of a government to maintain public finances at a credible and serviceable position over the long term” (OECD, 2015b). Fiscal sustainability does not preclude increases in government spending on health. Societies often express a willingness to contribute more for health care than other areas of government spending, reflecting the value placed on health. Such an increase, however, has to be considered in relation to other domains of government interventions, as well as to the willingness of the population to pay higher taxes or other contributions for health care.

Within public health spending, many governments set budgets by line item, such as pharmaceuticals or hospital care. Such “silo budgeting” often leads policy makers and analysts to narrowly focus on individual sectors, such as hospital care or pharmaceutical care. Any change in pharmaceutical spending, however, be it an increase, should not be considered in a vacuum. As shown in the first section of this chapter, spending more in pharmaceutical treatments sometimes offsets spending in other parts of the health system. Without well-informed predictions on trends in prevalence, upcoming new treatments and their potential impact, and on patent expiries and opportunities for increased competition, it is difficult to determine a desirable level for future pharmaceutical spending.

### High prices are not always correlated with measurable health improvements

In some market segments, typically oncology and medicines used to treat rare diseases, high prices are not always correlated with measurable health improvement. As early as 2004, a study on medicines used in colorectal cancer showed that new treatments had significantly improved outcomes for patients with metastatic disease, nearly doubling the median survival time from 12 to 21 months, but had also increased costs 340-fold in 40 years (Schrag, 2004).21 More recently, a study looking at oncology medicines approved between 1995 and 2013 in the United States found that the average survival benefit was a little less than six months. Over the same period, the list price paid for an additional year of life increased in real terms from USD 54 100 in 1995 to USD 139 100 in 2005 and USD 207 000 in 2013. These costs do not include the costs of other medicines or treatments used in combination nor the costs of dealing with adverse effects (Howard et al., 2015). Drawing on a sample of 63 oncology medicines approved since the mid 1990s, a replication of this study in the French context showed that growth in the average launch prices of oncology medicines had given rise to increasing costs per life year gained, from EUR 20 700 in 1996 to EUR 175 968 in 2016.22 This represents an eight fold increase or compound annual growth of 11%, consistent with the growth observed in unpublished net confidential prices (CNAMTS, 2017).

Bae and Mullins (2014) compared incremental costs par QALY of oncology and non-oncology pharmaceutical treatments in a sample of studies published between 2003 and 2013 and adopting the perspective of a US payer. The average ICER for oncology medicines was about USD 138 600 (vs USD 50 000 for other products). ICERs were above USD 50 000 in 55% of cases and above USD 100 000 in 30%.. Even if prices in other countries are often lower than in the United States, new oncology medicines often do not meet ICER thresholds in use.23 In England, for instance, the proportion of oncology medicines not recommended by NICE has increased over time. While this proportion was of 31% for the whole period 2000-2016, it was 51% for indications approved by the EMA since 2007, in spite of an increase in the ICER threshold for these treatments during the period (Polton, 2015).

The issue of increasing prices for small improvements in health is also prevalent for orphan medicines (some of which are oncology treatments). A study looked at the level of evidence available for 74 approved drugs (2 of which have been withdrawn after approval) approved by the EMA for 63 orphan conditions, between 2002 and 2014. Of the 74 drugs, 5 (6.8%) were granted conditional approval, while 15 (20%) were granted approval under “exceptional” circumstances. Based on the GRADE criteria,24 the overall quality of evidence could be rated as moderate in 54 (73.0%) drugs, low in 16 (21.6%) and very low in 4 (5.4%). While 85% of these drugs showed significant clinical effects, serious adverse events were reported in 86.5% of them. The annual cost in the United Kingdom ranged from GBP 726 to GBP 378 000, with a median cost of GBP 31 012.While the study could not systematically compare the association between costs and benefits of all drugs because of differences in outcome measurements, it noted that the five drugs approved for pulmonary hypertension had comparable benefits but very different price levels (a 1:2 ratio for the two more recent approvals) and found similar results for drugs approved in cancer. Within indications, the study showed an impact of the date of approval on prices (Onakpoya et al., 2015). A systematic review of economic evaluations of ultra-orphan medicines showed high levels of uncertainty in effectiveness leading to wide ranges in ICERs, ranging from EUR 351 622 to EUR 3 282 252 per QALY for Fabry disease, from EUR 153 405 to EUR 1 043 868 per QALY for Pompe disease, and from EUR 43 532 to EUR 432 540 for Gaucher disease (Schuller, Hollak and Biegstraaten, 2015).

Many observers, experts and stakeholders suggest that observable health improvements (in length and quality of life) may not be the only parameters that payers and societies value when defining their willingness to pay. They advocate for a wider framework and/or for the use of multi-criteria decision analysis to better take into account all dimensions valued by patients and societies, especially for orphan and oncology medicines (Angelis and Kanavos, 2016; Carrera and IJzerman, 2016; Eurordis, 2017). The disconnection between prices paid and health improvements, however, raises questions about the efficiency of pharmaceutical spending. A New Zealand study which analysed population health gains foregone from unfunded cancer medicines found that funding more new cancer medicines in order to achieve numerical parity with Australia or other countries would not result in substantive health improvement and would cost significantly more. The authors suggested that selective funding of new medicines that demonstrate clear clinical benefit and that are cost-effective and affordable would be a sensible approach to deliver the best health outcomes for all New Zealanders (Evans et al., 2016). The disconnect between prices paid and measurable health benefits also raises questions in terms of signals being sent to industry for the development of new treatments.

### Countries may be ill-prepared for the arrival of new treatment for highly prevalent diseases

The launch of a breakthrough treatment of hepatitis C in 2013 created a shock in many countries. Gilead’s direct antiviral agents (DAAs) sofosbuvir entered the US market in 2013 at a list price of USD 84000. Prices were lower in other markets, as a result of the tiered-pricing strategy defined by Gilead, linking the price of its medicine to income level and prevalence of the disease in different countries. The therapeutic value of the new treatment was immediately recognised and it was found cost-effective, even at its high price. Payers, however, had not anticipated this launch and felt overwhelmed by the budget impact of treating the whole affected population at that price. For countries with the highest prevalence, it was considered unaffordable (Iyengar et al., 2016). In several OECD countries, payers initially limited access to the latest hepatitis C treatments to the most severely affected patients (CNAMTS, 2016), creating frustration for patients and clinicians.

Many payers negotiated lower prices and confidential discounts leading to reduced net prices (IMS Institute for Healthcare Informatics, 2016a), but prices really began to decrease when competition kicked in. Between August 2013 and August 2017, five new single-component DAAs and six new fixed-dose combinations received their first worldwide regulatory approval (Unitaid and WHO, 2017). Competition helped reducing prices and the last competitor entered the US market in 2017 with a list price of USD 26 400 per treatment course (before discounts), a monthly price 30% to 60% lower than that of competitors (Sagonowsky, 2017). The most recent negotiations in OECD countries where prices are regulated also achieved lower prices and expanded population coverage (see Box 1.7).

If public payers had better anticipated first the arrival of the sofosbuvir and second, the arrival of competitors, they might have responded differently. First, they could have planned and justified a prioritisation of treatment of the most severely affected patients and populations with high risks, in order to spread the costs over several years. In some countries, such an approach would have been anyway justified by the system’s capacity to initiate treatments. They would have also benefited from the impact of competition on prices for a deployment of DAAs for the whole affected population.

This exceptional case also shows that the “willingness to pay” for a new product treating a highly prevalent disease does not only depend on its intrinsic value (or cost-effectiveness), but also on the total bill. Since production costs are relatively low, volumes are an important element of the company’s returns on investments and caps on spending may well be acceptable for the company if volumes are high.

Box 1.7. Coverage and pricing of hepatitis treatments - examples

In France, the committee in charge of price regulation negotiated prices and confidential arrangements, while the Parliament enacted spending caps and ex post rebates (CEPS, 2016).

• From September 2013 (prior to the marketing authorisation obtained in 2014 from EMA), Sovaldi® (sofosbuvir) was available in France through a temporary authorisation for use (financed by health insurance, at the price set by the manufacturer). The Transparency Commission issued a positive recommendation for coverage in June 2014 for a restricted population and the Economic Committee for Health Products (CEPS) negotiated a price, informed by an economic evaluation undertaken by the Haute Autorité de Santé (HAS). Sovaldi® was included in the positive list in November 2014.

• Prior to its inclusion in the positive list, Sovaldi was accessible to patients for a standard price of treatment, set by Gilead at EUR 56 000, fully covered by health insurance. In addition, the CEPS negotiated an agreement with ex post rebates linked to volumes of sales and performance of the product observed in a large cohort of patients. The net price was substantially lower than the list price.

• Two other DAAs, Daklinza® (daclatasvir) from Bristol-Myers Squibb and Olysio® (simeprevir) from Janssen were included in the positive list in May 2015, with list prices of EUR 46,000 per treatment course with managed entry agreements (MEAs) similar to those of Solvaldi. In August 2015, the listing of two other products from Abbvie, Exviera® and Viekirax®, used in association allowed a further price reduction.

• In addition, two provisions were included in the Social Security Finance Acts of 2014 and 2015: 1) the restitution by Gilead of the difference between the price set by the CEPS and price used by the company before listing, multiplied by the volume of sales, for the whole period prior to listing; and 2) the implementation of an expenditure ceiling on DAAs for 2014 and 2015 (respectively EUR 450 million and EUR 700 million), beyond which companies had to pay rebates. Rebates paid by individual companies can be defined through individual agreements or computed using the default rule set by law. Ex post rebates for DAAs amounted to EUR 21.4 million for the two years 2014 and 2015.

In May 2016, the French Minister of Health announced the expansion of treatment to all infected patients. The pricing committee negotiated again, first with MSD and then with Gilead, to further lower prices. From April 2017, the cost of a treatment with Sovaldi has been lower than EUR 28 700. All treatments are fully reimbursed by health insurance (Ministère des affaires sociales et de la santé, 2017).

In Australia, DAAs for hepatitis C have been reimbursed since early 2016 (more than 18 months after the approval of the first entrant) without any restrictions in terms of population. This broad access for Australians with hepatitis C arises from the close interplay between the role of health technology assessment (HTA) and the role of the payer.

• First, the statutory HTA expert advisory committee advising government (the Pharmaceutical Benefits Advisory Committee or PBAC) paid close attention to the appraisal of cost-effectiveness. For example, the PBAC’s assessment of sofosbuvir published in March 2015 set a broad context in order to provide its advice to the Minister for Health; noted the gains in clinical effectiveness and safety and the shorter duration of the new treatment regimens; and justified the use of a (lower than usual) ICER/QALY threshold of AUD 15 000 as the basis for back-calculating a cost per course per patient and thus a price across the proposed regimens (PBAC, 2015). PBAC assessments for other DAAs also influenced the negotiations between the Australian government and sponsor companies.

• Second, the government adopted a clear approach to price and expenditure negotiations across the competing sponsors of hepatitis C medicines. Finalisation of these negotiations was backed by formal Deeds of Agreement comprising special (confidential) pricing arrangements and (financial) risk sharing arrangements.

• According to public statements, the Australian government committed over AUD 1bn in funding to an HCV treatment programme over a five-year period for an estimated treatment population of 62 000 (Unitaid and WHO, 2017). A key objective of the scheme was to achieve budget certainty through an annual budget cap above which manufacturers would be required to rebate treatment costs. The programme subsidises unrestricted access for patients regardless of route of transmission or disease stage. In addition, GPs are allowed to prescribe DAAs (Australian Government Department of Health, 2017).

• While PBAC’s assessments produced an initial estimate of 6 600 patients in year one of the programme, an estimated 32 400 individuals initiated DAA treatment in 2016 (from March to December), or 14% of the people living with chronic HCV infection in Australia (Kirby Institute, 2017).

In countries outside the OECD, new DAAs and fixed-dose combinations have been made available at lower prices through multiple mechanisms including voluntary licenses granted by companies marketing these products and the production of generic products where patent protection does not exist (Unitaid and WHO, 2017). For example:

• In May 2014, Egypt, the country with the highest HCV prevalence, concluded an agreement with Gilead to purchase a 12 weeks’ course of sofosbuvir for USD 900. Gilead has since extended this price to 101 low- and middle-income countries included in its voluntary licences, and subsequently set the price of sofosbuvir/ledipasvir for these countries at USD 1 200 for 12 weeks. In September 2016, Gilead reduced its prices to USD 750 for 12 weeks of sofosbuvir and USD 900 for 12 weeks of treatment with either sofosbuvir/ledipasvir or sofosbuvir/velpatasvir. Four additional countries (Armenia, Georgia, Moldova and Ukraine) are now eligible to procure at these prices. Other countries like Argentina or Brazil are not eligible and pay substantially higher prices (more than USD 6 000 for sofosbuvir).

• In September 2014, Gilead signed voluntary licences for sofosbuvir and ledipasvir with a number of major generic producers in India. These companies have the right to manufacture generic versions of sofosbuvir and ledipasvir, and supply them to the 101 countries included in the licence. In January 2015, these licences were amended to include velpatasvir. The voluntary licence does not include middle-income countries such as Brazil and China.

• Generic prices of sofosbuvir and sofosbuvir/ledipasvir fell well below “access price” levels set by Gilead and prices as low as USD 66 for a 12-week course of treatment for sofosbuvir and USD 191 for sofosbuvir/ledipasvir in India in 2016-2017 (Unitaid and WHO, 2017).

### The prices of some existing off-patent medicines have increased rapidly

A few cases of sharp price increases in off-patent drugs have received substantial media attention recently. For instance, Mylan came under public scrutiny in August 2016 after reports that since acquiring rights to EpiPen in 2007, the company had implemented a series of gradual price increases in the United States, inflating the price of the drug from USD 56.64 to USD 317.82, a 461% increase. In 2015, Turing Pharmaceuticals increased the price of its HIV medicine Daraprim by 5 000%. The price of the drug was raised from USD 13.50 to USD 750 overnight (Pollack, 2015).

Increases in off-patent medicines are not only a US phenomenon. From 2005 to 2015, the prices of 10 generic drugs increased by 1 000% or more in the United Kingdom. The pain medication and antidepressant doxepin saw the biggest rise. In 2005, its list price was GBP 2.36 (USD 2.89). By 2015, the price had increased to GBP 124.56, a change of more than 5 000%. The prices of six other generic drugs have increased by at least 2 000% over the same period. In response to a query launched with the Pharmaceutical Pricing and Reimbursement Information (PPRI) network, Austria reported a price increase of 2 584% for a sodium chloride infusion solution, as well as price increases of more than 100% for about 50 medicines since 2012, of which 16 were for medicines covered by social health insurance (OECD, 2017e).

An exhaustive search of all cases of price increases brought to competition or anti-trust authorities and reported in the PaRR database between 1 January 1 2007 and 28 February 2017 was undertaken as part of this project.25 Twenty cases were reported in the United States, two in the United Kingdom and one in Italy, all filed between 2013 and 2017, for price increases ranging from +80% to 8 281%. On 15 May 2017, the European Commission initiated formal anti-trust proceedings against Aspen for a suspected breach of EU rules prohibiting abuse of market dominance (European Commission, 2017).

Price increases among off-patent drugs have only occurred for a small number of products and are unlikely to significantly increase health spending. In Canada, for instance, prices of single-source non-patented drugs26 have increased on average by 18% between 2010 and 2016 but the budget impact remains limited: these drugs represented only 2.5% of total expenditure of public drug plans in Canada (PMPRB, 2017). In the United Kingdom, however, after a decade in which the annual change in the average price of generics ranged from -17% to 5.1%, in 2015 the figure shot up to 14.4% (RAPS, 2016). In the United States, a study showed that one in five generics marketed for the whole 2010-15 period experienced extraordinary price increases (higher than 100%), which moderated the price erosion of “established” generics (GAO, 2016).

The common element in all these international cases of price hikes in off-patent medicines is that the providers have a monopoly on the supply of the medicine in question. There is, in fact, significant market concentration in the generics market. A recent study of trends in US generic markets between 2004 and 2016 showed that 40% of generic molecules were supplied by a single manufacturer; another 40% are supplied by more than four manufacturers; the remaining 20% molecules being supplied by two or three manufacturers. The number of generic manufacturers has been decreasing recently and generic prices have risen faster than inflation, especially in markets with few competitors (Berndt, Conti and Murphy, 2017). Recently, new measures have been introduced to facilitate more efficient generic drug review by the FDA, with the objective of enhancing competition and thereby lowering prices (FDA, 2018).

### Innovative medicines are not always available to patients who need them, either because of cost - or for other reasons

Several studies have shown that innovative medicines are not always available to patients who need them. High prices are not the only cause of poor access (see Box 1.8 for a discussion of access). Sequential launch strategies, delays imposed by health technology assessment and price negotiations, and characteristics of health systems (e.g. low capacity or inefficiencies) can also delay or impede access to treatment.

One study compared the availability and coverage of 45 oncology drug indications approved in the United States between 1 January 2009, and 31 December 2013 in the United States, Canada, Australia, and the United Kingdom. The US Medicare programme covered all indications in June 2014. Of 30 indications already approved by the European Medicines Agency, 87% were covered in the United Kingdom. In Canada, 54% of the 24 indications approved were recommended by the Committee in charge of assessing oncology drugs to inform coverage decisions (which public plans remain free to follow or not). In Australia, 46% of the 24 indications approved were publicly covered (Zhang, Chantel Hueser and Hernandez, 2017).

A study of the European Society for Medical Oncology (ESMO) evaluated the formulary availability of licensed anti-neoplastic medicines in 49 European countries; patient out-of-pocket costs for the medications and their actual availability for patients with valid prescriptions. It showed that oncology treatments are more often available and covered in Western European countries than in Eastern European countries, where many drugs are either not available or only available at full cost to patients. Most of the medicines included in the 2015 updated version of the WHO Model List of Essential Medicines, however, were available with no out-of-pocket cost to patients, even in Eastern European countries, and were reported as being always or usually available. Problems impeding actual availability were reported in some Eastern European countries, including Hungary, Latvia, Slovenia, and the Slovak Republic (Cherny et al., 2016).

A recent study assessed the availability and accessibility27 of all orphan medical products (OMPs) approved by the EMA to mid-2016 in several EU countries (Zamora et al., 2017). Since the introduction of the orphan drug legislation in Europe, 143 orphan drugs have obtained a marketing authorisation according to the centralised procedure (for 145 indications). Of these, nearly 40% (56) were products used in oncology. This report shows that availability varies from a low of 55% in Spain to 97% in Germany, and access (i.e. coverage) from 33% in Wales to 93% in Germany (see Figure 1.10). However, these factors do not guarantee patients will be treated with these medicines, and information on diffusion and uptake – together with information on prevalence – would be useful to complement these data. Marketing authorisation, coverage and price are not the only potential barriers in access to care for patients with rare diseases. Delayed diagnoses, limited access to resources, and absence of specific therapies often preclude patients from receiving proper, timely care (Dharssi et al., 2017). In the United States, a recent study of the coverage of 138 orphan medicines marketed between 2000 and 2016 by 20 leading commercial payers revealed that 80% of orphan medicines are covered by at least 75% of payers and one third are covered by all payers. About 44% of these medicines are in formularies’ tier 4, with the highest levels of cost-sharing. The authors also noted an increase over time in the number of coverage denials, in patient cost-sharing, and in use of prior authorisation and quantity limits (Cohen and Awatin, 2017).

Box 1.8. Accessibility and affordability of medicines

For health care in general, access can be defined as the ability to obtain health care services based on medical need and irrespective of factors that are not related to need, such as physical location, socio-economic status, or income and ability to pay (Oliver and Mossialos, 2004). For medicines more specifically, this requires that they are marketed in a country, affordable (no financial barriers for patients or for the health system) and that they are physically accessible at a reasonable inconvenience or cost. For purposes of monitoring achievement of the Millennium Development Goals, the United Nations Development Group defined access as “having medicines continuously available and affordable at public or private health facilities or medicine outlets that are within one hour’s walk of the population” (UNDP, 2003).

In addition to general availability through marketing of a medicine in a country, the existence and magnitude of financial access barriers can serve as one measure of access because the full prices of health services paid to providers or manufacturers of medicines are usually beyond the financial means of all but the wealthiest patients in need of treatment. Therefore, indicators of access include the coverage of medicines by publicly funded health care or insurance schemes, levels of public spending or medicine prices.

Medicine prices determine affordability not only from the point of view of individual patients but also for publicly funded health care systems. If prices of medicines are very high, they may be excluded from public coverage or restricted to only the most severe cases. The latter has been a problem in resource constrained settings for some time but more recently also in high-income countries. It is difficult to define affordability in general because it is an inherently normative concept dependent on local constraints or preferences (Niëns and Brouwer, 2013). Some practical definitions have been developed with reference to incomes, wages or wealth in the population or levels of expenditure that would lead to impoverishment (ibid.).

### Incentives for the development of orphan medicines may not always be closely aligned with their intended policy goals

Between 5 000 and 8 000 rare diseases have been identified to date. Many of them are degenerative and life-threatening, 80% have a genetic origin, and half have a childhood onset, causing about one-third of child deaths before one year of age (López-Bastida et al., 2016). Since the population affected by each of disease is small, commercial incentives to develop medicines for patients with rare diseases were seen as inadequate. To encourage the development of such treatments, several countries (United States, European Union, Australia and Japan) have introduced specific “orphan” drug policies. While eligibility criteria and incentives differ across countries, they have undoubtedly helped foster the development of drugs for orphan conditions (Giannuzzi et al., 2017). A number of trends, both in science and in pharmaceutical markets, however, have raised questions about the ability of existing policy frameworks to support the objective of encouraging the development of therapies that would otherwise be considered commercially unattractive.

First, a few orphan drugs have reached the market at high prices in the past decade. A recent report by Quintiles IMS shows that in 2016 approximately one quarter of orphan medicines in the United States had annual costs of more than USD 100 000, accounting for 35% of all sales for orphan indications (see Figure 1.11). The report shows that medicines for conditions with very low prevalence are likely to command higher prices, but does not provide any information on price trends, even though the sample of products spans 33 years of approvals. In 2014 in the United Kingdom, 18% of orphan medicines had annual treatment costs higher than GBP 100 000 (about USD 130 000) (Onakpoya et al., 2015).

Second, if companies obtain orphan indications for which they charge high prices, they may subsequently develop non-orphan indications for which they are able to maintain this price. This strategy, however, can only be used in the United States, where a brand-name medicine can be sold for both orphan and non-orphan indications. Of 449 orphan drugs approved by the FDA between 1983 and 2016, 98 (22%) have both orphan and non-orphan indications, among which 34 obtained an orphan indication first and 10 orphan and non-orphan indications simultaneously (QuintilesIMS Institute, 2017b). In Europe, medicines used in the treatment of orphan and non-orphan indications must be marketed with different brand names and are likely to be priced differently.

Third, incentives for the development of orphan medicines are being provided even for drugs with revenues exceeding USD 1 billion. Côté et al. (2012) showed that 43 brand-name medicines with at least one orphan indication had annual global sales of exceeding this figure in 2008. Among them, 18 products only had orphan indications. A more recent study observed that a number of top selling drugs had one or more orphan designations: in 2015, seven out of the ten top-selling drugs worldwide had orphan indications, with global sales ranging from USD 5.2 billion to USD 14.1 billion (Daniel et al., 2016). In interpreting these figures, however, several points need to be taken into account:

• First, this problem mainly seems to concern US-specific incentives for orphan drugs. EU orphan drug legislation is very different and the number of approvals of orphan indications is much lower. Between 2001 (first year of implementation of the EU legislation) and 2015, 339 orphan drugs were approved by the FDA and only 117 by the EMA (OECD, 2017b).

• Second, two cases should be highlighted. In some cases, so-called blockbuster medicines have both orphan and non-orphan indications. For Humira®, for instance, the company developed six orphan indications, four of which have been approved, but orphan indications account for less than 4% of the product’s sales in the United States (QuintilesIMS Institute, 2017b). The question is then: would the company have developed these indications without the specific 50% tax-credit on R&D spending and extended exclusivity for these indications? In other cases, these blockbusters only have orphan indications. This was the case for 18 products among 43 blockbusters with orphan indications identified in 2008 (Côté and Keating, 2012). For example imatinib (Gleevec®) first obtained FDA orphan designation and marketing approval as a breakthrough treatment for patients with CML in 2001. Subsequently, it obtained a further eight orphan indications through 2014,28 and had global sales of USD 4.65 billion in 2015 - just prior to generic entry in the United States (Pharmaceutical-Technology, 2016).

• With the development of precision medicines and targeted therapies, many oncology products may be proposed for a large number of indications with small populations. PD-1 inhibitors have such profiles. Nivolumab for instance (Opdivo®), first approved in melanoma and lung cancer, has been tested in more than 50 trials - as a monotherapy or in combination with other therapies - for multiple tumour types while pembrolizumab (Keytruda®) has been tested in 85 clinical trials for 30 tumour types (Gibney, 2017). Nivolumab has FDA orphan drug designations (as monotherapy or in combination) in 14 indications and pembrolizumab has 11.29 With high unit prices, Keytruda® and Opdivo® achieved respectively USD 3.8 billion and USD 4.9 billion in global sales in 2017, 60% of which were in the United States (Bristol-Myers Squibb Company, 2018; Merck & Co. Inc., 2018). In such cases, the question is whether current incentives are sufficient for these companies to develop as many indications as possible.

Finally, a few new orphan products are “repurposed” medicines, previously prescribed off-label to patients at low prices and launched with an approval for an orphan indication at a high price (see Box 1.9). While it is necessary to compensate developers for the costs of generating data to demonstrate the safety and efficacy of a product in a new indication, appropriate policies may be needed to ensure that patients with orphan diseases can continue to access these drugs.

Box 1.9. Examples of repurposing at high cost to patients and payers

Murphy et al. provide a few examples where orphan drug legislation in the United States did not improve patient access (Murphy, Puwanant and Griggs, 2012):

• Tetrabenazine was approved as orphan drug in 2008 for treatment of chorea associated with Huntington’s disease. This medicine has been used for decades in the United Kingdom and many other countries in this indication. Prior to FDA approval, patients could try to import the product from abroad with a prescription from their doctor, based on FDA’s Personal Importation Policy, but had to pay the full cost themselves. In 2012, the cost per month for the initial dose of 25mg/day was estimated at USD 2 055 in the United States, while the same dose was available in the United Kingdom for approximately USD 43.

• Adrenocorticotropic hormone (ACTH, corticoptropin) has been used as a first-line treatment for infantile spasms in many countries - including the United States - since the 1950s, albeit without FDA approval for this indication. In 2007, the price for one vial increased from USD 1 650 to USD 23 000, when the company applying for the orphan drug indication adopted an “orphan-style pricing model”. In October 2010, the FDA approved the product for the treatment of infantile spasms in children under two years, a very small target population estimated at 800 children per year. The cost per vial in the United Kingdom was still under USD 523 in 2012. The manufacturer subsequently obtained other indications, including multiple sclerosis, for which it originally kept the same price in the United States. The current US list price is approximately USD 39 000 (Hartung et al., 2018). As a high-priced drug, it is likely to be placed on a higher or specialty tier in insurance formularies and associated with a significant co-payment or co-insurance amount, and/or subject to prior authorization. While the manufacturer provides a co-pay programme, assistance is limited. Finally, the exceptionally high cost of the drug to third-party payers is likely to influence treatment choices and to have flow-on effects for the magnitude of premiums and deductibles (Drash, 2018).

Incentives for the development of orphan drugs have to be sustained. Although the number of therapies available to patients with rare diseases has increased overtime, it is important to stress that treatments are available for less than 5% of the 6 000-8 000 rare diseases. More efforts are also needed to make sure that available therapies are available in different countries. A recent study showed that 64 orphan drugs approved either in the United States or in the European Union were not approved in the other region, with or without an orphan drug designation/status (Giannuzzi et al., 2017). The question raised in this section is whether incentives are still well targeted to encourage the development of medicines that would not occur in the absence of these policies.

## References

Abboud, C. et al. (2013), “The price of drugs for chronic myeloid leukemia (CML) is a reflection of the unsustainable prices of cancer drugs: From the perspective of a large group of CML experts”, Blood, 121(22), pp. 4439–4442, http://dx.doi.org/10.1182/blood-2013-03-490003.

ACS CAN (2013), Understanding Biologic Medicines From the Cancer Patient Perspective, American Cancer Society Cancer Action Network, Washington DC.

Aggarwal, A., Ginsburg, O. and Fojo, T. (2014), “Cancer economics, policy and politics: What informs the debate? Perspectives from the EU, Canada and US”, Journal of Cancer Policy, Elsevier Ltd., 2(1), pp. 1–11, http://dx.doi.org/10.1016/j.jcpo.2014.02.002.

AHIP (2016), High-Price Drugs Based Upon Estimated Annual Per-Patient Expenditure High-Priced Drugs: Estimates of Annual Per-Patient Expenditures for 150 Specialty Medications, America’s Health Insurance Plans, Washington.

Aitken, M. et al. (2016), “Has The Era Of Slow Growth For Prescription Drug Spending Ended?”, Health affairs (Project Hope), 35(9), pp. 1595–1603, http://dx.doi.org/10.1377/hlthaff.2015.1636.

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. http://dx.doi.org/10.1007/s40273-015-0370-z.

Arcidiacono, P. et al. (2013), “Pharmaceutical followers”, International Journal of Industrial Organization. Elsevier B.V., 31(5), pp. 538–553, http://dx.doi.org/10.1016/j.ijindorg.2013.10.005.

ASPE (2016), “Observations on Trends in Prescription Drug Spending”, ASPE Issue Brief, http://dx.doi.org/10.1017/CBO9781107415324.004.

Asselah, T., Marcellin, P. and Schinazi, R. F. (2018), “Treatment of hepatitis C virus infection with direct-acting antiviral agents: 100% cure?”, Liver International, 38 (November 2017), pp. 7–13, http://dx.doi.org/10.1111/liv.13673.

Auraaen, A. et al. (2016), How OECD health systems define the range of good and services to be financed collectively, OECD Health Working Papers, No. 90, OECD Publishing, Paris, http://dx.doi.org/10.1787/5jlnb59ll80x-en.

Australian Government Department of Health (2016), Other supply arrangements outside the Pharmaceutical Benefits Scheme (PBS), Australian Government Department of Health, http://www.health.gov.au/LSDP.

Australian Government Department of Health (2017), “Hep C medicines - The facts”, http://www.pbs.gov.au/publication/factsheets/hep-c/files/hepc-factsheet-community-based-prescribers.pdf, accessed 31 Oct. 2018.

Bach, P. B. (2009), “Limits on Medicare’s Ability to Control Rising Spending on Cancer Drugs”, New England Journal of Medicine, 360(6), pp. 626–633, http://dx.doi.org/10.1056/NEJMhpr0807774.

Bae, Y. H. J. and Mullins, D. (2014), “Do Value Thresholds for Oncology Drugs Differ from Nononcology Drugs?”, Journal of Managed Care and Specialty Pharmacy, 20(11), pp. 1086–1092, http://dx.doi.org/10.18553/jmcp.2014.20.11.1086.

Barnett, J. C. and Berchick, E. R. (2017), “Health Insurance Coverage in the United States: 2016”, United States Census Bureau, Washington, DC, pp. 7–10.

Belloni, A., Morgan, D. and Paris, V. (2016), Pharmaceutical Expenditure And Policies: Past Trends And Future Challenges, OECD Health Working Papers, No. 87, OECD Publishing, Paris, http://dx.doi.org/10.1787/5jm0q1f4cdq7-en.

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, Working Paper No. 23640.

Van Den Brink, R. (2014), “Reimbursement of orphan drugs: The Pompe and Fabry case in the Netherlands”, Orphanet Journal of Rare Diseases, 9 (Suppl 1), p. 2014, http://dx.doi.org/10.1186/1750-1172-9-S1-O17.

Bristol-Myers Squibb Company (2018), Annual report 2017 - Form 10-K, Annual Report Pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 For the Fiscal Year Ended December 31, 2017, US Securities and Exchange Commission, Washington.

Bruen, B. et al. (2016), The impact of reimbursement policies and practices on healthcare technology innovation, Final report prepared with the support of the U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation; Washington DC, https://aspe.hhs.gov/system/files/pdf/188741/ImpactofReimbursementonInnovation.pdf.

Campillo-Artero, C., Garcia-Armesto, S. and Bernal-Delgado, E. (2016), “The merry-go-round of approval, pricing and reimbursement of drugs against the Hepatitis C virus infection in Spain”, Health Policy, Elsevier Ireland Ltd., 120(9), pp. 975–981, http://dx.doi.org/10.1016/j.healthpol.2016.07.005.

Carrera, P. and IJzerman, M. J. (2016), “Are current ICER thresholds outdated? Valuing medicines in the era of personalized healthcare”, Expert Review of Pharmacoeconomics and Outcomes Research, 16(4), pp. 435–437, http://dx.doi.org/10.1080/14737167.2016.1180980.

CEPS (2016), Rapport d’activité 2015, Comité Economique des Produits de Santé, Paris.

Chaplin, S. (2015), “Health Survey for England 2013: The use of prescribed medicines”, Prescriber, Vol. 26/4, pp. 16-19.

Cherny, N. et al. (2016), “ESMO European Consortium Study on the availability, out-of-pocket costs and accessibility of antineoplastic medicines in Europe”, Annals of Oncology, 27, pp. 1423–1443, http://dx.doi.org/10.1093/annonc/mdw213.

CMS (2018), National Health Expenditure Data, Centers for Medicare & Medicaid Services, U.S. Centers for Medicare & Medicaid Services, Washington DC, https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData/NationalHealthAccountsHistorical.html (accessed 12 July 2018).

CMS (2016), Medicare Prescription Drug Benefit Manual Chapter 6 – Part D Drugs and Formulary Requirements Transmittals for Chapter 6, U.S. Centers for Medicare & Medicaid Services, Washington DC, https://www.cms.gov/Medicare/Prescription-Drug-Coverage/PrescriptionDrugCovContra/Downloads/Part-D-Benefits-Manual-Chapter-6.pdf (accessed 2 November 2018)

CNAMTS (2016), Améliorer la qualité du système de santé et maîtriser les dépenses: Propositions de l’Assurance Maladie pour 2017, Caisse Nationale de l’Assurance Maladie, Paris, https://assurance-maladie.ameli.fr/sites/default/files/rapport-activite-charges-produits-17_assurance-maladie.pdf

Cohen, J. P. and Awatin, J. G. (2017), “Patient access to orphan drugs”, Expert Opinion on Orphan Drugs, 5(12), pp. 923–932, http://dx.doi.org/10.1080/21678707.2017.1402676.

Côté, A. and Keating, B. (2012), “What is wrong with orphan drug policies?”, Value in Health, 15(8), pp. 1185–1191, http://dx.doi.org/10.1016/j.jval.2012.09.004.

Council of the European Union (2016), “Council conclusions on strengthening the balance in the pharmaceutical systems in the EU and its Member States - Consilium”, Press release 17/06/2016, http://www.consilium.europa.eu/en/press/press-releases/2016/06/17/epsco-conclusions-balance-pharmaceutical-system/ (accessed 19 November 2017).

Crédit Suisse (2016), Global Pharmaceuticals, Industry Overview/Analysis, Crédit Suisse, Equity Research Global.

Culyer, A. J. (2016), “Cost-effectiveness thresholds in health care: a bookshelf guide to their meaning and use”, Health Economics, Policy and Law, 11(4), pp. 415–432, http://dx.doi.org/10.1017/S1744133116000049.

Cutler, D., McClellan, M. and Newhouse, J. (1998), The costs and benefits of intensive treatment for cardiovascular disease, National Bureau of Economic Research, Working Paper No. 6514.

Daniel, M. G. et al. (2016), “The Orphan Act - Restoring the mission to rare diseases”, American Journal of Clinical Oncology, 39(2), pp. 210–213.

Danzon, P. M. and Towse, A. (2003), “Differential pricing for pharmaceuticals: reconciling access, R&D and patents.”, International journal of health care finance and economics, 3(3), pp. 183–205.

Dharssi, S. et al. (2017), “Review of 11 national policies for rare diseases in the context of key patient needs”, Orphanet Journal of Rare Diseases, 12(1), p. 63, http://dx.doi.org/10.1186/s13023-017-0618-0.

Drash, W. (2018), Anatomy of a 97,000% drug price hike: One family’s fight to save their son, CNN, https://edition.cnn.com/2018/06/29/health/acthar-mallinckrodt-questcor-price-hike-trevor-foltz/index.html (accessed 13 July 2018).

Drummond, M. and Towse, A. (2014), “Orphan drugs policies: A suitable case for treatment”, European Journal of Health Economics, 15(4), pp. 335–340, http://dx.doi.org/10.1007/s10198-014-0560-1.

Dubois, P. et al. (2016), The Effects of Pharmaceutical Innovation on Cancer Mortality Rates, Toulouse School of Economics, Working Papers, No. TSE‐688.

Ehreth, J. (2003), “The global value of vaccination.”, Vaccine, 21(7–8), pp. 596–600.

European Commission (2017), “Antitrust: Commission opens formal investigation into Aspen Pharma’s pricing practices for cancer medicines”, Press release - 15 May, http://europa.eu/rapid/press-release_IP-17-1323_en.htm (accessed 3 November 2017).

Eurordis (2017), Mechanism of Coordinated Access to orphan medicinal products (MoCA), https://www.eurordis.org/content/Moca (accessed 24 November 2017).

Eurostat (2016), HEDIC - Health expenditures by diseases and conditions, Edition 2016, Statistical Working paper, European Union, Luxembourg https://ec.europa.eu/eurostat/documents/3888793/7605571/KS-TC-16-008-EN-N.pdf/6cb33aa4-2e65-4df7-9b2b-1ff171eb1fba

EvaluatePharma (2016), World Preview 2016, Outlook to 2022, Evaluate, London

Evans, J. et al. (2016), “Mind the gap: An analysis of foregone health gains from unfunded cancer medicines in New Zealand”, http://dx.doi.org/10.1053/j.seminoncol.2016.10.004.

FDA (2018), Press Announcements - Statement from FDA Commissioner Scott Gottlieb, M.D. on new steps to facilitate efficient generic drug review to enhance competition, promote access and lower drug prices, US Food and Drug Administration, Office of the Commissioner, Washington DC, https://www.fda.gov/newsevents/newsroom/pressannouncements/ucm591184.htm (accessed 2 July 2018).

FDA (2015), Sandoz Launches Zarxio at 15 Percent Lower Price Than Neupogen, FDANews, US Food and Drug Administration, Washington DC, http://www.fdanews.com/articles/173036-sandoz-launches-zarxio-at-15-percent-lower-price-than-neupogen (accessed 12 October 2017).

Frank, R. G. (2012), Prescription Drug Procurement and the Federal Budget, Medicare Policy Issue Brief, The Henry J. Kaiser Family Foundation, Menlo Park, CA.

GAO (2016), Generic drugs under Medicare - Part D Generic Drug Prices Declined Overall, but Some Had Extraordinary Price Increases, Report to Congressional Requesters, Report GAO-16-706, United States Government Accountability Office, Washington DC

Gaviria, A. (2016), Strengthening Governments Capacity to Discern Value: The Need to Address Technological Pressure on Health Expenditure, Breve 13, Based on a presentation at Criteria by Alejandro Gaviria, Minister of Health of Colombia, December 2015, Criteria, Inter-American Development Bank, https://publications.iadb.org/handle/11319/7763?locale-attribute=en&

Giannuzzi, V. et al. (2017), “Orphan medicinal products in Europe and United States to cover needs of patients with rare diseases: an increased common effort is to be foreseen”, Orphanet Journal of Rare Diseases, 12(1), p. 64, http://dx.doi.org/10.1186/s13023-017-0617-1.

Gibney, M. (2017), “Opdivo”, FiercePharma, https://www.fiercepharma.com/special-report/3-opdivo (accessed 26 November 2017).

Harris, R. (2016), Small Savings For Drugs Made To Mimic Biotech Blockbusters, NPR, http://www.npr.org/sections/health-shots/2016/10/19/498559386/small-savings-for-drugs-made-to-mimic-biotech-blockbusters (accessed 12 October 2017).

Hartung, D. M. et al. (2018), “Industry Payments to Physician Specialists Who Prescribe Repository Corticotropin”, JAMA Network Open, 1(2), p. e180482, http://dx.doi.org/10.1001/jamanetworkopen.2018.0482.

Hartung, D. M. (2017), “Economics and Cost-Effectiveness of Multiple Sclerosis Therapies in the USA”, Neurotherapeutics, 14(4), pp. 1018–1026, http://dx.doi.org/10.1007/s13311-017-0566-3.

Hartung, D. M. et al. (2015), “The cost of multiple sclerosis drugs in the US and the pharmaceutical industry: Too big to fail?”, Neurology, 84(21), pp. 2185–2192, http://dx.doi.org/10.1212/WNL.0000000000001608.

Harvey, R. and Boer, R. de (2015), Growth in expenditure on high cost drugs in Australia, Parliamentary Library Research Paper, Parliament of Australia, Canberra.

Home, P. et al. (2014), “Insulin therapy in people with type 2 diabetes: Opportunities and challenges?”, Diabetes care, American Diabetes Association, 37(6), pp. 1499–508, http://dx.doi.org/10.2337/dc13-2743.

Howard, D. H. et al. (2015), “Pricing in the Market for Anticancer Drugs”, Journal of Economic Perspectives, 29(1)pp. 139–162.

Hutchings, A. et al. (2014), “Estimating the budget impact of orphan drugs in Sweden and France 2013–2020”, Orphanet Journal of Rare Diseases, 9(22), pp. 1–9, http://dx.doi.org/10.1186/1750-1172-9-22.

IMS Health (2017), Oncology spending as share of pharma spending by country 2015 | Statistic, Statista, https://www.statista.com/statistics/696055/oncology-spending-as-share-of-pharma-spending-by-country/ (accessed 21 November 2017).

IMS Institute for Healthcare Informatics (2016a), Comparison of hepatitis V treatment costs - Estimates of net prices and usage in U.S. and other major markets, IMS Institute for Healthcare Informatics, Parsippany.

IMS Institute for Healthcare Informatics (2016b), Delivering on the Potential of Biosimilar Medicines - The Role of Functioning Competitive Markets, IMS Institute for Healthcare Informatics, Parsippany.

IMS Institute for HealthCare Informatics (2013), Avoidable costs in US Healthcare - The USD 200 billion opportunity from using medicines more responsibly, IMS Institute for Healthcare Informatics, Parsippany.

Iuga, A. O. and McGuire, M. J. (2014), “Adherence and health care costs.”, Risk management and healthcare policy, Dove Press, 7, pp. 35–44, http://dx.doi.org/10.2147/RMHP.S19801.

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, http://dx.doi.org/10.1371/journal.pmed.1002032.

Jönsson, B. et al. (2016), “The cost and burden of cancer in the European Union 1995-2014”, European Journal of Cancer, 66, pp. 162–170, http://dx.doi.org/10.1016/j.ejca.2016.06.022.

Jørgensen, K. K. et al. (2017), “Switching from originator infliximab to biosimilar CT-P13 compared with maintained treatment with originator infliximab (NOR-SWITCH): a 52-week, randomised, double-blind, non-inferiority trial”, The Lancet, 389(10086), pp. 2304–2316, http://dx.doi.org/10.1016/S0140-6736(17)30068-5.

Kabiri, M. et al. (2014), “The Changing Burden of Hepatitis C Virus Infection in the United States: Model-Based Predictions’, Annals of Internal Medicine, 161(3), pp. 170-U120, http://dx.doi.org/10.7326/m14-0095.

Kaltenboeck, A. and Bach, P. B. (2017), A $475,000 price tag for a new cancer drug: crazy or meh?, Statnews, https://www.statnews.com/2017/08/31/475000-price-tag-new-cancer-drug-crazy-meh/ (accessed 12 July 2018). Kantarjian, H. et al. (2012), “Improved survival in chronic myeloid leukemia since the introduction of imatinib therapy: a single-institution historical experience”, Blood, American Society of Hematology, 119(9), pp. 1981–7, http://dx.doi.org/10.1182/blood-2011-08-358135. Kanters, T. A., Steenhoek, A. and Hakkaart, L. (2014), “Orphan drugs expenditure in the Netherlands in the period 2006–2012”, Orphanet Journal of Rare Diseases, 9(1), p. 154, http://dx.doi.org/10.1186/s13023-014-0154-0. Khan, R. and K. Socha-Dietrich (2018), "Investing in medication adherence improves health outcomes and health system efficiency: Adherence to medicines for diabetes, hypertension, and hyperlipidaemia", OECD Health Working Papers, No. 105, OECD Publishing, Paris, https://doi.org/10.1787/8178962c-en. Kim, B. J., Kim, J. H. and Kim, H. S. (2017), “Survival benefit of immune checkpoint inhibitors according to the histology in non-small-cell lung cancer: A meta-analysis and review”, Oncotarget, Impact Journals, LLC, 8(31), pp. 51779–51785, http://dx.doi.org/10.18632/oncotarget.17213. Kirby Institute (2017), Monitoring hepatitis C treatment uptake in Australia, Kirby Institute, Univeristy of South Wales Australia. Lacey, M. J. et al. (2014), Impact of Pharmaceutical Innovation in HIV/AIDS Treatment during the Highly Active Antiretroviral Therapy (HAART) Era in the US, 1987-2010: An Epidemiologic and Cost-impact Modeling Case Study, White Paper, Truven Health Analytics Lamprea, E. (2014), “Colombia’s Right to Health Litigation in a Context of Health Care Reform”, in Flood, C. M. and Gross, A. (eds.), The Right to Health at the Public/Private Divide. A Global Comparative Study, Cambridge University Press, Cambridge, United Kingdom. Leigh, S. and Granby, P. (2016), “A Tale of Two Thresholds: A Framework for Prioritization within the Cancer Drugs Fund”, Value in Health, 19, pp. 567–576, http://dx.doi.org/10.1016/j.jval.2016.02.016. Leopold, C., Chambers, J. D. and Wagner, A. K. (2016), “Thirty years of media coverage on high drug prices in the United States - A never-ending story or a time for change?”, Value in Health, 19(1), pp. 14–16, http://dx.doi.org/10.1016/j.jval.2015.10.008. Levinson, D. R. (2012), Least costly alternative policies: Impact on prostate cancer drugs under Medicare Part B, US Department of Health and Human Services, Office of Inspector General, Report OEI-12-12-00210, Washington DC. Lexchin, J. (2006), “Do manufacturers of brand-name drugs engage in price competition? An analysis of introductory prices”, Cmaj, 174(8), pp. 1120–1121, http://dx.doi.org/10.1503/cmaj.051687. López-Bastida, J. et al. (2016), “Social/economic costs and health-related quality of life in patients with rare diseases in Europe”, European Journal of Health Economics, 17, pp. 1-5, http://dx.doi.org/10.1007/s10198-016-0780-7. Mack, J. W. et al. (2015), “Patient beliefs that chemotherapy may be curative and care received at the end of life among patients with metastatic lung and colorectal cancer”, Cancer, 121(11), pp. 1891–1897, http://dx.doi.org/10.1002/cncr.29250. Mensah, G. A. et al. (2017), “Decline in Cardiovascular Mortality: Possible Causes and Implications”, Circulation Research, 120(2), pp. 366–380, http://dx.doi.org/10.1161/CIRCRESAHA.116.309115. Mentzakis, E., Stefanowska, P. and Hurley, J. (2011), “A discrete choice experiment investigating preferences for funding drugs used to treat orphan diseases: an exploratory study”, Health Economics, Policy and Law, 6(3), pp. 405–433, http://dx.doi.org/10.1017/S1744133110000344. Merck & Co. Inc. (2018), Annual report 2017 - Form 10-K, Annual Report Pursuant to Section 13 or 15(d), of the Securities Exchange Act of 1934 For the Fiscal Year Ended December 31, 2017, US Securities and Exchange Commission, Washington. Miller, R. G. et al. (2012), “Improvements in the life expectancy of type 1 diabetes: the Pittsburgh Epidemiology of Diabetes Complications study cohort.”, Diabetes, American Diabetes Association, 61(11), pp. 2987–92, http://dx.doi.org/10.2337/db11-1625. Ministère des affaires sociales et de la santé (2017), Accès universel aux traitements innovants contre l’hépatite C, http://solidarites-sante.gouv.fr/IMG/pdf/17_03_31_-_cp_concretisation_acces_universel_hepatite_c.pdf (accessed 31 October 2017). Moorkens, E. et al. (2016), “Overcoming Barriers to the Market Access of Biosimilars in the European Union: The Case of Biosimilar Monoclonal Antibodies”, Frontiers in Pharmacology Front. Pharmacol, 7(7), p. 193, http://dx.doi.org/10.3389/fphar.2016.00193. Morgan, S., Lopert, R. and Greyson, D. (2008), “Toward a definition of pharmaceutical innovation.”, Open medicine, 2(1), pp. e4-7. Morgan, S. G., Vogler, S. and Wagner, A. K. (2017), “Payers’ experiences with confidential pharmaceutical price discounts: A survey of public and statutory health systems in North America, Europe, and Australasia”, Health Policy. Elsevier Ireland Ltd., http://dx.doi.org/10.1016/j.healthpol.2017.02.002. Mueller, M. T. and Frenzel, A. (2015), “Competitive pricing within pharmaceutical classes: evidence on “follow-on” drugs in Germany 1993–2008”, The European Journal of Health Economics, 16(1), pp. 73–82, http://dx.doi.org/10.1007/s10198-013-0555-3. Murphy, S. M., Puwanant, A. and Griggs, R. C. (2012), “Unintended effects of orphan product designation for rare neurological diseases”, Annals of Neurology, 72(4), pp. 481–490, http://dx.doi.org/10.1002/ana.23672. National Academies of Sciences, Engineering and Medicine (2018), Making medicines affordable: A national imperative, The National Academies Press, Washington, DC, http://dx.doi.org/10.17226/24946. National Center for Health Statistics (2017), Health, United States, 2016: With Chartbook on Long-term Trends in Health, National Center for Health Statistics, Hyattsville, MD. New England Health Care Institute (2009), Thinking Outside the Pillbox A System-wide Approach to Improving Patient Medication Adherence for Chronic Disease, New England Health Care Institute Research Brief, https://www.nehi.net/writable/publication_files/file/pa_issue_brief_final.pdf NHS England (2016), Appraisal and Funding of Cancer Drugs from July 2016, including the new Cancer Drugs Fund. A new deal for patients , taxpayers and industry, https://www.england.nhs.uk/wp-content/uploads/2013/04/cdf-sop.pdf. Niëns, L. M. and Brouwer, W. B. F. (2013), “Measuring the affordability of medicines: Importance and challenges”, Health Policy. Elsevier Ireland Ltd, 112(1–2), pp. 45–52, http://dx.doi.org/10.1016/j.healthpol.2013.05.018. OECD (2018), OECD Health Statistics (database), http://dx.doi.org/10.1787/health-data-en. OECD (2017a), Health at a Glance 2017: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/health_glance-2017-en. OECD (2017b), New Health Technologies: Managing Access, Value and Sustainability, OECD Publishing, Paris, https://doi.org/10.1787/9789264266438-en. OECD (2017c), New Health Technologies: Managing Access, Value and Sustainability, OECD Publishing, Paris, https://doi.org/10.1787/9789264266438-en. OECD (2017d), OECD Health Statistics, OECD, Paris, http://dx.doi.org/10.1787/health-data-en. OECD (2017e), “Reducing ineffective health care spending on pharmaceuticals”, in Tackling Wasteful Spending on Health, OECD Publishing, Paris, pp. 161–191, https://doi.org/10.1787/9789264266414-7-en. OECD (2017f), “Sharp increases in drug prices, query launched by the OECD Health Division”, Vienna: Pharmaceutical Pricing and Reimbursement Information (PPRI) network. OECD (2015a), Cardiovascular Disease and Diabetes: Policies for Better Health and Quality of Care, OECD Health Policy Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264233010-en. OECD (2015b), Fiscal Sustainability of Health Systems: Bridging Health and Finance Perspectives, OECD Publishing, Paris, http://dx.doi.org/http://dx.doi.org/10.1787/9789264233386-en. OECD (2008), Pharmaceutical Pricing Policies in a Global Market, OECD Health Policy Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264044159-en. Oliver, A. and Mossialos, E. (2004), “Equity of access to health care: outlining the foundations for action”, Journal of Epidemiology & Community Health, 58(8), pp. 655–658, http://dx.doi.org/10.1136/jech.2003.017731. Onakpoya, I. J. et al. (2015), “Effectiveness, safety and costs of orphan drugs: an evidence-based review”, BMJ Open, 5(6), pp. e007199–e007199, http://dx.doi.org/10.1136/bmjopen-2014-007199. Ozierański, P., McKee, M. and King, L. (2012), “Pharmaceutical lobbying under postcommunism: universal or country-specific methods of securing state drug reimbursement in Poland?”, Health Economics, Policy and Law, 7(2), pp. 175–195, http://dx.doi.org/10.1017/S1744133111000168. Palnoch, D. (2016), “Cancer Drugs: NICE and the CDF”, Department of Health, London. Panteli, D. et al. (2016), Pharmaceutical regulation in 15 European countries, WHO Regional Office for Europe, Copenhagen. Paris, V. and Belloni, A. (2014), “Value in Pharmaceutical Pricing: Country Profile Australia”, OECD, Paris. Paris, V. and Belloni, A. (2013), “Value in Pharmaceutical Pricing”, OECD Health Working Papers, Paris, http://dx.doi.org/http://dx.doi.org/10.1787/5k43jc9v6knx-en. PBAC (2015), Public Summary Document - March 2015 PBAC Meeting – Sofosbuvir, Pharmaceutical Benefits Advisory Committee (Australia), https://www.pbs.gov.au/industry/listing/elements/pbac-meetings/psd/2015-03/Files/ledipasvir-sofosbuvir-psd-march-2015.pdf . Pharmaceutical-Technology (2016), The world’s most sold cancer drugs in 2015, pharmaceutical-technology.com, https://www.pharmaceutical-technology.com/features/featurethe-worlds-most-sold-cancer-drugs-in-2015-4852126/ (accessed 26 November 2017). PMPRB (2017), NPDUIS CompassRx, 3rd edition: Annual Public Drug Plan Expenditure Report 2015/16, Patented Medicine Prices Review Board, Ottawa. PMPRB (2016), Market Intelligence Report: Biologic Response Modifier Agents 2015, Patented Medicine Prices Review Board, Ottawa. Pollack, A. (2015), “Drug Goes From$13.50 a Tablet to $750, Overnight”, The New York Times, https://www.nytimes.com/2015/09/21/business/a-huge-overnight-increase-in-a-drugs-price-raises-protests.html (accessed on 2 November 2018). Polton, D. (2015), Rapport sur la réforme des modalités d’évaluation des médicaments, Paris. https://solidarites-sante.gouv.fr/IMG/pdf/rapport_polton_-_evaluation_medicaments.pdf Pophali, P. and Patnaik, M. (2016), “The Role of New Tyrosine Kinase Inhibitors in Chronic Myeloid Leukemia Priyanka”, The Cancer Journal, 22(1), pp. 40–50, http://dx.doi.org/10.1016/j.bbamem.2015.02.010.Cationic. Pulido, T. et al. (2013), “Macitentan and Morbidity and Mortality in Pulmonary Arterial Hypertension”, New England Journal of Medicine, Massachusetts Medical Society, 369(9), pp. 809–818, http://dx.doi.org/10.1056/NEJMoa1213917. QuintilesIMS (2017), Medicines use and spending in the U.S. - A Review of 2016 and Outlook to 2021, QuintilesIMS Institute, Parsippany. QuintilesIMS Institute (2017a), Global Oncology Trends 2017 - Advances, Complexity and Cost, QuintilesIMS Institute, Parsippany. QuintilesIMS Institute (2017b), Orphan Drugs in the United States - Providing context for use and cost, QuintilesIMS Institute, Parsippany. QuintilesIMS Institute (2016), Outlook for Global Medicines through 2021 - Balancing costs and value, QuintilesIMS Institute, Parsippany. Quittner, A. et al. (2015), “Effect of ivacaftor treatment in patients with cystic fibrosis and the G551D-CFTR mutation: patient-reported outcomes in the STRIVE randomized, controlled trial”, Health and Quality of Life Outcomes, BioMed Central, 13(1), p. 93, http://dx.doi.org/10.1186/s12955-015-0293-6. RAPS (2016), European Regulatory Roundup: UK Advances Bill to Stop Price Spikes on Generic Drugs, Regulatory Affairs Professionals Society, http://www.raps.org/Regulatory-Focus/News/2016/10/27/26080/European-Regulatory-Roundup-UK-Advances-Bill-to-Stop-Price-Spikes-on-Generic-Drugs-27-October-2016/. Rémuzat, C. et al. (2017), “Key drivers for market penetration of biosimilars in Europe”, Journal of Market Access & Health Policy, 5(1), p. 1272308, http://dx.doi.org/10.1080/20016689.2016.1272308. Robertson, J. et al. (2013), “Medicines and the media: news reports of medicines recommended for government reimbursement in Australia.”, BMC public health. BMC Public Health, 13(1), p. 489, http://dx.doi.org/10.1186/1471-2458-13-489. Roebuck, M. C. et al. (2011), “Medication Adherence Leads To Lower Health Care Use And Costs Despite Increased Drug Spending”, Health Affairs, 30(1), pp. 91–99, http://dx.doi.org/10.1377/hlthaff.2009.1087. Ross, J.S. and Kesselheim A. S. (2013), “Prescription-Drug Coupons — No Such Thing as a Free Lunch”, New England Journal of Medicine, 369(13), pp. 1188-1189. Sagonowsky, E. (2017), “AbbVie’s new pan-genotypic hepatitis C drug Mavyret deeply underprices the competition FiercePharma, https://www.fiercepharma.com/pharma/abbvie-s-new-pan-genotypic-hep-c-drug-mavyret-undercuts-competition (accessed 21 November 2017). Sawhney, R., Gordian, M. and Behnke, N. (2016), Pharma: The Market Access Dilemma, Bain & Company. Schrag, D. (2004), “The Price Tag on Progress — Chemotherapy for Colorectal Cancer”, New England Journal of Medicine, Massachusetts Medical Society, 351(4), pp. 317–319, http://dx.doi.org/10.1056/NEJMp048143. Schuller, Y., Hollak, C. E. M. and Biegstraaten, M. (2015), “The quality of economic evaluations of ultra-orphan drugs in Europe – a systematic review”, Orphanet Journal of Rare Diseases, 10(1), p. 92, http://dx.doi.org/10.1186/s13023-015-0305-y. Setoguchi, S. et al. (2008), “Improvements in Long-Term Mortality After Myocardial Infarction and Increased Use of Cardiovascular Drugs After Discharge - A 10-Year Trend Analysis”, Journal of the American College of Cardiology, 51(13), pp. 1247–1254, http://dx.doi.org/10.1016/j.jacc.2007.10.063. Shepherd, J. et al. (2004), “Pegylated interferon α-2a and -2b in combination with ribavirin in the treatment of chronic hepatitis C: a systematic review and economic evaluation”, Health Technology Assessment, 8(39), p. 500, http://dx.doi.org/10.1007/978-3-319-52311-8_1. Simon Kucher (2016), Payers’ price & market access policies supporting a sustainable biosimilar medicines market - final report, Simon, Kucher & Partners, Bonn. Sood, N. et al. (2017), The flow of money through the pharmaceutical distribution system, Schaeffer Center for Health Policy & Economics, University of Southern California, Los Angeles. Sun, E. et al. (2010), “The Contributions of Improved Therapy and Earlier Detection to Cancer Survival Gains , The Contributions of Improved Therapy and Earlier Detection to Cancer Survival Gains”, Forum for Health Economics & Policy, 13(2), pp. 1988–2000. Tefferi, A. et al. (2015), “In Support of a Patient-Driven Initiative and Petition to Lower the High Price of Cancer Drugs”, Mayo Clinic Proceedings, 90(8), pp. 996–1000, http://dx.doi.org/10.1016/j.mayocp.2015.06.001. The Alliance of Community Health Plans (2015), “The Spike in Drug Costs: Diabete”, Infographic Series, The Alliance of Community Health Plans, https://www.achp.org/wp-content/uploads//Diabetes_FINAL_Revised-12.7.15.pdf, accessed on 31 October 2018. Timmins, N., Rawlins, M. and Appleby, J. (2016), A terrible beauty - A short history of NICE, http://www.idsihealth.org/wp-content/uploads/2016/02/A-TERRIBLE-BEAUTY_resize.pdf Uhl, K. (2017), 2016: A Record-setting Year for Generic Drugs, FDA Voice, https://blogs.fda.gov/fdavoice/index.php/2017/02/2016-a-record-setting-year-for-generic-drugs/. Unal, B. et al. (2005), “Life-years gained from modern cardiological treatments and population risk factor changes in England and Wales, 1981-2000.”, American journal of public health, American Public Health Association, 95(1), pp. 103–8, http://dx.doi.org/10.2105/AJPH.2003.029579. Unal, B., Critchley, J. A. and Capewell, S. (2004), “Explaining the Decline in Coronary Heart Disease Mortality in England and Wales between 1981 and 2000”, Circulation, 109(9), pp. 1101–1107, http://dx.doi.org/10.1161/01.CIR.0000118498.35499.B2. UNDP (2003), Indicators for Monitoring the Millennium Development Goals: Definitions, Rationale, Concepts and Sources, UNDP, New York City, NY. UNITAID and WHO (2017), Technology and market landscape - Hepatitis C medicines - August 2017. US Court of Appeals- District of Columbia Circuit (2009), Hays v. Sebelius, No. 08-5508, http://caselaw.findlaw.com/us-dc-circuit/1498630.html (accessed 9 February 2018). van den Brink (2014), Reimbursement of orphan drugs: the Pompe and Fabry case in the Netherlands, Orphanet Journal of Rare Diseases, 9 (Suppl 1):O17. Vargas-Pelaez C.M. et al. (2014), Right to health, essential medicines, and lawsuits for access to medicinese: A scoping study, Social Science & Medicine, Vol. 121, pp. 48-55, Ventola, C. L. (2015), “The antibiotic resistance crisis: part 1: causes and threats.”, P & T 40(4), pp. 277–83. Visante (2011), How Copay Coupons Could Raise Prescription Drug Costs By$32 Billion Over the Next Decade, report prepared for the Pharmaceutical Care Management Association, Visante.

Visante (2017), Increasing Prices Set by Drugmakers Not Correlated With Rebates, an analysis prepared for the Pharmaceutical Care Management Association, Visante

Vogler, S. et al. (2015b), Study on enhanced cross-country coordination in the area of pharmaceutical product pricing - Final Report, http://dx.doi.org/10.2875/631265.

Vuorenkoski, L., Toiviainen, H. and Hemminki, E. (2003), “Drug reimbursement in Finland - A case of explicit prioritising in special categories”, Health Policy, 66(2), pp. 169–177, http://dx.doi.org/10.1016/S0168-8510(03)00042-3.

Walker, J. (2016), Drugmakers Point Finger at Middlemen for Rising Drug Prices - WSJ, The Wall Street Journal, https://www.wsj.com/articles/drugmakers-point-finger-at-middlemen-for-rising-drug-prices-1475443336 (accessed 17 November 2017).

Weeks, J. C. et al. (2012), “Expectations about Effects of Chemotherapy for Advanced Cancer”, New England Journal of Medicine, 367(17), pp. 1616–1625, http://dx.doi.org/10.1056/NEJMoa1204410.

Whiting, P. et al. (2014), “Ivacaftor for the treatment of patients with cystic fibrosis and the G551D mutation: a systematic review and cost-effectiveness analysis”, Health Technology Assessment 18(18), http://dx.doi.org/10.3310/hta18180.

Woessner, D. W., Lim, C. S. and Deininger, M. W. (2011), “Development of an effective therapy for chronic myelogenous leukemia.”, Cancer journal (Sudbury, Mass.) 17(6), pp. 477–86, http://dx.doi.org/10.1097/PPO.0b013e318237e5b7.

Zamora, B. et al. (2017), “Comparing Access to Orphan Medicinal Products ( OMPs ), in the United Kingdom and other European countries Comparing Access to Orphan Medicinal Products ( OMPs ) in the United Kingdom and other European countries”, OHE Consulting Report, (March), pp. 1–37.

Zhang, Y., Chantel Hueser, H. and Hernandez, I. (2017), “Comparing the Approval and Coverage Decisions of New Oncology Drugs in the United States and Other Selected Countries”, Journal of Managed Care & Specialty Pharmacy 23(2), pp. 247–254.

Annex 1.A. Supplementary table
Annex Table 1.A.1. Expenditure on prescribed medicines by financing schemes, OECD countries, 2016

Government schemes

Compulsory contributory health insurance schemes

Voluntary health care payment schemes

Household out-of-pocket payments

Australia*

75.9

..

1.0

23.2

Austria

..

88.1

0.3

11.6

Belgium

..

78.4

0.2

21.3

38.7

3.9

35.7

21.7

Czech Republic

..

75.6

..

24.4

Denmark

62.6

..

8.4

29.0

Estonia

0.2

65.5

0.4

33.8

Finland

..

66.1

0.5

33.4

Germany

5.1

88.7

0.4

5.7

Hungary

1.9

47.4

4.1

46.7

Iceland

63.3

..

2.8

33.9

Japan*

9.2

75.7

1.1

13.9

Korea

7.8

58.3

1.1

32.8

Latvia

51.6

..

0.2

48.3

Luxembourg

..

80.1

8.0

11.9

Mexico

..

..

4.0

96.0

Norway

..

54.1

..

45.9

Poland

1.5

68.3

0.4

29.8

Slovak Republic

..

71.3

..

28.7

Slovenia

..

64.1

34.3

1.6

Spain

84.7

4.3

..

11.0

Sweden

72.1

..

..

27.9

Switzerland

2.4

64.1

2.5

31.0

United States**

13.6

72.7

13.7

Note: * 2015; ** In the United States, payments from private health insurance cannot be split between compulsory and voluntary PHI.

Source: OECD (2018), OECD Health Statistics (database), http://dx.doi.org/10.1787/health-data-en.

## Notes

← 1. Overall cancer survival gains are higher than survival gains by individual cancer sites because it includes other types of cancers.

← 2. Pharmaceutical spending, as reported in the System of Health Accounts and OECD statistics, is net of discounts and rebates.

← 3. Although Quintiles IMS refers to “spending”, the Institute publishes estimates based on sales from wholesalers to their customers, including pharmacies and hospitals. These sales are in general reported at the invoice prices wholesalers charge to their customers. In some countries, these prices are exclusive of discounts and rebates paid to governments, private insurers or the specific purchasers, while in others, such discounts and rebates do not occur [see (QuintilesIMS Institute, 2016b); p.12). In this report, to avoid any confusion with OECD data on “pharmaceutical spending”, which includes all distribution margins and VAT where relevant but excludes discounts or rebates, data from QuintilesIMS, now IQVIA, are referred to as “pharmaceutical sales”. These, however, do not coincide with industry’s revenues, which are lower, due to discounts and rebates that are not accounted for.

← 4. See Note 2.

← 5. The period of observation predates the introduction of the new legislation in Germany, in 2011, which may have changed how markets function.

← 6. Data on Mexico do not include public spending and are thus not comparable.

← 7. Tiered Pricing is a means of supplying products at different price points in different markets. In this context if refers to different prices in different countries.

← 8. All new patented medicines introduced in the German market are subject to the evaluation and price negotiation process, except those with annual expenditure for statutory health insurance (SHI) below EUR 1 million. For orphan drugs, additional therapeutic benefit is assumed by virtue of marketing authorisation without reference to an appropriate comparator in Germany for as long as annual SHI expenditure for the drug remains below EUR 50 million.

← 9. At the time of the an OECD survey launched in 2014-2015, only five countries including Hungary, the Republic of Korea, Poland, the Slovak Republic and the United Kingdom, had published an ICER threshold range; 2 to 3 times GDP per capita in Hungary, GDP per capita in the Republic of Korea (may vary by disease), 3 times GDP per capita in Poland, EUR 18 000 to 26 500 in the Slovak Republic and GBP 20 000 to 30 000 in the United Kingdom.

← 10. Orphan drugs refer to medicines developed for rare conditions. Countries use different thresholds to consider if a disease is rare: “rare conditions” are those that affect less than 200 000 people in the United States (or about 1 in 1 500 people), less than 1 in 2 000 people in the European Union and less than 1 in 2500 people in Japan.

← 11. The estimates of population covered by type of coverage are not mutually exclusive; people can be covered by more than one type of health insurance during the year (Barnett and Berchick, 2017).

← 12. These percentages differ from those presented in Figure 1.8 because the latter takes into account OTC drugs and medical non-durables.

← 13. PBMs administer drug benefits for commercial health plans, self-insured employer plans, Medicare Part D plans, the Federal Employees Health Benefits Program, and state government employee plans.

← 14. Discounts refer to price reductions while rebates refer to funds returned to the payer after the transaction.

← 15. The unit rebate is the highest of (1) the Average Manufacturer Price (AMP) minus Best Price or (2) 23.1% of AMP. The AMP is defined as the average price paid to the manufacturer by wholesalers and retail pharmacies that purchase the drug directly from the manufacturer.

← 16. This is a sub-sample of the 449 medicines approved in the United States with at least one orphan indications, which were still marketed in 2016 and for which price information was available.

← 17. In the United States, a brand-name medicine can have both orphan and non-orphan indications. In Europe, a same product has to be marketed with a different brand-name for non-orphan indications.

← 18. Prices reported in this report are list prices, estimated from the MIDAS database, in retail and hospital market sectors. They do not account for potential confidential rebates.

← 19. This analysis covers the US non-Medicaid markets for these self-administered products.

← 20. This share only takes into account spending for orphan indications for drugs which have both orphan and non-orphan indications.

← 21. During the same period (1960-2002), the general inflation in the United States was +504.4% (a 6-fold increase) (https://data.bls.gov/cgi-bin/cpicalc.pl).

← 22. Prices presented in current Euros in the original study have been adjusted for inflation using the Consumer price index published by INSEE.

← 23. Many countries do not publish cost-effectiveness thresholds (Auraaen et al., 2016).

← 24. Grades of Recommendation, Assessment, Development and Evaluation (GRADE) criteria,18 which assesses the five domains: study design; consistency of evidence; directness of the evidence; precision; and reporting biases

← 25. The Policy and Regulatory Report (PaRR) competition database is a subscription-only proprietary tool that tracks competition developments from almost 200 agencies worldwide. The parameters for this search were pharmaceutical cases initiated between 1 January 2007 and 28 February 2017 in OECD member countries. The review of cases focused on instances of sudden price increases of patented or generic pharmaceuticals. Cases concerning frivolous patent lawsuits, so-called product hopping, pay-for-delay agreements, or other barriers to market entry of generics were not included. A companion document will be made available on the OECD website and is available upon request.

← 26. These are off-patent products that are only available from a single manufacturer.

← 27. Where “availability” is defined as the ability to prescribe the medicine and “access” is defined as full or partial coverage by a public payer.