M.L. (Marion) Hof January 20, 2020
The effect of pricing strategies on drug prices
The difference between value-based and cost-based pricing
The effect of pricing strategies on drug prices
The difference between value-based and cost-based pricing
University of Groningen Faculty of Economics and Business
MSc BA Health and Management accounting & controlling
January 20, 2020
Marion L. Hof m.l.hof.1@student.rug.nl
Thesis supervisors:
dr. K. Linke dr. O.P. Roemeling
Word count: 15,390
Abstract
Background: Healthcare costs are increasing, including drug expenditures. The government introduced market regulation system, which include patent protection and price controls. The currently used price controls are based on a value-based pricing policy.
Objective: This study investigates the effect of different pricing strategies on drug prices. The results of this study will be evaluated via a stakeholder and agency theory lens.
Method: Both a cost-based and a value-based model were built, to compare the results of these two pricing strategies on drug prices. Two sample drugs were selected: a common drug (sacubitril/valsartan) and an orphan drug (alfa alglucosidase).
Results: The results of this study show, that the market price of sacubitril/valsartan (€2.54) lies between the cost-based price (€2.30) and the value-based price (€3.31). Based on the cost-based model, the cost-based price of sacubitril/valsartan consists largely out of out-of-pocket failure costs (46%) and costs of capital (29%). The value-based price of sacubitril/valsartan is most sensitive to the survival rate and the quality of life of patients treated with sacubitril/valsartan. The market price of alfa alglucosidase (€575.25) is higher than both the cost-based price (€192.30) and the value-based price (€128.83). The cost-based price of alfa alglucosidase consists mainly out of production costs (59%), due to the complex production process with recombinant DNA technology of alfa alglucosidase.
The effect of alfa alglucosidase on the quality of life has the highest impact on the value-based price followed by the other costs related to the treatment with alfa alglucosidase.
Conclusion: The current market price of both the common drug and the orphan drug lies above the cost-based price, pharmaceutical companies earn a profit margin of more than 15%. The value-based price of the common drug is much higher than the current market price. Incentives which stimulate value-based pricing could increase drug prices. The value of the orphan drug is lower than the cost- based price. Orphan drugs will not be brought to the market against its value-based price.
Many countries are leaning more towards a system which stimulates value-based pricing. They introduce incentive systems which stimulate value-based pricing to limit healthcare costs and to improve the value for money in healthcare. As appears from this study, the use of value-based pricing strategies will probably increase drug prices and will therefore have the opposite effect.
Keywords: value-based pricing, cost-based pricing, drug prices, orphan drugs, drug development, agency theory, stakeholder theory.
Content
Abstract ... 2
1. Introduction ... 1
2. Theoretical Background ... 4
2.1 Stakeholders ... 4
2.2 Agency theory ... 5
2.3 Economic market for medical care ... 6
2.3.1 Mandatory health insurance ... 6
2.3.2 Patent protection ... 7
2.3.3 Price controls ... 7
2.4 Drug development process ... 8
2.5 Pricing mechanism ... 9
2.5.1. Value-based pricing ... 10
2.5.3. Cost-based pricing ... 12
2.6 Theoretical framework ... 13
3. Methodology ... 15
3.1 Sample selection... 15
3.2 Cost-based model ... 15
3.2.1 Time Horizon ... 16
3.2.2 Input parameters ... 16
3.3 Value based model ... 20
Common drug ... 20
Orphan drug ... 20
3.4 Value added tax ... 23
3.5 Analyses ... 23
4. Results ... 23
4.1 Common drug ... 23
4.1.1 Value-based price ... 24
4.1.2 Cost-based price ... 27
4.1.3 Comparison ... 29
4.2 Orphan drug ... 30
4.2.1 Value-based price ... 30
4.2.2 Cost-based price ... 33
4.2.3 Comparison different prices ... 35
5. Discussion and Conclusion ... 37
5.1 Common drug (sacubitril/valsartan) ... 37
5.2 Orphan drug (alfa alglucosidase) ... 37
5.3 Results compared to international market prices ... 38
5.4 Study results via a stakeholder and agency theory lens ... 39
5.5 Comparison with the original study ... 41
5.6 Limitations ... 42
5.7 Future research ... 43
5.8 Conclusion ... 43
References ... 45
Appendix E – Limitations of the study ... 54
1
1. Introduction
In the western society, there is an ongoing debate about rising healthcare costs and high drug prices (Lonkhuyzen & van Steenbergen, 2019). Yet, the number of expensive drugs which are available on the market is increasing. In 2017, € 2,08 billion were spent on expensive drugs in the Netherlands (Nederlandse Zorgautoriteit, 2019). This is an increase of 9% compared to the year before. Obviously, the development of new drugs which can improve the quality of life is of major importance, however, healthcare budgets are limited, and the distribution of these funds need to be considered carefully.
Drug development is a long-term and risky process which needs to be compensated in order to attract investors (Keyhani, Diener-West, & Powe, 2005). Generally, research and development (R&D) of new medication is considered an expensive endeavour. However, there is controversy about how expensive it actually is (Gupta Strategists, 2019). In this study, we seek to explore whether pharmaceutical companies are either achieving high profit margins for their stakeholders, or that these high drug prices are necessary to cover the costs of drug development.
The market for medical care, including the pharmaceutical market, differs from other markets. Mainly because health is different from other goods, and therefore seen as a ”special economic problem”
(Arrow, 1963; Bhattacharya, Hyde, & Tu, 2013). There are many different stakeholders involved in the pharmaceutical market, among them the government, health insurers, patients and hospitals.
Pharmaceutical companies often receive active, informed and critical stakeholder attention, because access to health care is assumed to be a fundamental human right (O'Riordan & Fairbrass, 2008;
Belinfante & de Reede, 2012). Besides the society, the government and patients rely on healthcare providers and the pharmaceutical industry to deliver health care. This relationship could be typified as an agency relationship, because some decision making and authority on the provision of healthcare services is delegated to healthcare providers and pharmaceutical companies (Jensen & Meckling, 1976). Due to the importance of their services, healthcare providers and pharmaceutical companies have a powerful position. They could (mis)use this powerful position to act in their own self-interest.
In order to stimulate healthcare providers and pharmaceutical companies to act in the interest of their stakeholders (e.g. patients, the society and the government), market regulation systems were introduced. In the current system, pharmaceutical companies can receive patent protection and drug prices are regulated by price controls (Bhattacharya, Hyde, & Tu, 2013). Patent protections are issued to stimulate innovation, and will provide a pharmaceutical company with the temporary right to exclusively produce a drug (Bhattacharya, Hyde, & Tu, 2013). In order to prevent pharmaceutical companies from charging very high prices, the government has introduced price controls. These price controls are used, to decide whether a drug will be reimbursed (Zorginstituut Nederland, 2014). Price
2 controls are used to achieve an optimal distribution of the limited healthcare funds, in order to gain the most health benefits for the society as a whole. Rules and regulations regarding the price of drugs play a central role in government policies. The effect of these rules and regulations on the market price of drugs is still underexplored.
The price controls in the Netherlands are currently based on a value-based pricing policy (Gandjour, 2015). In short, value-based pricing involves a valuation of the added value of an intervention compared with existing alternatives (Drummond, Sculpher, Claxton, Stoddart, & Torrance, 2015). A drug will only be reimbursed if it has clinical benefits and is regarded to be cost-effective. Currently, experiments are underway with a different pricing mechanism, called pay-for-performance (College voor Zorgverzekeringen, 2012). Based on this method a drug will only be reimbursed if it achieves the predetermined performance objectives (Cashin, Chi, Smith, Borowitz, & Thomson, 2014). The performance objectives are disease and intervention specific and need to be built on clear concepts concerning: a definition of the objectives it seeks to promote; quality metrics that influence the payment; rules regarding the provision of information; and arrangements which ensure the intended working of the system. The pay-for-performance pricing mechanism is strongly related to the value- based pricing mechanism, because the performance objectives need to be valued (Gandjour, 2015).
Currently, drugs are evaluated based on the value they deliver and independently of the costs they inquire.
Some scholars have their reservations about the suitability of the currently used value-based pricing mechanism (Neyt, 2018; Danzon, 2018; Pauly, 2017). One of the reasons is that it is not possible to calculate the cost-effectiveness for all interventions at the same time, and other elements outside of health maximisation are not included (Neyt, 2018). In addition to that, it could be doubted whether this economic evaluation method is suitable for all drug types, especially for orphan drugs (Drummond, 2008). These orphan drugs are used for rare diseases, with a small patient population. Therefore, the R&D costs need to be recovered by charging a much higher cost per patient than for drugs with large sales potential. Because of the high price involved, these drugs could often not be considered as cost- effective. But these drugs might have a high social value, due to the serious condition of patients, inadequacy of alternative treatments and perceived community needs. Drummond (2008) argues that the social value is not represented in the value-based price. Another reason is that there is not a single value-based price, the value based price depends on the assumed value of a quality-adjusted life-year (QALY) (Pauly, 2017). Besides there are some concerns about the affordability within the short-term budget (Danzon, 2018). If a drug cures a chronic disease which otherwise would entail the necessary annually costs, the value-based price would be quite high. Next to that, drugs will become unaffordable if we only consider the willingness-to-pay for a life without considering the limited budgetary resources (Neyt, 2018).
3 This study will investigate the effect of pricing mechanism on drug prices. The currently used value- based pricing mechanism will be compared with a new method which is not widely used in practice:
cost-based-pricing. This method adds a certain profit element to the manufacturing and development costs (Edmonds, Edmonds, Tsay, & Olds, 2014). The Dutch minister of Health, Welfare and Sport, Bruno Bruins, is lobbying in Europe for the introduction of such a pricing mechanism for orphan drugs (R.Cats, 2019). Minister Bruno Bruins believes that the introduction of such a pricing mechanism could help to keep those drugs affordable. Pay-for-performance is not included in this study, since there is much overlap between pay-for-performance and value-based pricing. Besides there is no standard measure for performance, which makes it difficult to compare. Previous studies only focused on the suitability of the value-based pricing mechanism. These studies did not compare the effects of different pricing mechanisms. As such, this study aims to explore what the influence of a different pricing mechanism will be on drug prices. The following research question was addressed in this study.
“How do different pricing strategies influence drug prices, and what are the main implications of each strategy?”
This thesis has the following structure. In the second chapter, relevant literature will be discussed in order to provide a theoretical background. The concepts of stakeholder theory and agency theory will be discussed to provide a theoretical lens for this study, which could be helpful to explain the findings of this study. Besides the concepts of market regulations and pricing strategies will be discussed. The methodology of this study will be explained in the third chapter and the results will be shown in the fourth chapter. Finally, the discussion, limitations, future research directions and the conclusion are presented in chapter six.
4
2. Theoretical Background
In this chapter, a literature review is provided. First, the concepts of stakeholder and agency theory will be discussed which will be used as a theoretical lens to interpret the findings of this study. Then the economic functioning of the market for medical care will be discussed, followed by a brief explanation of the drug development process. After that the pricing strategies which are central to this research will be discussed. In the final part the conceptual framework of this study will be explained.
2.1 Stakeholders
The modern stakeholder theory is based on the business science literature of Freeman (1984).
Freeman defined stakeholders as “any group or individual who can affect or is affected by the realization of the health innovation”. The stakeholder theory broadens the vision of managements role and responsibilities. Managements role is no longer only focused on the profit maximization for stockholders. The stakeholder theory acknowledges the responsibility of management towards other interest groups, who are affected by or can effect management decisions (Mitchell, Agle, & Wood, 1997).
Over the years a lot of research has been conducted into the field of stakeholder theory (Mainardes, Alves, & Raposo, 2011; Rowley, 1997; Pouloudi & Curry, 2016). Managers have to identify the stakeholders which they should take into account. Many different scholars have introduced frameworks which could help to identify and analyse stakeholders regarding different characteristics, for example: power, interest, legitimacy or engagement. (Savage, Nix, & Whitehead, 1991; Pouloudi &
Curry, 2016; Van Offenbeek, Boonstra, & Seo, 2013; Mitchell, Agle, & Wood, 1997).
There are many different stakeholders involved in the pharmaceutical market, among them the government, health insurers, patients, hospitals, society, pharmaceutical companies and their stockholders. Pharmaceutical companies are facing active, informed and critical stakeholder attention (O'Riordan & Fairbrass, 2008). They receive this amount of stakeholder attention due to the fact that many stakeholders who can affect or are affected by the pharmaceutical industry assume the access to health care as a fundamental human right (Belinfante & de Reede, 2012). Various stakeholder groups have different interests and also within the stakeholder group the interests are not always aligned. Shareholders would like to maximize their profit. Contrary, the society and the government, would like to keep the costs for pharmaceuticals as low as possible to ensure that the health costs remain affordable. At the same time, they want to stimulate the pharmaceutical industry to develop new drugs, which could be in favour of human health status and wellbeing.
Many different stakeholders have influence on the availability, development and price of drugs. This study will only focus on two powerful large stakeholders: pharmaceutical companies and the government. Pharmaceutical companies have the power to offer a drug at the market for a certain
5 price. They receive this power, because they have a valuable good in their hands; the society relies on them to develop new drugs which could increase the health benefits of the society as a whole. The government has the power to set rules and regulations to intervene in the pharmaceutical market.
These government interventions could have a major impact on the behaviour of pharmaceutical companies; they could influence both the availability and the price of drugs (Puig-Junoy, 2010; Kyle, 2007). This study will investigate how the current market price of drugs is related to different pricing strategies.
2.2 Agency theory
Agency theory is focussed on the relationship between two parties which rely on each other; the principal engages the agent to perform some service on their behalf, in which some decision making and authority is delegated to the agent (Jensen & Meckling, 1976). This delegation of authority is necessary to complete the task. Problems may arise, whenever the two parties have conflicting goals and the agent is not completely acting in the best interest of the principal (Mahaney & Lederer, 2003).
The agent could take an advantage of the information asymmetry (e.g. about quality) between the agent and the principal (Maestrini, Luzzini, Caniato, & Ronchi, 2018).
The pharmaceutical company is generally assumed to be well-informed while the government is less informed (Blomqvist & Léger, 2005). It is very difficult for the government to determine whether the pharmaceutical company has acted in its best interest. The information asymmetry and the conflict of interest may cause several problems in agency relationships. Two types of problems could occur:
adverse selection and moral hazard (Eisenhardt, 1989). Adverse selection appears when the agent concludes a contract based on wrong or incomplete information. Moral hazard arises when the principal only has information about the outcomes but has no information about the agent’s actions.
These problems could be reduced by the use of financial incentives, which could align the interest of the principal and the agent. On behalf of Shavell (1979), financial incentives could improve the quality and efficiency of healthcare.
This study could give some insight into the agency relationship between the pharmaceutical company and the government. If the current market price differs a lot from the cost-based price, this might be caused by information asymmetry and the effects of adverse selection and moral hazard. The government has incomplete information about the manufacturing and development cost of a drug.
This makes it difficult to indicate whether the pharmaceutical company has acted in the interest of the government or acted in their own self-interest (profit maximization). If moral hazard and adverse selection seem to be present at the pharmaceutical market, it could be wondered whether the current incentive systems stimulate the desired behaviour; good quality drugs for a low price and investment in R&D.
6
2.3 Economic market for medical care
Basic economic theories postulate that any competitive market will produce an efficient outcome where it is not possible to make anyone better off without making someone else worse off, if there are no externalities and information asymmetry (Bhattacharya, Hyde, & Tu, 2013). In 1963 Kenneth Arrow argued that this basic economic theory was not applicable to the healthcare market, because health is different from other goods and services and is from major importance to people. He defines health as a “special economic problem”, which involves a lot of uncertainty. The demand for healthcare is highly uncertain, because people don’t know if and when they will become ill. Besides there are externalities, people’s health choices can affect each other due to the transferability of various diseases. The Dutch government also emphasises the importance of people’s health and the equal access to healthcare. Access to healthcare is a fundamental social right in the Netherlands (Belinfante
& de Reede, 2012). Based on the beliefs of the western world, government bodies have the task to make provisions that promote a society in which citizens are enabled to build a dignified existence. In order to guarantee the equal access to healthcare, the government stepped in to regulate the market.
The Dutch government introduced mandatory health insurance, price control mechanisms and pharmaceutical companies can receive patent protection (Wammes, Jeurissen, & Westert, 2015).
These three types of government interventions will be discussed below, because they influence the functioning of the pharmaceutical mark and drug prices.
2.3.1 Mandatory health insurance
Insurance is used to pool risk and reduce uncertainty (Bhattacharya, Hyde, & Tu, 2013). As already stated before, health is a very uncertain good, because people don’t know if and when they will become ill. Most people are risk-averse if it comes to healthcare. Which means that people gain more utility from their expected income when they buy insurance, than the expected utility they will get from their actual (uncertain) income without insurance. This means that being insured could increase utility and welfare. But insurance could have some negative consequences on the behaviour of people.
Insurance could lead to adverse selection and moral hazard. Moral hazard occurs when people increase their risk-taking behaviour or consume more health services, because they do not have to bear the financial consequences due to insurance (Morrisey, 2008). Adverse selection could arise when there is information asymmetry between the insurer and the insured. Insurers do not have all the information about the health status of their insured. The premiums of insurance could be different if information asymmetry did not exist. Adverse selection will not occur in countries with mandatory health insurance (as in the Netherlands) or state funded health insurance, in which the premium does not depend on the health status of the insured. Besides, the conditions of a health insurance include often measures against moral hazard, such as a deductible.
7 Health insurance increases the demand for health care including the demand for drugs (Riphahn, Wambach, & Million, 2003). The increased demand for drugs, could affect the price and costs of a drug.
If the number of drugs sold increases, the R&D costs per unit will decrease. Therefore, it is expected that drugs with a higher sales volume have a lower price.
2.3.2 Patent protection
Pharmaceutical companies can receive a government-induced monopoly position for the duration of the patent. Patent protections are issued to stimulate innovation, and will provide a pharmaceutical company with the temporary right to exclusively produce a drug (Bhattacharya, Hyde, & Tu, 2013).
This makes it possible for pharmaceutical companies to recover their high R&D investments. Under the current law, patent protection holds 20 years from time of patent filling (Berger, Dunn, Johnson, Karst,
& Shear, 2016). Much of this time is spend on product development and regulatory review, which leaves an effective patent life of 7 to 10 years. There is no market competition during the patent period, if there is no alternative drug on the market. After this patent period, generics could enter the market.
This means that R&D costs need to be recouped within this useful patent period of 7 to 10 years.
Patent protection could also have negative consequences on the behaviour of pharmaceutical companies, because it could induce moral hazard. Moral hazard occurs when pharmaceutical companies abuse their contemporary monopoly position to ask much higher prices than necessary to recoup their investment. The results of this study will provide insight into the profit margins achieved by pharmaceutical companies. If the profit margins of pharmaceutical companies turn out to be very high, this could be caused by the patents they obtain. In a free market, new providers will enter the market if a product turns out to be very profitable (Geroski & Jacquemin, 2013). However, this is not possible on the pharmaceutical market, due to patent protection.
2.3.3 Price controls
Price controls are used to prevent pharmaceutical companies from asking high prices (Bhattacharya, Hyde, & Tu, 2013). A commonly used price control is a price ceiling. Price controls mitigate the effects of patents, but at the same time they limit the incentives for drug development. Price controls could lead to adverse selection and moral hazard. The government could lower drug prices through price controls, but the products that are more expensive to produce will not be offered on the market. The price controls are based on the value of certain product groups, if the high value products are no longer available on the market, the price ceilings will go down to represent the value of the goods which are available on the market. On behalf of Akerlofs’ (1978) Nobel prize winning theory, this could lead to a market where eventually only products with a low value will offered. Due to information asymmetry the government could not predict which products will not be developed or offered on the market as a result of price controls. Pharmaceutical companies have more information about the value of their
8 products and their production costs. The opacity of pharmaceutical companies concerning their R&D and production costs, will make it hard to estimate the value of a drug and the appropriate price controls.
Next to that, price controls could induce moral hazard. Pharmaceutical companies might be incentivized to set a price at the price ceiling, if they know the maximum price a country is willing to pay. Without the price ceiling, the price might have been lower. This means that price ceilings could have an adverse effect and could increase drug prices.
Several years ago, most European countries have introduced price controls for innovative drugs (Gandjour, 2015). Two policies became very popular: value-based pricing and international reference pricing. In short, value-based pricing involves a valuation of the added value of an intervention compared with existing alternatives, while international reference pricing uses the prices of a drug in one or several countries to derive at a benchmark of reference prices to set or negotiate a price (Gandjour, 2015; Neyt, 2018). Price controls of different countries could affect each other (Danzon, Wang, & Wang, 2005). A low price in one country could spill-over to other countries, through parallel trade and external referencing. Therefore, pharmaceutical companies may rationally prefer to launch a product later or not at all in a certain country, instead of accepting a low price (Kyle, 2007). The use of price controls could thus have a major impact on the availability of drugs and healthcare in a certain country.
2.4 Drug development process
Drug development is a long-term sequential procedure. Pharmaceutical companies will regularly review the status of drug development and decide whether to continue or not (DiMasi, Hansen, Grabowski, & Lasagna, 1991). Generally, this decision is based on the potential therapeutic benefits, the expected frequency and severity of adverse reactions, projected additional development, marketing, distribution, and production costs and the estimate future revenue streams (DiMasi, Hansen, Grabowski, & Lasagna, 1991). The development process starts with basic research, to find the cause of a disease or its symptoms (Schweitzer & Lu, 2018). Based on the findings of this research, they will start to develop a drug, this drug will first be tested in the preclinical phase. This phase is followed by three clinical trials, in which the drug will be tested on humans. These clinical trials have to prove that the drug is safe and effective.
In the first phase, the drug will be tested on a small group of healthy humans (Vereniging innovatieve geneesmiddelen, 2019). These trials need to establish that the drug is not detrimental to the health of human beings. If the drug passed the first trial, the drug could enter the second clinical trial. This is a clinical trial with patients who are actually suffering from the disease. A part of the patients will receive the new drug and another part will receive a placebo or the drug which is perceived to be the standard
9 treatment at that time (Schweitzer & Lu, 2018). The effectiveness of the drug is tested in this phase, even as the short-term side-effects and other risks (Vereniging innovatieve geneesmiddelen, 2019). If the drug seems promising the third phase could start. In this phase the drug will be tested on large scale. Many patients all over the world are involved in this trial. In this phase data about safety and effectiveness will be gathered. Once the drug has entered the market, pharmaceutical companies have to conduct after-market studies which are focussed on the safety and long-term side-effects. A schematic overview of the drug development phases is shown in figure 1.
Figure 1. Schematic overview of drug development process.
The drug development process consists of many phases and is an expensive process which will take a long time. The process will take approximately 10 till 15 years and only 3% - 11% of the drugs in the preclinical phase will eventually reach the market (Gupta Strategists, 2019). The total development costs are estimated between 0.5 and 6.5 billion dollars, depending on the type of drug involved (Gupta Strategists, 2019). Most of the cost are caused by the long-lasting development period, the low success rate and the costs of capital involved, which combined involve approximately 53% of total costs.
The costs related to this research and development process need to be recouped during the patent period. Depending on the size of these costs and the number of drugs sold, these research and development costs could have a high impact on drug prices. It is expected that the research and development costs weigh more heavily on the price of medicines for rare diseases, since these drugs only have a small target group.
2.5 Pricing mechanism
A pricing mechanism is a method which is used to establish the price of a product or service. Most companies will set their prices at a level which maximizes profit. According to the basic economic
10 theory companies will set their price at the demand curve where the marginal revenues equal the marginal costs (Nagle & Müller, 2017). It is difficult or even impossible to set a price at this level in the real world, because of the uncertainty regarding consumer behaviour and the difficulty to identify the costs. In the next sections two different pricing mechanisms will be discussed: value-based pricing and cost-based pricing. Value-based pricing is currently used for innovative drugs while cost-based pricing is not commonly used in this field.
2.5.1. Value-based pricing
Value-based pricing aims to price pharmaceuticals according to their value, that is, the therapeutic benefits to patients (Hughes, 2011). To be able to determine a value-based price, benefits need to be identified, measured and valued (Sussex, Towse, & Devlin, 2013). The aim of pharmaceuticals is to increase patients’ health. Health benefits are mostly measured in QALYs (Neyt, 2018). QALYs are arrived by adjusting the length of time affected by the intervention with the value of a given health state on a scale from 0 to 1 (Drummond, Sculpher, Claxton, Stoddart, & Torrance, 2015). There are several health status classification systems, which could be used to determine the QALYs gained by a certain intervention. The three most widely used systems are EuroQol-5D (EQ-5D), Health Utility Index (HUI) and Short Form 6D (SF-6D). According to these classification systems, the quality of life will be determined based on several factors among which: mobility, self-care and pain.
New interventions will be compared with the current standard intervention, to examine their added medical benefits. The additional QALYs gained need to be valued. The value of a gained QALY depends on the valuation of the buyer; the willingness to pay. The willingness to pay for a QALY might differ between individuals and countries. Some countries have a formal willingness to pay threshold which is used to determine whether a drug will be reimbursed or not (Gandjour, 2015). Countries will calculate the incremental cost-effectiveness ratio (ICER), which offsets the additional costs against the additional health gains of the new intervention compared to the current standard intervention (ICER=∆costs/∆effects). The ICER will be used to determine the maximum price a country is willing to pay for a certain drug. In the European Union, only the UK set a formal willingness to pay threshold, they set their threshold at £20,000–£30,000 per QALY. Which means that a drug will only be reimbursed if the additional costs per QALY gained by an intervention compared to the existing alternative does not exceed this threshold.
The Netherlands does not have a certain formal willingness to pay threshold, but there are some informal guidelines. The informal willingness to pay threshold is set on €20,000 for a low disease burden, €50,000 for an average disease burden and €80,000 for a high disease burden (Zorginstituut Nederland, 2014). The disease burden is measured in Disability-Adjusted Life Years (DALYs)
11 (Zorginstituut Nederland, 2018). DALYs emphasize the health that is lost within a specific condition.
The DALY is the opposite of the QALY, this is illustrated in figure 2 below.
Figure 2. Schematic overview of the relationship between QALY and DALY (Zorginstituut Nederland, 2018).
The informal willingness to pay threshold thus depends on the impact a disease has on someone’s life.
In table 1. below, a schematic overview can be found of the informal willingness to pay thresholds and the related disease burden accompanied with an example.
Disease burden Willingness to pay €/QALY Example disease
Low (0,1 - 0,4) €20.000/QALY Eczema1
Medium (0,41 – 0,7) €50.000/QALY Dementia1
High (0,71 – 1,0) €80.000/QALY Lung cancer2
Table 1. Schematic overview of the informal willingness to pay threshold in the Netherland.
Formal or informal willingness to pay thresholds can affect drug prices. Due to moral hazard, pharmaceutical companies will be seduced to value their drug at the upper limit of the willingness to pay threshold or just above, to create some negotiation space (Kyle, 2007). Without the willingness to pay threshold drug prices might have been lower. The difference in willingness to pay threshold between countries could result in early availability in countries with a high threshold and later or no availability in countries with a lower threshold. Due to the fact that a low price in one country could spill-over to other countries, through parallel trade and international reference pricing (Danzon, Wang,
& Wang, 2005).
Value-based pricing has both advantages and disadvantages for both parties; the pharmaceutical company and the government. It provides rules to value certain health benefits. A clear willingness to
1 https://www.volksgezondheidenzorg.info/wegingsfactoren-voor-ziektelastberekening#node-definitie-wegingsfactor
2 https://www.volksgezondheidenzorg.info/ziektelast-nederland#node-ziektelast-naar-afzonderlijke-ziekten
12 pay threshold could lead to lower prices if it is used to determine whether a drug will be reimbursed or not. Whether a drug is reimbursed or not will affect the sales volume. Therefor pharmaceutical companies will set their price at the willingness to pay threshold or just below. On the other hand, a clear price ceiling could increase drug prices. Pharmaceutical companies know exactly what the maximum price is they could ask for a drug. Besides such price ceilings are helpful to prioritize healthcare spending in a way that maximizes the total health benefits which could be achieved with a certain budget.
2.5.3. Cost-based pricing
Cost-based pricing is a pricing strategy where the selling price depends on the cost price of a product.
A profit element is added to the cost price. A product or service is priced to yield a fair return over all costs (Nagle & Müller, 2017). The profit element could be added to the variable costs or to the full costs (variable costs plus an allocated proportion of the fixed costs). The difficulty of the full cost method is the uncertainty with regard to the sales volume, the useful patent period and the chance of prevalent drugs entering the market. The sales volume and the effective patent period affect the amount of fixed costs which are allocated to a unit. Cost-based pricing is a method which is commonly used to determine the transfer price for inter-company transactions (Edmonds, Edmonds, Tsay, & Olds, 2014). The full cost method will be most suitable for the determination of drug prices, while the fixed R&D costs are a major part of the costs involved in the production of a new drug. Nevertheless, the uncertainty with regard to the sales volume and the effective patent period need to be considered.
In order to make a calculation of the assumed cost-based price, it is important to have an indication of the costs related to R&D and manufacturing of a drug. Currently, there is a lack of transparency about the R&D and manufacturing costs. It is estimated that the R&D costs of drug developing lie between 0.5 and 6.5 billion USD, depending on the type of drug involved (Gupta Strategists, 2019).
Basing drug prices on the cost incurred will increase the transparency concerning the costs of drugs and the attained profit margin. In society there is a debate about the justification of the assumed high profit margins, which are attained by pharmaceutical companies (Fleming, 2019). Drugs have a high value for society because they could save lives or make them bearable. Profits should be high enough, to stimulate innovation. The investment in drug development is very risky, because the outcome is unsure, and the investments are very high and long-term. Only 3% to 11% of the developed drugs will eventually reach the market (Gupta Strategists, 2019). This uncertainty must be compensated in the profit margin.
The difference between the production costs of a drug and the market price could differ enormously.
In 2019 the University hospital of Amsterdam started to produce a medicine for the disease cerebrotendineuze xanthomatose (NU.nl, 2019). The market price of this drug was approximately
13
€200,000, the hospital was able to produce this drug for “only” € 25,000. This example shows how big the difference between the production costs and the market price of drugs can be. A cost-based pricing mechanism could lead to more transparency about the attained profit margins.
Using a cost-based pricing mechanism has also a down-side. Using the costs as a benchmark will stimulate inefficiency (Edmonds, Edmonds, Tsay, & Olds, 2014). The incentive to control costs is diminished, which could potentially lead to higher drug prices.
2.6 Theoretical framework
As appeared from the literature above, there are many stakeholders who can affect or are affected by drug prices. All these different interests need to be considered. This study will focus on two main stakeholders: pharmaceutical companies and the government. Those two parties can directly affect drug prices. The government can influence drug prices via the introduction of rules and regulations.
Pharmaceutical companies decide the market price of their drugs. The interaction between those two parties is from main importance in this study. This study will explore the effect of pricing mechanisms on drug prices.
Next to the variety of different stakeholders involved in the pharmaceutical market, a lot of agency relationships exist between those stakeholders, which makes it a complex field. While the pharmaceutical company is aiming for profit maximizing, the government tries to keep healthcare affordable and stimulates innovation. This study will indicate how drug prices are influenced by different pricing mechanisms. This could be helpful to determine which behaviour should be incentivised. A simplified overview of the main stakeholders and their agency relationships is provided in figure 3 below.
Figure 3. Theoretical framework: overview of the main stakeholders and agency relationships.
14 In order to investigate the effects of different pricing mechanisms, two models will be built for a certain drug: a value-based model and a cost-based model. The drug price will be calculated based on the assumptions related to those pricing mechanisms. The effect of those different pricing mechanisms on the drug price will be indicated for two different drug types: a drug for a mass disease and a drug for a rare disease. This distinction is made because it will provide insight into the influence of economies of scale. A schematic overview of this study is provided in the figure 4 below.
Figure 4. Schematic overview of the study
This study focusses on the influence of pricing mechanisms on drug prices. The results of this study give insight in the profit margin which is attained by pharmaceutical companies. Besides this study will provide insight in the major cost drivers of a drug. Next to this, the findings of this study could be used to provide an incentive system which stimulates a certain pricing mechanism. The implementation of a certain incentive system could be beneficial for the government to keep healthcare affordable. There are no previous studies conducted on this topic, which makes it a rather underexplored field. This study will contribute to the academic literature, because it provides insights into the effects of pricing strategies on drug prices.
Drug price Pricing-mechanism
Cost-based Value-based
Rare disease Mass-disease
15
3. Methodology
In this section, the methodology of this study will be discussed. This study has an explorative character and no hypotheses will be tested. The aim of this study is to explore whether the use of different pricing mechanisms will influence drug prices. The two pricing mechanisms, value-based and cost- based pricing, are compared to understand their effect on drug pricing. This study investigates the effect of pricing mechanism on two types of drugs with a different purpose and patient target group.
A drug which is on the market for a mass disease (common drug) and a drug for a rare disease (orphan drug). This distinction is made because of the effect of economies of scale on the production costs and the distribution of R&D costs. Orphan drugs have a smaller sales volume, but their R&D costs are much lower due to the small size of their clinical trials, shorter clinical trial time and higher rates of regulatory success (Meekings, Williams, & Arrowsmith, 2012).3
3.1 Sample selection
The samples were selected based on the characteristics of the drugs and the availability of data. Only drugs of publicly listed pharmaceutical companies are included in this study, due to the availability of financial information in their published financial statements. Besides, data should be available about the clinical outcomes of the drug. Applying these criteria, two different drugs were selected.
For the mass disease the drug sacubitril/valsartan (Entresto ®) was modelled. This drug has been on the market since 2015 to treat patients with heart failure. The drug is developed and marketed by Novartis, a company which is listed on different stock markets among which the SWX stock market in Switzerland. Data about the clinical effects was gathered from previous studies which were conducted for this drug (van der Pol, Degener, Postma, & Vemer, 2017).
For the rare disease the drug alfa alglucosidase (Myozyme®) was modelled. This drug has been on the market since 2006. It is used to treat patients with infantile Pompe disease. Pompe is a rare, hereditary, progressive and fatal metabolic disease (U.S. National Library of Medicine, 2019). Alfa alglucosidase is currently marketed by Sanofi. Sanofi is publicly listed on the stock market Euronext Paris. Information about the clinical effects of alfa alglucosidase was published by Kanters et al. (2014).
3.2 Cost-based model
The cost-based model is mainly built on the information which was published in the financial statements and a research report of Gupta strategists. The report of Gupta strategists was commissioned by Fair Medicine. The study of Gupta strategists was based on academic literature and the database of EveluatePharma. The models were built around a timeline in Microsoft Excel 2016.
3 A disease is considered to be an orphan disease if it has an incidence at birth of less than 1 in 2000. The patients populations of rare diseases represent maximum 6-8% of the world population (Sharma, Jacob, Tandon, &
Kumar, 2010).
16 Costs and benefits were linked to the different development and market stages. The model for the common drug (sacubitril/valsartan) can be found in appendix A and the model for the orphan drug (alfa alglucosidase) can be found in appendix C.
3.2.1 Time Horizon
For this model, a time-horizon of 20 years was used. Because pharmaceutical companies could obtain a patent for twenty years (Berger, Dunn, Johnson, Karst, & Shear, 2016). In this patent period, the drug needs to be developed and tested. Once a drug is approved, pharmaceutical companies will benefit from their patent protection for approximately 7 to 10 years, dependent on the type of drug involved.
Orphan drugs often benefit longer from their patent protection due to their shorter clinical trial time (Meekings, Williams, & Arrowsmith, 2012). The R&D expenses need to be recouped within this period of patent protection. After this time horizon, generics could enter the market which will drastically lower the price. If generics enter the market, R&D investments will no longer influence the drug price.
3.2.2 Input parameters
Development phases
The drug development process consists of four stages as described in the theory section. The pre- clinical phase and three phases of clinical trials (Vereniging innovatieve geneesmiddelen, 2019). During the pre-clinical phase, the drug will be tested on animals. In the first phase of clinical trials, the drug will be tested on healthy humans to ensure that the drug is not detrimental to the health of human beings. During the second phase of clinical trials, the drug will be tested on a group of patients suffering from the disease. A part of the patients will receive the new drug and another part will receive a placebo or the current standard treatment. During the third phase the drug will be tested on a large scale; many patients all over the world are involved in this trial.
Common drug
The length of the different development phases was partly found in the financial statements of Novartis. This information was only available from the year 2007 and on. In the year 2007 the development of this drug was already halfway phase two, therefor the length of the pre-clinical trial and the phase 1 and 2 clinical trials were estimated based on the study of Gupta strategies (2019).
Orphan drug
The length of the different development phases was found in the financial statements of Genzyme.
Genzyme developed alfa alglucosidase and was merged by Sanofi in 2009. The data in the financial statement only included the development stage per year. The length of those cycles in months was estimated based on the study of Gupta strategists (2019).
17 R&D costs
Common drug
The R&D costs of the development of sacubitril/valsartan were not specified in the financial statement of Novartis. Therefor general information about the R&D costs of the development of a cardiovascular drug was used to estimate the R&D costs of this drug (Gupta Strategists, 2019).
Orphan drug
The R&D costs for the development of this orphan drug were not specified in the financial statement of Genzyme nor in the financial statement of Sanofi. The acquisition price of Genzyme could not be used either, because Genzyme produced many different drugs. General information about the R&D costs of drug development of biological orphan drugs was used to estimate the R&D costs. This data was gathered from Gupta Strategists (2019).
Failure costs
The costs of failure were calculated using the success costs and the success rate. The success costs and success rates of different development stages in the R&D process of cardiovascular drugs and orphan drugs were published in the study of Gupta Strategists (2019). A timeline was conducted to calculate the accumulated success rate of the different R&D phases.
Market introduction and products sold Common drug
The market introduction and the number of drugs sold per year were estimated based on the net sales which were published in the financial statements. The net sales in the United States and the rest of the world were separately stated in the financial statement. The net sales were divided by the estimated drug price in those areas. The drug price of sacubitril/valsartan in the United states was estimated to be $544.75 a month (Entresto, 2019). The number of drugs sold in the rest of the world were estimated based on the net sales divided by the Dutch list price of sacubitril/valsartan of € 5.08 a day and the daily dose contains two pills (Zorginstituut Nederland, 2019). The Dutch list price is used because this price is comparable to the price in other western European countries, due to international reference pricing (Wet geneesmiddelenprijzen). The actual price which is paid for those drugs is unsure, because parties such as the Department of Health, health insurers and hospitals may still negotiate lower prices.
The outcomes of these negotiations are not public, but it appears that the Dutch government negotiated prices which were on average 36% lower than their market price in 2018 (Rijksoverheid, 2019). The European price was converted into a dollar prices, the average conversion rates of 2017 and 2018 were used.
18 Orphan drug
The sales volume of Alfa alglucosidase was estimated based on the data which was published in the financial statements of Genzyme and Sanofi. Alfa alglucosidase is on the market since 2006. The yearly sales were divided by the Dutch market price, because this price is representative for the price in the western world, due to international reference pricing. The Dutch market price of Alfa alglucosidase is currently € 575,25 (Zorginstituut Nederland, 2019). It is questionable whether this is the price, which is really paid for this product, because drug prices will often be negotiated.
Cost of capital
The weighted average cost of capital was calculated based on the information which was published in the financial statements. Both the amount of debt and interest were stated in the financial statements.
These numbers were used to calculate the yearly interest rate. The costs of equity were calculated, using the capital asset pricing model. This model considers the riskiness of the investment relative to the market (Berk & DeMarzo, Corporate finance: the core, 2016). This model is based on the following formula:
𝐸[𝑅𝑖] = 𝑟𝑓+ 𝛽𝑖(𝐸[𝑅𝑀𝑘𝑡] − 𝑟𝑓) 𝐸[𝑅𝑖] = Expected return on investment
𝑟𝑓 = Risk-free rate of return 𝛽𝑖 = Beta of investment
𝐸[𝑅𝑀𝑘𝑡]= Expected market return
The risk-free rate was based on the AAA rated bonds of the ECB. The expected market return was set on 6% (KPMG Meijburg, 2018; PWC, 2019). The Beta’s of Novartis AG respectively Sanofi SA were based on information which was published by Reuters, and were estimated to be 0.61 respectively 0.72 (ReutersA, 2019; ReutersB, 2019).
Common drug
The market capitalization of equity was stated in the financial statement of Novartis since 2009.
Therefor the period of 2009 until 2018 was used to calculate the weighted average costs of capital.
Orphan drug
Both the amount of debt and the market capitalization of the firm were stated in the financial statement of Sanofi since 2009, this is also the year in which Sanofi merged with Genzyme. Therefor the period of 2009 until 2018 was used to calculate the weighted average costs of capital.
19 Profit margin
The after-tax profit margin is set at 15%, this profit margin was suggested for pharmaceutical products by the Organisation for Economic Co-operation and Development (OECD) (Danzon & Towse, 2003).
The pre-tax profit margin was calculated, based on the average effective corporate tax rate of Novartis AG respectively Sanofi SA in the past ten years. The pre-tax profit margin was used to calculate the price which delivers an after-tax profit margin of 15% (pre-tax profit margin=after-tax profit margin/(1- effective corporate tax rate).
Discount rate
The costs were discounted with 1.91%. This discount rate is based on the average yearly inflation in the Netherlands over the past 20 years (CBS Statline, 2020).
Value added tax (VAT)
A value-added tax rate of 9% was used. This is the Dutch value-added tax-rate for pharmaceutical products (Wet op de Omzetbelasting 1968).
Production cost Common drug
Sacubitril/valsartan is administered in the form of a tablet, which contains the active ingredient sacubitril/valsartan. There is already a drug on the market which contains valsartan. This product is no longer protected by patents, and therefor generics have entered the market. This will result in drug prices which is closely related to the production costs. Just as sacubitril/valsartan this drug is also administered in the form of a tablet. The list price of the generic drug of valsartan was € 0,08 per tablet (Rijksoverheid, 2019). It is expected that the production costs of sacubitril/valsartan will be comparable to the price of valsartan, because the drug sacubitril/valsartan also consists of one molecule which makes the production process comparable (Gu, et al., 2010). This assumption was discussed with a pharmacist, who agreed that these production processes are of a similar nature.
Orphan drug
Alfa alglucosidase is a powder which need to be dissolved and administered via intravenous therapy.
It contains an enzyme replacement therapy for patients suffering from Pompe disease. The drug is produced by recombinant DNA technology on Chinese Hamster Ovary (CHO) cell lines (EMA, 2006).
Based on the opinion of an expert from the biotechnical pharmaceutical industry, the production costs of 50 mg of alfa alglucosidase were estimated to be €110.
20
3.3 Value based model
Common drug
A model about the cost-effectiveness of sacubitril/valsartan in the Netherlands was published by van der Pol et al. (2017). This model was used in this study to determine the value-based price of sacubitril/valsartan. A healthcare payer perspective was adopted in this study. This means that only medical costs which were actually paid by the healthcare payer were included in this study (Drummond, Sculpher, Claxton, Stoddart, & Torrance, 2015). The original data of this study comes from the PARADIGM-HF trial, in which the clinical effects of sacubitril/valsartan were compared to enalapril (McMurray, et al., 2014).
The model originally focused on the calculation of the ICER but has been extended to calculate the net monetary benefit (NMB). The NMB is calculated as the value of the benefits of a therapy net of all costs (Shafrin, et al., 2016). The NMB calculates the difference between the ICER and the willingness to pay threshold (NMB = (health effect x willingness to pay threshold) – treatment costs). To calculate the maximum value-based price, the NMB should be equal to 0. The total drug costs which result from the calculations based on the Markov model, where divided by the market price of € 5.08. The adjusted model can be found in appendix B.
Orphan drug
The value-based model was based on the cost-effectiveness study of Kanters et al. (2014). The study of Kanters et al. compared the cost and effects of the treatment with alfa alglucosidase with the cost and effects of the supportive therapy (ST) on patients suffering from classic-infantile Pompe disease.
This study adopted a societal perspective which implies that all costs were included, no matter to whom they accrued.
Model design
A Markov model was developed to simulate a cohort of 1.000 patients suffering from Pompe disease.
This type of model is widely used for economic evaluation of healthcare. A Markov model is based on a series of states that a patient can occupy at a given point in time (Drummond et al., 2015). Data about survival, utilities and costs were used. Cycles have a duration of 6 months for a lifetime period (figure 1.). The model is a simplified representation of reality and contains two health states.
1. Alive: patients who are alive and suffering from Pompe disease.
2. Death: patients who passed away.
A distinction could have been made in the progression of the disease (e.g. use of ventilation support or being dependent on a wheelchair), but the available data was insufficient. Therefor the model contained only two states, all patients started in the alive state.
21 Figure 5. Schematic representation of the Markov model. The circles represent the two states of the model, and the arrows indicate the possible transitions.
Time horizon
A time horizon of 90 years was used as a lifetime time horizon.
Effects
Survival
Survival data for patients treated with the standard therapy was retrieved from an international study on the natural course of infantile Pompe disease (Kishnani, et al., 2006; Zorginstituut Nederland, 2006).
Survival rates at 12, 18, 24 and 36 months were used to indicate the 6 months mortality rate. Survival rates at these ages where plotted and a trendline with a logarithmic function was found to indicate the mortality rate. A study about the early treatment of Pompe disease with Alfa Alglucosidase showed, that this drug reduced the risk of death by 95% (Kishnani, et al., 2009). Survival probabilities were corrected with the general mortality rate of the Dutch population. Data about life-expectancy and mortality rates of the general population were gathered from CBS Stateline (CBS Statline, 2019;
CBS Statline, 2019).
Quality of life
Quality of life was based on data which was available in the study of Kanters (2014). This study assed the quality of life of Dutch patients using the Euroqol-5D (EQ-5D). The average quality of life of patients with Pompe disease was estimated to be 0.62 patients who were treated with Alfa alglucosidase were assumed to experience the same quality of life.
Costs
Costs were calculated from a societal perspective. This implies that also non-healthcare costs were included in this assessment. Total costs for patients treated with Alfa alglucosidase consisted of three components: medicine costs, infusion costs and other costs. Patients who receive the standard therapy did only incur other costs.
