Pharmaceutical expenditure control in Europe: time to move from pricing to budgeting?

Price regulation schemes have long been the typical policy response to control public pharmaceutical expenditure by European health authorities before recurring more recently to direct negotiation.¹ On the other side, high prices in all countries are considered a success for the private pharmaceutical industry − nowadays predominantly multinational − crucial for maximising worldwide profits.² However, the present era of austerity in public funding has made resources really scarce in all European countries, including the wealthier ones.

Past and present situation

Price setting in the western continental countries

In the very first attempts in big continental countries (e.g. France, Italy and Spain), prices were set by posting to each drug the main cost items borne by the pharmaceutical industry (e.g. research and development, manufacturing, marketing).³ These schemes were progressively abandoned because of the difficulty of estimating single cost items per product in this research-based industry.

Grouping similar drugs for pricing was introduced in Germany 30 years ago,³ followed by the Netherlands in 1990s, and is now applied in various forms in many European countries. The same ‘reference price’ for drugs considered therapeutically overlapping can be based on either domestic (‘internal’) or foreign (‘external’) prices of already marketed products. The sensible strategy of clustering products with similar efficacy under the same price suffers two practical limits for widespread application. The first is implicit in the scheme, which requires already available prices for similar drugs domestically or abroad. The second is a consequence of the current lack of transparency on many real prices at both national and international levels (see below).

A more straightforward approach to exploit the purchasing power of health authorities to the full is competitive tendering. Recently extended from hospital services to community care in some countries (e.g. Germany, Italy and the Netherlands), bids have to be designed in such a way that many manufacturers can tender and lots include many products.⁴ So, the crucial issue is where to ‘draw the line’ for bidding similar drugs. A very likely drawback is that when big lots are tendered and the turnover is substantial, ‘losers’ will probably raise legal questions. This is why tenders are still limited to off-patent drugs (generics and biosimilars) in most European countries.

A more recent approach for pricing new drugs is to ‘monetise’ their therapeutic added value on the basis of economic models populated by short-term efficacy data and long-term cost estimates.¹ Announced some years ago in the United Kingdom, value-based pricing is a form of extension of reimbursement criteria based on economic evaluations, with the drug price the unknown in the economic model. According to the lack of clinical evidence for new drugs at market approval, in the end these exercises are very open to authors’ discretion and may appear as much an art as a science.⁵

The negotiation trend

The leading tendency in Europe is to negotiate prices directly with the pharmaceutical industry.¹ Many domestic authorities strive for savings through price caps, paybacks and confidential discounts. Some small countries (e.g. Austria, Belgium, Ireland, Luxembourg and the Netherlands) are even trying to increase their single purchasing power and negotiate lower prices together with pharmaceutical companies.⁶ However, all these strategies are administratively burdensome, possibly with costs offsetting savings if poorly managed (e.g. performance-based agreements in Italy).⁷ Yet they can lead to different real prices for the same drug in the same country, making international price comparisons a puzzle. Last but not least, any trade negotiation implies some confidentiality to be really effective and thus lacks transparency by definition. For instance, the recent ‘patient access schemes’ negotiated with companies by the National Institute for Health and Care Excellence (on behalf of the British National Health Service) are commercial-in-confidence contracts whose financial results are not made public.⁸

Lessons from the British Pharmaceutical Price Regulation Scheme

In the UK, public authorities have never controlled pharmaceutical prices directly. In fact, despite its name, the Pharmaceutical Price Regulation Scheme does not imply price setting, but more in general controls the broad profitability − related to sales to the NHS − of the branded pharmaceuticals at a company level.³ To trigger an industrial ‘virtuous circle’, the Pharmaceutical Price Regulation Scheme has historically aimed at discouraging promotional expenditure on pharmaceuticals, while simultaneously rewarding local research investments to spur a domestic ‘research-oriented’ industry. Delocalisation and globalisation are the main reasons for the present crisis of the Pharmaceutical Price Regulation Scheme too.¹ Despite the successful historical performance, the control of companies’ profitability has been inevitably undermined, as multinationals tend to delocalise manufacturing and research investments, replacing many accounted cost items (e.g. licensing costs) with internal ‘transfer prices’, which hardly mirror the real costs. Not by chance, the NHS recurs more and more to direct negotiations, which have been extended to cover all new drugs in the last Pharmaceutical Price Regulation Scheme,⁹ started this year. However, a general lesson that can still be drawn from the Pharmaceutical Price Regulation Scheme is that pharmaceutical price setting per product is a very arguable exercise.

The unsustainable price trend

Pharmaceutical expenditure, like anything else, is the product of prices and volumes. While nowadays the latter are easier for health authorities to monitor through big databases, the former seem increasingly beyond control in a ‘market failure’ situation like pharmaceuticals.¹⁰ This happens today in most developed countries, with Europe no exception. The main goal for health authorities of highly developed countries is universal access to essential drugs, if possible with prompt access for new, better ones.⁶ The main (obvious) goal for the pharmaceutical industry is to legally maximise profits, so as to guarantee high returns on investments and push up the value of stocks and shares.¹¹ So multinational companies are keen to maximise overall turnovers throughout countries with similar prices,⁶ minimising global loss of money due to parallel imports.

While the advocates for the pharmaceutical industry claim that high prices are needed to support the enormous expense of researching and developing new drugs,² critics argue that the present huge profits generated by too many new sky-high priced drugs can no longer be justified in the long run.¹² The pro-industry argument seems harder to accept in the light of the increasing shift by big companies from in-house discovery of new drugs to acquisition of small start-ups. This somehow shifts the risk from industry, investing mainly in drug development, to governments and investors backing basic research. Not by chance, the recent cost estimates for developing a new drug vary almost ten times from the lowest to the highest.¹³ Then, too aggressive marketing is still required to sell expensive medicines that often differ only marginally, with fashionable ‘personalised medicine’ helping to create an ideal setting for pricing new similar drugs differently.¹⁴ Anti-cancer therapies are an emblematic example of untenable prices.¹⁵ Pharmaceutical companies have a clear incentive to invest in these drugs and health authorities spend an increasing share of pharmaceutical expenditure on very expensive end-of-life treatments, regardless of their low impact on survival (a few extra months at best) and quality of life (very severe side effects, without much relief of symptoms in life’s final phases).^10,11 Last but not least, anti-cancer therapies are often more effective in combination, so most new drugs result in a massive add-on cost.

An alternative scenario

Starting from patenting and market approval

We believe that pharmaceutical policy in the European Union needs drastic changes from the start of a drug’s life-cycle, to (re)establish an acceptable trade-off between public and private interests.

First, a specific European agency for health products could be introduced to tame the ‘jungle’ of pharmaceutical patents.¹⁶ The European Patent Office is a big executive body in no way legally bound to the European Union and entirely funded by patent fees, so it is totally beyond control in a public health perspective. The excess of discretion by the pharmaceutical industry in filing patents often generates an ‘invention cascade’ aimed at dramatically prolonging them. These ‘ever greening’ strategies give rise to many costly litigations, with eventually a waste of money indirectly funded by public pharmaceutical expenditures. A specific EU agency dedicated to ‘merit goods’ for health − pharmaceuticals included − could limit this widespread malpractice, somehow allowed by the present patent regulation.

Second, the regulatory tasks of the European Medicines Agency should be expanded far beyond the present scope of preliminary assessments of efficacy and safety of new drugs,¹⁷ to cover at least their added therapeutic value compared to existing drugs. Since a fast, inexpensive and unbiased study design for these assessments simply does not exist, the return to head-to-head Phase III trials for market approval might be highly recommendable. This would help make national reimbursement policies more rational and homogeneous, with a substantial reduction in workloads for both regulatory authorities and pharmaceutical companies. Moreover, the European Medicines Agency might conduct ex ante priority-setting studies to highlight the ‘unmet needs’ that really matter in terms of public health in Europe, as was written in its ‘roadmap’ to 2020. This would help steer the industry’s research investments in advance towards priority therapeutic areas.

And, finally, from pricing to budgeting for reimbursed drugs

After having decided what drugs are eligible for reimbursement according to clinical evidence, the payment procedures for pharmaceuticals – pricing included − could be drastically revised,¹⁸ and phased in five steps (Box 1). First, reimbursable drugs could be listed in a limited number of therapeutic classes according to international validated classifications. Second, the first year’s budgets for each therapeutic class could basically coincide with the historical expenditures of the previous year, to avoid any start-up shock. Over the subsequent years, national authorities might introduce smooth changes according to pharmaceutical policy decisions based on local variables. Third, the listed drugs could be classified in four groups by patent protection (off-/on-patent) and manufacturing procedure (chemical/biological) for each therapeutic class, so as to differentiate the unit prices reimbursed. These are the only two criteria that can be objectively applied for differentiating the drugs’ unit costs for reimbursement. Fourth, national authorities could reimburse all the daily doses prescribed, regardless of the size of the packs in which the doses are dispensed, and their unit prices could be modified during the year to respect the budgets per therapeutic class if the final volumes differ substantially from those initially budgeted. Finally, health authorities could pay − out of the budget for each therapeutic class − a flat fee-for-service per prescription to community pharmacies, to recognise only their public service for drug delivery. At the same time, pharmaceutical companies would be free to trade distribution margins with private wholesalers and pharmacies through commercial negotiations, as has happened for decades in countries like the Netherlands and the United Kingdom.¹⁹

Box 1.

Payment procedures for pharmaceuticals.

Price scheme steps	Hypothetical examples
(1) Reimbursable drugs are classified in a limited number of therapeutic classes.	According to the second level of the ATC classification,a around 90 classes could be considered for reimbursement.
(2) The first-year budget for each of these classes is based on the historical expenditure of the previous year.	Smooth changes to the first year’s budget (e.g. ±10%) could be made by national authorities on a yearly basis, taking account of future drug listing/delisting and inflation/deflation rates.
(3) The unit prices of the listed drugs in each therapeutic class are classified in four groups by (i) patent protection (off-/on-patent) and (ii) manufacturing procedure (chemical/ biological).	A 50% decrease after patent expiry could be reasonable for patenting, while manufacturing would require a public survey to calculate an average difference for the two types of drugs. Assuming a hypothetical difference of +20% for manufacturing biologics, the other three weights would range from 0.5 for a chemical generic to 1.2 for an on-patent biologic, with an intermediate value of 0.6 for biosimilars.
(4) National authorities reimburse all the daily doses prescribed, applying the four unit prices for each therapeutic class, regardless of the size of the packs in which doses were dispensed. If volumes vary substantially, the unit prices per dose will be modified during the year, to respect the initially planned yearly budget.	The DDDsb could be the variable used to quantify volumes. To make DDDs fully comparable per form, they should be adjusted according to specific procedures. If DDDs fell or rose substantially on a monthly basis compared to the previous year, the unit prices per dose of each therapeutic class should be proportionally raised or lowered every quarter.
(5) National authorities pay community pharmacies a flat fee-for-service per prescription out of the national budgets per therapeutic class.	A flat fee per prescription (e.g. €3) should be enough to acknowledge the pharmacies’ public service for drug delivery.

ATC: anatomical therapeutic chemical classification; DDD: defined daily dose.

^aATC Structure and Principles. WHO Collaborating Centre for Drug Statistics Methodology. Available at: https://www.whocc.no/atc/structure_and_principles/. Last accessed: May 2019.

^bDDD Definition and general considerations. WHO Collaborating Centre for Drug Statistics Methodology. Available at: https://www.whocc.no/ddd/definition_and_general_considera/. Last accessed: May 2019.

To conclude, pharmaceutical pricing has become increasingly challenging and politically unsustainable even in high-income European countries.²⁰ While the pharmaceutical industry has always been ready to adapt promptly to new market scenarios, health authorities still have difficulty in steering companies’ marketing strategies. We fear that sound strategies like reference pricing and tendering are not enough to invert the present trend. But more recent strategies (e.g. value-based pricing and outcome-based agreements) have already shown evident limits, while full transparency of pharmaceutical costs is not achievable and that of negotiated prices is hardly effective.

The underlying rationale of our proposal is to stop setting arbitrary prices and to drastically reduce their impact on pharmaceutical expenditures. Prices can hardly (if ever) be really right in a ‘market failure’ context like health,²¹ so their effect must be minimised to limit the distortion of allocation of resources, from upstream research investments to downstream healthcare expenditures. Since there is no way to rank pathologies according to their importance, we can just assume that all drugs judged eligible for reimbursement are equally essential for the population’s health. So rational budgeting should be given priority over irrational pricing, hopefully steering the drug consumption only towards health needs. We are convinced that such a pharmaceutical policy should not necessarily be worse than the present ones in driving research investments towards the still unmet needs of patients.

Declarations

Competing Interests

None declared.

Funding

None declared.

Ethics approval

Not required.

Guarantor

LG.

Contributorship

Authors contributed equally to the article.

Acknowledgements

None.

Provenance

Not commissioned; peer-reviewed by Riccardo Roni.

ORCID iD

Livio Garattini https://orcid.org/0000-0002-8952-9968