CIAO DATE: 11/04

Safety Standards: A Race to the bottom of EU Regulatory Policy?

Christa Altenstetter
Professor of Political Science
Ph.D. Program in Political Science
CUNY Graduate Center

Occasional Paper Series No. 56

May 6, 2004

European Union Studies Center

Today’s topic: Safety Standards: A Race to the Bottom of EU Regulatory Policy tackles a problem of interdisciplinary interest and significance for policymakers and researchers in countries around the world. It transcends cultures, societies, ideologies, political systems, and economies. Above all, it is a topic which concerns all of us because we are all potential patients, not just voters, patriots, citizens or residents. Had I added ‘public health and patient safety’ to the title, it would have become immediately apparent that the topic is construed in a much broader framework and is not limited to a discussion of safety standards of products for regional trade.

With an aging population in the industrialized world, we can expect a rise in chronic diseases, physical handicaps and multi-morbidity and should therefore anticipate an increase in the demand for health services and, concomitantly, the use of medical devices. This is the case regardless of who delivers care and where services are delivered (in hospitals or doctors’ offices, nursing homes, at home, etc), regardless of the risk level of medical devices, and regardless of the country of origin. Medical products, whether authorized for the market in the EU, Japan, and the US (the three regional global regulatory systems), or manufactured locally elsewhere, are sold on many markets around the globe and are used in clinical practice on a daily basis. Given the well-known ratio of wealth to healthcare capabilities, medical devices are used in the advanced industrialized countries more than in developing countries. Their use is higher in the old European Union of the fifteen than in the ten new member states.

What follows in the 45 minutes at my disposal is in no way an exhaustive discussion of all ‘significant’ regulatory issues raised by EU regulation of medical products nor is the topic limited to the facilitation of trade. Rather, it is an attempt to sketch a broader set of issues having profound implications of EU regulatory policy for national healthcare systems. Historically, the member states have jealously guarded exclusive authority and control over health protection schemes, healthcare delivery systems, and the development of national health and medical care policy. I don’t need to tell you that in reality, the economic dynamics of the single market is spilling over into healthcare systems and blurring legal boundaries between EU powers and member state powers in impressive ways. The same applies to the rulings by the European Court of Justice which have begun to reach deep into member states’ jurisdiction over national health protection schemes…

Trade policy, which clearly and indisputably is an exclusive EU power, turns into something very different at the national level from what it is at the European level. At the European level, the regulation of medical products is approached as part of trade policy. Once trade policy is translated into the context of nation-states, it touches on several distinct policy areas, which are central to domestic public policy and politics, and which are typically treated as distinct policy sectors. The three distinct policy sectors are: medical and healthcare policy, industrial policy, and public health policy. Examining the interface between these three distinct policy sectors is not the topic of my talk today. Today, I will focus on trade aspects. However, I will approach the topic from a somewhat broader take than economists or lawyers may want to do.

The regulation of medical products matters. But why, how and to what extent?

Briefly and upfront let me say that most of us have rather naïve notions about the safety, quality and efficacy of instruments used in clinical practice. We assume that they are safe and bear no risks once they are approved for the market. However, experience tells us that this is not always the case. For this reason, experts differentiate between three major sources of risks:

1. Known side effects

2. Device Error (by a physician, a nurse)

3. Product Defects (implies the liability of manufacturers)

All three kinds of problems can lead to ‘adverse events’ in clinical practice, that is, they may do harm to a patient, and, more importantly, they may lead to injury or death. Obviously, issues # 1 and 2 are manifest in processes internal to the health care system while issue #3 about product defects are processes external to the health care system.

The issues of risks to humans, risk analysis and risk management are receiving considerable attention by legal scholars (e.g. Christian Joerges, visiting professor at NYU law school and his team at the Institute in Florence). However, there is a problem in that different scholarly communities are addressing different aspects. Joerges et al. focus on EU developments; an entirely different group of scholars is interested in the impact of EU legislation on national law concerning the delivery of health care and medical services while still others concentrate on processes internal to the healthcare systems in each country. Dialogue across all three communities is limited in most member states.

Medical products carry special risks when they are unregulated or inappropriately regulated, wrongly classified, dumped on foreign markets where regulatory requirements may be lower; or they simply can be fraudulent products where actual performance of the product fall short of the promised claims in labeling/naming and advertising. Counterfeit devices are sold and distributed on many markets around the globe. Risks are not only due to product defects but arise when clinicians engage in off-label use of high risk medical devices and wildly experiment with a particular surgical-medical procedure. For all these reasons, medical devices are classified by four risk levels and can be reclassified upwards or downwards. Some of the criteria for classification are duration of contact with the body, the degree of invasiveness, and the anatomy affected by the use of a device, just to mention a few.

The key issue really is this. EU medical device regulation is very recent. It began in 1990 with a first directive addressing heart transplants (AIMD); a second directive adopted in 1993 covered all medical products not covered in the first and third directive. The third directive adopted in 1998 includes all diagnostic products. (Historically, a few countries which regulated IVD products prior to EU regulation did so under the pharmaceutical regime. This raises particularly thorny adjustment problems to EU legislation).

Given that regulation of medical products is entirely embedded in the single market project of 1985 (prior to 1985 only a few countries regulated medical products) the question about ‘a race to the bottom’ is linked to trade, building a Europe-wide market, opening new markets in Russia, China and Latin America, as well as strengthening the interest of companies and increasing profits for shareholders. This is the conventional reaction which many of us have when we learn that an issue is tied up in trade. We expect a ‘downturn spiraling effect’ to the bottom. Alternatively, is the goal of regulation also about minimizing risks to humans and patients? In light of the illustrations I gave earlier, there can be no doubt that the regulation of medical products is, and should be, about protecting people and patients from harm first and about creating a level playing field second.

The argument about a “race to the bottom” is primarily an ideological argument. It may apply in many cases but the question here is: is the argument about a ‘race to the bottom’ in line with empirical evidence? This clash between claim and reality is the puzzle which I will attempt to unravel in this talk.

Let me clarify how I use the concept regulatory policy. I am using it in the tradition of political science rather than according to the legal language of EU regulation. In this language a decision of the Commission and decisions by the European Court of Justice have direct legal effect on those to whom they are addressed, for example, firms, individuals, national governments, regional or local governments. Nor do I use the concept regulation as defined by economists who define regulation in different ways.

I treat regulatory policy as Lowi suggested some 40 years ago. According to him, regulatory policy is a specific type of policy distinct from distributive, redistributive and constitutional policy. Each policy type is assumed to produce a particular pattern of politics. When applied to the European Union, we are dealing with two levels of government: the EU level and the national level. Keeping in mind that directives leave most monitoring and enforcement functions to the member states while retaining regulatory powers of EU institutions for formulating the content and substance of directives, I will argue that regulatory policy plays itself out differently at the two levels of government.

One additional clarification is useful. Rather than viewing EU regulation of medical devices solely as the result of international competitiveness and global trade, as some scholars do, or solely as an attempt by EU member states to influence the content and form of EU regulation which they have to enforce as other scholars do, I treat EU regulation as being the result of the two types of influences—global and national, which impact upon the content and specifications of directives simultaneously.[2]

In my research I follow the OECD Working Party on Biotechnology (1998), which defines regulation as

The exercise of sustained control over an activity by a public agency (and including both authorization of release of products, and post-release monitoring) can be seen as an instrument for reducing the protection of people from risks linked to such uncertainty.

As previously mentioned, EU regulatory policy takes the form of directives, including the medical device policy sector. According to Professor Majone (2001), a leading interpreter of EU regulatory policy-making, the majority of EU regulatory policies take the legal form of directives. The cost of regulation is shifted to firms, individuals, and institutions, and no EU or national taxes are used for implementation.

The enforcement, the monitoring of compliance with EU regulation, the implementation of directives, all remain within the jurisdiction of the member states, in other words, of national officials and offices. This fact alone provides powerful ammunition to argue that too much is made of an unqualified ‘loss of sovereignty’ without due regard to the issue of what policy area and what governmental function is transferred to the EU level. I do not question that regulatory functions have not been transferred to the EU level. I am arguing that under certain circumstances the power of the member states over implementation, and hence the possibility for ‘flexible’ implementation, is equally real. In implementation, member states are found to re-interpret, modify, ignore or block the implementation of EU intentions. Sometimes these domestic strategies and tactics more than compensate for the ‘loss of sovereignty’ for formulating regulatory content.

By endorsing the single market project, it is true the member states voluntarily abdicated regulatory policy-making functions. Given the way decision-making on regulatory policy works in the EU, this is a loss of formal powers but not in actual influence. Why? National officials serve in a dual capacity: they participate in EU-decision-making of the so-called ‘comitology’ and other committees by which regulatory specifications are decided on and, second, they monitor the implementation of EU directives in the member states.

In order to appreciate why the regulation of medical devices matters, and why medical devices raise very sensitive public health issues which justify a higher level of regulatory protection of medical products than normally given to average consumer goods, let us look at the entire range of medical products falling under the category of medical devices, as defined by the medical device-specific EU directives. Three or four directives are directly pertinent to the topic, in addition to the entire spectrum of single market legislation and judicial policy-making by the European Court of Justice.

Medical Device Product Ranges

• Aids for disabled persons (e.g., wheelchair, crutch, standing support, electrical bed, hearing aid)

• Active and non-active implantable devices (stent, cardiac pacemaker, hip implant, neurostimulator, insulin pump)

• Anesthetic/respiratory equipment (oxygen mask, anesthesia breathing circuit, gas delivery unit)

• Orthopedic equipment (knee prosthesis, orthopedic shoe, spinal corset)

• Dental devices (dentistry tools and drills, alloys and resins, dental floss, tooth brush)

• Electro-medical and imaging equipment (x-ray machine, scanner, electrocardiograph, monitor, laser, microscope)

• In vitro diagnostics (devices for clinical chemistry, microbiology, and immunology; genetic tests)

• Ophthalmic equipments (contact lenses, optometer, optical lens, eyeglasses, ophthalmoscope)

• Surgical instruments (scalpel, surgical drill, forceps, tube, drain, sutures, masks)

• Biotechnological products (tissue engineered bone, cartilage and skin)

• Medical disposables (bandages, dressing, syringe)

EUCOMED Medical Technology Brief. Pamphlet A Saving Life...Improving the Quality of Life, 2000b, p.3.

The range of products used in medical practice on a daily basis is huge. The information in the table also suggests why it is important not to confuse medical devices with pharmaceuticals, and why generalizing from the politics of regulating pharmaceuticals to the politics of regulating medical devices may be misleading. The alleged ‘loss of sovereignty’ differs for the medical device regulatory regime and for the pharmaceutical regime, and plays itself out differently in the member states. For example, there is no European-wide regulatory agency for medical devices like the European Medicinal and Evaluation Agency in London, or not yet. Most member states have not created a separate regulatory agency distinct from the ministry of health, except France, Germany and the UK. Most regulatory enforcement functions remain national.

Furthermore, medical products are far more heterogeneous products than prescription drugs. Over 95% of devices are incremental improvements over pre-existing products or medical and surgical procedures. Only 5% are genuinely innovations and medical advances, (Rozynski 2000). Unlike pharmaceuticals, the lifetime of a medical device for exploiting a patent is relatively short (except for medical equipment), 3 to 5 years as opposed to 5 to 20 years for pharmaceuticals.

Why France, Germany and the UK have been the leaders and promoters of EU regulation of medical products is easy to understand if we follow a line of argument and evidence offered by empirical and legal scholarship. For example, Adrienne Héritier (1996:282, the director of the Max-Planck-Project Group, “Common Goods: Law, Politics and Economics” in Bonn) gives four reasons why countries with a high level of regulation prior to EU regulation promote regulation at the European level. Her findings stem from the environmental policy sector but they equally apply to the medical products sector.

(i) Each Member State seeks to impose its own regulatory style and regulatory philosophy on the other Member States in order to reduce the costs of legal adaptation to European legislation.

(ii) At the same time each Member State seeks to present the national industry with a certain regulatory stability because the absolute costs of environmental investments are less important than the stability and predictability of those costs.

(iii) High-regulating Member States seek to establish the same level of regulatory strictness on the European level in order not to jeopardize the competitive situation of their own industries.

(iv) Governmental actors in Member States with high regulatory standards are interested in raising standards on the European level to their own because lower European standards reduce the bargaining power in dealing with industry in the national network.

These explanations are powerful. They are convincing as to why the ‘big three,’ France, Germany and the UK promoted EU medical device regulation along with globally operating multinational companies and the European Commission while the remaining member states initially took a back seat in this policy sector. The three countries have had strong regulatory capacities in the past, that is prior to 1986. According to experts, the EU regulation in this field incorporates policy preferences and regulatory practices of France, Germany and the UK and a few elements arising from efforts at global harmonization.

Let’s have a brief look at an unusual industry. Prior to 1985, these three countries and a few others regulated medical products, but they put weight on different aspects of regulation, for example, Germany and some of the new member states regulated the safety and performance of equipment, France placed her efforts on health and patient safety, and the UK did a little bit of both.

The industry is relatively small. It generates few jobs. The majority of the jobs in the European Union are spread among these three countries as follows: (Germany about 32%, the UK about 14% and France about 10% of total jobs in the European Union). Switzerland ranks fourth in Europe with 8.5% of the European jobs (Altenstetter 2004, ch.2). Worldwide, there are about 700,000 highly skilled and specialized workers, with about half in the EU and half in the US. Japan employs another approximately 7,000 people.

World Medical Technologies and Devices Market Data

Country	Market Size 2000 € billion	% of World Market 2000	Medical Devices Expenditure per Capita €	% Market Growth Rate 2000
EU	41	25.6%	66	5.5%
USA	60	41.5%	125	7 %
Japan	24.5	15 %	116	4 %
Rest of the World	34.5	18%	—	15 %
World	160	100%	—

Source: EUCOMED and AdvaMed (formerly HIMA) estimates that ”parity was assumed between the US $ and € at that time of these calculations.“ In EUCOMED 2001, p.13.

The table on the top leaders conveys an important message that is very similar to what David Vogel called the ‘California effect’ in environmental regulation. A ‘California effect’ manifests itself when a country strengthens its regulatory standards as a result of another’s trade measure. This leads to the raising of regulatory standards rather than the lowering of standards, as is often assumed. Since the EU and the US are the two most important regions with distinct regulatory regimes on medical products–the third being Japan—, and since the healthcare systems in Europe offer laboratories for U.S. subsidiaries, we can easily see why EU trade policy may have had the effect of raising rather than lowering standards over time. I will come back to this issue later.

Top leaders

1. Johnson & Johnson (USA)

2. General Electrics (USA)

3. Baxter International (USA)

4. Tyco International (USA)

5. Siemens (Germany)

6. Medtronic (USA)

7. Fresenius (Germany)

8. Philips (NL)

9. Becton Dickinson (USA)

10. Abbott Laboratories

11. 3M (USA)

12. Guidant (USA)

13. Boston Scientific (USA)

14. Gambro (Sweden)

15. Stryker (USA)

Source: Espicom Business Intelligence (EBI). The World Medical Market Report 2003. Chichester: EBI Publications Ltd.

Unlike the pharmaceutical industry, the medical device industry is highly sectorized and fragmented. Generally, one can distinguish four manufacturing sectors:

• Electro-medical devices and equipment

• Non-electrical products;

• Diagnostic products

• Implants and more recently human cells and tissues

Despite a frenzy of takeovers, mergers, and ownership changes in the late 1990s and

Which is on-going, the structural profile of the industry per se has not changed. 10% of all firms are MNCs while 90% are small and medium-sized companies. This is in striking contrast to the resource-rich, powerful and highly concentrated pharmaceutical industry.

Why does EU regulatory policy on medical devices matter?

I can offer several explanations.

First, with the creation of the single European market a new era in regulatory policy was launched. Historically, and prior to 1985, public health authorities regulated the medical devices primarily in the interest of public health. Now EU regulation is centrally connected to market creation rather than with market correction. In this new era, firms seek regulatory clarity, and governments in countries which have a significant export-oriented medical device industry seek regulatory harmonization as a process of creating and facilitating cross-border, ‘global’ markets for medical devices. This, in turn, allows firms to consolidate and establish market dominance, increase sales volume and maximize profitability.

Second, EU regulation of medical devices matters because the use of these products is at the heart of the delivery of health care at the beginning of life, during all stages of life, and in particular at the end of life. Patients are recipients of medical devices, prescription drugs, and implants and transplanted tissues. Medical devices can incorporate drugs; drugs sometimes need a medical device for delivery to the patient (for example, an instrument to inject a sedative, an asthma inhaler, etc). EU regulation matters, third, because transplants are also medical devices, and medical devices can incorporate human cells, biologically or pharmacologically active substances, or synthetic device-like structures. This product diversity requires regulatory differentiation and stricter regulatory requirements than average consumer products and highly specialized and scientific knowledge and resources.

Fourth, the topic is salient because national payers of health care, that is, public (NHS and NHI) and commercial insurance companies, have become interested in this product sector as a result of rising healthcare costs. The spending for these products has gone up everywhere over the last twenty years. This, in turn, has attracted the attention of the buyers of medical services, budget cutters in the treasury, and healthcare reformers who promote health technology assessment and evidence-based medicine as a cost containment measure in order to contain aggregate spending on health. Pay-for-performance of providers may become the latest cost containment tool in Europe, as it is in the U.S.

Fifth, looking at medical devices from the perspective of risks generated by the use of such products in medical and clinical practice, the topic no longer remains limited to product liability issues and obligations of the manufacturers. The latter are of concern to legal scholars and law firms in Washington and Brussels. But the topic concerns health policy-makers whether they are in government and public administration or in delivery systems. Unfortunately, the channels of communication across these groups are not necessarily open.

The EU directives specifically recognize that ‘adverse incidents’ can have multiple causes and may be due to shortcomings in:

• The device itself

• Instructions for use

• User practices, including training

• Servicing and maintenance

• Management procedures

• The environment in which a medical device is used or stored

• Locally initiated modifications or adjustments

• Incorrect prescriptions or ‘off-label’ use of a device

• Environmental conditions (e.g. the dangers posed to someone with a pacemaker who must undergo security checks at the airport.

The member states control most of these issues; decision-making on some of these topics is overlapping and shared with EU institutions. The end result is an astonishing maze of regulatory details that only legal specialists and a few officials are able to master.

Power Sharing Arrangements Regarding Medical Devices

Policy Issues Within EU Jurisdiction	Policy Issues Within Sole Member State Jurisdiction
Trade	Health care
Commerce	Health and safety
International Competitiveness	Professional and lay users
AIMD, MDD, IVD	Coverage
Blood Safety and other directive	Pricing
Pharmaceutical regime	Reimbursement
Data Protection Directive (95/46/EEC)	Clinical investigation/evaluation
Cross-border health care	Laws on labeling
Professional mobility	Medical institutions & health facilities
Requirements on instructions for use/leaflets	Post-market controls & surveillance
Packaging requirements	Distribution
Advertisements	Installation
EU product licensing through the CE mark	Vigilance
Conformity with standards	Notification/registration for market access
Patent protection	Unregulated medical devices
Parallel trade
Wholesale distribution

How does EU regulation matter, and what is the extent to which it matters?

In the EU’s multi-level governance system, politically sensitive health and safety issues must be approved by the Council of Ministers leaving regulatory details to experts and specialists from national offices, academia and multinational companies, all under the umbrella of ‘comitology’. Regulating in EU committees should not be confused with standard-setting by European standardization organizations, CEN and CENELEC. This process is as obscure as the process centering on ‘comitology.’ It would require a separate research project to map empirically who is representing whom in committees convened by CEN or CENELEC, and who from which multinational company, serves on which of the eleven working groups involved in setting standards for medical devices.

In order to further clarify whether EU regulation is a case of a ‘race to the bottom’, as critics of the EU approach argued at first, or ‘a race to the top’, as we suggested earlier, we need to look at the central features of EU regulation of medical devices, as they initially evolved.

In addition to the “hard” rules (such as regulation, directives and decisions of the Commission), “soft” law and informal rules through Commission-supported guiding documents (so-called MEDDEVs and European NB Recommendations for certification bodies and others) secure EU regulation of medical devices. Guidance documents are legally non-binding except when they specify technical aspects. While guidance documents carry little weight in court, they do spell out the substance and essence of the scientific and technical criteria over which specialists agree, though not necessarily national regulators and legal consultants.[3]

After 1987, all products needed to show the CE-mark based on a conformity assessment procedure issued by a third party, mostly commercial, entities. Contrary to widespread understanding, CE does not stand for the French term of European Community. Rather it means conformity with a particular set of EU rules, norms, and standards.

The CE-mark is the passport to regional and global trade. As of May 1, 2004, it became valid in 29 countries (the EU 25 and four EFTA countries). The same 29 countries use a single nomenclature for medical devices, the Global Medical Device Nomenclature (GMDN). As to the new member states, an industry spokesman[4] estimated that the medical device market in the new member states of the European Union could rise by about 30% by 2010 and accelerate on-going healthcare reform in these countries.

For critics of the EU regulatory regime on medical devices, which initially included France, the U.S. FDA and a few other member states, the CE-mark was and still is an insufficient test of safety and quality for these products. The most spectacular case of faulty or deficient certification of a medical product by a German certification body involved a product for hip replacements manufactured by Sulzer Medica. However, Sulzer Medica is not alone. Empirical evidence suggests that the most prestigious and reputable companies are caught up in investigations by the relevant regulatory agency when one of their products is involved in a hazard, near hazard, or death. Withdrawal from the market is the most dramatic step a regulator can take.

After a ten-year period the critics have conceded that the CE-mark may be sufficient for testing the performance of equipment, machines, and throw-away disposables. But they strongly feel that it is an entirely inappropriate procedure for allowing high-risk medical devices on the market. This is easy to understand. All diagnostic products (used in testing for pregnancy, HIV, and Hepatitis C), all implants (that is, breast, knee and joint implants, and human tissues and cells) are high-risk products. Today after a ten-year period of argument and counterargument, when France was very isolated in the EU, practically all member states now have come around to France’s initial critical position. The member states who promoted not only product safety but also health and patient safety won, despite the initial opposition of the Commission and the industry. The latter argued that upgrading implants to class III (highest risk) would create technical barriers to trade and hence would be in violation of single market legislation. However, the member states and the advocates of public health, who were very much supported by the European Parliament, defeated the industry and the Commission. It is noteworthy that U.S. multinationals provide the greatest volume of highest risk devices.

Even the FDA now agrees that the EU approach to medical devices has considerable merit and is appropriate for 90% of medical devices. However, the FDA is not convinced about the value of all aspects of EU regulation and thus reserves its prerogative to stipulate whether and under which conditions a CE-marked product can enter the U.S.

The major disagreement concerned about 10% of devices in class IIb and III. France, with the support of the UK, was able to introduce in the EU regulatory discourse the tougher requirements for clinical evaluations and evaluating the efficacy of medical devices on clinical outcomes.[5] This position now is endorsed by all member state governments. In July 2003, the European Commission accepted a report submitted by Medical Device Expert Group (MDEG), the most important decision-making body comprising national officials and industry representatives. The report stated that the medical device directives provided an appropriate legal framework for safety aspects and technological evolution. However the report pointed out two structural deficiencies:

• considerable weaknesses in domestic implementation and

• weaknesses in regulatory requirements which demand the further strengthening of weak points in EU regulation.

In other words, the issue of a ‘race to the bottom’ no longer touches solely on the quality of regulatory stipulations and standards. Rather, the key issue which determines whether patients are benefiting from EU regulation lies in weak implementation capacities in the member states. There are considerable “regulatory gaps” in domestic implementation, even in the rich countries which promoted the EU directives a decade ago.

Regulators distinguish between reactive and proactive enforcement. Reactive enforcement tends to be the rule in Europe. Reactive enforcement in the US is triggered by vigilance reports involving a medical product, user reports, information from competitors, and/or networking with overseas regulators. Reactive enforcement extends to all risk classes of devices regardless of the manufacturer’s location (that is, inside or outside the EU) or the location of the notified body that certified conformity with essential requirements (ERs) under the EU regime. In contrast, proactive enforcement means that a regulator may select a manufacturer from the register of self-certificated devices regardless of the location of the manufacturer. Proactive enforcement may start with a document review supplemented by site visits.[6]

State capacities for reactive and proactive enforcement are very uneven across the member states and have become more uneven as of May 1, 2004. It is likely to increase still further when Romania and Bulgaria join in 2007. Clearly, EU-wide rules, statutes, and procedures are a first step to gear up trade with medical supplies and products. However, to turn EU regulatory goals into effective practice on the ground takes more than statutes, norms and rules. There is a need to have highly trained scientific staff which is knowledgeable about the differences between pharmaceuticals and medical devices. There is a need for health human resources, and there is need for financial resources. These are not always available even in rich countries.

From the start when the Commission began to keep implementation scores for each member state, implementation deficiencies in the member states have been viewed as the weakest link in the EU regulatory chain. Types of problems encountered in adverse incident reporting by MDA vary greatly. For example:

In case of endoscopes: third party servicing companies may not always use spare parts from the original equipment manufacturer

• Stretcher lock fails in ambulance crash

• Wheelchairs suffer castor failures

• Radioactive leaks in protective clothing

• Critical care devices: electrical shock from a disconnected device

• Catheter tip fault causes patient death

• Orthopedic implants: suspension of CE mark on a bone substitute product (Sulzer Medica).

For the new members the major challenge is institution-building, creating national capacities and adapting to the acquis communeautaire, a collection of 2,084 EU directives and 22,454 Community regulations. All require transposition into national law, as well as compliance with rulings of the ECJ.[7]

Medical devices were subject to “screening,” a process by which the acquis was examined by the European Commission with each candidate country. The new members face a rather novel situation in that none of them differentiated between a certification body and a so-called competent authority or lead ministry. Building on the practices in Germany, France, and the UK, these differentiations were introduced by EU regulation and binding on the member states. Another concern is how long local certificates issued, for example, in Hungary or Poland for medical devices will be valid under EU law. Under EU law, non-CE-marked products are illegal. Will existing certification bodies become notified bodies automatically without changes? Will they need recertification based on EU law? The answers will probably be affirmative.

The EU Commission, and in particular European trade associations, has held several workshops designed to facilitate the phasing in of the EU regulatory medical device framework. In 2003, a Steering Group comprised of the Commission and equal representation from candidate countries was set up. For the ten new members the key issue now is whether they are in any position to extensively and effectively report on risks and adverse events with medical products, given the manpower and financial resources at hand. Obviously, their priority is the struggle to stop the on-going decline in health in every single former CEE country. This decline can be documented along practically all important health status indicators when compared to the former EU15 (WHO and OECD).

The reason that the concern with the domestic implementation of EU regulatory policy is intimately linked to the issue of ‘a race to the bottom’, results from lessons learned in cross-national field research and practical experience. The key issue is this: is information on risks to patients obtained from what I call the product-regulatory “track,” and mandated by EU directives, communicated to the health care “track”, that is, to those who regulate health professionals and health facilities as well as all providers of healthcare. In my field research I found that the power to regulate health professionals and health sites was rarely in the hands of those in charge of regulating products and the domestic market. Nor are these powers housed under the same public regulatory authority (e.g. France, UK and Germany). Depending on the member state, separate institutional arrangements coexist which are controlled by the ministry of health, trade or yet another ministry. In other countries, regulatory responsibilities partially overlap; and in a few countries, all regulatory responsibilities over products and medical services converge in the national health bureaucracy, as in the Scandinavian countries.

EU trade policy does not pose the greatest threat to lowering standards thus bringing on a ‘race to the bottom’ and its concomitant risks to health and patient safety. The threat is actually twofold: first, the effects of healthcare reform by micro-managing the delivery of care and services in each member state are ominous and far reaching. Second, the weak capability of hospitals, nursing homes, and research centers to live up to EU intentions and standards, as well as their own professional codes of behavior.

The report of the EU expert group, mentioned earlier, did recognize the seriousness of implementation deficiencies in the member states and suggests several ways to improve domestic implementation through:

• Better coordination of post-market surveillance (PMS) and exchange of medical vigilance data, in other words, bringing together the regulatory framework of the products with the regulatory framework for overseeing medical and nursing care. The latter is entirely a member state responsibility.

• Improving the regulatory data base (EUDAMED pilot project, the EU Database for Medical Devices);

• A more consistent approach to clinical investigations;

• Improved conformity and assessment practices by commercial certification bodies;

• More transparency and trust (trust is fine, but ‘to verify’ or monitor is better);

• Linking medical device regulation with pharmaceutical regulation, for example, for combination products;

• Better cooperation between the Commission and member states;

• The creation of yet another “high level committee for medical devices” at the EU level.

Finally, in 2003 several decisions were reached, which merit mention. They add strength and empirical evidence to the argument that EU regulation of medical devices is NOT a ‘race to the bottom’. Rather, it has been and continues to be ‘a race to the top’, that is, EU regulation of medical devices has improved the situation when compared to what existed in individual countries prior to EU regulation on medical devices.

The first decision concerns drug-device borderline products, which have been a contentious issue for over a decade. On December 17, 2003, a decision was made that all borderline and combination products would be treated like drugs and therefore be subject to the stricter pharmaceutical regime in the future. The draft of this decision is in its first reading by the European Parliament.

The second factor is WHO’s proposed new international standard for hepatitis B antigen (HBsAg) intended for use in the calibration of secondary reference materials. Eventually, these requirements will be integrated into the EU regime which has produced several common specifications (CTs) for reagents and reagent products including calibrators and control materials for developing blood groups and the markers for HIV infection. WHO and the Commission are in contact with each other over this issue.

Third, the industry, with the support of Germany, Belgium and Netherlands, has been pushing for E-labeling for medical device and diagnostic instructions for use (IFU). [They are similar to instructions to patients you find when you buy prescription drugs]. This would require amending the wording in all directives, and it would mean that countries whose hospitals and health sites are equipped with limited e-capabilities may be disadvantaged. This certainly would include resource-poor healthcare systems in the Mediterranean basin and the new Members.

A fourth decision promises a more a consumer friendly orientation in regulation. The Commission is supporting initiatives to make the EU evaluation and registration system more transparent to consumers.

Finally, a new regulation will enter into force on July 1, 2004, which sets the conditions for taking action against any goods that infringe upon intellectual property rights (IPRs) including patents, other protections, country of origin, counterfeit and pirated goods, and other materials infringing against trademarks. Obviously, this last point serves trade interests rather than public health or healthcare interests.

Conclusion

In conclusion, while in the early 1990s, trade-related concerns dominated the EU agenda on medical devices, over a period of fifteen years public health, health and patient safety concerns have moved to the forefront of concerns and exchange of information across the member states and between them and the Commission. Although the EU regulatory regime on medical devices has introduced a great deal of regional harmonization and convergence and has raised the quality and safety standards upwards in the European Union, at least in terms of law, there are considerable ‘regulatory gaps’ in domestic implementation. In addition, national practices and capacities considerably diverge, depending on the product sector and across all member states. This divergence without any exaggeration has substantially increased since May 1, 2004.

[1] This lecture draws on my on-going research and publications, for example, EU and Member State Medical Device Regulation. International Journal of Technology Assessment in Health Care, Vol. 18, # 4, December, pp.1035-1055. International Collaboration in Medical Device Regulation: Issues, Problems, and Stakeholders. Prepared for the UNRISD international conference on Commercialization of Health Care: Global and Local Dynamics and Policy Responses, Helsinki, Finland, March 15-17; Bridging EU and Member State Implementation: The Case of Medical Goods, In Vitro Diagnostics and Equipment. In: The Europeanization of Health Policies edited by M. Steffen, W. Lamping and J.Lehto. Routledge (forthcoming).

[2] I do not share the theoretical differentiation offered by Scharpf (1999) and others (e.g. Héritier who argue that one has to decide on one or the other interpretation).

[3] Other “soft laws” include resolutions which are usually of a political nature, communications which have no legal effect, guidelines which cover technical topics published by the Commission that must be complied with when directives are implemented, and recommendations and opinions which have no legal force.

[4] Philippe Verdonck of Baxter, speaking on behalf of the European medical devices industry at the AdvaMed International Conference in Washington, DC, September 2003.

[5] It is possible that when Sweden, Finland and Austria joined in 1995 this helped to redraw power relations in the Medical Device Expert Group (the most important group consisting of national officials and Commission officials) and possibly in the eleven working groups in favor of France.

[6] Interview with Alan Kent, the former head of the British regulatory agency in October 2001.

[7] SIGMA 1998, p. 75.