I’ve taken a look at the new EU medical device regulation including the clinical evidence required, the postmarketing data needed and the proposed device database and how it might be interpreted.
Carl Heneghan
The International Consortium of Investigative Journalist (ICIJ) has published the “Implant Files.” Their global investigation, which I’ve been a part of, reveals the problems with weak regulations and dangerous devices – many patients have been seriously harmed, some have died.
Simon Bowers important article on ‘How Lobbying Blocked European Safety Checks For Dangerous Medical Implants‘ quotes me saying: “the new device regulations give the impression of creating safer devices, but more regulation does not necessarily mean better regulation.” He said they amounted to “a hundred pages of smokescreen”.
Regulation (EU) 2017/745, published on 5th April 2017, aims to improve the approval and post-marketing of medical devices. This new EU regulation for medical devices (MDR) makes many changes to the previous directives and fully come into force on May 26th 2020. However, does it solve the structural problems with past regulations, does it protect patients and improve safety, or, does it provide more work for manufacturers, more work for notified bodies and false reassurance – is the MDR a smokescreen?
(the bracketed numbers refer to the relevant regulations or articles within the MDR)
Clinical evidence required
To ensure a high level of safety and performance, class III devices, as a general rule, should be sourced from clinical investigations that have been carried out under the responsibility of a sponsor (point 63). The rules on clinical investigations should be in line with ISO 14155:2011.
Article 2 set out the definitions: clinical evidence means clinical data and clinical evaluation results of a sufficient amount and quality to allow a qualified assessment of whether the device is safe and achieves its intended clinical beneficial (2.51) The sponsor also should submit a summary of the results with the clinical investigation report. If this is not possible – which will happen – then the sponsor has to justify the reason as to why it hasn’t. (70)
Chapter VI. Article 61 sets out the requirements of these clinical evaluations and clinical investigations. Most of the requirements for evidence fall upon the manufacturer to ‘specify’ and ‘justify’, and still permits the use of ‘equivalence’, which was a significant contributor to the metal hip and surgical mesh design faults that caused many harms.
61.3. A clinical evaluation shall follow a defined and methodologically sound procedure based on the following:
(a) a critical evaluation of the relevant scientific literature currently available relating to the safety, performance, design characteristics and intended purpose of the device, where the following conditions are satisfied:
—it is demonstrated that the device subject to clinical evaluation for the intended purpose is equivalent to the device to which the data relate, in accordance with Section 3 of Annex XIV, and
— the data adequately demonstrate compliance with the relevant general safety and performance requirements;
The MDR allows manufacturer exceptions to clinical investigations if the device is a modification of a device already on the market made by the same manufacturer, or if the devices are equivalent in accordance with Section 3 of Annex XIV, (61.4) or the manufacturer has a contract for ongoing access to the technical documentation of another manufacturers device. (61.5)
61.5 A manufacturer of a device demonstrated to be equivalent to an already marketed device not manufactured by him, may also rely on paragraph 4 in order not to perform a clinical investigation provided that the following conditions are fulfilled in addition to what is required in that paragraph:
—the two manufacturers have a contract in place that explicitly allows the manufacturer of the second device full access to the technical documentation on an ongoing basis, and
—the original clinical evaluation has been performed in compliance with the requirements of this Regulation,
and the manufacturer of the second device provides clear evidence thereof to the notified body.
Devices already on the market require ‘sufficient clinical data’ (61.6) to be approved under the new MDR.
61.6. The requirement to perform clinical investigations pursuant to paragraph 4 shall not apply to implantable devices and class III devices:
(a)which have been lawfully placed on the market or put into service in accordance with Directive 90/385/EEC or Directive 93/42/EEC and for which the clinical evaluation:
—is based on sufficient clinical data, and
—is in compliance with the relevant product-specific CS for the clinical evaluation of that kind of device, where such a CS is available;…
It is not clear what sufficient means; it is not clear what ‘as a general rule’ refers to; it is not clear what ‘adequate data’ is; it is not clear what a ‘sound procedure’ is; it is not clear…. I think by now you should have got my point. The language is vague and therefore open to interpretation. This lack of clarity allows a race to the bottom for notified body approval, and it may take time to recognise this problem within the MDR.
Notified bodies assessments of a manufacturers documentation on clinical evaluation, should be critically evaluated by the authority responsible for notified bodies. (51) However, this request is based on only a sampling of the documentation. What happens if some of that documentation is found to be inadequate is not clear.
For implantable devices, manufacturers should summarise the main safety and performance aspects of the device and the outcome of the clinical evaluation in a document that should be publicly available. (48) This sounds like a good point. However, evidence can be omitted leading to problems of publication bias. There is, no requirement to publish all data whether positive or negative: harmful data can go unreported.
An electronic system is also supposed to be set up to record every clinical investigation and report it in a publicly accessible database (67). To achieve this, you could require clinical trials that comply with the EU trial directives and, therefore, require registering and reporting of results. However, as we’ll see later, this publicly accessible database may never happen.
Postmarketing evidence
Manufacturers are required to actively gather information from post-marketing experience and update their technical documentation accordingly. (74) They are also required to play an active role during the post-market phase by systematically and actively gathering information from post-market experience with their devices in order to update their technical documentation and cooperate with the national competent authorities in charge of vigilance and market surveillance activities.
I consider there is only one way to do achieve this; through the use of national registries that necessitate the collection of the data. Otherwise. We’ll see the same inconsistent, uncoordinated approach to post-marketing that has plagued the past and rendered this regulatory requirement largely ineffective.
Eudamed (Article 33)
- The Commission, after consulting the MDCG, shall set up, maintain and manage the European database on medical devices (‘Eudamed’) for the following purposes:
(a) to enable the public to be adequately informed about devices placed on the market, the corresponding certificates issued by notified bodies and about the relevant economic operators;
Eudamed is a big ask and a big task to deliver on. The functional specifications have to be drawn up, and an implementation plan and to be fully functional by 25th March 2020. (34.1) But, if circumstances prevail that could not reasonably have been foreseen that prevent Eudamed being fully operational on 26 May 2020, then the date will move to six months after the date of publication of a notice in the Official Journal of the European Union.
And until Eudamed is fully functional, the corresponding provisions of the previous Directives shall continue to apply. (Article 123). So, if a new fully functioning database is not forthcoming, we’ll default to the old regulations. On May the 26th, in 2020, will l be writing about the lack of a device database – highly likely! – the dates in my diary
Notified Bodies and the experts
Notified bodies should be strengthened including their rights to carry out unannounced on-site audits and to conduct tests on devices to ensure continuous compliance (52) This point addresses some of the issues that led to the PIP breast implant scandal, which went unnoticed by the regulators: the change from medical to cheaper industrial grade Silicone could have been detected with unannounced audits and testing of the silicone implant. No mention though of how often, how they should be reported and what happens if there are problems found.
Implantable devices should, except in certain cases (not clear what the exceptions are) be obliged to request expert panels to scrutinise their clinical evaluation assessment report (56) This point aims to address the lack of medical expertise in notified bodies. It is also not clear to me what constitutes ‘certain cases,’ and whether the panels will be impartial in their assessments – I suspect they won’t, not least because a manufacturer should be able to consult voluntarily an expert panel for clinical evaluations, investigations or its development strategy. (57) Conflicts of interest could also have a significant role in approval decisions because of the need for experts.
Article 15 sets out the necessary skills required of the person responsible for regulatory compliance within a company. It ain’t much: the person responsible requires a diploma, certificate or other evidence of formal qualification in law, medicine, pharmacy, engineering or another relevant scientific discipline and at least one year of professional experience in regulatory affairs or in quality management systems relating to medical devices. Not much then.
What can we conclude
Sloppy language within regulations leads to different interpretations, which can lead to a race to the bottom. Discussions turn to the minimum level of evidence required for regulatory purposes: what can we get away with. These discussions are already underway. The threat of unannounced audits and the potential need to place information in the public domain means companies will have to provide more paperwork- the MDR will create more work. My thoughts are the Eudamed won’t come into force within the timeframe required. But, if you think changes are going to be made to this new MDR, then think again, it will be 2027 before they are evaluated.
BMJ Evidence-Based Medicine – original evidence-based research, insights and opinion
Read more in the Welcome to BMJ Evidence-Based Medicine Editorial.
Competing interests
Carl Heneghan participated in the ICIJ investigation but received no payments. Carl is Director of the Centre for Evidence-Based Medicine at the University of Oxford, and Editor in Chief of BMJ Evidence-Based Medicine. He has received expenses and fees for his media work including BBC Inside Health. He holds grant funding from the NIHR, the NIHR School of Primary Care Research, The NIHR Oxford BRC and the WHO. He has also received income from the publication of a series of toolkit books. CEBM jointly runs the EvidenceLive Conference with the BMJ and the Overdiagnosis Conference with some international partners which are based on a non-profit model.