Intended for healthcare professionals


Commentary: Europe needs a central, transparent, and evidence based regulation process for devices

BMJ 2013; 346 doi: (Published 07 May 2013) Cite this as: BMJ 2013;346:f2771
  1. Michaela Eikermann, head of department of evidence based medicine1,
  2. Christian Gluud, head of department2,
  3. Matthias Perleth, chairman3,
  4. Claudia Wild, director4,
  5. Stefan Sauerland, head of department5,
  6. Iñaki Gutierrez-Ibarluzea, knowledge manager6,
  7. Sunya-Lee Antoine, research associate1,
  8. Jacques Demotes-Mainard, project coordinator7,
  9. Edmund A M Neugebauer, director1
  10. On behalf of signatories of our open letter to the European Union*
  1. 1Institute for Research in Operative Medicine, Faculty of Health, School of Medicine, Witten/Herdecke University, D-51109 Cologne, Germany
  2. 2Copenhagen Trial Unit, Centre for Clinical Intervention Research, Copenhagen University Hospital, Copenhagen, Denmark
  3. 3German Society for Technology Assessment in Health Care, Berlin, Germany
  4. 4Ludwig Boltzmann Institute of Health Technology Assessment, Vienna, Austria
  5. 5Non-Drug Interventions, Institute for Quality and Efficiency in Healthcare (IQWiG), Cologne, Germany
  6. 6Osteba, Basque Office for Health Technology Assessment, Vitoria-Gasteiz, Spain
  7. 7INSERM, Institut Thematique Sante Publique, Paris, France
  1. Correspondence to: E A M Neugebauer Edmund.Neugebauer{at}

Last September, the European Commission published proposals to update regulations for medical devices in order to improve patient safety.1 The proposals are being discussed by the European parliament where critical debate is being led by the Committee on Environment, Public Health and Food Safety.

Medical devices range from bandages to life support machines, and manufacturers classify them into four risk categories from low (such as urine drainage bags) to high risk (such as drug eluting cardiac stents) according to EU rules.2 The risk associated with the device depends on the duration of contact with the body, invasiveness, and whether it has a local or a systemic effect. Medium and high risk devices must be certified by one of the notified bodies, organisations that are accredited to assess a product’s compliance with EU legislation (CE mark).

At a first glance this seems a structured process. But examples such as the withdrawal of Poly Implant Prothèse (PIP) breast implants show that the process cannot guarantee safe and effective patient care.3 The certification process is inconsistent, opaque, and operates in the interests of manufacturers. It requires insufficient evidence of efficacy of the devices and no long term follow-up of patients.4 5

Campbell and colleagues’ analysis of the problems with the EU regulations points out the usefulness of registries for collecting postmarket surveillance data.4 They criticise the current EU proposals for failing to include measures to ensure complete and transparent documentation and exchange of the data. Storz-Pfenning and colleagues argue that the new proposals do not sufficiently strengthen the requirement for reliable evidence before market access.5 They recommend that regulations should require better evidence before new technologies are approved or that use of new devices after approval is restricted to trials.

Last month a group of experts, including ourselves, submitted a petition to the European Commission, European Parliament, and European Council asking them to enforce the rigorous clinical evaluation of medical devices.6 We set out three main requirements for an updated European regulation process:

Centralisation of the regulation process and independent assessment—Under the proposed legislation manufacturers of devices will still be free to choose to which of the 84 notified bodies they submit their application. We believe that approval of high and medium risk devices (category III and IIb) as well as in vitro diagnostic devices should be done by a new public body similar to the European Medicines Agency or that the EMA is given an extended mandate to carry out these assessments.

Requirement for high quality evidence of patient relevant benefits—Currently, there is no requirement that approval of high and medium risk devices should be based on high quality evidence of benefits that are relevant to patients. We recommended that patient safety should be improved by requiring assessment of short and long term benefits and harms in well designed randomised clinical trials and other high quality clinical studies. Post-marketing surveillance should also be compulsory to ensure that benefits and harms of the device in real world settings are similar to those shown in clinical trials.

Transparency of the assessment process and results—Currently, there is no publicly available summary describing the basis for granting a CE mark. Members of the public therefore cannot find out information about the process (place of application, requirements for application, pre-defined criteria for decisions) or the rationale for an approval (efficacy and safety data). We called for this information to be available on a publicly accessible website for all approved devices, irrespective of risk class. The European Databank on Medical Devices (Eudamed) should be publicly accessible and should include all relevant information concerning assessment, including data on safety, performance, and incidents. With regard to vigilance and market surveillance, clear criteria for reporting of and reacting to incidents have to be established and immediately reported to Eudamed to provide reliable statistics on incidents and their consequences.


Cite this as: BMJ 2013;346:f2771



View Abstract