International collaboration in the use of registries for new devices and procedures

The evaluation of new medical devices and interventional procedures (including surgical operations and transluminal interventions) is more challenging than the evaluation of new drugs. Fewer regulatory demands mean that the evidence base for devices is typically much more limited than for pharmaceuticals. There are no legal frameworks for regulation of procedures, although some countries have systems for producing guidance on their use1. The fact that some devices require novel procedures for their use, and that some novel procedures may be done using a variety of different devices whose design may be evolving, complicates the issue still further. Postmarketing surveillance of devices is variable and may fail to provide timely and reliable information about the incidence of serious adverse events2. Diffusion of procedures in the absence of good evidence increases the risk of poor outcomes for patients, and may have damaging effects on health services by disturbing established care pathways, altering use of resources and adding unexpected requirements for training3. Gathering good observational data would enhance the typically poor evidence base related to devices and procedures, and allow better planning of their use4. This is a well recognized need but has been achieved only in limited areas and on a limited scale5,6. International collaboration in collection and collation of data would be the ideal, to accrue information about real-world use in the shortest possible time. For international initiatives to be successful, there needs to be a shared view about which devices and procedures to prioritize for the allocation of resources. Registers need to be well conceived, with an agreed common purpose and compatible data items. This demands close links between clinicians in the countries involved (typically through their professional organizations) and significant infrastructure for support. Regulators such as the Medicines and Healthcare products Regulatory Agency in the UK and the Food and Drugs Administration (FDA) in the USA, and guidance-producing organizations such as the National Institute for Health and Clinical Excellence have considerable interest in the organization and success of registers. Some surgical specialties have already established registers that cross national boundaries, such as the European Cardiac Surgical Registry7. In the USA, the FDA recently held an exploratory meeting of those supervising joint replacement registers in several countries and is developing the capacity to collate data from these (MDEpiNet mission and strategic goals). The European network for health technology assessment (EUnetHTA) has developed the EVIDENT database (http://www. eunethta.eu), which aims to support collaboration between health technology assessment agencies and reduce duplication of work on additional evidence generation, while the Agency for Healthcare, Research and Quality in the USA is developing a ‘Registry of Patient Registers’8. Postmarket surveillance data from manufacturers could also be a valuable resource. Manufacturers hold unique information about all the devices they have sold and they often span national boundaries. For manufacturers’ data to be sufficiently useful and reliable for use in national assessments, their registers would need to be overseen by independent health professionals, scientists and others free from conflicts of interest, with all data on all patients available. Clinicians play a fundamental role in all stages of register development. Their engagement is vital to achieve comprehensive and valid submission of data9. Features of any register that help secure this engagement include a small data set, a ‘user friendly’ web-based data entry system that allows easy data submission, and easy retrieval of their own data for local audit and review of practice. Some registers (for example for joint replacements) are likely to continue in perpetuity. However, for any register there should be a clear plan about how long data submission will continue, in order to fulfil the specific purpose for which it was established. The burden of data submission and long-term followup may be reduced, and the likelihood of capturing certain outcomes may be enhanced, by linking registry data with other healthcare databases and national statistics systems. Such linkages may complicate data governance and international collaborations, but these issues will