The International Medical Device Regulators Forum (IMDRF) has opened for consultation until 1 December a third document on examining the use of registries to support medical device regulatory decision making.
Registries, which can capture outcomes associated with medical device use, were identified by IMDRF in 2014 as having "a significant gap" in optimal use for regulatory decision making.
This gap led to the creation of IMDRF Registry Working Group that produced two documents to guide alignment and use of registries to generate data for regulatory decision making: (1) Principles of International System of Registries Linked to Other Data Sources and Tools, (2) Methodological Principles in the Use of International Medical Device Registry Data.
Regulatory decisions based on registries can include: primary device approvals, expanded/broadened indications, post-market studies, post-market surveillance, development of objective performance criteria (OPCs)/performance goals (PGs) and device tracking/ field safety correction actions.
Tool for Assessing Usability of Data
IMDRF says the purpose of this third document is to provide advice to device regulators and registry organizations on the use of registry-generated data in support of regulatory decisions.
This document identifies key processes and features to be considered in assessing the usability of registry data for regulatory purposes, encompassing both (a) data produced by registries and (b) data produced by linkability to other sources (including other registries) to enrich the evidence available for regulatory decision-making.
"The authors of this document recognize that data produced by a registry may be suitable for making one type of regulatory decision but not others. Individual country regulators are expected to both (a) assess independently the suitability of registry-generated data for regulatory purposes and (b) decide what actions to take based on applicable national and regional regulations," the document says.
For instance, IMDRF says that using registry data to win approval of a primary indication for a device "might require accurate and reliable patient data capture, using robust study designs, at clinically relevant time intervals throughout the appropriate portions of the device lifecycle, and data should be analyzed with appropriate statistical methods for addressing the pertinent scientific questions relevant for the decision making. On the other hand, registry-generated data not meeting requirements for primary indication might be able to support broadening the indications for use of already approved devices. Further, registry data may serve as a postmarket control suitable for providing ongoing information for safety surveillance and for effectiveness."
The Forum also highlights the need for appropriate transparency through the establishment and continuous maintenance of a public website that can describe the aims of the registry, explain how to participate and disclose how the registry is funded.
In terms of the variables necessary to satisfy regulatory purpose or decision making, IMDRF says that in general, the list "should include demographic factors, medical history/co-morbidities, procedure/device information, operators/physicians, follow-up information and outcomes of interest."
And to be used for regulatory purposes, IMDRF says registries should demonstrate their ability to detect:
- The overall percentage of patient exposure to a device and representativeness of the registry population to the treated population;
- The extent to which exposed patients within the scope of the registry are actually consecutively captured (i.e., minimization of selection bias);
- Extent of follow-up available at important durations of time following the index procedure (which IMDRF says, if inadequate, ability to link to additional datasets may potentially be a good surrogate);
- Qualifications of data entry personnel if direct data quality/validation is not possible;
- Adherence to source verification procedures and data collection and recording procedures for completeness and consistency;
- Completeness (i.e., minimized missing or out of range values) of data necessary for specified analyses, including variables required for adjustment/confounding factors;
- Data consistency across sites and over time;
- Evaluation of on-going training programs for data collection and use of data dictionaries at participating sites;
- Evaluation of site and data monitoring practices.
Tools for Assessing the Usability of Registries in Support of Regulatory Decision-Making