Cindy Henrich, Regulatory Specialist
Clinical Evaluation Reports (CERs) are required to support the CE marking of medical devices in the European Community and should be conducted in accordance with Annex X to Directive 93/42/EEC or with Annex 7 to Directive 90/385/EEC and after 2020 will need to comply with MDR. They are also required to support regulatory submissions in other worldwide geographies, i.e., Canada, China, and Australia. The clinical evaluation report is “a methodologically sound ongoing procedure to collect, appraise and analyse clinical data pertaining to a medical device and to evaluate whether there is sufficient clinical evidence to confirm compliance with relevant essential requirements for safety and performance when using the device according to the manufacturer’s Instructions for Use.”3
The information resulting from literature searches is integral for establishing the safety and performance of the device and to inform the current knowledge and/or state of the art (SOA) sections. In turn, these two sections should integrate and support the evaluation and synthesis of other aspects of a CER; clinical data, worldwide post-market surveillance (e.g., sales, complaints, field actions, CAPAs), and risk assessments, to reach conclusions about the clinical safety and performance of the device.
But how does one get the necessary data? According to MEDDEV 2.7/1 rev 4, one should determine “The current knowledge/state of the art in the corresponding medical field, such as applicable standards and guidance documents, information relating to the medical condition managed with the device and its natural course, benchmark devices, other devices and medical alternatives available to the target population.”3
Getting the Most from Literature Searches
In the CER scoping step (Stage 0), first develop a literature search protocol which enables planning the search before execution of the search. This not only helps clarify issues, but aligns with the current guidelines requiring that the search strategy be thorough, objective and reproducible as well as documented and justified.
Additionally, the guidelines state, “the clinical evaluation should be conducted by a suitably qualified individual or a team.” 3 The CER writer is most likely not the subject device expert. This is when one needs to employ the team. After a thorough reading of the available information such as technical files, on-line information, and other data supplied by the manufacturer, convene a fact-finding meeting with the CER team and other clinical experts. This “meeting” may be conducted via email, phone meetings or in person- depending on the complexities of the device.
During the planning stage one should:
• Obtain a good description of the device, its different names, other products within the line of devices, and any predecessor devices
• Determine if the CER is for pre-market, post-market, PMS or another objective.
• Determine which databases to use. Note, that “databases” is plural here; the guidelines state, “a comprehensive search strategy is required, normally involving multiple databases.”3 These databases must provide adequate coverage of devices and therapies in use in Europe and identify relevant clinical trials and publications. The most commonly used databases are: PubMed, Embase, Google Scholar, and guidelines from pertinent societies and agencies.
• Determine date ranges for the searches. The SOA date range may not be the same range as for safety and performance measurements.
• Searches for devices with previous CERs should start about 3-6 months before the end date of the last search.
For instance, if the previous CER literature search ended 01-Jun-2015, the current search might start on 01-Feb-2015. This approach should catch any publications that weren’t identified in the search engine at that time.
• Because many previous CERs were based on searches from only one database, the date range for any new database queried may need to reach back further than the previous CER searches.
• In cases of a first CER, the time span may need to be longer, such as when the device or its predecessor was first utilized or studied.
• Typically SOA searches go back three to five years. In the case of the initial CER, five years is fairly standard, but for updates it can be much shorter. This depends on the rate of change in this field, the length of time the device has been on the market, the technology and other issues.
• Safety and performance date ranges are determined by the date of the last search, when the subject device was first marketed or developed, and manufacturer preferences.
• Always use an end date so the search can be replicated.
• List the target device’s indication(s) and population.
• Determine the performance expectations, acceptable risks, benefits and side-effects, and claims of the device or therapy.
• List comparator methods or devices. For instance, if the targeted device is over-the-counter bandages, list other methods used to close a wound: stiches, liquid bandages, or no action, etc.
• Determine if there are new developments in the area, new data or new knowledge as to potential hazards and risks, previous literature reports (e.g., pre-market evaluations or possible guidelines from professional societies), and what data the manufacturer already holds.
• Determine the selection criteria to be applied to published literature and document this justification for these choices along with strategies for addressing potential duplicate data across multiple publications.
• Define data management practices to ensure data integrity during extraction (e.g. quality control/second review of extracted data by additional reviewer).
• Define the appraisal plan, which defines the methods for appraising each publication, including the relevance of the data to the intended clinical use and the methodological quality of the data (see Appendix of MEDDEV 2.7/1 revision 4).
• Define the analysis plan, which defines the methods for analyzing the data including data processing and transformation.
Proof the Plan
Once the plan is developed, ask that the clinical experts review and comment on the search protocol and resultant search terms. Document these decisions in your search protocol because this information will inform your literature search terms. Conducting a few experimental searches will help determine if you are on the right track and if the terms need to be refined.
Once the searches have been executed, the adequacy of the searches should be verified and a literature search report should be compiled to present details of the execution, any deviations from the literature search protocol, and the results of the search.
The most effective literature searches to support CERs start with planning. Establish a protocol for the search, just as you would for any research endeavor. This plan should include not only expected items for a literature search such as terms, time frames and analysis methods, but also the composition of the team for writing and reviewing the CER. Ensure you have the right team assembled to participate in the development, interpretation and summarization of evidence.
If you have questions about writing CERs or conducting literature searches, contact Samantha Spence at firstname.lastname@example.org or 952-224-2260 to be put in touch with an RCRI expert.
Directive 93/42/EEC: COUNCIL DIRECTIVE 93/42/EEC OF 14 JUNE 1993 CONCERNING MEDICAL DEVICES
MEDDEV 2.7/1 revision 4, June 2016; CLINICAL EVALUATION: A GUIDE FOR MANUFACTURERS AND NOTIFIED BODIES UNDER DIRECTIVES 93/42/EEC and 90/385/EEC.