Regulatory Open Forum

 View Only
  • 1.  Clinical Evaluation, article screening, Class IIa

    This message was posted by a user wishing to remain anonymous
    Posted 26-Jan-2022 16:56
    This message was posted by a user wishing to remain anonymous

    I am wondering how many articles people are generally reviewing in CERs for Class IIa devices. Thank you!



  • 2.  RE: Clinical Evaluation, article screening, Class IIa

    Posted 27-Jan-2022 04:11
    There is no magic number. It really depends on your product and the amount of articles that you will find with your key words.

    ------------------------------
    Anne-Sophie Grell
    RA lead
    Antwerpen
    Belgium
    ------------------------------



  • 3.  RE: Clinical Evaluation, article screening, Class IIa

    Posted 27-Jan-2022 07:58
    Edited by Sara VanWyk 27-Jan-2022 07:58
    Hello Anon,

    I would agree with Ms. Grell; there is certainly no magic number, and your total screening obligation can depend on a variety of factors, such as the history of your device and device technology as well as the methods (e.g., key words and phrases) used to detect relevant articles. In a recent article on the notified body perspective (linked below), it is recommended that searches for the device under evaluation (DUE), equivalent device, and device technology (i.e., state of the art; SotA) be described separately. As such, the RQM+ team has been developing separate research questions (i.e., PICO tables) and, accordingly, search strings for each of these (DUE and SotA) search efforts. I generally expect the screening obligation to vary by search effort, with the SotA search often returning more articles than the DUE search; although, this can depend on the market history of the DUE. To my knowledge, RQM+ has not received any notified body feedback on the number of articles returned by either search; however, we have responded to feedback regarding the search string justification, where there was concern that the results of a search were not comprehensive. To avert this type of feedback, the RQM+ team validates its search strings, as a best-practice, using gold standard articles (identified by previous reports or by scoping searches of full-text databases). In doing so, we equip notified body reviewers to accept the search result by consideration of our methods, rather than number of returned articles (so long as the number returned isn't suspiciously low, given what is known about the device); these methods are generally applied across literature search efforts, regardless of device risk classification.

    Article link: https://www.bsigroup.com/globalassets/meddev/localfiles/en-gb/documents/bsi-md-jmdr-feb-2021-en-gb.pdf

    ------------------------------
    Sara VanWyk, MPH, CCRP, RAC, MWC
    Principal Regulatory Scientist, RQM+
    ------------------------------



  • 4.  RE: Clinical Evaluation, article screening, Class IIa

    Posted 28-Jan-2022 08:28
    I would disagree and say we do have "magic numbers" for clinical data searches in CERS. Basically "zero" means you need to work harder and ~200 means we need to work smarter. For literature searches, look to published meta analyses and systematic reviews for examples. For complaints, use grouping and triage. For trial data, follow the protocol when analyzing ARs and other clinical data.

    ------------------------------
    Joy Frestedt PHD, CPI, RAC, FRAPS, FACRP
    President and CEO
    Frestedt Incorporated (www.frestedt.com)
    Saint Louis Park MN
    United States
    612-219-9982
    jf@frestedt.com
    ------------------------------