Regulatory Open Forum

 View Only
  • 1.  CER Literature Searches

    Posted 21-Dec-2020 19:03

    Hi Everyone, I'm wondering if anyone has insight on limiting the results of a literature search and how to determine when the search is too limited versus too broad, or when it can be limited? 



    ------------------------------
    Elizabeth Tuite
    RA/CA Technical Writer
    Hopkinton MA
    United States
    ------------------------------


  • 2.  RE: CER Literature Searches

    Posted 22-Dec-2020 01:15

    Hi Elizabeth

    I don't know if there are general rules... there's definitely a level of judgment involved in choosing a "right" focus. If you haven't read it already, try MEDDEV 2.7/1. There's a nice compilation of suggestions in the appendices that might give you some inspiration.



    ------------------------------
    Anne LeBlanc
    Manager, Regulatory Affairs
    United States
    ------------------------------



  • 3.  RE: CER Literature Searches

    Posted 22-Dec-2020 04:43

    Hello Elizabeth,

    Agree with Anne there is definitely a level of judgement/adjustments which are made during a clinical literature review.  The advice maybe to provide is ensure you establish a good Clinical Evaluation Plan with search strategy or a literature search protocol with detailed information on how to the literature search is made.  The better establishes, the closer you will get to a nice balance of articles.  This of course all depends on the type of device because well established products which have been on the market for a long time usually will not find many results from the device under evaluation.  The important aspect too for any device under evaluation or state of the art search is if changes are made to the inclusion/exclusion, to search terms, to search strategy, document these well in the CER.  We usually have a separate section in the CER which is identified as "deviations" where changes to any part of the search strategy is documented.  Then when the next literature search is conducted there is more refinement of the process.  As Anne mentioned, the MEDDEV has some information, but there are also some nice sources out there helping with literature search strategies.



    ------------------------------
    Richard Vincins RAC
    Vice President Global Regulatory Affairs
    ------------------------------



  • 4.  RE: CER Literature Searches

    Posted 22-Dec-2020 09:07

    Searching is certainly a balancing act - making sure you've gotten the appropriate articles while at the same time limiting those that are not relevant.

    Some things to keep in mind.  If every article retrieved is relevant then you probably want to broaden the search.  You want to have some non-relevant articles to ensure you've gotten all the relevant ones.  It's along the lines of removing cancerous cells - you want to ensure you've removed all the cancer so you take away some healthy tissue.

    It also helps to have some 'known' articles going into the search - articles you know will answer the question.  If your search retrieves ALL the known articles then you can feel fairly confident the search is getting what you need.  If you are missing ANY known article you need to figure out why and adjust the search accordingly.  You need to find the known articles in EACH database searched.  

    I'd also encourage you to enlist the assistance of an information professional for literature searching, similar to using a biostatistician to analyze research data.



    ------------------------------
    Linda Hartman
    Clinical Research Specialist
    Monroeville PA
    United States
    ------------------------------



  • 5.  RE: CER Literature Searches

    Posted 22-Dec-2020 10:13

    Thank you Linda, for this feedback. It is very helpful!

    -Liz



    ------------------------------
    Elizabeth Tuite
    RA/CA Technical Writer
    Hopkinton MA
    United States
    ------------------------------



  • 6.  RE: CER Literature Searches

    Posted 23-Dec-2020 02:11
    Edited by Stephanie Grassmann 23-Dec-2020 15:21

    Dear Elizabeth

    For literature searches for most products, especially well established technologies, limiting the publication year to the last 10 years is reasonable. Older than 10 years are considered to be common/published knowledge (i.e. textbooks). 

    To limit the searches further and depending on the medical device application, focus strictly on "human studies" using e.g. the words "patient", "human", "subject", etc or using the words "in vitro" or "animal" etc. as exceptions (in the case that the medical device is for direct human use only).

    There are guidances, e.g. MEDDEV 2.7/1 which provide further information and assistance in terms of conducting searches. 

    During the literature search exercise, I would ensure that the key terms within your intended use are covered/ included in your search phrases or search results.  

    Best Regards,

    Stephanie



    ------------------------------
    Stephanie Grassmann
    Managing Director of MedTechXperts Ltd
    Biberstein
    Switzerland
    ------------------------------



  • 7.  RE: CER Literature Searches

    Posted 23-Dec-2020 13:43
    Edited by Linda Hartman 23-Dec-2020 13:43

    Caution when using the Humans Filter in most databases.  In PubMed for example, the Humans filter uses the Humans MeSH (subject heading).  This means you will miss out on the newest records.  (Records are placed in PubMed when they are supplied by the publisher.  It takes a while for the subject headings to be assigned.)  There is a priority list for journals which means some get their MeSH headings assigned sooner than others.  

    It might be tempting to use "NOT animals" in your search, but this can cause articles that look at both animals and humans to be eliminated.

    You are better off eliminating the true animal studies after the search is done, during the screening process.



    ------------------------------
    Linda Hartman
    Research Librarian
    Monroeville PA
    United States
    ------------------------------



  • 8.  RE: CER Literature Searches

    Posted 24-Dec-2020 03:18

    Dear Elizabeth

     

    To clarify my earlier recommendations with regards to eliminating publications presenting results from animal studies or in vitro studies.

    • In the medical subject areas that I have been involved for the last 27 years, I have not found clinical studies of good to excellent clinical quality in which the authors supported their clinical results with data from animal studies or in vitro studies.
    • Only when the literature is extremely limited are studies with animal and/or in vitro results included in literature search results
    • Literature search results contain only peer-reviewed literature to avoid potential bias

     

    The goal of the literature search is to find a representative amount of literature - in an unbiased manner - to analyse, to discuss and to present the state of the art, risk and benefit, claims, intended use/purpose etc.  for your medical device with those devices/ therapies having the same intended use/purpose.

    Best Regards,

    Stephanie



    ------------------------------
    Stephanie Grassmann
    Managing Director of MedTechXperts Ltd
    Biberstein
    Switzerland
    ------------------------------



  • 9.  RE: CER Literature Searches

    Posted 24-Dec-2020 04:12

    Hi all

    I also wanted to point out that the literary search as mentioned by my colleague above is, in my view, fine.
    However, we must not forget that an equivalence study should also be performed with a device that is described / mentioned in the literature, ie whether it is an equivalent device to my device (see MEDDEV) or similar (see MDGG Clinic evaluation for legacy devices) or am I wrong?.



    ------------------------------
    Evangelos Tavandzis
    Lead Auditor, Consultant
    Praha
    Czech Republic
    ------------------------------



  • 10.  RE: CER Literature Searches

    Posted 24-Dec-2020 09:37

    There are 2 good webinars saved as videos in the RAPS learning portal. One is by Evidence Partners and very comprehensive with good tips. The other is by Criterion Edge which is on a higher level and focuses on lit screening. Free for RAPS members and worth watching.



    ------------------------------
    John Minier RAC
    Consultant, Principal
    ------------------------------