Regulatory Open Forum

 View Only

  • 1.  Statistical techniques-ISO 13485:2016

    Posted 13-Jul-2017 08:08

    Hi,


    Any idea on the statistical techniques to meet ISO 13485:2016 requirements.


    Regards


    Rashmi



  • 2.  RE: Statistical techniques-ISO 13485:2016

    Posted 14-Jul-2017 12:54
    Are you asking about the repeated references that ask for justification of sample sizes or are you asking about the mention of statistical techniques under analysis of data?

    ------------------------------
    Morgan Nalley
    ------------------------------

    Original Message


  • 3.  RE: Statistical techniques-ISO 13485:2016

    Posted 15-Jul-2017 21:19
    ​Hi

    The full universe of statistical methodology applies to ISO 14385:2016 with a design preference for frequentist methods (in contrast to adaptive and bayes methods) and an analysis preference for frequentist-based statistical inference, e.g. 95% confidence intervals (in contrast to bayes methods, e.g. 95% credible intervals).  Precedents matter.  Notified body and competent authority experience is also a factor.  Keep it simple.

    Phil Lavin PhD, FASA, FRAPS
    Lavin Consulting LLC


    ------------------------------
    Philip Lavin
    Principal
    Lavin Consulting LLC
    Framingham MA
    United States
    ------------------------------

    Original Message


  • 4.  RE: Statistical techniques-ISO 13485:2016

    Posted 16-Jul-2017 02:06
    Any guidance or standard that can be followed for reference?

    ------------------------------
    RashmiEllex MedicalAdelaideAustralia
    ------------------------------

    Original Message


  • 5.  RE: Statistical techniques-ISO 13485:2016

    Posted 15-Jul-2017 10:28
    In general, the application of statistical techniques is aimed at data analysis and providing valid results to draw conclusions or make decisions that have a solid base. This mirrors subpart O of 21CFR820.

    ------------------------------
    Regards,
    Mark Swanson, ASQ CBA, CMQ/OE, CQE MBA
    Becker MN
    United States
    ------------------------------

    Original Message


  • 6.  RE: Statistical techniques-ISO 13485:2016

    Posted 16-Jul-2017 01:36
    ISO/TR 10017 guidance on statistical techniques provides a good overview.
    Here is what it covers:

    • Descriptive Statistics
    • Design of Experiments
    • Hypothesis Testing
    • Measurement Analysis
    • Process Capability Analysis
    • Regression Analysis
    • Reliability Analysis
    • Sampling
    • Statistical Tolerancing
    • Time Series Analysis
    If you google ISO/TR 10017 you may come across the full version - tsk tsk.
    Here is a link to a legal <g class="gr_ gr_654 gr-alert gr_spell gr_inline_cards gr_run_anim ContextualSpelling" id="654" data-gr-id="654">preview</g>.

    Statistical sampling plans/standards issued by ISO, ASQ, etc. are also useful.
    I generally prefer to use this "C=0" sampling plan since it is easy to understand/apply and is generally accepted.
    Zero Acceptance Number Sampling Plans, Fifth Edition | ASQ
    Asq remove preview
    Zero Acceptance Number Sampling Plans, Fifth Edition | ASQ
    The zero acceptance number plans developed by the author were originally designed and used to provide equal or greater consumer protection with less overall inspection than the corresponding MIL-STD-105-E sampling plans. In 2000, the Department of Defense declared MIL-STD-105-E obsolete and recommended the c=0 plans in this book for use in place of them.
    View this on <g class="gr_ gr_1165 gr-alert gr_spell gr_inline_cards gr_run_anim ContextualSpelling ins-del multiReplace" id="1165" data-gr-id="1165">Asq</g> >


    Hope this helps.

    ------------------------------
    Sam Lazzara
    https://mdqc.blog/
    sam@lazzara.net
    ------------------------------

    Original Message


Related Content