Regulatory Open Forum

 View Only

  • 1.  MDR Article 88

    Posted 08-Nov-2019 14:34
    ​(EU) 2017/745 Article 88, Trend reporting, states "The manufacturer shall specify how to manage the incident referred to in the first subparagraph and the methodology used for determining any statistically significant increase in the frequency or severity of such incidents, as well as the observation period, in the post-market surveillance plan..."

    Is there a suggestion for specifying the methodology used for determining any statistically significant increase in the frequency of incidents (general) and when the frequency is extremely low, e.g. 0.007%? How is a significant increase in the severity of an incident calculated?

    ------------------------------
    Melissa M. Traylor
    Regulatory Affairs
    FzioMed, Inc.
    San Luis Obispo, California
    USA
    ------------------------------


  • 2.  RE: MDR Article 88

    Posted 08-Nov-2019 15:01

    There is one piece you need for the methodology I'll describe. Annex III(1.1)(b)(fifth indent) says the PMS plan includes, "methods and protocols to manage the events subject to the trend report as provided for in Article 88, including the methods and protocols to be used to establish any statistically significant increase in the frequency or severity of incidents as well as the observation period". The piece I'm looking for is the observation period.

    First, start with complaints.

    Classify each complaint as an incident or not, using the MDR definition of incident.

    Classify each incident as a serious incident or using the MDR definition of serious incident.

    Classify each not serious incident as subject to trend reporting or not using the criteria in article 88.

    Establish an observation period of 12 months.

    For each month in the observation period calculate the rate of not serious trend reporting incidents. This will be the number of qualifying incidents in the month divided by the number of uses of the device in the month.

    Plot a time series that covers the observation period.

    Calculate the slope of the linear regression line for the time period.

    Determine if the slope is statistically different from zero. Excel's Analysis Tool Pack has a method to do this. The default p value is 0.05, but you can change it. The Analysis Tool Pack provides the confidence interval surrounding a slope of zero. If the actual slope is above the upper confidence limit, then there is a statistically significant increase, so you must report.

    Repeat this on a regular basis such as monthly or quarterly.



    ------------------------------
    Dan O'Leary CQA, CQE
    Swanzey NH
    United States
    ------------------------------

    Original Message


Related Content