Start by thinking about the relationship between complaints, incident reporting in Art. 87, and the risk management hazard analysis.
You will be looking at complaints, and incidents over the time period you define in the PMS Plan. It is useful to tie it to the frequency of the PSUR.
Using the EU-MDR definition of an incident, classify each complaint as an incident or not an incident. Following the EU-MDR definition of a serious incident classify each incident as a serious incident or a not serious incident. Serious incident are reported following Art. 87. Not serious incidents at further analyzed following Art. 88 for potential reportability.
Let's start with the severity of not serious incidents. Each one should be in the hazard analysis for the device, which will identify its severity using the risk matrix developed in the Risk Management Plan. If you follow the traditional definitions of severity, then not serious incidents should be in one of the two lowest severity categories. This is the foreseeable severity. Classify each not serious incident using the risk management definitions. If it is in the same severity category as shown in the hazard analysis, then you don't have an increase in severity. If it is a higher severity category, then it is reportable under Art. 88.
Now consider frequency of occurrence (probability) for the same time period.
Start with a time series chart (in Excel) of the not serious incidents reported by month. This is a rate in which the numerator is the number of not serious incidents reported in the month and denominator is the number of device uses in the same month. Draw the Excel chart and add the linear trend line and its equation.
The easiest way to determine a statistically significant increase is to use the Analysis DataPak in Excel. It is built in, but you need to turn it on. Use the Regression tool. It will give you some tables. Look at the upper and lower confidence intervals for the slope of the regression line. If the slope is positive and 0.0 is not in the interval, then you have statistically significant increase and must report under Art. 88.
Next, compare the regression line with the estimated frequency of occurrence from the hazard analysis. Observe the regression line and determine the associated rate. If it is higher than the rate in the hazard analysis then this is a significant increase as compared to the established frequency.
This applies to devices including SaMD. You should analyze each device marketed in EU and include incidents from any geographic region with the same or similar device.
------------------------------
Dan O'Leary CQA, CQE
Swanzey NH
United States
------------------------------
Original Message:
Sent: 21-Feb-2024 09:36
From: Anonymous Member
Subject: Art. 88 MDR
This message was posted by a user wishing to remain anonymous
Dear RAPS Community,
According to Art. 88 of MDR manufacturer shall report based on trend reporting also not serious incidents, if a statistically significant increase in the frequency or severity of those incidents occurs. Following the art. 88 "...The significant increase shall be established in comparison to the foreseeable frequency or severity of such incidents in respect of the device, or category or group of devices, in question during a specific period as specified in the technical documentation and product information."
About severity: if the incidents are not- serious it gives an assumption of a low/moderate risk, and in case of IIa class device there is very low probability of significant increase of severity, as by default it is a low/moderate risk class device. So how to assess the significant increase in severity?
Considering frequency: are there any thresholds recommended in term of "statistically significant" if so, what would be the value and "specific period"? How to establish thresholds reasonably for class IIa device and class IIa SaMD? And finally, should statistically increase in frequency of those incidents be considered as an overall, or e.g per market area/country?
Thank you for sharing your experience!