Leo and his "traffic-lights-idea" – trend reporting under MDR
Do you remember Leo? In the past I wrote already some LinkedIn articles about him as a Quality Manager struggling in the fancy world medical device regulation. For sure Leo is not a real person but he maybe somehow representative of all of us trying to survive in the constantly changing regulations. This time, Leo will be thinking about trend reporting as per article 88 of (EU) 2017/745 Medical Device Regulation.
Disclaimer: The story, all names, characters, and incidents described in this article are fictitious. No identification with actual persons (living or deceased), places, buildings, and products is intended or should be inferred.
Late 2019, somewhere in Europe….Leo was exhausted from being scrutinized by the auditors in his first MDR audit. The initial certification to (EU) 2017/745 was actually a combined audit including ISO 13485:2016, MDSAP and the old MDD. MDR just came on top of the existing requirements. This approach was driven by the time constraints of Leo’s notified body and it was quite challenging for him. However, looking to the auditors in the closing meeting revealed that it was challenging for them as well. “They appeared like two old men after an intensive bachelor party, still awake due to tons of coffee and energy drinks”, Leo said. “The audit result was great, ending up in minor findings only. The auditors expected us to have contracts with our distributors regarding their duties under MDR. I was trying to argue that this is not a real requirement but finally I gave in”, Leo remembers. “The second finding was trickier”, Leo continues, “It was related to the trend reporting required under article 88 in the MDR.” When talking to Leo and his team, the auditors concluded, that written procedures did not establish transparent and objective thresholds when trend reporting would be needed.
Fact: The requirement for trend reporting is not really new. It was required and explained in MEDDEV 2.12-1 rev 8, dated January 2013. It goes back to the GHTF Guidance Adverse Event Reporting for Medical Devices (GHTF/SG2/N54R8:2006) and GHTF/SG2/N36 Manufacturers Trend Reporting of Adverse Events from 2003. Under (EU) 2017/745 MDR, trend reporting is required in article 88 for “any statistically significant increase in the frequency or severity of incidents that are not serious incidents or that are expected undesirable side- effects that could have a significant impact on the benefit-risk analysis … and which have led or may lead to risks to the health or safety of patients, users or other persons that are unacceptable when weighed against the intended benefits.”
Already during the audit, Leo started his thought process diving deep into the statistical methods given in those old GHTF guidance. He immediately started getting goose bumps and the scary feeling of not having any idea how the methods could be applied reasonably to his company. The complaint rate of the products manufactured by Leo’s company is quite low and there is no real baseline occurring in systematic patterns. The timelines and thresholds discussed in the GHTF guidance may not work for him. Leo also considered the good old quality control charts resulting in predictions for signal and noise. Although those charts were great to observe the overall complaint rate over longer periods, they may not be suitable to detect trends for different failure modes in shorter time periods. Leo gave in for the moment and called it a day. “I was actually driving on the highway when some traffic lights stopped me”, Leo said. “That’s it … traffic lights, red, yellow, green….Why not using the nice risk control chart from ISO 14971 for decision making on trend reporting?” Leo got excited and wanted to make use of all of the huge workload spent in a proper risk management in the past.
Fact: ISO 14971 requires manufacturers to establish acceptance criteria for risk management. The third edition going to be published in 2019 will emphasize the need for proper decision making. There are acceptance criteria for the individual risks and acceptance criteria for the overall residual risks. The acceptance of risk must be closely connected to the clinical benefits of the device. A benefit-risk analysis must be conducted.
During that night, Leo was thinking a lot about his idea of using the risk control chart for trend reporting. Close to midnight his thoughts were fading, probably well deserved…..
Back in office next morning, he was fighting with the coffee machine, the same fight as every morning. Overruling the alarms did not work and finally the machine won….as always. Leo cleaned the equipment as instructed and enjoyed some cups of coffee. “Great, let’s get started!” He took out an empty piece of paper and came up with a colorized flip-chart.
“Trend reports are needed for non-serious incidents, i.e. severity 1-4 in my chart” Leo said. “If the severity would increase, we would need to report. Trend reporting if the situation would become unacceptable, ideally already before it would really hurt anybody.“ Leo marked those threshold with blue stickers. “Entering blue means that we will have to decide if article 88 applies, i.e. if trend reporting is needed. That’s not too difficult, if we pass the severity-3-threshold, we will report”, Leo concluded. “If occurrence would increase, it’ll be also a trend report ideally early enough, before we would become unacceptable. Looking to my chart, there are only two thresholds to be considered, again marked in blue.” Leo was somehow happy thinking that this may work. He went to the coffee machine again. “OMG, alarms all over the places….”, Leo got frustrated and read the display “Replace filters in 1 day, preventive maintenance needed”. Okay, this could have been worse, and Leo got his coffee. Back in office…”Preventive maintenance needed?”, he thought, “why not taking this into my chart? I don’t want to be too late and do lots of trend reporting if I could do actions already earlier….”. Leo added green thresholds and decided that it may be a good idea, starting preventive actions one level before a trend report will be needed. “Clause 8.5.3 of ISO 13485 must have a reason, maybe that’s the one?!”
Suddenly, there was noise in the office and Paul, his General Manager, rushed in with chocolates and champagne celebrating the MDR audit success. “Leo, stop doing fancy charts!” Paul said, “come-on, let’s have some fun…” Leo stopped thinking and took the chart aside….
Being at home that night, he started thinking again. “Would those trending ideas be feasible...?”
Regulatory Affairs and QMS for regulated medical systems
3 年Great Blogpost, Leo's idea sound really appealing. (and his struggles are sooo real) But for my general understanding: This matrix I use normally to evaluate Risk-Benefit of individual Hazards, so specific Hazard has a severity and should not "appear" more often than xx / year depending on the Probability Class. Did Leo then asses this on an individual Hazard base? Or does he count all issues of non-reportable severity Hx have a yearly count above the expectation of Probability class Px?
Medical Device Expert at Michael Schaefer Quality Management
4 年Just a quick update. Leo's idea for trend reporting triggered by risk control chart was presented the first time during the MDR initial audit and the auditors welcomed the approach.
Medical and in-Vitro Diagnostics Products
5 年Very nice article but unfortunatly o final answerow to proceed.
Very interesting idea of coupling RM and trend reporting. I also like the simplicity of looking at these borders, as this is precisely how one should set up good risk management acceptance criteria. I would only note that following the wording "any statistically significant increase ... [of] undesirable side-effects that could have a significant impact on the benefit-risk analysis", one could argue that the blue and green line is not necessarily straight, but can be adapted accordingly.?For example, changing the image above not have a blue line at occurrence 6 between severity 2 and 3? We are reaching unacceptable there after all. ;) Just my two cents - great article and food for thought!
Experienced Quality System & Regulatory Affairs Professional f. Medical Devices & Combination Products - Passionate Engineer
5 年Interesting - What will Leo do next?