Informational How to perform a clinical evaluation of medical devices – Part 2 – Level of clinical evidence and what sufficient clinical evidence means

[Marcelo]

This is the second installment of my series on medical device clinical evaluation. I suggest reading the first part (How to perform a clinical evaluation of medical devices – Part 1 – Overview and sample of activities – How to perform a clinical evaluation of medical devices – Part 1 – Overview and sample of activities – Medical Device Expert News to have a better understanding of the concepts so the understanding of this part is easier).

One of the fundamental discussion related to clinical evaluation is: Level of clinical evidence and what sufficient clinical evidence means? The attached document provides a discussion of the scientific basis to determine level and sufficiency of clinical evidence.


EU-MDR-Level-of-clinical-evidence-and-what-sufficient-clinical-evidence-meansDownload

Posted at Marcelo's Medicaldevice.expert website...

You may comment and discuss in this discussion thread.

 

[Weeder]

This is the second installment of my series on medical device clinical evaluation. I suggest reading the first part (How to perform a clinical evaluation of medical devices – Part 1 – Overview and sample of activities – How to perform a clinical evaluation of medical devices – Part 1 – Overview and sample of activities – Medical Device Expert News to have a better understanding of the concepts so the understanding of this part is easier).

One of the fundamental discussion related to clinical evaluation is: Level of clinical evidence and what sufficient clinical evidence means? The attached document provides a discussion of the scientific basis to determine level and sufficiency of clinical evidence.


EU-MDR-Level-of-clinical-evidence-and-what-sufficient-clinical-evidence-meansDownload

Posted at Marcelo's Medicaldevice.expert website...

You may comment and discuss in this discussion thread.

Hi Marcelo, I have been struggling with the question of sufficient clinical evidence (I have read your above mentioned article). In my case, all the data I have is directly relevant to the device and it is high quality data and comes from published articles in respectable journals. The question is if this number of articles is say over 200, is there a need to in appraise each and everyone of them? or can you select from among them say 50 to evaluate? Are you required to appraise each and every article that your search finds? Some clarification would be helpful.

 

[Marc]

Marcello moved on a while back so he will not be replying. Maybe someone else here will be able to help you.

 

[Watchcat]

I don't know about "required." If you don't appraise all of them, how would you know that one or more didn't report data critical to the safety and/or performance of your device of which you are totally unaware? And if you don't appraise all of them, how would the reviewer know that you didn't select the articles that had the most positive results and omit those with less positive results?

 

[Rincewind]

In general I would agree with Watchcat but I could see omitting some papers based on actuality. What was your period of time you looked papers up? Is there any way to reasonably shorten that time period to reduce the amount of papers?

That said, if that is the first CER for this product or product group I don't really see a way around all of the papers but the good thing is for the updates of that report you would just have to look of papers released in the time between the last CER update and today.

 

[Weeder]

I don't know about "required." If you don't appraise all of them, how would you know that one or more didn't report data critical to the safety and/or performance of your device of which you are totally unaware? And if you don't appraise all of them, how would the reviewer know that you didn't select the articles that had the most positive results and omit those with less positive results?

Thanks Watchcat, I think that is a good point that how would the reviewer know what you selected? I think some more clarification on my part is needed. First this is not the first revision of the CER and since the last revision a lot more papers have been published. All (100%) of them show positive results and there has never been any safety related issue with the device.

In this case you just keep adding more and more appraisals and paperwork to the CER (now closing in on 200 publications and increasing). So what is the point of doing all that? Shouldn't there be a provision to explain that you selected a reasonable number of papers and then provide justification for why you did not include all of them. As long as the papers you eliminated did not show less positive results for safety or performance, it should be okay. Is that something acceptable?

 

[Watchcat]

It sounds like you've already "appraised" them. I'm wondering what you envision as an "appraisal." Perhaps there is something high level you could do as a first-tier appraisal, confirming that which you have already said: all directly relevant to the device and it is high quality data and comes from published articles in respectable journals, and show show positive results. Then select a sample to appraise in more depth?

 

[Weeder]

Thanks Watchcat,
That is an idea worth exploring. Maybe a tiered approach to appraisals. I will look into that.

 

[Ed Panek]

We first defined our search terms and performed searches in PubMed etc. We limited it from 2018 to current as our previous review covered previous entries. We then reviewed the papers and determined if they were in the same subcategory as ours. Then we filtered that list to pivotal papers and wrote our report about them.

 

[Weeder]

Thanks Ed, it is just wonderful how many ideas come out of these discussions and I am very grateful for all those who responded and provided valuable feedback. I can definitely tailor these ideas to my own situation. Thanks everyone.