This year's PEAC meeting got off to something of a wobbly start with me. The Federal Register notice indicated that the Committee would discuss and make recommendations on the topic "Connected and Empowered Patients: E-Platforms Potentially Expanding the Definition of Scientific Evidence."
I wonder a lot about the extent to which CDRH knows what it is doing; I regrettably chalk this one small piece of evidence into the "Science-No" column. I'm of the strong opinion that the definition of scientific evidence falls within the domain of science, not within the domain of government regulation of medical products, and certainly not within the domain of e-anything.
On the other hand, it's just an FR notice; could be just a verbal hiccup. Maybe I'm making too much of it.
Next I wonder whether CDRH has a definition of scientific evidence, since I take it as a given that you can't expand upon a definition that you don't have. Thus, I am further discouraged when CDRH distributes a Glossary of Terms to inform the Committee's discussion on expanding the definition of scientific evidence, and "scientific evidence" is nowhere to be found among them.
But then I find some comfort in the introduction to the questions that CDRH wants the Advisory Committee to address, as it seems to be driven by the regulations, which include a definition, not of scientific evidence, but of "valid scientific evidence." 21CFR 860.7(b)(2)
I am happy enough for CDRH to take ownership of what it considers "valid" scientific evidence in its regulations, because I think "valid" is sort of like "significant" as used to refer to a statistical outcome, in that you can define it however you like for your purposes. (Hopefully with careful consideration of how well your definition will serve that purpose, but perhaps I can only expect so much.) However, when the term "valid" is dropped…not happy.
Scientific evidence, like the difference between two groups in a clinical study, is what it is. Neither the actual ("real world"?) nor the statistical difference between two study groups changes just because you change the level you will accept as significant. Nor does the difference between the two populations the study may have attempted to sample. Similarly, evidence does not become "scientific" or "un-scientific" just because someone says so.
Next I look at the questions FDA wants the Committee to address. It seems that the issue has now been reformulated a bit:
Some challenges with the use of patient-generated health data include the statutory requirement that information used in regulatory decision-making be valid scientific evidence. To address some concerns about whether this information could be used as valid scientific evidence, FDA is requesting your recommendations about whether patient-generated health data could be fit-for-purpose in a regulatory context.
I am not exactly sure what FDA is trying to say here, but I think it wants to know if, and perhaps under what circumstances, patient-generated data might meet the statutory definition of valid scientific evidence, or if the statutory definition would have to be revised to allow FDA to consider these data as scientifically valid. Or, who knows, maybe do away with the constraint altogether and just let FDA call it from its gut.
FDA's questions for the Committee are not put to it until fairly late in the day, after a lot of presentations and other activities. When they are, things take a strange turn. A lot of the questions are not related to scientific evidence, valid or otherwise, nor even to "fit-for-purpose," which I haven't figured out is supposed to be an alternative to valid scientific evidence, or another, maybe broader, way of saying it. The full discussion is documented in a little over 40 pages of the transcript, starting on page 137. On page 145, a Committee member comments that "I guess we are beyond the question of whether this is scientifically based data." I would say mostly they skipped right past it, and that, after that point, they never really came back to it.
I'm not sure what to make of this. I'm inclined to think CDRH is still finding its way when it comes to how to work productively with this type of Advisory Committee, including figuring out what expertise should be included among its members and what questions can be reasonably put to it.
It was my impression that relatively few of the Committee members had the background needed to tackle the topic of valid scientific evidence. Maybe CDRH concluded the same thing between the time it crafted the topic of the meeting and the time it sat down to develop its list of questions for the Advisory Committee to address, and therefore it took its questions in a different direction. As a result of this shift, I think the few members who had the background opportunity to bring that background to bear on the topic. Since this was my interest, I found it a bit frustrating, and I imagine those Committee members may have found it frustrating as well. I worry that they won't come back next year.
All that said, the fact that I found the topic of interest to me personally doesn't mean that it was the best topic for FDA to bring to this Committee. I will be interested to see what topic it settles on for this year's meeting, and if that seems to be a better fit.
------------------------------
Julie Omohundro, ex-RAC (US, GS), still an MBA
Principal Consultant
Class Three, LLC
Mebane, North Carolina, USA
919-544-3366 (T)
434-964-1614 (C)
julie@class3devices.com------------------------------