At the end of 2022, FDA published a draft guidance on Voluntary Malfunction Summary Reporting (VMSR) Program for Manufacturers. The draft guidance explains several aspects of the VMSR Program, including FDA’s approach to determining the eligibility of product codes for the program. Consistent with the goals outlined in the Medical Device User Fee Amendments of 2017 (MDUFA IV) Commitment Letter, the VMSR Program streamlines reporting of device malfunctions. The program began in 2018 when FDA issued an order granting an alternative reporting approach under 21 CFR 803.19. The alternative approach permits manufacturers of devices in eligible product codes to report certain device malfunction medical device reports (MDRs) in summary form on a quarterly basis instead of within the 30-day window, subject to certain conditions.
My question is, what product codes are the focus of this alternative reporting and how did FDA select them?
Results
To learn something about FDA’s criteria for selecting product codes, I sorted the product codes by therapeutic area, both regarding what percentage of product codes are eligible and also the absolute number, since the number of product codes varies considerably from one therapeutic area to another.
Voluntary Malfunction Summary Reporting (VMSR) Program Product Codes
Methodology
This month did not require me to write much code that required any judgment, as the device classification database has all the information for this VMSR Program. All I had to do for the first visualization is simply divide the number of qualifying product codes within a given therapeutic area by all the product codes in that area. The second visualization is just the raw numerator.
Interpretation
In the draft guidance, FDA explains that each year, FDA receives over two million MDRs for suspected device-related deaths, serious injuries, and malfunctions. The vast majority of those reports are for malfunctions.
Manufacturers are under significant time pressure to file MDRs for malfunctions, having to meet the 30-day time limit. [21 CFR section 803.50] FDA, for its part, must handle literally millions of these reports. And frankly, for many individual MDRs, the safety significance isn’t that great.
According to FDA in the draft guidance,
“As explained when it proposed the VMSR Program (82 FR 60922, and consistent with our VMSR Program experience to date, FDA believes that bundling “like events” together into a single summary report description has benefits for manufacturers, FDA, and the public. For many manufacturers, we expect this approach will greatly reduce the volume of reports that the manufacturer needs to submit to FDA. As more information is received in a streamlined manner, it can facilitate a more efficient understanding by FDA of malfunction issues. For the public, summary reports may make malfunction event trends for a particular device more readily transparent. We believe increased manufacturer participation in the program will enhance these benefits.”
While the program has many elements, in this post I only want to focus on how FDA selects the product codes that are eligible for this summary reporting. My motivation for doing so is that I consider the selection process for determining these product codes to be frankly opaque.
In the Federal Register notice accompanying the proposed rule in 2017, FDA explained its approach to selecting the initial list of product codes as follows:
An important part of this proposed voluntary program is providing clarification to manufacturers regarding the product codes eligible for the program. FDA is currently in the process of evaluating device product codes to determine which ones should be eligible. The Agency is requesting comments on the product codes that should be eligible for this proposed Voluntary Malfunction Summary Reporting Program, including for combination products. FDA will consider the proposed list of eligible product codes submitted by industry along with any comments received on this proposal in determining the product codes that would be included in the proposed alternative granted to permit summary malfunction reporting.
Consistent with the MDUFA IV Commitment Letter (Ref. 4), when this proposed voluntary program is finalized through publication of a Federal Register document granting the alternative under § 803.19, FDA will identify on its website a list of device product codes that are eligible for the Voluntary Malfunction Summary Reporting Program as part of granting the alternative.
In other words, FDA did not propose a list of product codes, but rather in its proposal simply mentioned that it had its own private list and invited the public to comment without putting forth the agency’s planned list. The reference list “submitted by industry” was apparently a list submitted by those involved in the user fee negotiations.
In my opinion, FDA must move away from referring to that as “industry.” That is not some democratically elected group of industry representatives, but rather representatives of certain specific industry trade associations invited to participate in the discussions. They don’t speak for everyone.
The point here is that FDA never proposed a list for the public to see and comment on. One could argue that the agency’s approach does not satisfy the legal requirement for notice and comment. FDA presumably would say that the list is not part of the rule, and thus did not need to be part of the rulemaking. Maybe, but the list does seem to constitute FDA guidance at least, and under the Good Guidance Practices regulations, FDA is likewise supposed to engage in notice and comment. [21 CFR section 10.115]
Some of the folks who commented on the 2017 Federal Register proposal expressed concern that a list of the product codes was not included. Trying to justify its decision not to publish such a list, in the 2018 final notice FDA said:
(Response 4) FDA disagrees that it should publish another Federal Register notice for public comment listing product codes that would be eligible or ineligible for the program. Among other reasons, the Agency expressly requested comment on the product codes that should be eligible for the proposed program, and many commenters submitted proposed lists of eligible product codes or identified specific devices about which they had concerns. FDA has considered these comments and has also conducted an extensive review of all product codes, regardless of device class, to determine whether each product code would be eligible.
FDA was simply not transparent about how they selected in that first-round decision-making the particular product codes that are eligible. The problem with the agency’s approach is that the public could not see what others were proposing in time to offer their comments on such proposals, let alone comment on what FDA itself or the industry trade associations proposed.
In its 2022 draft guidance, in a couple places, FDA states the negative --what product codes are ineligible. For example, “product codes that have been in existence for fewer than two years generally are not eligible, unless the new product code was created solely for administrative reasons.” That helps, but it is certainly still less than clear how FDA arrived at its list.
FDA describes also why the agency might remove a particular product code from the approved list. In the 2022 draft, FDA explains that the agency:
“intends to periodically assess and update the eligibility of product codes for the VMSR Program. As part of determining eligibility of product codes, FDA intends to consider the device’s benefit-risk profile and available postmarket safety information, particularly related to device malfunctions. The Agency generally considers whether quarterly, summary reporting of device malfunctions, in accordance with the conditions of the VMSR Program, would allow FDA to timely identify potential new or increased safety concerns for devices within the product code at issue. If FDA determines that a product code is eligible, FDA intends to update the Product Classification Database accordingly.”
The agency goes on in the draft guidance:
In analyzing available postmarket safety information for devices within a certain product code, the Agency also intends to consider, among other things, the frequency of reported serious injuries and deaths, the number of 5-day reports, and whether the product code has any class I or II recalls. The Agency may also consider the types of malfunctions that occur in a given product code, the complexity of those malfunctions, and the ability for FDA to understand their root cause. FDA may also consider whether the product code is associated with recent, ongoing, or potential public health issues that may necessitate the detail and frequency of individual malfunction reporting for FDA to identify and better characterize new or persistent safety issues. When a public health issue necessitates close monitoring of individual adverse events associated with certain devices, the Agency may determine that summary reporting under the VMSR Program is not appropriate for product codes for those devices.
I gather this means that the public will only know that product code has been removed if, when you consult the classification database, the product code is no longer listed as part of the program. No explanation will be given.
While at least certain of the criteria are stated in the draft guidance, the application of those criteria to a particular product code is not at all transparent. We don’t know why XYZ code was included or excluded. It is a black box. The program involves FDA moving very much behind the scenes to add and subtract product codes as eligible.
Manufacturers do have the option of asking for a specific product code to be added to the list. Manufacturers can send a request under 21 CFR 803.19(b). Presumably, if they get turned down, FDA will at least provide some general reason for its decision, but I don’t know that for sure. I have not seen the results of such an inquiry. And to my knowledge, they are not public.
When I look at the high-level depiction of the product codes that FDA has selected, I can see some logic to it. Apparently dental and physical medicine, for example, have a high percentage of low-risk product codes. But at the same time, I wonder why general hospital would not be higher up the list, as well as gastroenterology and urology. Neurology has barely 50% of its product codes in the program. I understand that there are a lot of high-risk neurology devices, but that still seems quite low. And there is simply not enough substance made public for me to comment intelligently on the FDA’s thought process.
Conclusion
Obviously, we can go through the individual product codes and guess why they are or are not included in the VMSR Program. The goal of this post was to look at the data at a high level to see if it made sense given the criteria, particularly since the selection process has been so opaque. The therapeutic areas at the top of the list make sense to me, but it seems as though there are at least a few categories at the bottom where FDA is being pretty conservative in excluding a very large number of product codes from eligibility in the VMSR program.