Division of Dockets Management
Food and Drug Administration
Department of Health and Human Services
5630 Fishers Lane, Room 1061
Rockville, MD 20852
Comments of the National Center for Health Research
on the draft guidance
‘‘Factors to Consider Regarding Benefit-Risk in Medical Device Product
Availability, Compliance, and Enforcement Decisions’’
[Docket No. FDA–2016–D–1495]
The National Center for Health Research does not support the current version of the draft guidance, “Factors to Consider Regarding Benefit-Risk in Medical Device Product Availability, Compliance, and Enforcement Decisions” because it would clearly lower FDA safety standards for Premarket Approval (PMA) medical devices, which are the highest-risk devices (the guidance does not apply to 510(k), de novo, and other device classifications).
This guidance is designed to assist medical device makers in making decisions regarding “nonconforming product or regulatory compliance issues” (recalls or market withdrawals). It is also written to help the FDA address shortage situations that may occur following an inspection where “non-compliance was observed.” The guidance clarifies the “benefit and risk factors” FDA and industry may consider regarding “medical device quality and patient safety.”
The guidance states that FDA may consider benefit-risk factors on a “case by case basis,” and “factors may be weighted differently.” But then it contradicts those statements by saying that FDA intends “to improve consistency and transparency” regarding the factors it considers. Case-by-case considerations do not lead to consistency; on the contrary, they tend to be subjective and as such are also labor and resource-intensive. Since they would likely be subject to bias, we oppose them. However, if the FDA believes they are needed, industry should pay for them through generous user fees. Such reviews should not be funded by appropriations that are already inadequate for CDRH.
The guidance also states that “product availability and other medical device compliance and enforcement decisions are generally fact-specific.” Providing fact-specific data is the best way to provide consistency in reviewing medical devices, which is why we are concerned about some of the other sources of information that the guidance promotes. Our specific concerns are listed below:
Real World Evidence
The guidance states that benefit-risk factors will be assessed using real world evidence but it does not define real world evidence. Anecdotal information from patients and clinicians should not be used, unless the FDA specifically reaches out to gain information from patients who were harmed according to MAUDE reports, and not just to patients affiliated with industry. Registries can provide useful information, but only if the registries are providing data on representative patients and healthcare providers. Registries that are only used by physicians that are most actively treating specific types of patients are not able to provide data that is generalizable to the patient or provider population as a whole.
Patients’ Perspective
The guidance states that FDA and industry “can help maximize benefit and reduce risk to patients” by considering patients’ perspectives regarding regulatory non-compliance or device nonconformity. It states that “FDA may consider…patient perspectives on what constitutes meaningful benefit, what constitutes risk and what tradeoffs patients are willing to accept.” However, the guidance does not state how FDA will obtain patients’ perspectives. It states that FDA will consider “reliable patient preference information” but it does not define “reliable.” The track record for FDA is that it primarily depends on the perspectives of patients who were recruited by sponsors. Those sponsors directly or indirectly influence patients by funding their nonprofit organizations. The FDA rarely relies on information from patients or consumers who were harmed by medical devices. They have very different perspectives from the patients recruited and encouraged by the device companies. Similarly, the Medical Device Innovation Consortium (MDIC) does not include small, independent patient organizations because they are not able to afford the annual fee to join.
Off-Label
The guidance states that “clinicians may find unanticipated ways to use the medical device and additional types of benefit.” In other words, they will be using the devices off-label. We have heard from many patients who were harmed by off label uses and who were never informed by their physicians that a use was off-label. FDA should ensure that such off-label uses are accompanied by a clear oral and written informed consent, informing the patient that the device is not approved for the off label indication. In addition, the FDA should require databases that clearly provide data supporting the safety and effectiveness of these off-label uses. A standardized database is important to provide uniform, consistent, and reliable data.
Nonconformities
The guidance states that “medical device nonconformities may directly increase risk or introduce new risks.” We agree. However, the guidance also states that nonconformity can “include the failure of a medical device to meet its performance specifications even though the device still performs adequately.” On what scientific evidence does FDA base this statement? In the “likelihood of risk” section of the guidance, the FDA states that “reliable data” may not exist.
Understanding risks
In the “patient tolerance of risk” section of the guidance, FDA notes that “patients may not understand device-related risks.” We strongly agree. Unfortunately, FDA increases the likelihood that the patient will receive biased information about those risks when the guidance states “a patient’s assessment of risk would be appropriately informed by information from his or her clinician” (emphasis added). Most clinicians receive information about medical devices primarily from the manufacturer, and many also receive consulting fees or other gifts from the manufacturers. In addition, it is not unusual for a clinician to design a device that they then encourage patients to use. For those reasons, many clinicians may not be able to provide objective or appropriate information to their patients.
Safety Standards Lowered
The guidance states that FDA should use the “outcome of a benefit-risk assessment to inform decisions related to product availability” due to a recall, safety signal, or medical device nonconformity.
The four bullets listed in this section all suggest lowering safety standards:
- When should a correction be used instead of a removal?
- What actionsif any should FDA take regarding nonconforming devices in shortage situations (emphasis added)?
- When should FDA grant a variance from Quality System regulations during a PMA pre-approval inspection?
- When can FDA “exercise enforcement discretion” and not take action when a device maker makes a significant change or modification prior to obtain a clearance?
All these questions revolve around issues that would weaken safety standards. These are not the questions CDRH should be asking. Where are the questions regarding how to strengthen safety standards? The list of medical devices that could have benefited from stronger standards is long and includes power morcellators, duodenoscopes, metal-on-metal hips, Essure, and surgical mesh to name a few. And yet, this guidance document states that “If FDA’s benefit-risk assessment indicates high benefit to patients with little risk, FDA may decide to work with the manufacturers to address the underlying issues without initiating a formal compliance or enforcement action.” Unfortunately, because CDRH requires so little data on safety and effectiveness, it lacks adequate information to make this determination on the basis of scientific evidence. The adverse medical device reporting system (MAUDE) greatly undercounts device problems, devices are constantly being changed before required post-market studies are completed, and devices are not yet part of the Sentinel program. This clearly puts patients at unnecessary risks.
Also, it is unacceptable that the guidance states “if FDA’s benefit-risk assessment indicates low benefit to patients with high risk, FDA would likely take formal compliance or enforcement action to address the problem” (emphasis added). The word “likely” should be removed from that sentence.
Conclusions
For the above reasons, we strongly oppose this version of the draft guidance. It does not enhance FDA’s goal of “protecting and promoting the public health.” We are particularly concerned that FDA is lowering its standards to address possible “shortage situations,” and it is putting patients at risk by allowing device manufacturers to use benefit-risk factors to allow “nonconforming” devices or devices with regulatory compliance issues to remain on the market instead of being recalled.