William Hyman: What Do FDA Inspections of Hospitals for MDR Compliance Tell Us?

Medical device reporting (MDR) is mandatory for hospitals and others captured under the definition of “device use facilities,” which does not include physician offices. Also subject to mandatory reporting are manufacturers for which this is a well-entrenched activity. The three categories of device-related events that must be reported are deaths, serious injuries, and, for manufacturers, malfunctions, which if repeated could cause death or serious injury. Other entities and individuals can file voluntary MDR reports. The reporting requirements for hospitals is summarized in a 1996 guidance document entitled Medical Device Reporting for User Facilities. There are other reporting mechanisms such as the FDA’s MedSun program and the federal Patient Safety Organization (PSO) program, but participation in either or both of these does not substitute for MDR compliance.

The FDA inspects hospitals for MDR compliance, and it recently released a summary of adverse findings for 15 hospitals inspected during the period December 2015 through April 2016. In addition to the tabular summary, there are links to the original FDA 483 forms. These forms are post-inspection records of findings provided to the inspected organization. Some of the hospitals in this summary are well recognized. Two additional hospitals are listed without adverse findings or 483s. It is not clear why these two hospitals are included. The summary says that the FDA had telephone meetings to discuss “significant” deviations” from the MDR regulation when it was determined that the facility had not provided an adequate response.

As an observer, I have been curious about how a facility or a manufacturer can have a negative inspection finding, and then be given a chance to correct the issues but still fail to do so. The report does not tell us how many other hospitals, if any, were inspected and therefore we do not know the failure rate. Regardless of rates and numbers, if a hospital is publicly reported as failing both the inspection and the opportunity to fix the problem, then those facts stand on their own no matter how many other hospitals passed or failed.

Inspection findings and inadequate responses can also result in a warning letter, but they are not mentioned in this report, and I did not find any such letters to hospitals for MDR issues. However, there are FDA warning letters to hospitals concerning adverse findings with respect to two other FDA mediated issues: mammography standards and institutional review boards (IRBs).

The adverse inspection findings in the summary include not having written MDR procedures, failing to report one or more “reportable” events, and not maintaining adequate MDR files. The 483s in some cases have specific information such as “10 of 10 deaths of patients who underwent endoscopic retrograde cholangiopancreatography (ERCP) procedures with a potential infected duodenoscope were not reported.” For other hospitals, the issue was more procedure-oriented, as in: “Your two-page written MDR procedure, entitled “Biomedical Equipment Safe Medical Device Act,” dated 07/22/2010, does not address or refer to the following elements: A) The use of FDA Form 3500A or electronic equivalent for mandatory device adverse event reporting, B) The definition of a serious injury, C) Your user facility reporting number.” I find this one curious in that the critique seems to be the absence of including the required and the obvious which are available elsewhere in an internal procedure, i.e., the definition of serious injury is dictated by the FDA, but apparently it still has to be copied and pasted into the local procedure. A lack of documentation of following procedures was also cited, as in: “You have no documentation of the deliberations and decision-making process used to determine if the death of E.M.F. … was or was not reportable.” Here I note that if EMF is the patient’s initials, the inclusion of the initials in the report might readily identify the patient to others—at least locally. In some cases, parts of the 483s are heavily redacted making them incomprehensible in this summary. Also interesting are the entries for “dates of inspection,” which in many cases are multiple days (or perhaps parts of a day) spread over multiple months. It seems unlikely that this is an efficient use of either the inspector’s or facility’s time, and it is certainly disruptive.

Problems that other facilities have had with MDR inspections can be highly educational in reviewing one’s own processes and compliance. They therefore make a good read for anyone involved in MDRs, which should include HTM with respect to at-a-minimum answering technical questions about what happened and what caused it. The FDA3500 reporting form for MDRs and its code entries also should be reviewed in this regard. I have noted before that this form presents opportunities to describe the event and assess responsibility for it, including confession and blame. In particular, it is the facility that assigns an Event Code, but the manufacturer provides the Evaluation Conclusion Code. Of particular interest to HTM might be a manufacturer’s conclusion of “maintenance deficiency.” In addition, there is the ever-popular manufacturer finding of “no failure detected, device operated within specification.” It is of interest that the manufacturer applies its analysis after the hospital has filled in its part of the form and sent it to the manufacturer. There is no direct reporting back to the facility of what the manufacturer concludes and tells the FDA. The above conclusions come from the “new” reporting codes which were introduced several years ago but which do not appear in the FDA’s MAUDE database, which we have been told for years would soon be replaced.

Beyond MDRs, the FDA has announced that it is taking a new look at how useful information can be garnered from hospitals. A public workshop entitled The Role of Hospitals in Modernizing Evidence Generation for Device Evaluation: Harnessing the Digital Revolution for Surveillance is scheduled for Dec. 5, 2016. According to the announcement, “Recently, the role of hospital reporting of device-related adverse events in device surveillance and, more generally, device evaluation, has garnered increased scrutiny. This public workshop will further explore the critical role of hospitals in the evolution of device surveillance and in creating more robust surveillance capabilities.” It is not clear here whose increased scrutiny is being referred to, but it might mean the FDA’s own efforts.

William Hyman, ScD, is an adjunct professor of biomedical engineering at The Cooper Union and a consultant in New York. He is also a professor emeritus of biomedical engineering at Texas A&M University.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s