By Patrick Lynch, CBET, CCE, CHTM, CPHIMS, FACCE
2016, for the most part, was spent waiting to see whether the U.S. FDA would increase regulations for medical equipment servicers. After all, since March 2015, the FDA has been collecting information and receiving factual and anecdotal comments concerning whether there is a public health and safety issue that warrants increased regulation and oversight of medical device companies and individuals.
This issue culminated in October’s two-day workshop of presentations to the FDA .
Dividing Lines
Attendees of the workshop could be classified into several groups—manufacturers, government, independent service organizations (ISOs)/independent servicers, standards organizations, hospitals and end users, and trade associations.Administrators accounted for the majority of attendees, with their job titles spanning from president, vice president, general counsel, general manager to director. (This speaks to the perceived importance of the issues at hand.)
Patrick Lynch
Manufacturers making formal presentations were Bayer, GE Healthcare, Hitachi America, Johnson & Johnson, the Medical Device Manufacturers Association (MDMA), Medtronic, Pentax America, Philips Healthcare, Siemens, and Varian Medical Systems.
The only government agency that presented formally was the U.S. Federal Trade Commission (FTC). During the FTC’s presentation, officials admonished the FDA and referred to President Obama’s April 2016 Executive Order, in which he urged all federal agencies to make no decisions that would harm competition without clear data to support the need for such rulings.
The ISOs and independent servicers that presented formally were the Afya Foundation, Alpha Source, First Source, Integrated Medical Systems, Mobile Instrument Service and Repair, Oxford Instruments, Paragon Services, Tri-Imaging, and TriMedX, while AAMI and Cambridge Polymer Group accounted for the standards organizations making such presentations.
Further, the hospitals and end users making formal presentations were the American College of Clinical Engineers, the U.S. Department of Veterans Affairs (VA), ECRI Institute, Healthcare Technology Consulting, ISS Solutions, the James J. Peters VA Medical Center, Massachusetts General Hospital, TriMedX, the Clement J. Zablocki VA Medical Center, and the VA National Center for Patient Safety.
Advamed, the International Association of Medical Equipment Remarketers and Servicers, MDMA, and the Medical Imaging & Technology Alliance (MITA) accounted for the trade associations that formally presented.
During the workshop, representatives from several of the groups presented their respective positions via four formal panels. (Such presentations are available on the FDA’s website). Opportunities for audience comments then followed.
The conference was about not only servicers, but also concerned the five “Rs”: refurbishers, reconditioners, rebuilders, remarketers, and remanufacturers. Curiously, the discussion and presentations were all about the service side of the house while the five “Rs” were largely ignored.
With little exception, each major group represented had a consistent set of beliefs and reasons that the FDA should act in a certain way. Even so, all manufacturers who were also ISOs only took the manufacturers’ stance and ignored that of the dedicated ISOs.
Furthermore, the end users, hospitals, ECRI Institute, the VA, and the ISOs/independent servicers all made the case that there is no record of any pattern of, or statistically significant numbers of adverse events or patient injuries associated with the repair of medical equipment, either by the manufacturer or by anyone else.
In fact, ECRI’s analysis of the FDA’s own patient injuries database revealed that maintenance-related injuries occurred in only 0.005% of reported patient injuries. Each person and company that testified to the FDA reported that these percentages were consistent with their own experiences.
Even so, the manufacturers—and the associations that represented them—believed that there are many more patient injuries than the data shows. They urged the FDA to ignore the data of 3.5 million reported events and, instead, use their reported evidence. And the manufacturers were all very careful to preface their comments by saying that they were only concerned with safety issues and had no financial interest in increased regulation.
What’s the Problem?
Here is my problem with the evidence presented by the manufacturers’ lobbyists: The MDMA showed only four (not 4,000, not 400—not even 40; just four!) alleged improper servicing events, complete with photographs and labels to point out the problem areas. None of these problems cited actually resulted in any patient injury or error, although they “could have”.
Let me note that there is no chain of evidence; they can’t say who made these alleged service errors. All we know is what they allege in their photographed labels and statements. Could their own people have made the mistakes? Could the equipment have been doctored to impress the FDA? I just know that these photographs would not have been accepted in a court of law without some sort of proof about where they came from and who did the work.
So there we have it. The facts presented demonstrated that nobody has been harmed. One group—which serves to gain financially—urged us to ignore the facts and believe their miniscule amount of self-discovered problems. And the FDA is being asked to impose a myriad of regulation and oversight to cure a problem that does not exist.
Who will bear the burden of the cost of the increased time, trouble, reporting, and inspection? We all know that the companies and healthcare institutions themselves will absorb much of it—but eventually, it will only increase the cost of an already out-of-control healthcare system, while providing no benefit for anyone.
What harm will it do? Well, it will certainly put some low-cost medical equipment service providers out of business, thus decreasing access to quality, affordable service in some of the neediest and remotest areas of the country. And decreasing competition will allow some of the remaining players (and some manufacturers) to increase their rates, costs, and take even more advantage of their place in the food chain since fewer businesses will provide the services.
Here’s an idea that I heard toward the end of the conference that caused the manufacturers to pause and think. (And it’s one I support). If ECRI-provided statistics show that service doesn’t cause injuries to patients, why not remove FDA oversight and reporting from all servicers—manufacturers and all?
But until we hear something else from those in charge, we will just have to wait. After all, there’s really no telling what the FDA will do. Will they follow the facts and regulate based on evidence? Or will lobbyists and capital-heavy manufacturers inordinately influence them? Again, we will just have to wait and see.
For more information about this subject, download my PowerPoint presentation from the December Florida Biomedical Society symposium at www.htma-sc.org. It’s free to join as a member.
Patrick Lynch, CBET, CCE, CHTM, CPHIMS, FACCE, is a biomedical manager with 40 years’ experience.
During the recent FDA Docket workshop on 3rd Party Repair the attendees, of which I was one (Acertara is a member of MITA and the MITA Service committee), were treated to an amazingly dubious finding based on the search of a database where the information they were looking for could not reasonably be expected to contain. Nonetheless the number that was presented soon managed to take on the central focus of the workshop. That number was 0.005%. Instantly 0.005% was standing alongside the pantheons of numbers such as π, and Avogadros number as an indisputable value that explained the gravitational constant of the medical device service universe. Every panel presentation by independent service organizations (ISO) referenced this number as proof positive that the FDA should leave ISOs alone; nothing to see here, time to move on. This number, 0.005%, became, in a sense, the ISO holy grail. Only one problem; this number is bogus, and even if it were in some manner partially accurate it was being applied in a wholly unscientific way – and frankly was irrelevant to the topic being discussed. The FDA Maude Database upon which 0.005% hangs its metaphorical hat, was never intended to be a reliable source for fettering out specific service events, or even capturing events that did not rise to the level of “death or a serious injury” as explained in the following quote from the FDA: “Although MDRs are a valuable source of information, this passive surveillance system has limitations, including the potential submission of incomplete, inaccurate, untimely, unverified, or biased data. In addition, the incidence or prevalence of an event cannot be determined from this reporting system alone due to potential under-reporting of events and lack of information about frequency of device use. Because of this, MDRs comprise only one of the FDA’s several important postmarket surveillance data sources.” – italics added for emphasis. Additionally, the FDA does not list or even mention 3rd party services providers as entities that are mandated to report MDR events: “Device user facilities include hospitals, outpatient diagnostic or treatment facilities, nursing homes and ambulatory surgical facilities. Device user facilities must submit reports when they become aware of information that reasonably suggests that a device may have caused or contributed to a death or serious injury of a patient in their facility. Death reports must be sent to the FDA and the manufacturer, if known. Serious injury reports must be sent to the manufacturer or to the FDA, if the manufacturer is not known.” Drawing conclusions from researching a database with the foregoing caveats seems very risky at best and certainly misleading.Although a lot of various topics were brought up during the workshop I maintain the core question remains; “Is it in the best interest of patient safety for only half of the providers of medical device service to be regulated (OEMs), while the other half is not (3rd Party providers)?” The reluctance shown by the panelist representing 3rd parties to submit to FDA oversight was not reassuring. Additionally, the focus on patient safety, while critically important, did not allow time for discussing the other main facet of a proper functioning medical device; specifically – is it effective for its intended use? An MRI is at its safest when it is turned off, it is safe obviously, but not very effective. The flawed study of the Maude data base is a case in point – this data base does not capture the frequency of occurrence of sub-optimally performing devices, and the reasons for their lack of performance. The lack of this information skews the results to conclude there is no problem with service. I look forward to the next steps by the FDA in reconciling the bifurcation that has occurred in the service market with only ½ of the providers being regulated.
Wayne, While you cite some “limitations” to using the FDA data you offer no suggestions for a better dataset to use for analysis. Even given its limitations it is the best available data. You also fail to address the fact that the organizations making the push for more regulation failed to produce any factual data supporting their hypothesis. In fact as Pat points out of the 4 examples that they provided there were no patient injuries reported or referenced. One can assume that if they discovered these issues that they would have also followed up to see if there were any injuries. If as they claim a serious patient safety issue exists one would have expected to find the proponents of increased regulation reporting numerous patient injuries caused by all of the substandard maintenance that requires more oversight. In fact they failed to produce one. There are millions of repair events that take place each year so millions of opportunities for this problem to surface and be recognized.
I also found your comment that “The reluctance shown by the panelist representing 3rd parties to submit to FDA oversight was not reassuring” to be faulty reasoning. Not even the OEMs were able to buttress their support for more oversight with factual substantiated examples. In fact as was pointed out they should have used the data to make the argument that there is more data to support the argument for less oversight. Of course that would not result in more profits would it?
Outstanding article Patrick. You presented the facts perfectly. When you need evidence of OEM service failings like expired labels, poor quality preventive maintenance, etc…, let me know. I have many, many examples as I follow behind what used to be an outsourced HTM program through an OEM. As I mentioned in a previous post, regardless whether or not you are a OEM representative, in-house representative or a third party representative, it always comes down to the person that gets hired. If you are the kind of employer that doesn’t pay attention to the kind of people you are hiring or think it isn’t important what the character is of the person you hire, then you are doomed to putting the wrong person in the wrong place at the wrong time.
The FDA Maude Database upon which 0.005% hangs its metaphorical hat, was never intended to be a reliable source for fettering out specific service events, or even capturing events that did not rise to the level of “death or a serious injury” as explained in the following quote from the FDA: “Although MDRs are a valuable source of information, this passive surveillance system has limitations, including the potential submission of incomplete, inaccurate, untimely, unverified, or biased data. In addition, the incidence or prevalence of an event cannot be determined from this reporting system alone due to potential under-reporting of events and lack of information about frequency of device use. Because of this, MDRs comprise only one of the FDA’s several important postmarket surveillance data sources.” – italics added for emphasis. Additionally, the FDA does not list or even mention 3rd party services providers as entities that are mandated to report MDR events: “Device user facilities include hospitals, outpatient diagnostic or treatment facilities, nursing homes and ambulatory surgical facilities. Device user facilities must submit reports when they become aware of information that reasonably suggests that a device may have caused or contributed to a death or serious injury of a patient in their facility. Death reports must be sent to the FDA and the manufacturer, if known. Serious injury reports must be sent to the manufacturer or to the FDA, if the manufacturer is not known.” Drawing conclusions from researching a database with the foregoing caveats seems very risky at best and certainly misleading.
Patrick, I agree with you completely. I would only make one comment: It could be of value for third parties to at least become familiar with the Quality System Regulations:
* to gain an appreciation for their content * to know what can and cannot be taken for granted concerning what manufacturers do to provide for quality and risk management
* to serve as a potential source of ideas to informs medical equipment management in the field
* to discern QSR processes where increased interactions between third parties and manufacturers could be of service to patients and caregivers
Is there evidence that the Quality Systems Regulations result in quality systems that in turn result in actual quality? I once asked an FDAer this and he responded, in the spirit of my question: “that’s our position and we are sticking to it”. On the other hand one should always be open to self evaluation and new ideas.
Also, OEM service failure anecdotes are no more informative than non-OEM anecdotes, since neither should be the basis for increased regulation.
Patrick, you mentioned that OEMs identified only 4 alleged improper servicing events. I would ask the question, if a driver has only 4 speeding tickets over the last 10 years, did he only exceed the speed limit 4 times? Or was he simply caught 4 times, and possibly exceeds the speed limit every day, or 3,650 times over 10 years?
In reference to only 4 examples of alleged improper servicing events I would respond with a what might be a compare-able example. The police make a traffic stop for a driver who was driving above the speed limit. The officer runs his license and discovers that he has been sited 4 times over the last 4 years for speeding. This is how many times he was “caught” putting others at risk with his driving; however, might it be safe to assume that he could have been speeding the other 1,457 days during that same period (including a leap year), putting others at risk but was not caught? Food for thought.