Statement of Diana Zuckerman, Ph.D., President, at the FDA Public Meeting on the Reauthorization of the Medical Device User Fee Act (MDUFA)

September 14, 2010

Thank you for the opportunity to comment on MDUFA. The National Research Center for Women & Families is dedicated to improving the health and safety of adults and children, and our Cancer Prevention and Treatment Fund is focused on improving the early diagnosis and treatment of cancer.  My perspective is as a researcher trained in epidemiology and public health at Yale Medical School, with experience doing research at Vassar and Harvard, as a fellow at the University of Pennsylvania Center for Bioethics, and as a former Congressional staffer focused on FDA issues.

1. What is our assessment of the overall performance of the medical device user fee program?

We believe that the medical device user fees are too low to adequately support the important work of the Center for Devices and Radiological Health (CDRH).  The user fees for 510(k) reviews and for PMAs are a small fraction of the actual cost of the reviews, and an even smaller fraction of what it would cost to do a more thorough review that includes effective and timely post-market surveillance.  We support sliding fee scales for smaller device companies, but the fees must be adequate to provide a meaningful contribution of resources for the work involved.  And, there is no reason why large corporations should pay so much less for each medical device application than they (or comparable companies) pay for an application for approval of a prescription drug.

The inadequacy of the fees for the 510(k) is especially egregious.  There is no way that the FDA can have the resources needed to assure the safety and effectiveness of medical devices reviewed through the 510(k) process with user fees of only a few thousand dollars each.  The largest companies should pay the entire cost of the 510(k) review, as well as additional substantial user fees for every new ad that the company directs at consumers.  And with the possible exception of the tiniest companies, all device companies should be able to afford a 510(k) user fee of $10,000 or more.

Even the much larger PMA user fees are disproportionately low compared to comparable reviews for prescription drugs.  The user fees for PMAs and NDAs should be very similar.

Our Center recently conducted a study of high-risk recalls of medical devices for the past 10 years, based on FDA’s own data.  We expected that most high risk recalls – those that FDA deemed necessary to save lives and prevent serious harm – would be devices approved through the PMA process.  After all, the PMA process is for high risk devices, whereas the 510(k) process is for low to moderate risk devices.  So, imagine our surprise when we found that most of the high risk recalls involved devices that had been cleared through the 510(k) process.  Many of these devices were responsible for patients dying from devices that didn’t work, although some were unsafe even when they did work.

If a device can kill you if it fails, or if it doesn’t work accurately, then it should be considered a high-risk device, not just a high-risk recall.  If it is a high-risk device, by law it should be Class III and be submitted for a PMA review.  To protect patients, high-risk devices should never go through the 510(k) process.  That especially includes implants and diagnostic devices for serious diseases or those that emit radiation.

2. What aspects of the program should be retained, changed, or discontinued to strengthen and improve the program?

It is essential that new MDUFA legislation:

  • Substantially increase FDA resources for both 510(k) reviews and PMAs
  • Revise current law to ensure that devices that can potentially sustain life, save lives, or harm lives are held to higher standards of safety and effectiveness, at least as high as current standards for prescription drugs.
  • Ensure that the FDA does not continue to approve dangerous devices that are similar to other dangerous devices already on the market.  The recent New York Times article on deaths that result from misconnecting IV tubes and other medical tubes is a perfect example of a system that is not working.  The big issue here is: safety needs to be measured by how a device functions in the real world, not in an ideal world.  Even if the risks are caused by human error, that doesn’t excuse the device or the FDA.  It should be the FDA’s job to require that devices be easy enough to use that human error is a rare problem, not a common occurrence..
  • In addition, changes in the law and user fees are needed to support post-market surveillance of all medical devices, inspections of all medical devices, and pre-reviews of direct to consumer advertising of medical devices.

MDUFA should protect the public health by improving the safety and efficacy of all medical products, whether PMA or 510(k).  That requires higher user fees, performance standards that focus on the quality of the review and not just the speed, and reviews that provide patients, consumers, and healthcare providers with information about how well the product works compared to other similar products on the market.

Currently, more than 90% of new medical devices are not required to undergo clinical trials to prove that they are safe and effective.  And let me be specific: what’s needed for moderate and high-risk devices are well-designed studies, not anecdotal clinical data.  Currently, clinical trials conducted on devices are often too small or too short-term to provide meaningful safety and effectiveness data.    The FDA must demand better data to protect the public health.  At the same time, they should give priority to reviewing applications for breakthrough products, and for products that provide comparative effectiveness data that tells patients and providers how this new device compares to others already on the market.

We agree with Consumers Union that it is important to poll your CDRH staff to obtain confidential responses to key questions about the current processes.  We would suggest questions such as: “Are devices being cleared where you have doubts about whether they are proven safe or proven effective?

FDA laws and regulations should require the same standards of safety and efficacy for devices as for drugs.  I don’t believe a Class IIb category should be added to the 510(k) process, unless it is a relatively small category.  We need a much larger percentage of new devices to be reviewed under the PMA system.  If a device can cause serious harm if it fails – even if it is not life sustaining or life saving if it works-it should be reviewed by an improved version of the PMA process.

Inspections

Inspections of the manufacturing facilities of devices are crucial.  It is outrageous that 510(k) devices are not subject to FDA inspections.  That must be fixed immediately, and let me be clear that third party inspections are not adequate.  The need for more resources for inspections of PMA devices is also urgent.

Post-Market

In addition to the need for more resources and improved quality of pre-market reviews of devices and the addition of inspections, user fees are needed to support post-market studies and surveillance.  First, the law needs to be changed to allow conditions of approval of 510(k) devices that include post-market studies.  It is ridiculous that the process that requires the least scrutiny before approval allows no post-market requirements.  FDA must fix that immediately.

Although most urgent for 510(k) devices, postmarket studies and surveillance of PMA devices are also urgently needed.  The clinical trials used as the basis for medical devices are almost always less well designed and executed than those for pharmaceuticals.

One of the ways to improve post-market surveillance is for FDA to finally implement a system of unique identifiers for medical devices and patient registries, especially for implants.  As someone who recently had my hip joint replaced, I can tell you that I was given no information about what specific device was used.  When I heard about several hip joint recalls in the weeks after my surgery, I had to contact my surgeon to find out whether I had one of the recalled devices.  (Fortunately, I don’t).  I was told which company made my hip implant, but still don’t know the model or any specific information about it.  In contrast, my toaster, TVs, and DVR all have unique identification numbers that I can easily read.

And finally, if a product is on the market and found to be ineffective or unsafe (such as ReGen mesh or medical tubes that are easily misconnected), it is FDA’s responsibility to take the unsafe or ineffective devices off the market.  I don’t want to pick on ReGen mesh, because there are so many 510(k) medical devices that apparently don’t work – not surprising, since clinical trials were not required.  Vagus Nerve Stimulation, the NeuroStar TMS depression device, and laser treatments of lymphedema are just a few of the many medical devices that have wasted millions of healthcare dollars.  If user fees are high enough to substantially offset the cost of the reviews of new devices, that should mean more appropriations available for such post-market safety decisions, so that they can be made in a more timely manner.

We agree with the recent CDRH 510(k) Working Group report regarding the need for a useable, searchable database of 510(k) predicates.  And, part of the MDUFA improvements should be the assurance that a device will not be cleared as “substantially equivalent” to a previously recalled device, or to a device that was cleared on the basis of a predicate device that was substantially equivalent to a recalled device:

We agree with device companies that since they need to report adverse reactions in a timely manner, the FDA should analyze adverse event data in a timely manner. User fees should be used to make that a reality.

User Fees, Not Payment for Services

We will urge Congress to consider again legislation like Rep. Maurice Hinchey’s HR 4816 that breaks the connection between user fees and specific performance levels. The FDA should not be restricted by performance timeframes that undermine public health performance standards

PDURF IV and FDAAA improved safeguards for pharmaceuticals, but not for devices.  We believe that the new leadership at CDRH and at FDA can do better.

Thank you for the important work that you do on behalf of the public health, and thank you for your consideration of our views.