Utah International Medical Devices Congress and National Policy Summits
Remarks by
David A. Kessler, M.D.
Commissioner of Food and Drugs
Utah International Medical Devices Congress
and
National Policy Summits
Salt Lake City, Utah
August 20, 1994

Good morning. I'm pleased to participate in this third annual Utah International Medical Device Congress. I am happy to be in Utah home of my good friend and teacher Dr. ____ These meetings provide an important opportunity, a critically important opportunity, for those who manufacture medical devices and those who regulate them to discuss areas of common interest and concern.

After all, we do have a common goal -- to see to it that the American people receive the benefits of new medical technology, while at the same time ensuring that new devices are safe and effective.

But sometimes that shared goal, that objective, is obscured by the problems we both experience in trying to carry out our job. Experiences that don't always tell the whole story.

From our vantage point, we have seen some companies press for approval of their devices despite the fact that they don't have sufficient data to demonstrate safety and effectiveness. And we've seen others promoting their devices for indications that go well beyond the approved indications.

From the other side of the street, there's also truth in your perception of our shortcomings. Our process for reviewing new medical devices has been too slow -- more so than we want, and more so than you want. And, I might add, more so than the Congress wants.

We all have a stake in fixing this situation. A more prompt, more responsive review system would allow us to fulfill our public health mandate more effectively. And it would allow you to carry out your research, development and marketing activities more efficiently.

So how do we arrive at the solution? How do we get new device applications in the door and out again within a reasonable time period? And how do we do this without compromising our mandate to protect people against unsafe or ineffective products?

The first step in arriving at an answer, really require us, I believe, to examine the causes of the problem. One of those is sheer workload.

[SLIDE 1] The first slide shows our workload from 1977 to 1994, and projects the workload several years into the future. As you can see, the increase has been dramatic, from approximately 2,000 submissions in 1977 to 11,000 projected for 1994. Over the past few years we've had a steady growth of about 4 percent per year, and by the end of the decade we expect to be dealing with 14,000 submissions for new devices each year.

And that's just quantity. In addition, the complexity of the submissions is increasing even faster, because of the explosive growth of new technologies. In a way, that's testimony to the highly robust nature of this industry; it is one of the fastest growing in the United States. Last year the American device industry supplied 46 percent of the world's technology needs and generated a trade surplus of $4.7 billion.

But this expansion of the industry, beneficial as it may be, has had a profound impact on the FDA. The simple truth is that the industry's growth has caused us to fall further behind that we would like in carrying out one of our most important tasks -- evaluating the safety and effectiveness of new devices before they are allowed on the market. For example, the inventory of pending 510(k) submissions increased from about 2,000 in September 1990 to over 5,000 in September 1993. Those are cold facts and figures, but they take on special meaning for the manufacturer of a new device who is waiting in line for his product to be reviewed.

We have tried our best to surmount these problems over the last year by a combination of management improvements and shifts in resources. Dr. Bruce Burlington, the director of our Center for Devices and Radiological Health, worked with Dr. Alpert to make some innovative changes in the device review program, starting about a year ago. You can see these in the next slide.

[SLIDE 2] A year ago, CDRH initiated an expedited review policy for life-saving and clinically significant devices. We did this so that important breakthroughs would not languish amid the "me-too" devices. We've had 20 expedited submissions thus far -- including leads for pacemaker-defibrillators that can spare patients from having to go through chest surgery in order to have these devices implanted.

We also shifted resources to those 510(k) submissions needing the most intensive review, in what we call the "tier system." This means that the most complex and risky devices get intensive scrutiny, while the lowest risk devices get less attention.

And we began a "refuse to accept" policy for incomplete 510(k) submissions. This means that we're simply not wasting time on inadequate applications. It also means that the good ones no longer have to wait in line behind the poor ones.

We've also taken steps to improve our communication with the industry. We want you to understand what we expect, so you're better equipped to submit a quality application. For example, we've disseminated new guidance on what constitutes an acceptable clinical study, in terms of statistical design and methodology. And we're working with advisory panels to develop guidelines for specific types of products.

We're finalizing guidance on a long-standing point of confusion for manufacturers -- how to determine whether a particular change in a marketed device is significant enough to require a new 510(k)s. We know how important that guidance is to the Industry.

We're also developing special new tools to communicate with industry. Electronic dockets now provide wide-scale access to our policy documents and guidance, and a new computerized system allows manufacturers to find out where their 510(k) stands in the review queue, by fax. And we've embarked on a series of quarterly video teleconferences for the device industry, with literally thousands of participants from coast to coast.

We've made more recent improvements, as you can see from this slide

[SLIDE 3] This fiscal year, we transferred some of the device review workload to our laboratory staff as a temporary expedient while we recruited some additional reviewers.

And we're now proposing to exempt the lowest-risk devices from the 510(k)s process, which will result in 10 percent fewer applications. Later this year, we will be proposing further exemptions for low-risk products, which should result in an additional 10 percent reduction in applications.

Taking these steps has made a difference.

[SLIDE 4] For example, we processed about 400 510(k) submissions per month from mid-1991 to mid-1993. But in June 1993, following implementation of the management initiatives I just discussed, the number of 510(k)s processed rose from 400 per month to 600 per month.

I cannot stand here, however, and tell you that these gains are sustainable over the long term. The reduction in the backlog of 510(k)s was achieved at the expense of other types of submissions. For example, actions on PMAs remained static and actions on PMA supplements have actually gone down.

In fact, we anticipate that without fundamental changes, we will begin losing ground again before long. This is illustrated in the next slide.

[SLIDE 5] The shaded area on the left shows our inventory of 510(k) submissions from 1990 through the present. Notice that the inventory peaked in September 1993. Then, as a result of our management initiatives and a modest increase in staff, it began to come down. So, at least for now, we've turned the corner on the mounting 510(k)s backlog.

But now notice the right side of the chart. The solid line shows that we will make some further gains over the next two years. But by 1997, the inventory of pending 510(k)s will start to rise again -- because of the rapid growth of the industry and its burgeoning technology.

This means that despite our management improvements, and despite our temporary reshuffling of resources, and despite a budget increase this past year, our basic problem remains the same. We simply cannot carry out an effective, timely and well- balanced program of evaluating new devices unless we achieve a better balance between our workload and our resources.

And this brings me to the dotted line on the right side of the chart, which shows how user fees can dramatically alter that course, bringing the backlog down and keeping it down.

Let me acknowledge that I recognize that some manufacturers feel uneasy when they think about the prospect of a user fee program for devices. Even some in the industry who have voiced support for user fees have done so reluctantly. Part of the unease is a general feeling that regulatory agencies should not have to bill the industry in order to get their work done. The other source of the reluctance is far more specific -- a fear that the collected fees might drop into some sort of government black hole, never to be seen again. Or that we would not be held accountable for producing tangible results. Or that the fees might place an unduly heavy financial burden on manufacturers, particuarly the smaller ones.

Let me address both types of concerns. With respect to the general unease with the idea, I think we can all agree that in an ideal world -- one in which we had the resources necessary to keep pace with our obligations -- there would be no need for user fees. But we do not live in such a world. And, given the continued growth in our workload, it doesn't seem at all likely that we'll soon see that kind of world. So user fees for medical devices may be an idea whose time has come. And that's not necessarily a bad idea. After all, user fees are based on an equity principle of economics, one that I think makes sense -- namely, that it is fair for the beneficiary of a government service to help pay for providing that service.

As to the industry's more specific concerns, let me share with you the elements of a user fee program which are supported by this administration. I readily acknowledge significant contribution of Chairman Dingell to the Central Principles listed here

[SLIDE 6] First, the user fees must be additive to FDA's medical device program. That means that the fees cannot be used to support other parts of the government, or even other parts of FDA.

Second, the user fees must be dedicated to premarket review functions. That means that the fees cannot be used for other aspects of FDA's medical device program, such as research or enforcement.

Third, the user fees must be linked to specific performance goals. That means that we will be required to establish specific benchmarks for our device evaluation program, and we must be accountable to the Congress in meeting those benchmarks.

Fourth, the user fees should be billed per application, not per establishment. That means that manufacturers who do not use the system will not have to pay for it.

And fifth, the user fees should must be tailored to accommodate small business. That means we must take into account the fact that the device industry is dominated by very small companies, and design a fee structure that will not pose an unreasonable burden on these companies.

The fees that we anticipate for devices are shown in the next slide

[SLIDE 7] Note that the fees for devices are significantly smaller than for drugs. And that they remain level for five years, whereas the drug fees increase annually. These fees will come to about $23 million and will allow us to add about 200 additional staff for five years.

I also want to address the issue of performance goals, which are shown here

SLIDE 8 Within 24 months, our goal is to eliminate 95 percent of the 510(k) backlog and 90 percent of the PMA backlog.

On a long-term basis, we're aiming for final action on 95 percent of 510(k)s within 90 days, and comprehensive review of 90 percent of PMAs within 180 days.

I should stress that these goals will require us to process submissions for new devices at a faster rate than at any time in the past decade -- and with high quality. We are willing to do this because that is what's needed to spur innovation. And throughout the process, we will be accountable to the Congress.

I feel strongly that the FDA must keep pace with the Industry's growth. We owe it to you to produce timely and predictable reviews of new medical devices -- and we owe it to the American people to get safe and effective medical devices to patients as soon as possible. A user fee program represents an enormously important way to get there.

Thank you. ####

(Hypertext updated by jch 1998-MAY-22)