Archived Content

The content on this page is provided for reference purposes only. This content has not been altered or updated since it was archived.

Medical Devices

Transcript: Identifying Unmet Public Health Needs and Facilitating Innovation in Medical Device Development, June 24, 2010




+ + +


+ + +



+ + +

June 24, 2010

8:00 a.m.

Hilton Washington DC North

620 Perry Parkway

Gaithersburg , MD 20877













Research Associate, Consumer & Patient Alliance Coordinator

Center for Medical Technology Policy, Baltimore, MD


Medical Director, Diabetes Research Institute

Mills-Peninsula Health Services, San Mateo, CA


Chair, Department of Radiology

Mayo Clinic, Rochester, MN


Board of Directors, National Venture Capital Association (NVCA)

Past Chairman, Medical Industry Group, NVCA

Founding General Partner, The Vertical Group, Summit, NJ



Director, Quality Assurance & Regulatory Affairs

Intel Corp/Digital Health Group, Folsom, CA


Chief Medical and Science Officer

West Wireless Health Institute, La Jolla, CA


Senior Director for Global Regulatory Affairs

Medtronic, Inc., Minneapolis, MN


ROBERT CAMPBELL, M.D., Children's Hospital of Philadelphia, The Need for Better Metrics for Pediatric Device Evaluation.

MARK SOBEL, M.D., Ph.D., Association for Molecular Pathology, Barriers to Innovative Diagnostics: The AMP Perspective.

CYNTHIA RICE, Juvenile Diabetes Research Foundation, Accelerating Improved Technologies to Treat Diabetes.

ZEN CHU, Wyss Institute, Harvard Medical School, Academic / Industrial Partnerships for Innovation & Commercial Translation.

KONSTANTIN GORANOV, SALUTARIS, Predicament of New Devices and Old Predicates.

RICHARD EATON, Medical Imaging & Technology Alliance (MITA), Barriers to Development and/or Redesign of Medical Devices.

DAN COHEN, Enteromedics, Inc., Systemic Change to the US Medical Device Approval Process.

CARL ROSNER, Cardiomag Imaging, Inc., EARLY, SAFE DETECTION OF HEART DISEASE POSSIBLE - Huge Savings In Lives and Money.

LISA SALBERG, HCMA - Hypertrophic Cardiomyopathy Association, Patients with Genetic Cardiac Disorders Requiring a Lifetime of Therapy - New Options Needed.

PEG GRAHAM, QUA INC., Self-Managed Bedside Commode: Avoidable Falls, Institutionalizations and Workers Comp Costs.

MARK DRUCKER, NewCare Medical




WELCOME - Jeffrey Shuren, M.D., J.D. 7

REVIEW OF MEETING OBJECTIVES - Jonathan Sackner-Bernstein, M.D. 9



Jonathan Sackner-Bernstein, M.D. 15

Louis Jacques, M.D., CMS 18

Jean Slutsky, AHRQ 20

Col John Bullock, PA-C, DOD 22

Gerald Cross, M.D., VA 24

Belinda Seto, NIH 26

Lt Col Daniel Wattendorf, M.D., DARPA 30




REAL WORLD EXPERIENCE - Justine Seidenfeld





Richard Ehman, M.D.






Robert Campbell, M.D. 135

Mark Sobel, M.D., Ph.D. 137

Cynthia Rice 139

Zen Chu 143

Konstantin Goranov 145

Richard Eaton 147

Dan Cohen 149

Carl Rosner 153

Lisa Salberg 156

Peg Graham 158

Mark Drucker 161

Mark Adlinger 161




FOR INNOVATION - Joseph Smith, M.D., Ph.D.






(8:00 a.m.)

DR. SHUREN: Good morning. I'm going to go ahead and get started. Can everyone hear me in the back? Perfect. Thank you.

Medical device innovation is responsible for significant public health benefits, including the prevention, treatment, diagnosis, and monitoring of serious or life-threatening diseases and improved quality of life. However, the path to patients for a medical device, particularly for more novel technologies, can be challenging. It can take years for a device to finally reach and help patients.

It starts with a new concept, moves through the rigors of research and development, and then undergoes a regulatory review. Even after it enters the market, the new device has to be covered by insurance, possibly receive a new code for billing purposes, and receive adequate reimbursement. It may take additional months, even years, before the new device becomes a standard of care. It must be widely adopted by clinicians and clearly understood from a risk/benefit perspective. All the while, the next generation of medical marvels is advancing through the pipeline.

The regulatory agency and third-party rules of the road attempt to achieve our public health goals by balancing the need to foster innovation with the need to assure that medical devices are safe and effective and provide meaningful value for payment.

While it's not possible to strike the perfect balance for all devices, we can be smarter about how we do it. The agencies represented on this council each play an important role in the life cycle of medical devices. The creation of this interagency council is an attempt to work together with industry, healthcare practitioners, patients, academia, and other members of the public to answer the question, can the federal government do more to encourage the development and bring to market the next generation of medical devices while protecting consumers by recalibrating the path to patients?

Today we want your input and ideas on the ways the federal government can minimize or even remove barriers that may hinder the development of medical devices that address unmet public health needs. Unmet public health needs are illnesses or injuries that meet the following criteria: First, they are serious or have moderate adverse impact on health, but affect many individuals. Second, they could be cured, significantly improved, or prevented by the development or redesign of a device. And third, the device is delayed or is not being developed or redesigned due to barriers that the federal government can directly or indirectly remove or minimize, and those barriers are out of proportion to what is warranted, based on public health needs.

We've already taken one step today. FDA Commissioner Hamburg and CMS Acting Administrator Tavenner have signed a Memorandum of Understanding that will allow for the first time routine and timely sharing of information and expertise between our two agencies to strengthen our ability to achieve our respective missions and creating new opportunities for collaboration.

One of the potential collaborations under the MOU that we are seriously exploring is the ability to start at a manufacturer's request the Medicare national coverage determination process while its medical device is still under pre-market review at the FDA, what we call parallel reviewing. Such an approach could reduce the time from FDA approval to CMS coverage and payment for some devices. It's an example of what FDA and CMS could do together to speed patient access without compromising patient safety.

Before we begin public comments, Dr. Jonathan Sackner-Bernstein, FDA's Medical Device Center's Associate Director for Postmarket Operations and the person coordinating this council, will discuss the logistics for the meeting. We'll then hear from each of the council members about actions each agency is taking or recently taken to facilitate medical device innovation.

I thank you in advance for what I anticipate will be a lively discussion and an informative day.

DR. SACKNER-BERNSTEIN: Thank you, Jeff. And I also would like to thank everyone who is here today for taking the time and applying their energies to this very important initiative on the part of all of these agencies.

We recognize that successful innovation is not something that can come into being merely by passing laws or issuing new regulations. Yet, part of the mission of the FDA is to foster advances in medical device performance and their impact, and this council, in parallel, is also charged with and has accepted the responsibility for looking at ways that we can facilitate and foster the development of innovative products that can impact public health.

But at the same time, these goals of fostering innovation and moving things faster must be balanced against the need for the regulation necessary to assure safe and effective products. It is in the context of what we all see around us of the increasing pace of technology and medicine advancing, that we feel the pressure and understand the urgency to find the right balance and enable the innovations to move forward.

If we look back in history, we can see many lines of evidence that show us the kind of forces that are in action right now, in terms of the pace of technology and medicine advancement. In the mid-1880s, Professor Semmelweis identified the basis for peripartum infections and the high mortality rate; followed soon thereafter by the work of Louis Pasteur, identifying the germ theory, showing that the germ theory as the cause of infection was, in fact, the basis.

In about a century, after the discovery of drugs like penicillin and the use of aseptic technique, we are able to see the mortality from bacterial infections slashed dramatically and the incidence of rheumatic heart disease virtually wiped out in the developed world.

A similar time course accompanies the evolution of medical imaging technologies. Professor Röntgen's discovery in 1895 of the utility of x-rays was followed by a series of developments in ultrasound, computerized tomographic x-ray imaging, MR imaging. And what we saw is, in the course of less than a century, the field, the technique of exploratory surgery, has virtually been eliminated.

The pace of innovation, though, increases dramatically over the next period of time after these developments. In the mid-1960s, the NIH established the artificial heart program. Barney Clark underwent the first successful implantation of a total artificial heart, and in 40 years from its inception, there was an approved total heart.

The strategy of assisting the heart instead of replacing it started at about the same time, and within a 30-year period, we went from its first successful use in the mid-'60s to the first approved device. The pace is increasing.

Research aiming to develop a therapy for treatment of patients who have suffered sudden death started at Johns Hopkins in 1969. In less than 30 years, there were randomized clinical trials showing that these devices improve survival.

And if we look outside of the medical developments and into technology development, we see the same pattern, and that's important because medical devices, in many ways, parallel the evolution of technology in general. We went from the first flight on a beach in North Carolina to the moon in about a century. The first computer was built during World War II. Of course, it took a room the size of an apartment and had less power than a handheld calculator of today, but we were able to move in 40 years to very, very powerful computers and then the work of DARPA to establish in the late '60s a distributed model of computing power. First used in 1969, who would have realized that only 30 or so years later we would see the development of the Internet and its impact?

And perhaps not at the same scale, but at the same pace, we've seen developments such as the impact of Google as a technology breakthrough that's changed the way we functioned, now happening in less than a decade, transforming our activities. And when the human genome project was started in 1990, who would have predicted that it would take only about ten years to basically be done with mapping the human genome.

So we're in the time where there is an incredible increase in the pace of technology. And while people here at the front tables and probably people in the audience don't necessarily know with certainty where the next breakthrough will be, one thing is for sure is that we have the opportunity to essentially see the application of Moore's Law from semiconductors apply to the development of medical technology. The pace of advancement is increasing. The impact, the possibilities, and the opportunities are getting greater, and we look forward to today as another important step in unifying these agencies so we can posture ourselves and prepare ourselves to be able to handle these kinds of technology developments.

With that said, what I'd like to do now is just talk a few moments about the logistics, which, of course, is what Jeff asked me to do, but I couldn't resist talking a little bit more. What we're going to do is move through a sequence that's on the agendas that were on your chairs, where we're going to give somewhere between a couple minutes, or 5 to 10-minute presentations, from each of the representatives of agencies, once again, as Jeff said, focusing on what we're doing now, what we'd like to do more of, and give you a sense of where we see some opportunities.

But the rest of the day is really going to be focused on us hearing from you. There will be dialogue, largely to try to clarify how some of the suggestions might be useful, how we can pursue them, as we then prepare to go into the next step of trying to figure out what the right paths are to pursue in the coming months.

There were many advance requests to speak at the Open Public Session. You'll see the list that's prepared there. We'll call on people one after another. I think we've got the list right from people who have requested. There were some people who decided not to make presentations. Based on the total number, it looks like we'll have about two minutes per person, so we'll ask you to keep your remarks to that period of time so that everyone who requested to speak can. And if there are others who would like to speak, please sign up at the front, and if there is time, we'll have you involved as well.

It may not feel quite right to only have two minutes, but there is a docket that is open that will remain open for a month where we will look and listen and read your comments, which we hope you will go ahead and submit to us, to help guide us as we take the next step going forward.

What I'll do now is I'll turn to the members of the council, and maybe we can each take a moment to introduce ourselves. Jean, I'll start with you.

MS. SLUTSKY: Good morning. I'm Jean Slutsky. I direct the Center for Outcomes and Evidence at the Agency for Healthcare Research and Quality.

MS. SETO: Good morning. I'm Belinda Seto. I'm the Deputy Director of the National Institute of Biomedical Imaging and Bioengineering, one of the 27 institutes of the National Institutes of Health.

DR. JACQUES: I'm Louis Jacques. I'm the Director of the Coverage and Analysis Group at CMS.

DR. SACKNER-BERNSTEIN: Jonathan Sackner-Bernstein from CDRH.

DR. SHUREN: Jeff Shuren. I'm the Director of the Center for Medical Devices and Radiological Health at the FDA.

DR. CROSS: I'm Gerald Cross. I'm the Deputy Chief for Patient Care Services at Department of Veterans Affairs.

Col BULLOCK: John Bullock from the Assistant Secretary of Defense's Health Affairs Defense Medical Research and Development Program Office.

Lt Col WATTENDORF: I'm Dan Wattendorf from the Defense Advanced Research Projects Agency.

DR. SACKNER-BERNSTEIN: Okay. What we're going to do now is move to the first presentation, which will be basically a presentation by the members of the Panel real briefly.

We all understand that the process for bringing medical devices and radiation-emitting products to market is one that many believe is somewhat of a black box. That's not because we want to keep it that way, but there are folks who have limited experience with us going down different pathways, and anyone of those pathways, which are going to be different based on the product being developed, are going to require different steps being taken.

In terms of the way the process works now, what I'm going to show you here on this slide is the way one path might evolve. And what I'm trying to show you here is that there are many steps. And by the way, since you can't read it -- there are many steps, as you'll see -- these slides we'll eventually go through, review according to 508 regulations, and be posted on our website, but it may take a few days. But you will be able to download them there and see them in detail.

I think the key message you should get from this slide is that along the lines from the idea generation through evaluation, marketing, all the way to product obsolescence, there are many places where industry is involved, there are many places where these federal agencies and others are involved. And a key responsibility for us on this Council is to try to understand where we may be hitting bottlenecks that are slowing down innovation or reducing the impact of innovation so that we can address those, as long as we make sure we don't sacrifice our assessments of safety and effectiveness and other requirements that each of the agencies have to make sure that the best medical care is being delivered.

You'll also notice that on this slide, I have not included practitioners, academic societies, or professional organizations. And that's not because they're not important but because the focus of today's discussion is really on steps that this group can take to facilitate innovation and impact public health. And as a reminder, those of you who will be speaking in the Public Session, we would hope you would keep that in mind and focus on targets that we can address as a group.

What I'd like to do now is shift to some of the current activities by these agencies, as was mentioned earlier. At CDRH, our objectives are listed here. They include our goals for predictability, the facilitation of innovation, which is the main focus of today's meeting, of advancing regulatory science and taking on these tasks in a transparent fashion.

Some of the activities that are ongoing now that address this include two task forces that Dr. Shuren set up as he became Center Director. Those reports are going through the review process, and we hope to have them available for public comment in the coming weeks. One is addressing the 510(k) process and the other on how science is used in regulatory decision-making.

We also have taken steps to proactively use the expertise we have in-house. In other words, instead of waiting passively for people to bring questions and issues to us, we're seeking out places where we can use our own expertise to help solve problems. And one example here is cited from the recent work by the Infusion Pump Working Group, which was released back at the end of April, that talked about how we can use our skills to help make sure the products that are in the marketplace are performing better.

This Council is an obvious part of the activity that we're using to facilitate innovation. We are continuing to develop new tools to assess safety and effectiveness. That includes different methods and analysis approaches as well as relationships. And then the FDA and CDRH transparency initiatives are going to include more postings of more of our actual decision-making on medical products.

I'll now turn to Louis to talk about CMS.

DR. JACQUES: Thanks, Jonathan. From CMS' perspective, let me talk a little bit about our context. Sometimes something new is clearly better. Sometimes something new is different. Sometimes something new is just new. And it can be very difficult to predict before there has been a significant amount of use that would generate an evidence base as to what category any particularly innovative device or drug may fall.

And while that's easy to do, maybe, retrospectively, it is hard to predict. And a big question that we struggle with is when is the right time to call the question. Those of you who have or have had teenagers have been through the SAT or the ACT, or whichever test your child took. Clearly, one can try to take that test too early and, in fact, fail, because it was premature to call the question. You wouldn't put a kindergartner in an SAT test environment, okay?

At the same time, you could also wait too late. You could wait till you're 30 years old to take the SAT. You'd probably have a fair amount of real-world experience that might help, but then you've probably missed out on many years of education and then career opportunities in that way. So we do struggle sometimes with when is the best time to call that question.

Our priorities essentially revolve around Medicare beneficiaries. We would like to base our coverage decisions on a review of all the relevant evidence, not just the convenient evidence or the readily available evidence. We do want to work with people to minimize significant evidence gaps. We do realize that there is a value of information calculus and that some evidence gaps may not be worth addressing compared to other, more pressing evidence gaps.

We do want people to focus on the real-world context in which Medicare beneficiaries use devices, which is often in the home with a beneficiary's family member who may be visually impaired, hearing impaired, or cognitively impaired and is trying to use a device that may have initially been tested with very high-level, technically skilled people in an inpatient hospital setting. We would like to encourage the development of technologies that really produce a meaningful benefit in the lives of beneficiaries and to help everybody do that a little bit more efficiently by finding those places where there is a synergy between CMS and FDA and our regulatory needs.

The way we have tried to do that is we're trying to be very public about evidence gaps, especially through some of our other avenues, for example, the MEDCAC, the Medicare Evidence Development and Coverage Advisory Committee, instead of only doing MEDCACs when we have an open national coverage determination. You may have noticed recently we've been doing a lot of MEDCACs, and there is not necessarily an NCD open at the time. There is not necessarily an NCD following soon on the heels of that MEDCAC. We do think that we can use our sort of public visibility to draw attention to technologies that may not have a natural stakeholder base, for example, chronic lymphedema.

We are increasingly doing post-coverage analysis to actually look at claims data and other data to see what has been the effect of national coverage determinations. Any of you who follow MEDPAC, the Medicare Payment Advisory Commission, may recall that this March they had a very complete discussion of coverage with evidence development, and their public report has just been published within the last couple weeks. We believe that when there are outstanding evidentiary questions, that CED, in one of its various forms, may be the best way to enhance access while at the same time making sure that we don't find out ten years later that early access simply meant an early opportunity for medical misadventure.

To that extent, we do try to collaborate with FDA, both in terms of Medicare coverage of IDE trials and in early engagement with stakeholders. We are continuing to work on our clinical research policy, which is clearly a work in progress, but we're very interested in this particular area.

MS. SLUTSKY: We're good. You know, Jonathan mentioned Dr. Semmelweis, and it brings to mind some actual research that I did on his early work on postpartum infection which shows how hard change is taken up and how new technology is adopted, particularly when it's beneficial. Dr. Semmelweis actually died in an insane asylum because he was rejected by his peers and ostracized and basically kicked out of the hospital in which he did the research.

So it's a sobering statistic that really getting information and evidence on the impact of technologies and making sure that that gets into the hands of people who need it, and particularly the patients who will benefit from it, is incredibly important. And we need to keep those implementation issues at the back of our mind at all times, which is very close to AHRQ's mission.

So I just wanted to briefly tell you some of the things that we're doing. This is a very small snapshot, but one of the things that we're trying to bring to decision-makers are sort of a snapshot of new innovations, new technologies -- and we define that very broadly, that it can be a surgical intervention, a device, a new biologic, a new radiological device -- to provide technical briefs or contextual descriptions of what this new innovation is even before there is any evidence as to what impact it has in different patient populations so that people are aware that this is potentially something that could be very important.

We also work through our legislative statute with CMS on coverage with evidence development. Often, we're the scientific arm of that effort, as well as paying for some of the studies. One of the most recent ones is actually continuing an evaluation of ICDs in the elderly, particularly concentrating on subgroups, to determine firing rates. And this is actually a very innovative collaboration between the American College of Cardiology, the HMO Research Network, NHLBI, and AHRQ. And that type of public-private partnership is really an important milestone, in my opinion, for how we can go forward with research projects of this type.

And then, finally, working with both CDRH and CMS, through their statutory requirements, oftentimes, the work that we do at AHRQ is complementary, both in terms of funding registry development and methods development on how to do technology assessment, methods on how to use observational studies in assessing technologies, and methods development on how you actually assess devices as opposed to pharmaceuticals or biologics, which are very different in how you evaluate them.

And then, finally, we often perform health technology assessments at the request of CMS as part of their national coverage decision process.


Col BULLOCK: DoD is primarily an end-user of devices, but we use them in an austere environment and often with users with varying levels of training. For example, looking up here, we're promoting standardization of medical supplies and equipment and stressing interoperability.

For example, if the soldier is down, wounded, he's picked up, taken care of by a medic who may put on a device at that time, and then gets picked up by an Army helicopter, taken to an Air Force facility and then finally air evac'd out, possibly out of the theater entirely, but several services may have touched the person during the treatment and evacuation, and up until recently, they may not all have had the same device as the native one that they trained on. So we're really working on interoperability and standardization, especially through Defense Medical Material Program Office, which is based at Fort Detrick.

We have a tri-service Deputy Surgeon General Board that collects information from the field. They're collecting information both from reports from the field and also looking at devices that are left on servicemen killed in action that come back into Dover, and through inference, in looking at that, seeing how the device was used, whether it was used as it was originally designed, whether people are improvising in the field, whether it's working as it was predicted to work, and we're starting to use that information and also getting some of that information back to the FDA and back to the manufacturers and making the devices more user-friendly and efficient.

And here is a slide that says we do that through collaboration with different DoD training agencies, the Committee for Tactical Combat Casualty Care, which is a standardization committee for combat casualty care specifically, combat developers, operational testers, researchers, and the logisticians, specifically educating the logisticians that cheaper isn't always the best choice.


DR. CROSS: Good morning, again. At the VA, we provide and offer comprehensive healthcare, integrated healthcare, for enrolled veterans across the United States and actually elsewhere. We see about a million patients a week.

The scope of services that the VA provides is very extensive, and it starts with prevention, with a foundation in primary care and going up through the -- on to surgical specialties, including multiple transplant programs within the VA. We're highly affiliated with universities and medical schools across the United States. I think it's over 110 medical schools at this time.

We're purchasers of devices, but we're also innovators of devices and vaccines and medications, based on our very strong research program that we're quite proud of and that many of you probably have had an opportunity to participate in.

An example of innovation, I think, in the VA of course would be, in terms of technology right now that we're probably well-known for, is the work that we've done on electronic health records, which has become a really strong foundation for our care, in primary care. Always the focus is on patient safety.

I wanted to alert you to a new initiative being developed and implemented at the VA, and we call it patient-centered medical home. And it's not just at the VA. It's certainly a national initiative as well, but we're very much moving forward with this in the VA also. The impact of this, in terms of medical devices, is that we want to become facility-dependent. We want to do more home monitoring. We want to do more testing at home. We want to do more telemedicine, reaching out into even remote communities to offer more sophisticated services to our veteran population.

In all of these situations, safety in monitoring and engagement with the Food and Drug Administration is of great importance. I wanted to give you one example of a program that we have recently developed where we do report back monitoring information on devices to the Food and Drug Administration. We call that CART-CL, C-A-R-T hyphen C-L. And it stands for cardiovascular assessment reporting and tracking system for cath labs. And on a periodic bases, probably monthly, we have meetings with the Food and Drug Administration to report back on our observations and findings regarding technology used in the many cardiac cath labs that we have across the country, providing information, as DoD reported, that could be of great value not just within the VA but I think outside the VA as well.

So I'm very much looking forward to hearing from the audience this morning, the speakers that are going to be presented, and I appreciate you all being here. Thank you.

MS. SETO: So good morning. I want to first recognize my partners on this Council, Dr. Guillermo Arreaza-Rubin from the National Institute of Diabetes and Digestive and Kidney Diseases, Dr. Tim Baldwin from the National Heart, Lung and Blood Institute, where I'm only speaking on behalf of all of the hard work that they have done for this workshop.

And so as Jonathan pointed out in his opening remark, NIH has had a long history in supporting research in the early development of devices. He pointed out the specific example of artificial heart, but there are many others, such as continuous glucose monitoring devices that we outline in this slide, and by nature or by inference, supporting the early development of devices, really rely on innovation. In fact, it's such a hallmark of what we do that it is part of the peer review criteria that NIH uses in evaluating applications.

I summarize on this slide a couple of examples, and I want to point out that these are really examples for illustrative purposes, and by no means these are exhaustive of what NIH supports in terms of medical devices. So in two broad category: the therapeutic device research versus the diagnostic devices research.

Under therapeutic, you will see the example, for example, for therapeutic ultrasound, tissue engineering, regenerative medicine, where we collaborate with the DoD in the AFIRM, Armed Forces Institute of Regenerative Medicine, and ventricular assist device, which clearly is an outcome from the National Heart, Lung and Blood Institute that began with an artificial heart and now expanded that program, and that's one example of their expansion; cochlear implant, which we collaborate with the Deafness Institute, artificial pancreas -- clearly this is within the mission of the diabetes, digestive and kidney diseases.

And under diagnostic devices, there are point-of-care -- and as to the comment made earlier about moving the paradigm of medicine from a hospital base to at the point-of-care, which may be at home, which may be in the village in the developing countries.

So one put that I want to make in terms of innovation is how do we measure the impact of that? And we talked about assessing the evidence, but it's also the global health and it's not just about developing devices for this country, but it's really, truly to develop for the world for developing country. And where point-of-care technology is really quite relevant for that arena.

And the last point, one of the -- we also want to address the issue of how do you actually provide a funding mechanism that is viable to move it in an expeditious manner from early development to application and translation. One that is used by all the federal agencies represented here is the Small Business Innovation Research program, the SBIR program, and I have just a few short slides to demonstrate some of the examples of research that has been funded by SBIR. And I will keep this very brief to make sure that I don't overuse my time.

So in terms of the SBIR, there is a partnership, companion program, which is STTR, which is a partnership between the small business firm and academic institution. There are two phases. The first phase is proof of concept. The second phase is really to really test the feasibility and the commercialization of such device in the marketplace.

Okay. We all recognize, too -- another point is that in developing devices, oftentimes the academicians primarily faces so-called "valley of death." How do actually translate the early development, the concept, into a commercializable, applied device. And NIH has a number of programs to assist these academicians, and we have the niche assessments that are applicable for the phase one awardees and also the commercialization assistance.

I will now give you some example of the SBIR program from my own institute, which addressed -- focuses on unmet needs of medically undeserved population. Again, I talked a little bit about measuring the impact of innovation, measuring the impact of devices in global health, but we also do not forget that in this country there are underserved populations, and within the SBIR program, one can issue specific requests for applications, which we did. And this is an initiative that we are in partnership with the NIH Center for Minority Health and Health Disparity. And we fund to the tune of just about $3.5 million that we set aside for this program.

So some of examples of the projects funded under this program. I already talked about the point of care. An example, a specific one, is here for isolating, identifying drug-resistant TB strains. How do you actually help and assist underserved population to read their drug label? And the telehealth, my VA colleague talked about telehealth, and again, for remote monitoring of Parkinson's disease in this specific example. Multi-functional self-service health screening kiosks, and this could be a kiosk in developing country. We actually have a grant -- which is you go up to the kiosk, and you could test for STD diagnostic. And so this is actually, again, another example of point-of-care funding technology for molecular diagnostics in low-resource setting.

As we understand more about the pathophysiology and the disease mechanism, it allow us to practice medicine not in the 20th century where you see and cure, but -- see and treat -- now you can actually apply molecular medicine knowing the biomarkers, knowing the molecular signature for disease.

And this is an example from NIDDK with their SBIR program. The intent of this is to promote technical innovation as the theme of this workshop, to promote pre-clinical and clinical testing of single or combined components of close-loop system for diabetes management.

There is a lot of information in this, so in interest of time, I'm going to skip this slide, but Jonathan has promised that these will all be posted.

This slide is provided by my colleague from NHLBI. We already touched on artificial heart, but there is innovative ventricular-assist systems, as shown in the first, to develop totally implantable ventricular-assist device. And this could also be used for pediatric -- being tested for a pediatric population, a circulatory support contract program. These are pumps for kids, for infants, for neonates. And lastly, grants for adult and pediatric mechanical circulatory systems. And we talked about some of these earlier. So I'm going to stop there. I'm sure I've run out of time. Thank you.

Lt Col WATTENDORF: Well, Dr. Shuren and Dr. Sackner-Bernstein, thank you for bringing us together, and DARPA is certainly happy to be here. And I think it's a testament to the idea of innovation by bringing DARPA along. It's perhaps illustrative, though, that we're not quite there yet when you look around the table and you see HHS is on one side of the table and DoD and VA is on the other.


Lt Col WATTENDORF: Wonder how long it will take us for full integration, although I will give that the slides looked like we're integrated with the different departments.

DARPA was started in 1957, and it was initiated after Sputnik. And there was a interesting challenge posed to the country at that time because the research study of why Sputnik happened brought to the fore that it wasn't because of the technological capability of the United States being behind the Soviet Union, and it wasn't really behind in the resources because we certainly had the resources. But it seemed to be because of the bureaucratic difficulties of our system and the posture of our bureaucracy at that time. And out of that came the Defense Advanced Research Projects Agency, which was called DARPA.

As you heard, the Internet was part of one of the initial successes. And there have been series of successes over the years and some of have even been in the medical space, to include the application of the understanding behind the science that led to medical ultrasound.

We're still in that situation today. DARPA still operates where they're trying to understand where, in the interest of national security, do we need new, innovative capabilities as opposed to answering to what are traditionally military requirements.

And one of the areas that I pick up on when I listen to the comments today is this notion of this point-of-care and the setting that seems -- the transformation that seems to be occurring where medicine is moving outside of the tertiary-care hospital. And we particularly see it in the Department of Defense where the number of our medical facilities that are tertiary-care facilities have been decreasing, and the increasing number of small clinics, and our need to provide healthcare around the world.

Particularly, there has been a change with the initiation of what is called stability operations, of which healthcare is part of. In the past, the military traditionally flexed to be able to handle natural disasters. We were called upon, for example, in Haiti, as you've seen, and we provide our care based on "out of hide," where commanders that are obligated and funded and have contingency plans to do traditionally military operations are providing their services to whatever the exigency of the day is.

Stability operations has actually changed the strategic focus of DoD to look at how can we plan for the exigencies that we would anticipate. And some of those include the need to provide healthcare in an emergency setting, but also to rebuild an entire country's infrastructure, as we've seen today in Iraq and Afghanistan.

With that becomes a need in new capabilities to provide medical care in, again, this distributed setting where we don't have these tertiary-care facilities for ourselves. And although most of DARPA's successes have come at the interfaces between traditional, basic sciences like physics, chemistry, engineering, and mathematics, what we are seeing is that while DARPA can demonstrate the technological capability, there is a massive chasm between getting from an understanding of a prototypic device in the medical device space to getting through the regulatory network to be able to bring that to our users in the field.

The other aspect of it I find particularly challenging is the need for a commercialization strategy for a company to be able to bring something to the FDA. With a military setting, again, if we have technologies that are working that are in the diagnostic space, for example, assays for infectious diseases that are not traditionally seen in the United States, what's the incentive for the commercial space to bring a test to bear through the FDA if the military is the only buyer? And so those are challenges that we continually face and are continually addressing at DARPA.

An example of some of the activities that have occurred that it also showed to illustrate how we cross traditional boundaries or silos within FDA, at least as defined by the centers, is some of the activities that are currently occurring. One is blood farming, where we're trying to do the development of red blood cells from regenerative cells outside the body, where that can be considered a medical device, a therapeutic and a biologic. Additionally, we have been revolutionizing prosthetics, where in that case we're trying to get direct neural control of the prosthetic device. Maintaining human compact performance, there are programs where we are looking at molecular markers or biomarkers of disease while an individual is asymptomatic. And while there have been significant scientific successes in that space, the ability to develop diagnostics when you have multivariate analysis is also currently a challenge, as we all know, even now in some of the discussions with IBDMIA.

And, finally, with biologically inspired platforms and systems, there was a program called MIMIC, where we were looking at the ability to do early detection of whether a vaccine may fail by using a non-animal model, by using a human model ex vivo.

There are many other examples of that. However, again, for DARPA, our traditional focus is to try to translate basic science rapidly into a device. And for us in the medical space, we definitely see a need in the direction of the point-of-care as well as a need in the ability to cross different regulatory pathways for one specific medical device.

Thank you for our time.

DR. SACKNER-BERNSTEIN: Thank you, to each of you, not just for the presentation but for all the time that's been invested already behind the scenes getting ready for today.

And in that light, there is one agency that is not represented here today, which is the Centers for Disease Control. Eric Sampson and his team have played a very important role in the preparations. Unfortunately, because of the situation in the Gulf, he and others have been deployed. So they're not part of this workshop but still remain a very active and committed part of this Council, as evidenced by a host of e-mails I received even last night.

So I'd just like to acknowledge their role especially as we talk about some of the issues, such as how we would measure the impact of the effectiveness of any measures that we might try to institute, or initiative, undertake since they're so tied into the way public health actually becomes manifest on a national and global scale.

With that said, I thought it might be worth a brief discussion amongst us about a couple of issues that have arisen in these presentations, one of which got to the question of interoperability of equipment. And I know that's something that's very important to many of us and to many of the people in the audience.

It's a very difficult thing to get a variety of players to come together to create a platform. What have been the steps that any of you have taken to try to make sure that your equipment does end up interfacing within its own environment successfully?

DR. CROSS: You know, we purchase a lot of equipment, all types of devices, and I'm not sure exactly if this is what you're looking for, Jonathan, but the safety features, the operative features, the training that goes with those devices, we have to engage -- we have to execute those programs on a nationwide basis at hundreds and hundreds of facilities with many different individuals. And, you know, I think for us, consistency is an asset, and moving forward with how devices function, work, and as we purchase different types of brands and pieces of equipment across the country.

There's many examples I could give, but the training, where we have to train technicians on multiple different ways to handle different devices, to reprocess devices, to -- how to use the devices in various environments presents an additional challenge for us. And so consistency I think is a remarkable benefit for an organization like ours, the VA, as we purchase medical devices.

DR. SACKNER-BERNSTEIN: So one of the things that has been touched on is that a device may look like it's an innovation in the eye of the beholder, but we don't really know until it's used in a more widespread fashion. Is it a barrier that we think is important to address this idea of interoperability?

In other words, a company comes forward. They develop enough evidence to show that they're safe and effective, so they get to market, but it's an electronic that has to interface with other systems. How would a company that's developing that, or an individual that brings that to market, be able to maneuver that under the current situation to make sure that they can actually get their device into practice, if it looks like it should be, and therefore eventually see what the impact would be?

MS. SETO: So maybe I could address that by way of an example. A number of institutes at the NIH are funding the Alzheimer disease's neuroimaging clinical trials, which involve multiple sites and I think something around 60 or 70 sites across the country. As you can imagine, you can then, just because NIH launched its clinical trial and say, "Thou shall use X MR or X GE, X PET scanners" -- that was a Freudian slip.


MS. SETO: But in the course of this clinical trial -- which is, by the way, is an observational natural history study; it was not to intervene with the medications or treatment of Alzheimer patients -- a number of observations emerged, one of which is that the volume of the hippocampus is reduced as individuals migrate from mildly cognitively impaired to having Alzheimer as a diagnosis.

So that relies on the fact that one has ability to quantitatively measure the volume shrinkage, and how do you do that when you have different machines, you have a different platform? And a number of activities we use to address that is, one, to standardize a phantom that I use across all the sites and, secondly, is in the image processing component. Once you acquire the data, there is informatics tool that one could use and standardize the image processing.

And it's in that activity that we, the academicians, the government, and the industry, have come together and partnered and form what we call the Quantitative Imaging Biomarker Consortium. And it's an activity that we fund. They, the industry, believe it or not, want to do the right thing and come together and work with the academicians and standardize protocol, standardize informatic in the image processing part. That is an ongoing activity.

I want to use that as an example, that while we address innovation, which could infer that you don't -- everyone should do their own thing and standards may be stifling, but there is a way to address so that you can have the cake and eat it, too, and you can have innovation and still have standards that everyone can comply with, within a study, within a clinical trial, such that you can exchange data and cross talk, and it would be meaningful to each other.

DR. SACKNER-BERNSTEIN: Perhaps that idea of interoperability is what would help us shrink the time to understanding what the impact would be because then it makes it easier to deploy and to gather information.

Another factor that was discussed was about the human device interface. We talked about it when it came to the use of devices in the home, and from the topics submitted for presentation, I think this will be discussed more later on today. We talked about the fact that there are lots of people that have to be trained, that people are trained in one device and then we're concerned there is another device, and of course, if there is not a good interface, there could be problems.

I don't think you need to look too far to figure that out. Most of us have electronic devices with us today. Those of us who use the old 8086-based microprocessors with a CPM interface know that it's not quite like using an iPhone. And so the ability to execute your goals is greatly enhanced by having a proper interface.

What are the steps that we're taking to try to make sure that we foster development of the right kinds of interfaces between people and devices?

DR. JACQUES: Louis Jacques from CMS. We're in a somewhat different position than some of the other people here in that we're not directly funding research and setting those sorts of agendas, so our influence is a little bit indirect. But when we would, for example, review the evidence on the use of a particular technology, we do look at it specifically in the context of typical use of the Medicare beneficiary. So is this particular device as usable by someone who is 80 years old with rheumatoid arthritis and ulnar deviation of their fingers as it is for a relatively healthy 20-year-old?

I think the realities of medical care payment over the last 30 years have forced members of the public who never thought of themselves as healthcare providers to, in fact, become healthcare providers. I remember 20 years ago a patient showed up in my office with a vial of insulin and said, "Well, my doctor always gave me my insulin." And I looked at that person like they were from the moon. I'm like, "What are you talking about? People give their own insulin."

But as patients get discharged from hospitals with non-healed wounds and family members are doing wound dressing changes, family members or patients are administering low molecular weight heparins, et cetera, et cetera, et cetera, combine that with an increasing diversity in the U.S. population where not everyone grew up speaking English in their home, the need to make sure that the people who actually will use these devices are not only sort of, in theory, capable of doing it but, in fact, in their, for lack of a better word, their bravery to look under the bandage and to actually do something, as opposed to, "Well, gee, that's kind of yucky looking. We'll just wait until you get to the doctor next week," and pay no attention to it, those human factors, I think, really do need to be addressed if you want us to be enthusiastic about your technology. We may be tolerant of your technology. You would rather have us be enthusiastic.

DR. CROSS: You know, the human interface is of paramount importance to us, and, you know, it has to be for me certainly quite simple. I'm sitting here with one of these "i" devices. I won't mention which brand, but I think everybody knows, and it had to be so simple that even I could use it.

Now, we're trying a new technology at the VA in regard to help individuals, help veterans check into our clinics. And the traditional model has been that they walk up to a clinic and they find a clerk sitting at a desk, and they engage with a clerk, but that clerk has many other things to do. They may be getting some request from a nurse or a doctor. They may be taking a phone call. They may be doing some paperwork. It's really multi-tasking. And we wondered if technology could help us with that. And so we were looking at kiosks.

The interface on the kiosks, the human factors, is vitally important in order to get patient acceptance because what we're finding at times, if that is not very, very engaging, including for individuals with poor eyesight, they bypass the kiosk and go straight to that individual still sitting at the desk. I think we all had similar experiences about this at airports where initially there was some resistance in using the kiosk, and many of us probably now prefer the kiosk.

We're looking at those kind of devices and trying to understand more specifically what factors make that acceptable, usable, and even pleasant to use for our population.

MS. SETO: So, Jonathan, when you first posed this topic, I thought of the interface as an implantable device within an individual, and I hope I would hear more about that in terms of, for example, implantable closed-loop glucose monitoring, in terms of neuroprosthesis, these smart devices that one could actually intelligently control for amputees, for example. So that's at one dimension what I thought of human interface with device.

At the other dimension, at more populational level -- in fact, interestingly enough, several institutes, including mine, are co-sponsoring a workshop yesterday, and as we speak, it still continues, on personal individual medical motion devices. These are clearly -- immediately you can think of the elderly population, and I was keenly interested, as I'm joining this population, but it's also for people who are -- who have motion difficulties, such as Parkinson's patients and so on.

There is very interesting finding in terms of sensor versus video for the elderly, and they clearly -- the results show that they would prefer to have sensors because in places where they're most vulnerable, such as a bathroom and the bedroom, they don't want to be videoed. And so there are also -- we're gathering data of how do you track the footprints, how do you track motions, and MIT, in particular, I heard they have actually developed this device that you could insert in the shoes and you could sense the pressure and that, from that pressure, developing a pattern recognition of is this person about to fall, is that person lightheaded because of some reasons.

So the science and the human interface, the science clearly will also involve not just the physical sciences and life sciences, but it's really about social and behavioral sciences. To actually get these individuals to participate in a trial, they have to see value in this, what is in it for me, and that's a behavioral and social dimension.

DR. SACKNER-BERNSTEIN: Well, perhaps I could pose a question and see if we can get each person to respond because, if you'll recall, the primary question that was asked on the Federal Register Notice was about addressing unmet public health needs. And so it's a tough question. How are we going to figure out where we should target, where we'd like other people to target? Perhaps we can go around and each speak for a moment about the ways that our agency prioritizes the public health focus, whether it be based on some system that's in place that's been tested that goes on, on an ongoing basis or whether it's needs driven, given the current environment?
As one example, hopefully speaking accurately on behalf of CDC, much of the work has to do with the global and national burden of infectious disease and lifestyle behavioral issues. I'm using that term lifestyle and behavioral issues rather loosely because I'm trying to encompass a broad range of initiatives that they're focused on, including diet, exercise, tobacco use, et cetera. But they've got a priority scale system that seems largely related to disease burden, as they measure it. And that's how they focus much of their work.

So, Jean, would you like to start?

MS. SLUTSKY: Sure. I think how we all answer this question really depends on where the focus of our respective agencies is. And for AHRQ, we're really at that interface between the patient and the healthcare system, whether it be a health plan, an individual provider, or another type of payer. So we live very much in the post-marketing arena.

And, you know, for example, you know, your discussion about diagnostics for Alzheimer's disease, one of the things that's of keen interest is the fact that there are really no best treatments for Alzheimer's disease, so we can become exquisitely more accurate in diagnosing it, but we still have a huge gap on best ways to treat it.

So from the standpoint of unmet needs, we need innovation where there are areas for which no very effective treatments. And you've actually brought up many of these conditions, Parkinson's disease, but we also have system-wide needs as well, you know, what is the best way to actually implement things across a broad system of healthcare that we know are effective. And the VA has been a great example. But outside of the VA or, you know, closed health plans -- and we recognize that most care in this country is actually provided by small practices of physicians from two to five physicians. So there is no shortage of areas where we need innovation. So that's why the turnout today is so incredibly important to me and to AHRQ.

MS. SETO: So while it is generally recognized that the NIH supports research both at the basic biomedical level and at the clinical level, we certainly do our research with an eye that many of these findings would impact health, public health. And I would use two examples to illustrate where NIH has taken leadership in public health arena. One is in the obesity initiative. This is an initiative that we launched about five or six years ago now; the NIDDK along with NHLBI are certainly leader in that arena.

Now, we addressed it from both understanding the basic disease -- basic mechanism, metabolism, but also from a standpoint with an eye toward a larger set of dimensions. Obesity isn't going to be addressed just by taking a pill. We have to change lifestyle, which Jonathan alluded to. We have to address it from the public school or the schools having to introduce physical ed, from urban planning, city planning, where there is actually a sidewalk where people could actually walk to grocery stores or the drug stores, from housing where there may be playground in the -- in community that children can go and play.

So while I recognize that NIH has a role to play, we also recognize the need to partner with many of the other constituents and stakeholders in addressing this really national public health problem.

So that's one example. The other example is smoking cessation, which the National Cancer Institute has for many, many years, decades, supported research, both in terms of what are the sequelaes and the effect of smoking, lung cancers, and now under the more -- the cancer genomic atlas. We have found that there are many different malignant subtypes of lung cancer that present biomarkers that we could address. We could also develop therapeutics that address these subtypes of patients with lung cancer. So that's at the molecular level, at the disease mechanism level, at the drug development level.

But there is a huge component that is behavioral, that is community-based. And, again, Cancer Institute has done many of these, in terms of public education, the message that we should send. And I should also mention that the National Institute of Drug Abuse Research has actually taken a lead to make sure that communities and states where we hold NIH conferences allow for no smoking policy. So by way of influencing that, we hope that the community as a whole will change, in terms of anti-smoking policies.

DR. JACQUES: I think the paradigm of CMS as a public health agency is a relatively new one, and I think it was first articulated by Mark McClellan, who was the FDA commissioner before he got a better job with CMS.


DR. JACQUES: Sorry. But the thought that CMS would be something other than sort of a plain insurance company and might sort of be more proactive in this space is relatively new.

That said, our processes, by statute, depend somewhat on public input. If someone comes to us with a request for a national coverage determination, the topic could be the most boring topic that we've seen in years. Nonetheless, we have to accord it the same sort of regard as we would if it were something that we thought was truly, you know, sort of marvelous and cutting edge, in terms of devoting time to go ahead and review it.

There has also historically been a sense that the Medicare program as an insurance company was focused specifically on paying certain types of healthcare professionals, be they physicians, hospitals, or other sorts of things, to do things to patients who were clearly identified as patients.

So even the idea that the Medicare program would wade into prevention and screening, for example, has been relatively recent and has been in response specifically to congressional language authorizing us to do something that we did not believe we otherwise had the authority to do.

We do find that much of the burden on Medicare beneficiaries relates to certain chronic diseases. There are many unglamorous things that will make your life miserable. And when one reaches a certain age, often it's not just one thing. It's a host of comorbidities and a host of sort of poly-pharmaceutical effects. And you may find that trying to fix one thing frankly leaves the patient as bad off as they were before because the thing you fixed wasn't the right limiting step, so sort of like that focus when we look at priorities.

Lt Col WATTENDORF: I think have a couple comments that's across that statement of unmet public health needs, and the first one that I thought of was the work that you had done regarding radiation exposure and medical imaging and how important that is. And it's a philosophical point, but it touches some of DARPA's interests and Department of Defense, and that is where is the regulatory aspect of safety and effectiveness. Where does it begin and where does it end? And is it a sequence of binary steps, as I have learned in medical school, particularly for the therapeutics as you go through the phases of a clinical trial, or is it a continuum that in the age of knowledge now, where knowledge is ubiquitous and immediate, why do we not have a stream of knowledge from the most -- the initiation of a basic concept all the way into the post-marketing space where we can understand what our medical devices are doing after they have been deployed and the differences between effectiveness and efficacy.

You know, as Department of Defense, we've 9.47 million beneficiaries that we care for, and it's distributed around the world. And every medical device you could imagine is being used in very, very different settings, scenarios of care from the elderly down to in hot environments, in cold environments, in vibration, and many, many others that you could immediately come to think of that were probably never even considered when they were part of their clinical trial.

But how do we access that information in a world where you can get information from your cell phone at any given second? Why is there not some push in the federal sector to try to capture this information in real time so that we can get aggregated reporting of what these devices are doing from the devices so that we have intelligent devices, and those devices can then inform what ways they need to be improved, for not necessarily what their intended use was, but for what their actual use was? I think that would be extraordinarily beneficial, and I think the technology -- it’s not a technology problem. It seems like it's much more of a policy and a regulatory challenge of changing the understanding of how to get a device to market.

The other one I feel like I can't come off of and I have to keep bringing to the fore is what to do about commercialization strategies when you talk about unmet public health needs. Who brings a product to the FDA where there is no commercial market for it? And that to me is what unmet public health needs are. So is there a way in the federal sector that we can look at the areas that are most pressing -- people have mentioned global health. I have mentioned certainly the DoD's need to do healthcare around the world for many diseases and conditions of which there is no market in the United States for. There is plenty of technology to improve the conditions of those people in the world, but if there is no U.S. market, we will never see those companies and vendors bringing their products to the FDA. It, to me, is paramount importance to this topic that I'm not sure I have the answers to.

And the last one that I wanted to mention is the ubiquity of molecular knowledge, whether it's genome projects or genome type or proteomics or metagenomics or any number of biomarkers that are out there that seem -- they are the commodity now. And the way we measure those now is no longer as a single task but as multiple tasks. And how do we figure out a way to -- we're so far behind in the power curve, in other words, to be able to capture these tests and provide them to patients in a meaningful way. The problem is not the biomarkers. The problem is in the diagnostics. What are our diagnostic platforms?

And, again, for DoD, the problem that we have seen, or at least from my perspective from DARPA, is that the laboratory-developed test has seen an increasingly larger and larger reference lab, and any intended use scenario that DoD is looking at where we have a turnaround time that involves FedExing a sample to a lab and back is 24 hours at least each way in the United States. If we were somewhere else around the world, for us to get to that reference lab means flying a medical -- a lab test somewhere to where there is a reference lab that is under -- at least from CMS' perspective, these are all high complexity molecular tests.

So how do we get those tests to the patient and have them simple enough to use when they're always considered high complexity molecular tests, for example, in the nucleic acid tests and laboratory-developed tests, but what if we were multiplexing them? And we know the capabilities there on every given journal, on every given Wednesday in the science section. How do we test those and bring those to the patients to make something different?

Col BULLOCK: That's good. DoD identifies gaps primarily through both current experience reports as well as long-term strategic planning and identifies capability gaps. Some recent ones are how to handle non-compressible hemorrhages, the post-blast injury assessment and care, things like that.

I did want to go back and talk about the questions on interoperability and the human device interface. And in DoD, we tend to do that by more aggressive operational testing and get a population of candidate devices and put them --


Col BULLOCK: Yeah. Get a population of candidate devices and put them through real-life operational testing with users that are identical to the people that are going to be the real-life end users, including the training materials, including some of the -- well, some of the scenarios that they'll actually be under. But it seems like unless there is a higher level of commitment to interoperability, where it's commercially attractive because the buyers have agreed that interoperability is important to them across a broad base, then that would be the ultimate driver to make it more popular.

DR. CROSS: You know, at VA, we have to and we should, of course, take a long-term view. And so what we do is a demographic projection that we spend lots of effort on, looking out for the years to come and for even decades to come what our population is going to look like, our population of veterans, our populations of patients, and what their needs are going to be.

We then factor in lots of other information, such as geographic shifts within the United States, where people are actually going to be living as opposed to where they live right now. Other factors, cohorts, exposures, environmental exposures, are also factored in as we get that information.

We then do gap analyses, looking at gaps in services, gaps in how we provide care, and part of that relates to the exact disease and the technology that's going to be used.

So when we're planning, using this information, we use it to plan for hospitals. We use it to plan for clinics, how many transplant programs we're going to need, how many providers of various types we're going to need, and so forth.

We do chronic disease management. We're trying to minimize the impact of chronic disease, long-term, on a population of individuals. A part of that, I can give you some specific examples. Glucose testing is well established, used at home by so many people, including in the VA, every day. The important thing for us, perhaps, is to make sure that that gets fed back into our electronic health record and that the providers who are caring for that patient are able to graph that out and understand how that patient is progressing.

Telemedicine. We have tens of thousands of patients right now that are getting some form of telemedicine intervention. We want to expand that by making that technology more effective for a wide variety of diseases.

Dialysis is a real challenge for all of us, not just in the VA, but everywhere. How do we do that in a way that's less facility-dependent, using mobile assets, home assets, home monitoring, and picking up the technology and pulling -- back into the electronic records.

Coumadin monitoring. We have so many patients on Coumadin. The prothrombin testing that could be done at home or in other environments could be of great value in that situation. Sleep apnea assessment and screening and monitoring is a real challenge for us and I think for medicine at large in the United States.

And then lastly, how do we take all of this information and give the patient real information that they're interested in and give them control, give them access, and a greater degree of control. And what I mean by that is giving our patients access to their medical records more and more so having them as a partner engaged in recording information about them, having them understand more how that's going to be used and understanding what information is there.

An example of how we're doing that is with one of our programs called My Healthy Vet, which is a personal website that a veteran can have to start developing their own electronic record. And if they then engage with a doctor in the community outside the VA for part of their care, enabling them to take that information and share it in a way that will then provide them better, less fragmented care in the future. And I think that's a point for satisfaction. I think patients want to be engaged, and I think that technology is going to be very valuable in the future.

DR. SHUREN: I think for FDA, we traditionally have been much more reactive when it comes to technology. We do engage in a horizon-scanning exercise every few years to see what's going to be coming on the market and then to try to prepare accordingly. Now, that's in response to what others are doing.

Secondly, we do engage in actually fairly extensive scientific research on the part of our labs, often in collaboration with others, on development of new assessment tools. Are there ways that we can be more precise in our understanding of the risk/benefit profile of a technology and to do so in a less costly way?

Examples include development of disease models, scientific computing. We just recently purchased a supercomputer for FDA that we'll be using for modeling purposes and computational analysis.

But third, more recently, we're looking at ways in which we can be more proactive, starting with a recognition about what is unique for the agency. And one thing that we do have is a certain expertise regarding medical device design. We see all moderate and high-risk devices before they come on the market in the U.S. We see proprietary data that's not available to others. A manufacturer has its data, but we see the data of its competitors. We see the data of similar kinds of devices. And we look at all these technologies, whereas some of our even regulatory counterparts don't because they rely on third parties. And a third party sees some devices, but not all, and that information isn't necessarily shared with the regulators themselves. And then we continue to monitor those devices when they're on the market.

What we have made a commitment to do is to share that expertise with manufacturers, with the healthcare community, with others, to not only identify problems but to try to solve them collaboratively. And I think the examples of reducing unnecessary radiation exposure when it comes to medical imaging technologies, from radiation therapy technologies, the work we're doing in external infusion pumps, and our efforts in trying to facilitate development of devices for use in the home all represent examples of our using our expertise to try to solve problems.

Fourth, I'd raise that there are some future directions for us to think about. Dan hit upon a very important one, and that's the development of smart devices, devices that can monitor themselves, that talk to other devices and can report back. And that truly is the wave of the future. You can kind of think about it. It's almost like a flight recorder is in the device. At one time, it's constantly reporting back, and then if things break down, we've got a record to try to understand what that problem is.

But it's also about smart healthcare systems because we are talking about networks within networks that are now collecting vast amounts of information, and FDA is going to try to leverage that by developing a unique device identification system so that we can actually connect the device with the individual experiences of patients. Right now, it is very hard to do that, and yet we're collecting all this information in an electronic environment.

Another challenge is the role of users. Very much we look at what manufacturers can do in development, but we need to be responsive to the needs of the healthcare practitioners and of the lay users. They have to have a greater role in driving the design of devices and what's important, what their needs are. So we truly go from not a bench to bedside model, but a bedside to bench to bedside model.

And then third and lastly is new models. Really think about new models that are based on our collective action rather than the individual actions of specific agencies. And one example is what I mentioned this morning, what we're thinking about along the lines of parallel review.

But I also think things for, let's say, FDA and CMS, would we consider reliance on coverage with evidence development as kind of a pathway, where we see that working together, in terms of what FDA's needs are? Could we rely a little bit more on a post-market data collection model?

Another is for thinking about, in works with the VA or Department of Defense, taking advantage of their healthcare system; would we think about test-driving certain technologies within a small subset of hospitals, and if they work there, then saying, okay, we'll purchase this for the entire system and we'll let that be used more widely on the market.


DR. SHUREN: Jonathan is like this around the office, too. He also cleans windows, and he is a great cook.


DR. SACKNER-BERNSTEIN: Thank you for that recognition.

So what I'd like to do as we transition into the public presentations to the Council is remind people that we're particularly interested in suggestions about how we can prioritize the unmet public health needs which should become our focus. This is a very difficult process, as you can tell. There are different approaches. And I don't think anyone here believes we have the best answer quite yet, so we're hoping to hear some good suggestions.

We did hear from many people in preparation for this Council meeting, for this workshop, about some specific areas that they wanted to talk about. And what we're really hoping is that when people do talk about specific areas, one particular treatment and one particular clinical scenario, it's done within the context of how that should be treated differently or approached different than other areas, or perhaps how the work in that area can be used as a model for how to approach different areas.

Remember that we all acknowledge there are many barriers, there are many steps in this process. It's complex. It involves many agencies, many scientific disciplines. Please also note the broad composition of this Council, and we would hope that the comments, even though those directed maybe just at CDRH would be welcome by Dr. Shuren and me, we really would rather have this a broad-based discussion so that we can move forward with this new model of a collaborative approach between multiple agencies.

The next couple slides have the questions that appeared in the Federal Register Notice, so you should be familiar with them, identifying areas of public health need and addressing barriers to development and/or redesign of medical devices.

And so with that, I'd like to turn to the first presentation. The presenters are listed with their affiliations and their titles on the agenda. I'll ask that when you come up, we're going to try to stay on the schedule as best as we can. There will be opportunity for question and answer, and I hope that the speakers don't get frustrated by the fact that you can't see the slides actually on the screen. You have to just -- I mean on the computer. You have to look at the screen. But sorry.

MS. SEIDENFELD: Good morning. My name is Justine, and I'm going to talk a bit about engaging stakeholders, particularly patients, providers, and payers, in priority setting efforts, especially as the Council seeks to identify important unmet public health needs.

So I work for an organization that's called the Center for Medical Technology Policy. We're a small non-profit in Baltimore, and most of our work involves engaging stakeholders in a variety of activities to facilitate new comparative effectiveness research. And these activities can range from priority setting efforts to the design of new clinical trials, to the development of implementation strategies.

And what I want to share with you a little bit is about why we think it's so important to engage stakeholders, particularly the patients, providers, and payers, as I mentioned earlier. What you're looking at is a figure describing some of the clinical practice guidelines that have been put out by the American College of Cardiology.

They went through and looked at all of their recommendations and rated whether they were based on A-level, B-level, or C-level evidence, A-level being strong evidence, usually at least one large randomized controlled trial, B-level evidence being a bit more conflicting, usually smaller RCTs or non-experimental studies, and C-level evidence is basically very limited, usually anecdotal or based on expert opinion. And what you can see here is that, you know, across a variety of cardiovascular conditions, a lot of the recommendations are still being based on C-level evidence, are being based on expert opinion.

So the question is why with, you know, over 18,000 RCTs being published in the literature every year and countless other non-experimental studies, why is it that so many of our practice recommendations are still being made on C-level evidence, on expert opinion? And you know this is true when you look at a lot of tech assessments or systematic reviews, they often tend to conclude something like, you know, the evidence is limited and we can't make any conclusions here.

So our hypothesis at my organization is that part of this stems from the fact that decision makers, and again, particularly patients who need to choose among interventions, providers who need to make recommendations about interventions, and payers who need to make decisions about reimbursement and coverage, have had very limited opportunities to influence the clinical research enterprise. And we think that's why there are so few A-level and B-level studies out there being incorporated into clinical practice guidelines.

And while what my organization does is work a lot in these sort of post-regulatory comparative effectiveness world, we think this hypothesis applies equally to the pre-regulatory world as well.

But, you know, this is our hypothesis, and no one is really going to believe us, and we're certainly not going to get funding without a molecular explanation for it, so I'm going to present the molecular basis of uncertainty.


MS. SEIDENFELD: So imagine, if you will, that you have a cell, and inside the cell are the decision-makers, again, patients, providers, policymakers, and, you know, as you can see, they have a very low affinity of receptors for evidence. And outside the cell, you have the world of scientific evidence and clinical research. And, you know, starting at the top of the pathway, you have intellectual curiosity driving the, you know, PI-driven model of the clinical research enterprise.

They publish lots of evidence in the literature, and it can diffuse over to the decision-makers in a couple of ways, one of which we're calling KT-1, or knowledge transfer one -- we're just trying to make it sound really scientific -- is a slower passive diffusion mechanism where these decision-makers go directly to the literature. And one of the other mechanisms that we're calling KT-2 is an active transport mechanism through health technology assessments or systematic reviews or meta-analyses of the data. And so, you know, the evidence diffuses into the cell, reaches the decision-makers, and they produce evidence gaps. They identify areas where they feel that the evidence is limited or of poor quality, and they can't answer the questions they need to know more information about.

And so what we see is the problem is even when they produce these gaps in evidence, there are very few ways for them to communicate them back to the clinical research enterprise. There is a defective transport mechanism, and our organization is basically trying to remedy that defective transport mechanism. So that's sort of the context and why we think it's so important to engage those end users of evidence, the patients, providers and payers, in all aspects of the clinical research enterprise.

So, again, that's what I'm here to talk about, priority setting. I'm going to run through the first three of these topics really quickly and talk a bit more in depth about engaging stakeholders and, again, particularly the end users.

So for methods of priority setting, there is a wide range of options. You can do everything from sort of a more rigorous population and economic-based analyses to public consensus forums. And at CMTP, what we like to use is sort of working with groups of expert stakeholders and using either sort of rating or voting methods or in-person discussions to elicit opinions about what are important topics that we need to concentrate on in the future. So we recommend methods that are structured to engage multiple stakeholders, and again, particularly these patients, providers, and payers because it's not all that frequent or all that easy.

In talking about scoping priority setting efforts, it's a little like Goldilocks and the three bears, not too big, not too small, just right. Not too big -- it's going to be hard to really elicit meaningful input from these end users of evidence if you're asking them to make comparisons among the entire world of unmet public health needs. And so what we have found to be really useful is at least to start scoping a little bit, we've done projects where we prioritize among cardiology interventions, among oncology interventions, and some that are more specific, including one I'll talk about, prioritizing among research questions for uterine fibroid disease. But again, not too small -- you can look at a range of intervention types from diagnostic and screening and treatment tools.

Priority setting criteria, you know, we've already talked -- the Council has already talked a little bit about it, but what I just want to say is it really depends on the goal of your project. We've looked at priority setting criteria for a number of sort of health policy or healthcare organizations, you know. There is a bunch that pop up pretty frequently. And, you know, they're ones that we heard about earlier in the discussion. But what we really decided is that not all of these criteria are going to apply to every single project that we have, so we tried to stratify them based on the goals of the project. And that's what I would recommend in terms of criteria. Think really carefully about the goals.

So this is sort of the meat of the presentation, engaging stakeholders, and I just also want to add that, you know, it's not just our organization kind of standing on the sidelines talking about this. I think a lot of other healthcare and health research entities are starting to realize the value of engaging patients and payers and providers as well.

And you know, one example that's been particularly heartwarming to us has been to see that when the IOM put together a committee to identify CER topics and recommendations for how to conduct CER, they did make a very specific recommendation saying that you need to involve consumers, patients, clinicians, payers, all of these end users in all aspects of the clinical -- or of the comparative effectiveness research enterprise, including priority setting.

So now I'm just going to talk about a couple of examples that our organization has done. This is sort of a model of the internal priority setting process that we use for our projects. But what I just want to say is we do a process of iterative rounds of voting or ranking followed by an in-person meeting among expert stakeholders.

And so as you can see here, for a process we did last year, prioritizing among cardiology topics, our stakeholder workgroup was very small, and it was composed exclusively of these end users, of clinicians and patients and payers. And we were really surprised at the richness of the discussion and the expertise of these people that we involved. And you know, just for example, sort of the patients really emphasized the research topics that had significant issues involving quality of life, safety, and adverse effects. The payers were very good about pointing out to us certain technology topics that are going to be very difficult for coverage and reimbursement decisions. And we think these are things that, you know, sort of the PI-driven model of picking clinical research topics they cannot identify by themselves. They need this input from people who are in the trenches, so to speak.

We also do priority setting efforts for external organizations. One of these, we were a subcontractor to Outcome Sciences, which is one of our DEcIDE centers, and as I mentioned earlier, we were helping them prioritize among uterine fibroid disease research questions. And as you can see, we had -- we put together a stakeholder committee with a broad representation from a variety of stakeholder groups.

And you know what I just want to emphasize here is that we feel that from our experiences, having the involvement of these stakeholders, particularly the patients, the payers, and the providers, really does change the outcome. It really does change the research topics that you are going to identify.

And so for this example, we came out with a much greater emphasis on the research questions that emphasized or had to do with how do you share decision-making techniques, and that wasn't something that we were expecting going in, so we felt it was really valuable to have the presence of some of these end users at the meeting.

One of the other projects that we've done like this is in collaboration with the Southwest Oncology Group, the University of Washington, and the Fred Hutchinson Cancer Research Center. We're helping them put together a center that's basically going to deal with genomic tests for cancer, everything from priority setting to the design of clinical trials, to the implementation and practice. And one of the first projects that we did with them is put together a stakeholder group to do priority setting among cancer genomic tests.

And again, you know, as you can see, it's a wide variety of participants. And again, they did change the outcome of the priority setting results. And, you know, one specific example is one of these genomic tests that's currently being used to monitor breast cancer in patients who currently have breast cancer is spreading into practice as a tool being used to test for predictability of recurrence. And there's a lot of uncertainties surrounding this use. And a lot of patients really emphasized that they felt that the uncertain environment surrounding, you know, monitoring for recurrence, they felt that was a very large unmet need, and if this test had any potential to fill it, they wanted more evidence on that.

So again, with all of these examples, my point is mostly that having patients and clinicians and providers and payers at the table really does make a difference.

One of the most complex groups to engage, I think, or at least what we have found from our experience are patients and consumers. So we at CMTP put together a patient and consumer advisory committee, people to review our processes for various activities, including priority setting, and give some recommendations about how to improve them.

And just some of the, I think most notable recommendations are, one, engage patients as early on in the process as you possibly can. They really specifically recommended going to patient advocacy groups within whatever area you're prioritizing among. They wanted us to have criteria that reflect values of the U.S. public, so we started emphasizing criteria about safety and risk a lot more heavily -- and also include at least two patient representatives in any group. It's basically, if you're going to do this, do it right. Don't have a token patient representative. So you need to make sure that you're doing a good job of engaging them.

And I don't want to end on a down note, but I just want to sort of offer some caveats and say that, you know, and stakeholder engagement is a complex process. First of all, there is this inherent tension with a PI-driven model, and you know, there aren't very many documented models to copy, and it can be very difficult to make sure that everybody is speaking the same language. We have had to really start doing a lot of work before these in-person meetings to make sure everybody is on the same page, and it does take a lot of time and resources.

And, you know, we don't know yet for sure whether this is going to fix the problem, if there is going to be more A and B-level evidence instead of all that C-level evidence now. But from our experiences, you know, we've had some really rich discussions. We've seen that they do change the prioritized agenda, and we are very optimistic, especially with the new focus on stakeholder engagement and a lot of research that's coming out of both AHRQ and the NIH.

So I would be happy to take any questions that anyone has. Thank you.

MS. SLUTSKY: Thanks very much. Justine, I just had a couple of -- well, a comment and then a question. You know, as you probably know, we've been working in this field a long time, and one of the things that I noticed on the slide, if you could put it up, where your PCAC I think -- yes -- there is an inherent tension in your first bullet between going to professional advocacy groups, so to speak, versus inviting a much broader perspective of individuals who may not, you know, actually be aligned with, you know, Washington-based advocacy groups. So I'd like you to comment on that.

And who makes up the PCAC? What's the makeup of that committee? I'd be curious about that.

MS. SEIDENFELD: Yeah. So the PCAC is made up of six members. We have two who directly participate in patient and consumer-engagement activities from the FDA patient representative program actually. And then we have four members who work for various patient and consumer advocacy organizations, including the American Heart Association, the Childbirth Connection, and the National Partnership for Women and Families. These are people who -- they don't come to us to advocate -- while we do have people who come from advocacy groups that have a specific disease focus, they're not there to advocate for their, you know, clinical area. They're there to just help us improve our processes. And we've tried to have a mix of people who do work for these organizations and people who do directly do advocacy work on various committees or act as, I guess, the patient representative on certain committees.

For talking about the kinds of stakeholders that we try to have participate and the first bullet point, as you were mentioning, it's definitely true. There is a variety of people who can fill sort of the public representative niche, and these include patients, both current and prospective, patient advocates who work for these organizations, you know, larger consumer representatives, these sorts of things. And they do all offer different perspectives. I would definitely acknowledge that.

But, for example, for the kinds of work that we've done, we've found, for priority setting efforts, it's especially useful to engage people who do work with patient advocacy groups because, so far, what they've done is go to their constituency or go on to their colleagues and, you know, sort of circulated the possible conditions and asked around. So it's nice to know that you're getting more than just one perspective when you go to that group.

For things like designing clinical trials, we've found it more useful to engage the spectrum of possible public representatives, including, you know, current and prospective patients, because it's very important to ask them, "Would you actually enroll in this clinical trial? Like, is it appealing enough to you or is it acceptable enough to you to do so?"

MS. SLUTSKY: Yeah. Just a comment. The first and second bullets do seem to be at a little bit of conflict when you don't really go out more broadly, but one of the things that we've tried to do is engage patients across the United States just to get that flavor of what the U.S. public really thinks -- and just to comment.

MS. SEIDENFELD: Right. Yeah. So that was a suggestion that was given to us by the PCAC, and we've put together some initial proposals for doing some kind of survey, larger survey instrument across the U.S., but unfortunately, we haven't had the resources at the moment to do that.

So when I was talking about emphasizing criteria like safety and risk, very specifically, you know, we had an in-person with the advisory committee, put up a slide with all of our criteria, and they sat there and talked about it and at the end concluded that we really didn't have anything that emphasized, you know, questions about safety and risk. And that's a very important research topic, or research topics that have a lot of questions about safety and risk are very important to patients, and we weren't doing a good enough job of that. So we've tried to remedy it.

DR. SACKNER-BERNSTEIN: I notice in the model that you have you talk about some of the receptor-mediated issues that are faced by both decision-makers and the clinical investigators. And yet part of this process is dependent on the use of experts and expert opinion to set the priorities. That's of course very important because expert opinion is valuable, but I wonder if you could give us a sense of whether -- how you go about selecting those experts. Are they the intracellular type, extracellular type, and what kind of receptor modification do they require?

MS. SEIDENFELD: Indeed. I would say most of the people who we -- all of our stakeholder committees tried to have both people inside and outside of the cell, essentially. You're not really going to get an adequate dialogue and you're not going to come out with any kind of meaningful and implementable conclusion if you just have people from inside the cell. You know, a lot of the times, you'll sort of -- even inside the cell, you'll come out with a research design or a list of priorities that, you know, are very ideal but aren't necessarily tempered with sort of the political and operational practicalities that, you know, a research enterprise is facing.

So for our priority study committees, we do tend to have just inside the cell because we're just asking them to identify topics rather than to design actual studies. But for the kinds of projects where we do where we try and come up with recommendations for clinical trial design, what kind of patient populations you should pick, what kind of outcomes, what kind of comparators, that sort of thing, we try and have both participants from inside and outside the cell.

But that's for our internal process. And then, you know, as I showed, for the AHRQ uterine fibroids and the CancerGen projects, we did have a mix of both internal and external participants.

Lt Col WATTENDORF: So I had a couple questions. One, I didn't see where reimbursability is on there, and since we have CMS here, what role they would play in that model. And, particularly, I noticed your construct is that you're looking at disease groups as opposed to the public. And I'm always wondering, in many of our federal settings, when we discuss health and diseases, that the focus always seems to be on disease and often not on health.

And what about our devices and our clinical practice guidelines and our evidence to determine who is healthy and how do we keep them healthy? How do we have them reimbursed for prevention, and how do we identify in this model -- what are the models and clinical practice guidelines to maintain health rather than avoid a specific disease?

MS. SEIDENFELD: So just to make sure I have this clear, you're asking how can this sort of view of evidence generation, but also how to, you know, complete the feedback loop, how can it apply to broader public health and chronic prevention issues as opposed to just specific disease categories?

Lt Col WATTENDORF: Well, let me make it maybe even more simple. Is this focused on disease constituency groups or is it focused on health because when I think about evidence and clinical practice guidelines, as a family physician, my primary goal is to maintain people's health. But it is often the clinical practice guidelines that I read, that I've experienced in my practice are just to risk stratify people to sort and treatment algorithms once they have the disease. And I wonder where the evidence generation is to maintain health, and how do I reimburse that? If I'm looking at the public groups, which you are, are they interested in reimbursement by the health community for maintaining health?

MS. SEIDENFELD: Okay. Hopefully, I can give you an adequate answer. I think patients or public representatives are aware of, you know, reimbursement issues. It is important to them. They do understand that that does influence whether they can or cannot gain access to a particular intervention. So I think during these dialogues -- I mean, that's why it is important to have everyone present at the table and, you know, from the payers that participate. And, you know, we do have CMS at the table sometimes, you know, to talk about sort of reimbursement issues. It does influence, you know, the ultimate outcomes.

As for how to incorporate sort of more questions of maintaining health rather than particular technologies, I think it's difficult, it's hard to -- again, it's hard to compare among such, you know, sort of broad public health questions. And that's, again, sort of why I think scoping is actually one of the more difficult aspects of any priority setting process.

DR. JACQUES: I guess to respond to the cellular model. Payment is ATP. We are ubiquitous. We are the energy that makes all of this work. Otherwise, all of it stops, okay?


DR. JACQUES: Now, that said, I do want to sort of follow up on Jean's comments a little bit because I think one of the challenges in engaging stakeholders, especially those in the lay public, is how do you do that in a methodologically rigorous way? It seems a little odd to say, well, we're going to be very evidence-based over on this side, and then we're just going to ask people kind of what they like, okay, in a very unsystematic manner. And it is sort of, like, well, if we're just going to ask people what we like, why bother to pretend we're talking about evidence generation and things along those lines.

And there have been challenges when one tries to engage the public because individual members of the public may have a focus that is almost entirely inward looking. And while they may be sympathetic that their neighbor has some other disease or someone else has some other condition, the issue is their kid, their mother, their father, or they themselves have something, you know, specific.

You know, how do you engage them in a way that they don't get run over by people who clearly have more training in this line of work? How do you make them somewhat immune to the pervasive influence either of, gee, all these people with a whole lot of letters after their name who, gee, seem to be smarter than I am -- at least they have college degrees to say so -- or the pervasive influence that may occur because some patient advocacy foundations really derive much of their funding from some of the extracellular folks over here? And that is a challenge. And I do know that AHRQ has been particularly, you know, interested in that focus. And we certainly would support that.

MS. SETO: So if I may, and I think what you're asking is how do we reimburse for healthy lifestyle, wellness, the prevention model of medicine rather than seeing and treating. And so in this context, I include payers broader than CMS. And clearly, the employers, many employers, the large corporations, have now put them in and have tried to incentivize their employees to adopt healthy lifestyles.

In terms of evidence, and NIDDK can speak better to that, in a Type II diabetes clinical trial, they showed that the 30 minutes of exercise and appropriate diet was better than medication with Metformin, and I think that clearly shows evidence that CMS would consider.

But I recognize the point that you're making as an important one, that how do you pay and reimburse for healthy lifestyle and wellness and prevent disease rather than reimburse for treating diseases?

DR. SACKNER-BERNSTEIN: Thank you very much, Ms. Seidenfeld, and also for the discussion. We're going to have our next presentation now. Dr. Klonoff?

DR. KLONOFF: Thank you very much. It's a pleasure to be here speaking to the FDA and the interagency committee.

I'm a Clinical Professor of Medicine at UC San Francisco. I'm an endocrinologist. I'm going to be speaking about standards and interoperability for better diabetes technology.

There are three issues that we're looking at here today at this FDA meeting. First is unmet public health needs, which could benefit from new devices. Second is barriers to development and use of new medical devices. Third, actions the federal government can take to remove barriers.

We're here to talk about healthcare, but I'm going to show one slide that's not healthcare, because if we can get these products on the market, it's actually good for the economy of this country. What this slide shows is a variety of medical technologies and where the United States is in terms of leadership. The blue is where the United States has clear leadership. The red is where we have leadership now but it is expected to lose it in the next five years. The yellow is where we've already lost leadership. This is from a Battelle survey two weeks ago.

We're losing our leadership. It's slipping. One of the ways to get back the leadership is to get more products commercialized through the FDA. We need it for health. We need it for the economy.

The rest of my talk will be devoted to healthcare, specifically diabetes. Diabetes is the fastest growing disease in the world, epidemic proportions in the U.S. There are 24 million Americans with diabetes, 57 million with pre-diabetes. Diabetes affects a person every day, all day long, every time a person eats, exercises, gets dressed. All day, the blood sugar is going up and down. People with diabetes need technology to help them manage the disease.

The first topic here is public health needs for diabetes devices. I'm going to list four types of technologies which are important for diabetes. First is glucose sensors. We need more accurate and less invasive sensors, whether it's blood sticks, whether it's continuous glucose, non-invasive, whether it's issues related even to cleanliness of the blood glucose testing equipment.

Second, we need better insulin delivery systems, more accurate, smaller, possibly intraperitoneal delivery.

Third, telemedicine and decision-support software. The word telemedicine comes up frequently. I'm going to try to define it here. It's one thing to measure glucose and send it wirelessly or measure blood pressure or weight or other features. That's telemonitoring. There is nothing that happens except it gets stored. What happens if you have a nurse that looks at that information and the nurse does her best to interpret it and send a message back to the patient? That's case management. That's good. But what this nurse needs is some decision-support software with information about what to do in each situation, including guidelines from national organizations. When you include telemonitoring and case management and decision-support software, now you truly have telemedicine.

Finally, the fourth technology that we're very interested in, in diabetes is closed-loop control, also known as the artificial pancreas. And these types of technologies certainly are useful for people with diabetes, but these are also -- every one of these is a model for certain other types of problems that affect other diseases. And since there are so many people with diabetes, if these products can become better commercialized and developed to perform better, there will definitely be spin-off into other diseases and into the military and into the space agency.

What these technologies have in common is that they all transmit data. And we've heard this theme from Dr. Shuren earlier. We're dealing with transmitting data so that we can get this information to electronic health record and so that we can act upon it with telemedicine. So the new technologies have data transmission.

The second topic is the barriers to development and use of new diabetes devices. So these are the devices, and each one has a major barrier, but they're multiple barriers. For example, glucose sensors, we need more funding, especially for breakthrough technologies like non-invasive monitoring.

For insulin delivery systems that are smaller, more accurate, we need better reimbursement. Traditionally, an insulin pump is like a box with a syringe that contains insulin. A little motor pushes the syringe and it delivers insulin through a catheter. The new generation of insulin pumps are patch pumps. There is no syringe. It's just like a little bag, and the patch pump is so small, it sits on the skin. There is no catheter that shows. This is the wave of the future. People want to use these, but they don't always get proper reimbursement. So I'm talking to CMS and other payers here in audience.

Telemedicine and decision-support software. On the one hand, this sounds like a no-brainer how good it could be, but there are licensing requirements and malpractice issues. For example, if a patient is in one state and the physician is in another state, are they licensed to treat that patient? The military certainly has to deal with this because they have Army doctors in one part of the country and patients in another part.

Also, malpractice coverage. The traditional practice of medicine requires that a doctor and patient be in the same room talking to each other, and it's not clear at this point, in many states, that telemedicine is safe for malpractice coverage and medical licensing. The State of Georgia is taking a lead on that. The state insurance commissioner there has worked with the insurance companies, the malpractice companies, and has created a very nice telemedicine system there in Georgia.

Finally, closed-loop control. We've got regulatory issues for sure. These studies are very complicated. One of the issues that comes up, and I'm here to put, I guess I'd say, put in the word from the artificial pancreas researchers is that when you're doing a closed-loop study, you have to modify the insulin delivery sort of "on the fly." You have an algorithm. You have a plan of how much insulin to give in every situation, but it's like having a roadmap to get somewhere. Sometimes you get off course and you have to get back on course, and there is a tension between the researchers and the FDA as to how specific the protocol has to be for the researchers to be able to deliver insulin in a certain way. The researchers are saying that they're being controlled too tightly and arbitrarily. The FDA says they need to know for the safety of the patient. So this is an area of controversy.

Now, another area is compatibility. An artificial pancreas contains many different types of devices. They have to be compatible. So I'll say more about compatibility in a moment.

But now the third question is federal government actions to reduce barriers to diabetes devices. I am listing three actions which I think the federal government should be working on which will help eliminate barriers.

First, to promote research and initiatives in regulatory science by FDA, NIH, other government agencies, academia, and industry.

Second, we need increased funding for the FDA because of the increasing complexity of medical science. For years, FDA was grossly underfunded. We can't do all these great tasks we're talking about unless there's funding.

Finally, we need to promote development of standards and interoperability to enhance compatibility and user uptake of medical devices. So I'm going to comment on each of these three types of initiatives from the federal government.

First, to promote regulatory science. Here is a press release from February 24th of this year. It says, "NIH and FDA announce a collaborative initiative to fast-track innovations to the public by promoting translational science and regulatory science." This is a really good move. And now we're hearing about this new Memorandum of Understanding between FDA and CMS, another really good move. When the agencies work together, things happen, and we need better regulatory science.

Second, increase funding for FDA. This is a press release from the Biotechnology Industry Organization, and in a blue square, I've circled -- it says, "In a recent statement about the Prescription Drug User Fee Act reauthorization, the organization said user fees which currently help support FDA were never intended to supplant a sound base of FDA appropriations." And they're concerned that appropriations for the drug review program has not kept pace with FDA's needs and workload. And it's true. And many other say the FDA needs greater funding.

Finally, to promote standards and interoperability. Now, clearly, the FDA is aware of this. Here is an example. This is a workshop that was put on a few months ago in Silver Spring called "The FDA Workshop on Medical Device Interoperability." So we know the FDA gets it on interoperability. They also have an excellent standards department run by Carol Herman.

So standards and interoperability are important, but what are standards and interoperability? Okay. Standards means that similar devices adhere to the same specifications. That means that similar devices will have similar performance, and this facilitates substitutions. Interoperability means a device functions with other products. In that case, products from different manufacturers can be combined into systems. And when you combine standards and interoperability, you have compatibility. Compatibility is a property of devices which can be substituted or combined.

Now, better compatibility is needed for all diabetes devices. I'm going to give some examples. I'm president of a non-profit organization, Diabetes Technology Society, showing how we've worked in some ways, but there's still a lot to be done.

First, continuous glucose monitoring. We worked with CLSI, Clinical Laboratory Standards Institute, to develop performance standards for continuous glucose monitors, and these were adopted by FDA this year.

We're also working with the Endocrine Society on clinical guidelines for continuous glucose monitors. The reason you need performance standards is because companies who are developing new products want to know how accurate does the device have to be before I go to the FDA, what types of data do I need to collect, what types of studies, what type of mix of glucose levels. And if a company doesn't know what type of standards they have to meet, they can waste a lot of time overdoing some studies, underdoing other studies. So it gets devices on the market more effectively if there are standards.

Then, once the products are approved, we want to get them into the hands of the patients. There, we need clinical guidelines. Unfortunately, money to pay for healthcare is not unlimited, so the best thing is to have the people who really need it the most receive the technology, and that's what the guidelines can be used for.

This is continuous glucose monitoring. We heard about infusion pumps. Well, at Diabetes Technology Society, we've been working with the FDA for a couple years now. The upper left corner, we put on an insulin pump safety meeting with the FDA one and a half years ago in Bethesda, and we started talking about issues related to insulin pump safety: software, hardware, human factors, specific issues about insulin such as bolus dosing.

Then we had a second meeting with the FDA last year, in San Mateo, California, the lower left corner, the second insulin pump safety meeting. We went into more detail. Again, we had experts in both meetings who were physicians and nurses and patients.

And then that led, in the lower right corner, to an article called "A Hazard Analysis for Generic Insulin Infusion Pump," by Paul Jones and his team at FDA. And this is a way of expressing all the different types of software that can go into an insulin pump. So if the company has a new pump, they can use that software and their software can be tested. So this is an advance in having standards for insulin pump safety.

We also need standards in blood glucose monitoring and in other devices. And I'll mention an area in blood glucose monitoring that possibly some people in this room have not thought of. When we think of standards for blood glucose monitoring, there are typically three areas that we think of as being important. One is the accuracy. Another is interfering substances. And a third is their performance in the hospital. And, in fact, those three topics were the core of a meeting that took place in this room three months ago that the FDA put on about blood glucose monitoring.

I'm going to introduce a fourth topic that I think the FDA must keep in mind with blood glucose monitoring, and that is to make sure that the devices are properly disinfected. I helped chair a meeting at CDC last month in Atlanta about blood-borne disease viral transmission and blood glucose monitoring. It turns out that hospitals sometimes, nursing homes sometimes, long-term facilities often do not have proper cleanliness of the blood glucose monitoring devices. It's not that they share lancets. That would be really obvious. But they share spring-loading lancing devices, which get blood on them from the lancet. Sometimes they share caps. They also don't disinfect the blood glucose monitors.

A study from UC Davis went to some hospitals. They tested the monitors for blood on the surface like they do on CSI Miami, and they found that 30% of them had blood on the monitor. And in hospitals, people share monitors. So it's very easy for a healthcare worker to transmit blood from a monitor to a patient and back and forth. The blood gets spread all over the place. And CDC is keeping track of these outbreaks of blood-borne viral diseases from blood glucose monitoring.

Now, the last technology that's really important that requires compatibility is an artificial pancreas. What is an artificial pancreas? It's a system which we think of traditionally as consisting of three parts: a continuous glucose monitor so we know what the blood sugar is; an insulin pump to push insulin in; and a type of software, and they're all linked together and they have a radio. So when you put those three pieces together, you have an artificial pancreas.

Well, I've seen a million slides at the diabetes meeting that show those three parts. I'm here to tell you that's obsolete. Why? Okay. You have a continuous glucose monitor. That's not the only kind of monitoring we're going to see pretty soon. We're going to see additional monitors of glucose. For example, we're going to see physiologic monitors of hypoglycemia. Those are going to be added. I'm explaining this to show how complex it's going to get.

We see just an insulin pump? No. We're going to see a glucagon pump also because if you have one system to lower the blood sugar, to protect the patient, you need another one to raise the blood sugar.

Okay. So you have two kinds of pumps. You have at least one type of software? No. You have two types because pretty soon -- we're seeing a trend that's about to begin where there'll be remote control of an artificial pancreas. Currently, we talk about local control. The patient can press buttons or the software is right there. But there is also a move toward keeping track of what's going on with the artificial pancreas so that if the patient is getting into trouble remotely, some information can be transmitted.

Two years ago, the FDA put a meeting on artificial pancreas at the National Library of Medicine. And I introduced the concept then about how there has to be remote control to monitor the safety of an artificial pancreas. So I talked about when you have an expensive car, you use OnStar. And OnStar monitors your performance, and if the car crashes, they'll send an ambulance to pick you up or a tow truck, whatever you need. Well, I used an acronym, which I call KlonStar -- of course that's my name -- and KlonStar will do the same thing. It will remotely monitor blood glucose levels. It'll have GPS. And if the glucose level falls, it will send an ambulance or whatever you need. And last year, a team of engineers from UC Santa Barbara actually wrote up a prototype device and talked about KlonStar. We're going to be seeing this.

So the point here is that you've got multiple inputs into this artificial pancreas, multiple glucose centers, multiple drugs being delivered, multiple ways of control. That's getting complicated for interoperability. But even that's not all because you've got these telemedicine type systems. You've got KlonStar coming in. And you've got data going out to the electronic medical record. And you've got other data coming in from diet and from exercise.

So with all this complicated system of the future artificial pancreas, you have to have standards, and you have to have interoperability. If you don't have those, the system will fall apart. Nobody is going to invest in a medical system if they don't have proper standards and if it's not interoperable.

And I'll leave you with one sort of anecdote about this. At the Diabetes Technology meeting every year, we cover topics of technology. So a few years ago, it seemed as if there was a need to get blood glucose monitoring companies to standardize the connection between the blood glucose monitor and the computer. That could be done wireless or it could be done with a cable.

So I invited a speaker there from Sunnyvale, in Silicon Valley, who is from the flat-screen TV industry. He got up there at the meeting and he said, "Okay. I want you to know I don't know a thing about diabetes. But I do know about computer chips. And when we introduced standards and interoperability into the flat-screen TV business, sales went up because now people knew they could buy a component, and if they wanted to replace a component, they could replace it; they wouldn't have to replace the whole thing." So if you're a manufacturer, you want to have standards and interoperability because it'll help sales, and if you're a patient, you'll have it because if you replace some of your equipment, you don't have to replace the whole thing. So the key here is compatibility is important. It's important in everything we do.

I don't know how that got in here. Never mind.


DR. KLONOFF: Okay. So my conclusions are for standards and interoperability for diabetes devices, first, the most innovative diabetes technologies to monitor and treat now transmit data. Second, regulatory barriers currently limit development and adoption of diabetes technology. Third, standards, interoperability, and compatibility are the greatest barriers. And, fourth, federal government policies can help remove current barriers to creating new diabetes technology.

Thank you.


DR. SACKNER-BERNSTEIN: Thank you, Dr. Klonoff. One of the things that was interesting that you touched on in standards and operability is something that we talked about a little bit. So it certainly makes sense that if there is a marketplace that has standards, like, in the TV example you gave, that that would give developers and manufacturers an entree into a bigger market.

So as I look across the table to two big buyers, DoD and VA, we talked a little bit about this idea of standards and interoperability. I asked this before, but maybe not directly enough, or maybe I just didn't understand the answer, but how do you make sure that there -- that the equipment that you're purchasing and using does have the interoperability that you need? I understood that you could test it and you do test it, but do you use other mechanisms so that people know what to create, such as purchasing controls or other approaches that are a little shy of standards but may communicate the same kind of information?

DR. CROSS: Jonathan, I think certainly one of the key things we work on over and over again in purchasing devices is the interaction with the electronic health record and whether that has to be developed after purchase or if it comes from a manufacturer with the ability to engage into our electronic healthcare system. That's always a key issue, you know, the interoperability in terms of different brands of devices, the training package, the reprocessing requirements.

Right now, we find a wide variety of requirements for each different device that we purchase. I mentioned earlier that that's a real challenge for us. Consistency would be a tremendous help.

But I think what I think we spend a great deal of time on is the communication part. The transfer of data into the electronic healthcare system is what that really entails.

DR. SACKNER-BERNSTEIN: And if I understand correctly, one of the ways that you communicate that need and the approach that a manufacturer might need to take is there is a version of the VA electronic health record system that's openly available under an open license to anyone. Is that correct?

DR. CROSS: There are private versions of the electronic VA system, yes.

DR. SACKNER-BERNSTEIN: Okay. So that's one way that people can get a sense even short of having the standards in place for at least one payer.

DR. JACQUES: Certainly an interesting presentation. Thank you. And as one who likes to buy consumer electronics, I certainly appreciate that part of it. Diabetes, though, is sort of an interesting paradigm for a couple of other reasons, which might explain, in some sense, why although diabetes is obviously a giant market -- I mean, one would wonder, well, why isn't this all fixed? And it's certainly not that there aren't enough diabetics to make it, you know, worth someone's while to do it.

And I think some of this relates to sort of evolution and the understanding of what diabetes is, I mean, and much more broadly or more detailed than just the course differentiation between a Type I and Type II diabetic or a juvenile and adult diabetic or whatever, you know, classification one wants to use. And the thought that diabetes may be sort of more than simply a pancreatic failure issue, especially in Type II diabetics, and the issues of peripheral insulin-resistant cellular starvation at the neurovascular level or whatever do create some challenges as one tries to develop these technologies as, you know, how do you define the scope of interested people because the broader you define that, arguably you're not going to necessarily address every bit of that.

And as we've found, for example, with KRAS expression and colorectal cancer, the historical, sort of phenotypic classification of things, even things like cancer, may start to break down when it turns out that even if you had four Type I diabetics, that there might be some meaningful difference between them. And there is a lot of science that I think still needs to get done.

And I just sort of throw that out as a comment, not necessarily, you know, as a question, unless you want to address it.

DR. KLONOFF: I'll take a stab at it. The technologies that I've described are all extremely worthwhile for Type I diabetes. There is variable need in Type II. I mean, you do need better glucose monitoring even for Type II. Many of them are not on insulin, but there could be a greater acceptance of insulin and faster absorption if you deliver insulin in better ways.

Telemedicine will actually go from diabetes and be expanded to congestive heart failure and other chronic diseases.

Closed-loop artificial pancreas is probably not for Type II diabetes at this point.

But I do agree with you that some recent studies have questioned whether blood glucose control is the most important element in Type II diabetes. And three very disappointing large studies appear that tight control in Type II may even be harmful.

But I'll point out that because of the issue that tight control may be harmful in outpatients and there is now controversy about tight control in the hospital, it makes it all the more important to have accurate blood glucose monitoring wherever you go so you can act appropriately.

DR. SHUREN: You raise this tension, which does exist, in terms of the need to have flexibility in an artificial pancreas, as we sort of figure out what is going to be the right algorithms to use and software and work that through, or the same token, the desire on the other hand to assure that there is patient safety.

You also mentioned that we're moving into a world, the ability where you can report back on the status of the patient, that information can be collected and sent back. I welcome your thoughts in terms of that role, in terms of the telemonitoring of the patient as a way to address these sort of tensions between allowing flexibility and protecting patients if you're allowed to sort of monitor remotely, particularly when we're talking about the need for testing in the home environment.

DR. KLONOFF: Well, I think telemonitoring is very useful in clinical trials because it provides information. I edited the first article that talked about using continuous glucose monitoring in a pharma trial of a diabetes drugs. Traditionally, diabetes drugs were -- with some blood glucose levels several times a day. The pharma companies found that with continuous glucose monitoring, they had a much better picture of what the blood sugars were doing. And certainly, if you're talking about a device or an artificial pancreas, the more monitoring that comes in, the better, so I think it's a good thing, especially in a clinical trial/research setting.

MS. SETO: So, David, you have a slide up there that talked about compatibility. Do you also include biocompatibility in the sense of the issue of a patient's fibrous around sensors and needles and so on? Is that broadly included?

DR. KLONOFF: I think biocompatibility is really an important issue. That limits implanting sensors for more than a certain period of time and possibly implanting insulin delivery systems. But in this case, I'm not using it to mean biocompatibility. I'm using it to mean that it achieves certain published standards and that the devices are interoperable with each other.

DR. SACKNER-BERNSTEIN: Okay. Thank you very much. We're a little behind schedule, but I think we'll be able to catch up later. We'll still take a break now. It's 10:35. We'll take a 15-minute break and be back at 10:50, please.

(Off the record at 10:35 a.m.)

(On the record at 10:50 a.m.)

DR. SACKNER-BERNSTEIN: The first presentation of this session is from Dr. Richard Ehman, and there is an error on his title, on the agenda, although the affiliation is correct. And just as a note, if you check back at the webpage on, not only will we correct your title, but once the presentations have gone through 508 clearance, they will be posted.

And you'll also notice that at the top of the agenda is a docket number. The docket is now open and will remain open for one month to have anyone in the public submit comments or submit materials for our review pertinent to the scope of this meeting. So we're hoping that people do go to the docket and submit information for our review.

So with that, Dr. Ehman?

DR. EHMAN: Thank you very much. Well, we'll improvise here. My name is Richard Ehman. I'm a physician from the Mayo Clinic. And perhaps unlike perhaps some of the other speakers here today, I'm not coming here with a specific issue. I've been encouraged as a physician and researcher/inventor involved and immersed really in technology development to come here and talk a little bit about it from that perspective.

So at the beginning, I'll just take a few minutes to talk a little bit about the journey that my team and I have been undertaking the last little while. Then I'll make a few observations that that has led to about the process of encouraging technology development and then -- this is Murphy's Law really at work here. There we go -- and then try to generalize that to some recommendations that pertain to the topic of the conference here.

So like most other things, this is not working either. So I think we'll just -- all righty.

So I have a disclosure and an assurance here for you to read. And just to say by background, I've been involved in technology development for about 25 years in the area of MRI. And in the mid-'80s, I was involved in developing some MR technologies that are now pretty much used in every MR examination on pretty much every scanner. One had to do with decreasing floor artifacts, and another one had to do with correcting for motion phenomena.

Fantastic. Okay. So with that, what I'm going to do is, as I say, is spend a little bit of time talking about this little application and then talk about some observations that's led to about the technology development process, the barriers there may be and so on.

And so at the beginning of the 1990s, my team and I started working on technologies to image what it is we assess with palpation. And that turns out to be a very sensitive and time-honored clinical technique that's actually very, very diagnostically powerful, and so developing imaging technologies that could assess the same properties have been a high priority, and there are a number of approaches for this.

In the mid-90s, my team and I discovered a way to image mechanical waves using MRI. Even though these waves might have amplitudes that would be on the order of the wavelength of light in amplitude, we could still image them in vivo and use that as a way of imaging the mechanical properties quantitatively. The basic idea is that if you put mechanical waves into a system, the wavelength reflects the stiffness of the materials in space. And so now we have a way, potentially, of non-invasively duplicating in the body what we do with palpation.

So just to show how this is now used, these are two patients with fatty liver disease. This is a condition that affects one in three Americans, is thought to probably be the most common reason for liver transplantation, for advanced liver failure leading to liver transplantation within a decade or so in the U.S., so a very serious problem leading to chronic liver disease.

You can look at both of these exams anatomically with CT, MRI, or conventional ultrasound, and there is no way to tell whether one of these people is getting into trouble with this disease. The liver is anatomically normal by every standard and by every sequence we could use with MRI. And yet, here, with this technique, this way of palpating by imaging, looking at the stiffness of the liver, the patient on your right has a very stiff liver whereas the patient on your left has a very soft liver, which is the normal condition. And that very stiff liver is because this patient has advanced fibrosis in the liver, something that you can really -- previously could only diagnose with a biopsy.

And multiple studies have now been done since this has been introduced as a research tool. And just to show you one coming out of Belgium, looking at the diagnostic performance, using something called an ROC curve, showing for various fibrosis stages how you go -- as you go from lower stage to highest stage. And you use a blood test here, the AST-to-platelet ratio index, and you can see the performance of this test, whereas 0.5 would be worthless and 1 would be 100% correct.

This is an ultrasound-based elastography technique, and you can see it performs better, and in these people's hands, this MR elastography technique, in this 2008 study, had very high performance, essentially perfect performance for high grades of liver fibrosis.

So just to show you how this test now is being used at the Mayo Clinic and other places -- this is still a developmental technology, though -- here is a patient, a 66-year-old male with chronic hepatitis C infection. About one-quarter to one-third of these people develop progressive liver fibrosis and liver failure. We don't know which ones, and you can't tell by looking at conventional imaging. This patient needs a biopsy because of some abnormal liver enzymes to rule out fibrosis, but the patient has hemophilia, a bleeding disorder. A biopsy would be very risky in this particular patient. And now we have an alternative to offer them in this non-invasive technique that only takes about 15 seconds of imaging time in a conventional MR examination, which gives us this new kind of information, showing here the stiffness of the liver is abnormally elevated and in fact it was -- and so it was consistent with fibrosis, actually consistent with Stage 3 fibrosis, and this patient was started on anti-viral therapy for hepatitis C. So that's the patient impact.

So the current status of this -- now I'll move on to the observations about how we got there, what's happening now, and what I see as obstacles to moving forward. It's really -- this technology is emerging as a useful, non-invasive method for diagnosing fibrosis. In that role, it seems to be -- it certainly is safer, it is less expensive, and it seems that we have evidence that it's potentially more accurate than liver biopsy because, of course, with liver biopsy, you're only taking a tiny little thread of liver tissue and using it to represent the whole liver, and you can see how that would lead to sampling error.

So there are many other potential applications that await exploration, but our real goal here is to determine whether this technology can provide a reliable, quantitative biomarker for diagnosing disease, assessing treatment, and studying the natural history of disease.

So have we reached that? The answer is no. So even though we started in the early '90s working on this and our first successful experiment was in the early '90s and our first publication was in the mid-'90s and we've had a series of patents since that time and a series of NIH grants that have been absolutely necessary for the further development of this technology, it really wasn't until the mid part of this decade that we started seeing this liver application develop into something that we could believe in. And, really, it came into use at the Mayo Clinic at about 2007 for clinical purposes, experimental but clinically used technique.

But we're still not there. It still has not been accepted as a validated biomarker. And I would now talk to several issues that sort of are affecting our ability to move forward in that area. I have five things to talk about. So I'll talk about five issues.

And I think the first of these is that there is a lack of standards and tools for presenting quantitative medical information. So I ask you this sort of rhetorical question. Can you identify the three patients with psoriasis here? I think you can very easily because of this very favorable type of color scale that's being used, that we've adopted for this technology which makes it very easy without putting a cursor on the image to identify advanced disease.

Now, we're using something called a rainbow color scale there. But the problem is that there is no standardization. There are no standards in the color scales that are used in quantitative imaging.

And here is another publication done by another group that's now doing MR elastography, and they've used an entirely different color scale. And you can see here that whether you like this or not, this is a different look. And if this was to end up going into the marketplace with several different color scales, one for each vendor, that would not be a very good result.

And so we might say why is it so hard to find these tools? Well, it's because even visualization packages that are available present us with a myriad of color scales that are useful. Very few of these have had any perceptual physiology applied to them, in terms of understanding how the human visual system uses these color scales. So this is something that's needed.

The second issue that I would raise is that there are a few incentives to standardize what's being measured with some new imaging technologies. So in this particular case, where we're looking at the mechanical properties of tissue, there would be several different ways we could express what it is we assess with palpation. We could quote something that you could call the shear stiffness or shear modulus, which is a simple calculation from the wave speed that we observe. You can also quote it as the Young's modulus, which is another type of mechanical stiffness. But for tissues, it turns out it differs just by a factor of 3 from the shear modulus. You could talk about something called the loss modulus, which is the real part of the complex shear modulus, and on and on and on. The problem is that every publication in the field now is using a different particular modulus, or many different groups are using different moduli in this area.

And this just shows why this is such a problem. Here is that paper again from the group in Belgium from 2008, and they were comparing the stiffness of the liver as measured with MR elastography, the technology that we're working on, and this ultrasound-based elasticity imaging technology, based on something called FibroScan. And you can see that the curves for the two sets of data look very similar, but, in fact, the scales are very different. In this case, the scale goes up to 7, and here the scale goes up to 80, and yet the units are the same, kilopascals. And that's because in the ultrasound field, they're using Young's modulus as the metric whereas in the MR elastography field, which I would say pre-dated the ultrasound field, we use the shear modulus. We need some way of dealing with that and getting this because you can see how confusing this will be to clinicians as this goes forward.

The third issue has to do with looming problems in optimally tying the R and D of technologies like these to the clinical environment. We know that that's very important. This shows the status that we had in 2003 with getting mechanical waves, which are generated by a driver system that's located outside the body into the liver here, and in fact, you can see that the waves are only -- you can only visualize them in a small part of the liver there. But in 2006, we had reached the point where we were able to generate waves very adequately throughout the upper part of the abdomen and get very good general elastograms of the entire cross-section of the abdomen.

Now, what made this possible? It was constant iterative research, where we basically -- where we tested in patient populations these new driver systems, and we had to be tied tightly into the clinical environment to do that. We simply could not have done that with phantoms and so on. So we ended up with a driver system now that's very robust and could be used for a variety of applications. And as I said, it took a multidisciplinary team. It took strong translational efforts. And it had this sort of iterative approach.

But that's not what I'm about to talk about. There is another looming problem for doing this kind of work. And this is this particular phrase. I'm not going to try to pronounce it, but this, in the legal system, is known as the rebuttable presumption. And this is an assumption that is made that will stand unless someone comes forward to contest it and prove otherwise.

Where do we hear that? We hear that in the area of conflict of interest management. And in, really, 2001, the AAMC came out with really an excellent report, which is really an excellent review of the entire approach for managing conflict of interest and scientific research. And they basically said that rebuttable presumption against financial interest in human subjects research means that the institution will presume, in order to assure that all potentially problematic circumstances are reviewed, that a financially interested individual may not conduct human subjects research.

What that means is that universities that are taking advantage of the BiDole Act, in which investigators will then end up being co-inventors on patents, will now be -- under this particular rubric will be facing very significant obstacles to being involved in clinical human subjects research. But the saving grace here is that the rule is not intended to be absolute in that you may rebut the presumption by demonstrating facts that in the opinion of a committee formed within the institution constitute compelling circumstances.

And so I think that we need to take on a little bit more of a nuanced approach to conflict of interest management that recognizes that we have a range of possible management measures, ranging from very minor things to much more extreme measures. We have a range of types of research, ranging from exploratory to multi-site clinical trials, where scientific integrity is obviously a greater issue. We have differences between technologies that are significant risk and those that aren't. And then we should have a nuanced approach to managing these conditions that's based on the consideration of all these factors. And, unfortunately, many times we see that that's not happening.

Item four. We know the importance of comparative effectiveness research. We were fortunate. My colleague, Jay Talwalker competed, and I draw it as a bulls-eye here, because he competed for one of the challenge grants for a comparative effectiveness study for MR elastography that in the space of just two years is going to generate tremendously valuable comparative effectiveness data that will allow us to move this project forward. But I will tell you that's half a million dollars, and I think that that's a very expensive way to get this kind of information, to do something like this, and I think that we need better ways, more efficient ways of generating this kind of information.

Finally, issue five. So if you want to move this forward as being qualified as a quantitative imaging biomarker, it will be contingent, according to the FDA, upon adequate standardization of the imaging process. And that's where we have this multi-organization alliance, the Quantitative Imaging Biomarkers Alliance, that's kind of focusing on this issue of how to ensure that, across platforms that run imaging technologies, in this case, diagnostic imaging technologies, how can we have standardization and so on.

And in the case of MR elastography, it's quite complicated because we have a situation where we have driver systems. We have imaging sequences. And then we have algorithms that convert those data into the final diagnostic images. So we got hardware, we've got imaging technology, and we've got processing technology, all of which has to be developed, and this has to be identical, essentially, across platforms. Fortunately in our case, because we have, in this particular case, have control over the IP and we will not license it to companies that won't implement it in a standardized way, we have this ability to make sure that we end up with the result we want. But normally that's not the case.

So I guess I'll just wrap it up, then, with this list of five observations that I've tried to broaden beyond these five specific points that I've listed that relate to the little research activity that we've been doing.

And so first, I would say that we would facilitate medical device technology development by establishing accessible, recommended tools for presentation of quantitative medical information. So, for instance, I don't think it should be hard to find a standard color lookup table. It shouldn't be that there are 15 different rainbow-colored lookup tables. It should be easy for some organization to put in place a recommended color lookup table so that we can all use the same one.

Secondly, create incentives to standardize among similar biomarkers, which might be something that is a process of negotiation, but then maybe suitable forums to accomplish that task so the investigators who are working in an area could somehow come together under the auspices of some organization at an early stage and say, gee, is it really necessary for us to be talking about different metrics of the same thing?

Instituting more precise, pragmatic, and focused approaches for recognizing and effectively managing conflicts of interest.

Finding new ways to accomplish comparative effectiveness studies in a more efficient fashion. Let's bring systems engineering to bear on this particular objective.

Finally, creating incentives for vendors to standardize technologies that go across all of their platforms.

And with that, I'll thank you for your attention.


MS. SLUTSKY: I'm looking to the woman in the first row. Is that better? Yeah. Thanks.

I wonder if I could ask a question about number two, the standardization of similar biomarkers. Are you also encouraging the commercialization of the biomarkers as well so that they're available to the treatment community?

DR. EHMAN: Right. Well, so we're talking -- when I talk about biomarkers, I'm talking about imaging-based biomarkers. So I'm talking about tissue properties that we can measure with an imaging technology. And we all believe that those are -- have -- there is tremendous potential in their quantitative use. And I really don't anticipate that, and I hope that we don't end up with a situation where they end up being proprietary so a single company would own that particular biomarker -- if I understand your question correctly?

MS. SLUTSKY: I guess what my question is, sometimes biomarkers are developed to get a product approved through the regulatory process --

DR. EHMAN: Yeah.

MS. SLUTSKY:  -- but then isn't available to the prescribing public so that you can adequately target the patient population that has been shown pre-market to benefit from it.

And then I just have another comment. One of the issues that we talked about earlier today was the increasing radiation exposure and the possibility of scope creep for what looks like a promising technology beyond a use for patients that are indicated to have this type of a scan.

DR. EHMAN: So that's a great question. Maybe I can take the second one first. So this would be an interesting view of imaging to say, gosh, we've developed this technology which has this application, and we would worry that people would start to use it for other things.

In fact, I think you could make the case that in the case of MRI, MRI is the technology that was invented by its users. It wasn't invented by the companies. So in other words, at the academic medical centers that have pioneered MRI, all of the new applications that we now enjoy, almost all of them were developed at academic medical centers. So I think that there is a role for exploring the -- there will be definitely a role for exploring the other applications of this. Now, not in a reimbursed mode, I think, would be true. So I hope I've kind of addressed that particular point.

Already, there is lots of work going on applying this technology in other areas. For instance, there's some early evidence now you can apply it in the brain. There is some early evidence that there is a distinct difference in brain material properties between patients with cognitively abnormal high plaque loads, Alzheimer's patients who have plaque, who are cognitively abnormal, versus patients who have high plaque loads who are not cognitively abnormal. So it's very intriguing and it will need to be explored.

DR. JACQUES: You've touched on several very important points. I want to sort of focus on number five a little bit.

When positron emission tomography with fluorodeoxyglucose was, in fact, in its adolescence, well beyond its infancy, I was at a meeting -- and actually, I think Dr. Woodcock may have been at the same meeting, it was sponsored by NCI -- with internationally leading researchers in this particular field. And they were describing how they might do studies in the future. And they couldn't agree on what was the dose of FDG, when would you administer it, how soon after administration would you start imaging, and for how long would you image. And on top of that, the conversation is clearly one that is beyond just the vendor. But it also includes how is that nuclear medicine physician going to use this, how, once you get out of the research community and into the community practice, which is particularly of interest to the Medicare program, how are you going to get any sort of comparability there. And the joke, the insider joke in that meeting was that an SUV was a silly useless value, not a standardized uptake value.

So in that light, we've been trying to spread that message -- and I see Houston Baker in the audience -- for optical imaging. They invited us to a meeting of researchers on optical imaging. And one of the questions that we asked was essentially, well, what exactly is white light? And people who are into this stuff realize that's actually a very complicated issue. And one of the things that I was gratified to see happen was that, at that very early stage, the people decided that they really had to have standards. Otherwise, they could do all the research they wanted to, but they were never really going to get where they wanted to get. So I thank you for bringing this to everybody's attention.

DR. EHMAN: Thank you for that. I guess I would say that I think that there's been a bit of a sea change as well. So it used to be that physicians -- and I know in my own specialty -- expressed their individuality by having different protocols and, you know, and everybody had their own special protocol. We all have, I think in most places now, there is rallying cry around standardization because we recognize that we can only learn if we apply standardized conditions to things.

And I think, actually, my sense now, from working with the companies in terms of the rollout of this technology, has been that they get it, too, now, and the major companies do understand that it will be an asset from the standpoint of the acceptability to the regulatory authorities and to the, eventually, the reimbursing authorities to have a procedure that it doesn't matter where you have it done.

DR. SACKNER-BERNSTEIN: I'd like to touch on point number three, the conflict of interest. I'm wondering how far your thinking has gone in this arena beyond the financial relationships that people in the research world can have. I think we're all aware of situations where people have certain intellectual commitments they've made, consciously and unconsciously, and how you think that can potentially be managed or whether you think that really is something that needs to be managed.

DR. EHMAN: It's a really, really good point. And although much of the approach for managing conflict of interest in the last few years has been mainly to identify significant financial interest, we all would recognize that there are many other motivating factors that affect people. And we have to accept that, in many cases, it will be governed by integrity. That's kind of what it amounts to.

So I think it's a very legitimate point. And, you know, if you are writing your next NIH grant application, you have a lot at stake in what your -- how you're reporting your results and so on. And some would argue that, in many cases, that's equally as potentially conflicting as would be some modest consulting income or something like that. So it's a really important point.

The point I'm trying to make about this is that we tend to assume right now that pure disclosure, what we call disclosure, which is that list of things, affiliations at the beginning of the presentation, that's kind of the extent of our conflict of interest management. That's really raw data. That has to be looked at by a group of people who will decide is there a conflict. And then they need to institute powerful, pointed measures which are available to manage the conflict in that particular case. And so I think -- I'm really arguing for stricter and stronger conflict of interest management methods, not less.

MS. SETO: Dick and Jonathan, I couldn't agree more that financial conflict of interest is really just one piece in this whole universe of conflict of interest. One example would be many of our investigators, not only are they PI on an NIH grant, they are also the care provider to patients, and it would be very tempting when they see patients and say, gee, I may be able to recruit these patients into my study. And so the role -- the fussiness of the role of the PI and the care provider then create the conflict.

DR. EHMAN: Yeah.

MS. SETO: And that often had occurred.

DR. EHMAN: There is another dimension to that as well, which is that one of the new frontiers that's being talked about in the conflict of interest management circles is the issue of conflict of interest in clinical practice, right? And so people are -- many institutions are instituting policies for conflict of interest in clinical practice.

But the point that's made by many people is that the practice of medicine itself, by nature of the way the reimbursement goes, is already rife with similar considerations, whereby doing more, in many cases, the income will be higher. And so you have to consider the entire setting, and these circles keep broadening, and I think it's highly relevant to the debate.

DR. SACKNER-BERNSTEIN: Moving on to the next presentation by Jack Lasersohn.

MR. LASERSOHN: Thank you, Jonathan. I'd like to thank FDA and particularly Jonathan for organizing this terrific meeting. I do have a lot of slides, so I'm going to try to go through them pretty quickly.

My name is Jack Lasersohn. I'm a venture capitalist myself. I'm also on the Board of the National Venture Capital Association and Past Chairman of the Medical Industry Group of the NVCA, and I'm speaking on behalf of the Venture Capital Association today.

This is one of the messages that I'd like to deliver. I also have some policy recommendations. The venture capital system does fund virtually all revolutionary medical device development in the United States and has for many years. Unfortunately, the VC systems is extremely sensitive to any perception of increased risk of any kind, particularly in the regulatory and payment system. VCs do believe that many parts of the current federal regulatory and payment system for medical device innovation is broken or severely damaged, and this is already having a significant effect on impairing the development of revolutionary device technology to meet critical public health needs. And I'll talk about this in more detail, and I always believe that if you're going to complain about something, you also have to have some suggestions for how to fix it, so I'll get to that as well.

The NVCA represents the entire venture capital industry. In 2009, venture firms were managing approximately $179 billion. That's down significantly from the peak, but it's still a great deal of money.

Many people know that venture capitalists invest in companies like Google and Apple and Microsoft. Many people don't know that, in fact, the venture system provides substantial capital for medical research. In fact, virtually all biotechnology commercial development and medical device development is funded by the venture capital system at the pioneering stages.

We do provide capital for early-stage companies. But more importantly, we often found and manage these companies. In fact, we always manage these companies. We're always represented on the boards of directors. Very often, we're the chief executive officers, temporarily, of those companies, in the early days. And most of us have actually been entrepreneurs, scientists, or technologists ourselves.

As I said, it's not well understood that most medical device innovation in the United States is, in fact, originally funded and pioneered by the venture capital system and by our entrepreneurial companies. This part of the innovation ecosystem, which we all depend on, is now seriously at risk.

Just by way of background, it's very hard to think of any significant medical device in the last 30 years that was not initially pioneered or developed by a venture capital-backed system. There are some notable exceptions to this, particularly in orthopedics, hips and knees and a couple of other areas. But by and large, if you think of a major medical device innovation, it started with a venture-backed entrepreneurial startup.

Unfortunately, the investment in these startups has declined precipitously over the last few years. That is not only a function of the financial crisis. It's also a function of our perception of increased risk of investing in these areas.

This is probably the most significant piece of data on that. These are the actual startups, the number of first-time startup companies that we fund in the medical device section each year. That has been going down now for four or five years, and it continues. The last one is a quarterly number. But if you annualize that, this first quarter is still significantly down from 2009, which itself was down precipitously from years before.

The reason for this is that venture capitalists look at all investments the way doctors look at treatment alternatives. We look at risks and benefits. In our case, the risks are the amount of capital that's required, the risk that the technology or the company will ultimately fail and we'll lose the investment, and how long it takes because that affects our rates of return. And on the other side, of course, we're looking for the ultimate financial rewards.

How do we assess risk? There are many, many, many factors. I'm not going to go into all of them. But suffice it to say that many parts of the federal government affect how we assess risk. In particular, FDA and CMS, although I will add that, for example, patent policy, university policy, NIH funding, many, many other factors are also very important here.

Another thing that's occurred more recently in terms of risk is that the capital system has failed to some extent, and as a result, there is not as much money available to support these companies. This has come at a time when the amount of money that's required to fund these companies has increased dramatically because of the needs for increased clinical trials and evidence-based medicine. So the two of these has created a serendipitous situation, and our companies are very, very squeezed. The result is that many novel ideas are, in fact, not being funded today as compared to just a few years ago.

This is the policy suggestion that I'd like to make to this Council. Evolutionary and incremental improvement in technology really can occur automatically. There isn't that much you need to do about it other than to have very reasonable cycle times on the regulation, have reasonable payment policies. We think the 510(k) system is very important to support evolutionary, incremental advances.

But unfortunately, this does not work for revolutionary advances. That's what this Council should focus on, we believe. The Council should focus on expediting the development of novel and potentially revolutionary technology for which the current system really is not functioning. These are mostly developed by small companies. They always raise lots of questions for the regulatory system and the payment system, but these are the technologies that really will affect and address important unmet clinical needs.

We have an entire agenda at the National Venture Capital Association in this regard. We do suggest ultimately appointing at the presidential level a medical innovation czar which will help to break many of these deadlocks. But we do have some specific recommendations for how many of the stakeholders here today, the regulatory stakeholders, can help, which involves, in particular, an approval process and payment reimbursement process and comparative effectiveness evaluation system for what we call novel technology.

I'd like to talk a little bit first about individual agency suggestions of initiatives and then move on to interagency collaborations.

With respect to the FDA -- and I don't want to pile on to the FDA -- the FDA's job is extraordinary difficult and is becoming more difficult because the complexity of medicine is increasing. But the FDA has become overly risk averse. As a result, FDA is adopting policies which are adversely affecting innovation. And, in particular, the FDA process for approving novel medical device technology really is perceived to be broken by the venture capital community, which, of course, increases our perception of risk and therefore reduces how much money we're willing to invest in this area. This has already happened. This is not a prediction. This is happening as we speak.

Novel technology poses a special challenge for FDA as well as for everybody else simply because nobody knows what to do with it. You've never seen it before. There are no precedents for novel technology. Most people's reaction to novel technology, I can tell you, as a venture capitalist for 30 years, is that will never work, that's a terrible idea, that's quackery, you know, you should forget about it. That's the typical, overwhelming reaction to anything new that we ever invest in. And, of course, it's a very human reaction and we understand that. But when it occurs at the FDA, it could really have a very harmful impact on our willingness to invest in these areas.

As we said, these are uniquely complex and difficult areas. We do recognize that. But the typical reaction at FDA to many of these -- most of this novel technology, we think, is to ask for impractical levels of evidence of safety and effectiveness. And we think that that is always a tension. We recognize it's a regulatory tension. But we believe that the pendulum has swung too far in the direction of risk aversion.

For example, we always would like to do randomized controlled clinical trials with sham placebo arms, and we are in fact increasingly doing it. In fact, many of our companies are voluntarily suggesting ab initio that we conduct 500 to 1,000-patient trials. In some cases, I have to say, the reaction to that has been to say thank you, but what we'd really like to see is something new called super superiority, and we'd really like you to do a 10,000-patient trial. This is not an exaggeration. We have been asked for trials like this. This will never occur. The medical device system can simply not support many trials of 10,000 patients.

We have had many discussions with FDA about a policy proposal that we call the Novel Technology Review Process. I'm not going to get into detail about it. Frankly, it mostly involves involvement of senior FDA officials at the earliest stage of review and management of a very small number of selected approval applications. We recognize management of FDA cannot possibly review every application, but certain applications really need to be picked out for their complexity and for their importance. And that's the basis of our recommendation.

With respect to CMS, I'd like to say that, in fact, the comment made that payment is the ATP of innovation is absolutely true. There is no question from a venture capital point of view that that is true. And CMS is very well aware of this. I would say that with respect to CMS, the most important thing that CMS can do is apparently already being done as of today or yesterday, which is to parallel path its coverage evaluations with FDA-approval decisions. This will be enormously challenging to implement. There are many issues that we have to face, but we think -- and we have recommended such a parallel path and collaboration for many years. In fact, it is one of our collaborative recommendations, among others.

There are many issues at CMS that can -- other issues at CMS that can be addressed. I'm not going to go through them in detail. I think CMS is well aware of them. We do think that, in general, focusing CMS on the adoption and publishing a critical path for reimbursement in the way the FDA has for many years published paths, critical paths for approval, would also be very valuable, and I know CMS is already working on such things.

With respect to NIH, there are already a couple of collaborations that are taking place which we think are terrific. The fast-track innovation collaboration is terrific. There is another thing that's in the new healthcare reform law which we hope will be implemented called the Cures Accelerated Network. We believe, in fact, that this could be a platform for interagency collaboration among all of the stakeholders in this room, and we would encourage that.

It is the case, though, I do invest in biotechnology as well as in medical devices, and it is -- my case, for example, and in the VC's case, every biotech company I ever invested in was really started by the NIH. I mean, it wouldn't exist but for NIH support at the research level. There are clearly exceptions to this on the device side -- I'm sorry -- that is, that on the device side, it's really just the opposite. No company I have ever invested in has been supported by NIH, and that includes all of the companies that were on the list before. There are notable exceptions to that, particularly in areas such as the development of LVADs. But we believe that NIH has an enormously important potential role to play here in sponsoring earlier-stage research in the device area, but frankly, much more important to us would be sponsoring trials at the pre-approval and approval stages, not at the post-market stages of device development.

With respect to comparative effectiveness, I would say that the comment made before that when to call the question is absolutely one of the most important issues that comparative effectiveness agencies are going to address with respect to innovation. When you ask a revolutionary technology to prove that it is in fact safe and effective for comparative effectiveness purposes can either make the technology successful or make it fail. If we had asked the wrong question at the wrong time of angioplasty, we would not have drug-eluting stents today. So understanding the life cycle of technology, when to ask the question, recognizing that new technology is often only marginally effective when it is first introduced and becomes much more effective over time is an extremely difficult methodological issue that comparative effectiveness will have to address.

Many of the consumers of healthcare who are here also have much to contribute. DoD, DARPA, and VA have done tremendous work in areas that they have particular interest in, for example, traumatic brain injury and prosthetics, wound management, which have had enormous public health benefits as a result of that in the broader economy. But on the other hand, things that affect the broader economy, healthcare in the broader economy obviously also affect the members of the armed forces either during active service or in retirement. So we think that DoD, DARPA, VA should be willing to participate perhaps a little more broadly in some of the less narrowly defined areas of interest that ultimately affect their consumers not necessarily on the battlefield but families of service people as well as elderly service people and retired service people.

With respect to collaboration, we believe that there is one -- we have one major recommendation which I'd like to talk about. As I said, we believe that FDA and CMS should create a novel technology approval path. They have already taken the first step towards collaboration, which again we applaud. We think this is terrific to have a parallel path on approval. There are many details that need to be worked out. We will be happy to work with the agencies on that.

But we think it should go further than that. We think that certain pieces of technology really do need to be pulled out for special, if you will, for special handling. These technologies should have access -- these approvals should have access to very high levels of review with inside of these agencies. We know that high-level management review is very resource limited, so we have to do this only on occasion, but it really is the only way to address novel and important technologies, which the agencies have historically not seen.

We also think that for such projects, NIH and DoD should establish rapid funding processes in support of pre-approval clinical trials of these technologies. In other words, we think that NIH, when such technologies are selected by CMS or FDA for special processes, that NIH, DoD, other funding agencies should give special consideration to supporting the clinical studies, financially supporting the approval studies or the CMS coverage studies for these devices. These studies are becoming increasingly expensive and are very difficult for the venture capital community by itself to conduct.

As another incentive for sponsors being willing to participate in these, which we recognize will be more burdensome in some ways for sponsors, we think that products approved by FDA under such a designation should have presumptive coding and coverage decision by the payment systems. I know this is obviously a very controversial suggestion, but it would be probably the single most important thing that we could do to create incentives for sponsors to participate in these kinds of studies.

And very importantly as well, any post-market comparative effectiveness studies of these devices should be based on methodology that has been explicitly developed to study new technology, recognizing that first-generation technology is very sensitive to how you analyze it and when you analyze it.

And I'll just conclude by putting up a chart that shows we are still investing. I hope we're still investing and starting companies in all of the areas that you would all recognize as being significant unmet health needs. And we will continue to do so, even if it is at a somewhat slower pace than before.

Thank you.


DR. JACQUES: Thank you. You did mention time. And if you hadn't mentioned time, I was going to mention time. Venture capitalists don't often come to us directly in coverage. Either that or they don't mention to us that they're also financing the company.

But I've sensed a tension between the venture capital community and the people whom you are actually funding. What I hear from venture capitalists is, "I need a quick answer. Give me a yes or give me a no." What I seem to hear from the people you are funding is they want to drag this out as long as possible because if they get a no early, then clearly they're out looking for a job. The benefit that might accrue to you to not pay them isn't necessarily communicated by them.

So we are aware of that. And sometimes the length of time is really not necessarily that we're the ones imposing it. It's that they simply want to drag it out and bleed things as long as possible. And that's sort of an issue, I guess, for you guys to talk about with those people.

But let me ask you this for a second. If CMS were to sort of prospectively, shall we say, take a leap of faith, whether that's through coverage payment coding, whatever, at an earlier stage before sort of the usual evidence base had been developed enough to have sort of more confidence, what would you think to a public payer would be sort of the acceptable win/loss rate in that? So if we did that for ten technologies and five of them turned out in the long term either to be harmful, wasteful, you know, red herrings, whatever, and five turned out to be good, or is it, you know, seven and three, you know, ten and zero? How would you articulate that?

MR. LASERSOHN: Well, I think that these would be -- not be reimbursed unless there was first-level evidence, high-level evidence of safety and efficacy. And, normally, our experience is that if you have demonstrated that, very few of these things turn out to be, you know, ten years later, to look like they were duds. It does happen occasionally, but I think that's because the earlier evidence was so weak. So I would not expect that you would have -- you will have duds. I mean, we'll learn. Everybody will do a study. We'll have evidence. We'll think that it works. You know, lavage for, you know, cartilage stuff, everybody believes it although there really weren't any studies for that, right? And then ten years later we'll discover it doesn't work. I think that that would really be quite rare if we had reasonable evidence up front.

The more subtle issues is understanding that the level of effectiveness of early-stage technology is small. The marginal difference in the first iteration of a technology is usually small and very hard to prove. I mean, I always use angioplasty as an example because it is such a perfect example. We actually invested in the very first angioplasty company. And you know, people forget that it was a treatment for angina. It had no impact on mortality or anything else. It took not only the invention of drug-eluting stents 20 years later but the invention of PCI and the change in the logistics system to actually have an impact on AMI mortality.

So I think that's the bigger challenge, which is at what point how much of an improvement in something do you think you're willing to pay for? What's the minimum you're willing to pay for?

DR. JACQUES: Okay. Because that does change the question a little bit, as you sort of alluded to calling the question. Right now, the question is yes/no. It sounds like, if I'm understanding you correctly, what you'd like to see is a question that has greater granularity so that, essentially, as the evidence of benefit became more robust, in fact, the payment would accompany that. So there would be some element of risk sharing, which currently would be a bit of a challenge under our current authorities around how we pay for things.

MR. LASERSOHN: Well, you're doing that, for example, to some extent in carotids, right, because you've done it not by saying we'll pay a different amount, but by limiting the indications. So you start with a more narrow indication. You can use it in this patient population. Now, we'd argue it should be a little broader than it is, but nonetheless, that's a very reasonable approach. And start with an indication where you have reasonably high confidence it works in the subpopulation without asking us to do a repeat subpopulation study, right, and then broaden the populations out from there over time. That's one possible approach.

MS. SLUTSKY: I wonder if I can ask you a question. When you talked about comparative effectiveness studies but also the need to get an early yes or no, what is your organization or what do you personally think about managed entry schemes?

MR. LASERSOHN: I didn't hear you. I'm sorry.

MS. SLUTSKY: What do you personally or your organization think of managed entry schemes?

MR. LASERSOHN: I don't know what that is.

MS. SLUTSKY: Well, in some health plans in other countries, what they do is give provisional approval as long as data is collected as it permeates in the health system.

MR. LASERSOHN: Right, right. So CER, coverage with evidence development type --

MS. SLUTSKY: It's a variant of coverage with evidence development.

MR. LASERSOHN: Right, right. We think that's a terrific idea. That's what I was just talking about. So there are two things that we are recommending. One is potentially actual deferrals of evaluations. In other words, give -- for particular technologies, recognize that the earliest iteration is not going to be the most effective, give us a year or two or three before there is a comparative effectiveness study to allow the bugs basically to be worked out of it. That's one possibility. And the second is managed entry, ideas like that. We think that would be a terrific way to approach this.

DR. CROSS: First of all, can you hear me way back there? Okay. I want to make a comment and then a question for the speaker. The comment relates to the previous discussion we had on how to interact with an organization such as mine, the VA, to move technology forward. And I did not mention one possibility, and that's CRADAs, cooperative research and development agreements, deriving from the Federal Technology Transfer Act of '86, and they tend to focus on pre-market work and offer protection for proprietary information. And I have an expert in the audience, Tim O'Leary, if you could stand up, if anyone has -- way in the back there -- if anyone has questions on the break.

And I wanted to ask, sir, about your experience. Have you had any experience with CRADAs and our organizations, in particular?

MR. LASERSOHN: Well, we've had some experience with CRADAs. We think those are, again, a very useful idea. FDA, there are some FDA industry collaborations with CRADAs because, you know, FDA actually has a research division as well.

I know that one company I'm familiar with has done a DoD CRADA, actually a DARPA one on traumatic brain injury. My impression is that these tend to be, again, in relatively narrow niches. We think expanding that would be very helpful.

You know, as an aside, for example, a related area is SBIR, with respect more to NIH. I mean, it's not well known that venture-backed companies are, in fact, not allowed to apply for SBIR grants on the kind of convoluted theory that if you're good enough to get venture capital, you don't need money from the government. And sort of what's the converse of that, which is if you're not, that's where you should get the money.

So I think there are some peculiarities in the grant systems that need to be addressed like that, but I think that that is another area of potentially enormous collaborative opportunity, yes.

Lt Col WATTENDORF: A couple questions. One, you had noted that the medical device companies and biotech are obviously so much smaller than big pharma, and in the VC scheme, are you looking to increase the investment, where the market is not entirely the U.S. market and looking at the global market or where the predominant market is the global market and the U.S. market may be the minority, and what is the investment strategy for that? Has that changed any in the past decade or so?

And the other is I didn't see any diagnostic investments on this. Is there no investment from your organization in the diagnostic space?

MR. LASERSOHN: So let me take that -- the second answer is much -- is very easy. The answer is, yes, we definitely invest in the diagnostic space. I could have put up a number of big developments in diagnostics. All of the personalized medicine companies that you're aware of have been founded, in many cases, by venture capital. So we're very interested in diagnostics.

With respect to the OUS/US, you know, one of the hard realities for Americans is to accept the idea that, in fact, the American medical system does essentially support and subsidize the entire innovation ecosystem in medicine for the entire world. And to some extent, the rest of the world is a free rider. Not completely, but most people recognize, most economists recognize that the rest of the world is a free rider on American research because we pay enough in our reimbursement system, our payment system, at least up until now, to allow for innovation to take place. Many other countries don't and, in fact, very explicitly free-ride on the system. So, in reality, we do not look at overseas markets as a key market.

Now, people do invest opportunistically in China. All of us take our technology after it's approved in the United States, or sometimes pre-approval because it's simply easier to get approval in Europe, to Europe first. But the reality is that it's very, very difficult to make enough of a financial return in those markets to justify it.

On top of the return you make in the United States, the marginal additional return is very, very profitable in those areas, but by themselves -- for example, if we shut down the American system or made it into a U.K.-like NICE system, there would be much, much less medical innovation in the world.

DR. SHUREN: You had raised two interesting roles for FDA and CMS. One was for CMS to get involved during the clinical trial phase, and the second was that when you have FDA approval, to have presumptive coding and payment for novel technologies. Be interested to hear what you sort of anticipate with CMS involvement in clinical trial design at that phase. Are you anticipating, then, the development of trials that would address the FDA standard of safety and effectiveness to get on the market and reasonable and necessary in order to get coverage, in that you are covering both, and therefore you have this presumptive coding and payment, or are you anticipating something different?

MR. LASERSOHN: Yes. In other words, we're expecting to cover the CMS-related questions in those trials. Now, in many cases, it's very easy, right? Just doing a trial that includes patients over 65 begins to address the question, and as you know, many trials, people just forget if they want CMS coverage, that they actually have to have some evidence that it works in the elderly -- in the Medicare population. So there is that level of sort of simple things to do.

It turns out that, at least in our interaction with CMS, because we now -- all of our companies now go to CMS sometimes before they go to the FDA to talk about what CMS wants to see -- that the CMS hurdle is really not impossible. It means you have to collect additional data. You have to do patients over the age of 65.

Most of the time, it's not really an enormous hurdle to add that at the earliest stages. There are exceptions to that. And one of the reasons it's useful to go to CMS is that you can be surprised. Obviously, in the case of disc arthroplasty, the fact that CMS didn't think that fusion was sort of a relevant comparator, namely, a relevant standard of care, came as a huge shock to the industry even though, you know, FDA obviously did think that that was the relevant standard of care comparator. Knowing that before people invested hundreds of millions of dollars would have really helped. And it may have even given us an opportunity to find other things to look at in these studies that would have persuaded CMS that these things actually worked.

So I think it is not going to create an enormous additional hurdle. It cannot. If it does, it will just reduce the investment even further because it will increase the cost and risk. But I don't think so, on average.

DR. SACKNER-BERNSTEIN: Okay. We're coming up on our scheduled time for lunch. I think in the interest of trying to have the afternoon run on schedule, we'll take a break now and we'll convene back at 1:15. When those of you who are scheduled to speak or think you're scheduled to speak head out of the room, please check at the counter to make sure you're on the list. It was part of the handout. And then also, anyone who would like to speak who has not asked to can submit a request on one of the index cards outside, and if we have time, we'll try to accommodate them as well. Thank you.

(Whereupon, a lunch recess was taken.)


(1:00 p.m.)

DR. SACKNER-BERNSTEIN:  -- the afternoon session. We're going to have one more person sitting at the Panel who has been very involved with the discussions, Guillermo Arreaza from NIH, the National Institute of Diabetes, Digestive Diseases, Kidney Diseases. Sorry. I'm not getting NIDDK right there, Guillermo. So he'll be joining us.

And as I told the people by e-mail who had registered, we have allotted two minutes per presentation. What we'll do is have you come up to the podium. There is a lavalier we'd like you to speak into. The list was listed publicly. I know there is at least one person who has decided not to present, and there was duplications. So we'll start with the presentations, and then every so often, we're going to break to see if members of the Council have any questions/clarifications.

Even though this is a short presentation, please remember that the docket is open. It's listed here on your agenda. So we would hope that anybody who has information they would like us to have for review would submit it. I know some of you have submitted information by e-mail already. Please, if you want it to be part of the consideration, submit it formally through the docket.

So with that said, we'll start with our first speaker, Robert Campbell from the Children's Hospital Philadelphia.

DR. CAMPBELL: Well, thanks a lot. I'm originally from Tennessee. I talk slow, but I will try to get my presentation out in the limit.


DR. CAMPBELL: I'm a practicing pediatric orthopedic surgeon at CHOP in Philadelphia. What I want to make some remarks about is the need for better metrics for pediatric device evaluation.

I have some personal experience. I have served as principal investigator for a pediatric device trial for 14 years, rumored to be the longest device trial in U.S. history, but we can't confirm it.


DR. CAMPBELL: Now, so what did we learn? I think when we were going through our device evaluations, same with the protocol, the problem was how do we measure things? That's what metrics is, to my understanding. And does it really measure what we need to know?

And you've heard a lot in the morning session about what hinders device development, standardization of outcome measures, metrics, also, affordability. You can't get a pediatric device developed unless there is an affordable regulatory pathway.

So if we had to do it all over again in a perfect world, what I would have done with our device is to meet with experts and actually have a set metric approach to it, have a expert in devices, expert in the diseases we're going to treat, and have them discuss what should you measure with this new device, get it out, discussed up front. And rather than do it the way it's done now, where you, you know, you go -- your new device, your company, you go to your reviewers, and then you just sort of argue out what's the best thing to do.

I'm an orthopedist. Our main outcome measure is something called a Cobb angle, which we measure on spine x-rays. That's 60 years old. That's our gold standard. It's very outdated. There are much better ways to do this, but it really hasn't been talked out.

So I think a proactive approach to this -- and I think it would be helpful for the FDA if clinical experts could come up front, talk about devices, what's being developed, and then develop a consensus on metrics, on how to measure this, and develop new metrics. And then I think it'd make it easier. You don't want to get in the situation when you're measuring everything because that gets very expensive.

So how we measure and how well you think it out before you start down the regulatory pathway really makes a difference as far as getting successful pediatric products to market because we aren't doing very well. We're trying to do better, and I commend the FDA for all their efforts in pediatric device development.

Thank you.


DR. SACKNER-BERNSTEIN: Thank you. Robert Granier?

(No response.)


DR. SOBEL: The Association for Molecular Pathology commends the Council for working to identify and remove barriers to transitioning research findings into clinical practice. Serious barriers do exist that can impede the path to approval of molecular diagnostic tests, and that can reduce the motivation to submit them for review.

Our written comments list several barriers, including the paucity of standard reference materials, which inhibits the production of appropriate control materials and methods, and the difficulty of obtaining rare specimens for studies to support applications for the approval of new indications for currently approved tests.

I would like to highlight the fact that test manufacturers perceive that their submissions are subject to inconsistent and unclear requirements.

Diagnostic device manufacturers face uncertainty and inconsistency in the review of submissions, in enforcement discretion, in device classification, in requirements for acceptable analytical and clinical validations, and in requirements changing from the time of pre-IDE meetings through mid-trial.

This uncertain regulatory environment limits their efforts to anticipate regulatory requirements and appropriately amend their business models. AMP is concerned that this greatly deters submissions of tests for review.

We present several current issues that illustrate this particular barrier. First, requirements for tests for the same analyte have substantially increased for subsequent admissions. The FDA has reviewed and cleared several tests for the detection of mutations, for example, associated with cystic fibrosis and blood coagulation disorders. The manufacturers of subsequent tests are required to perform testing on a greater number of clinical samples for each mutation, and the scientific evidence supporting these escalating requirements has not been communicated.

Second, comprehensive studies to demonstrate the clinical utility of tests that monitor therapy for viral infections are required even though clinical utility for such testing has been well established.

AMP believes that the FDA can take several steps that would improve the regulatory process for molecular diagnostic tests without impinging upon an appropriate review to ensure that the public is protected. FDA could collaborate with relevant agencies and organizations to establish a bio repository of clinically relevant infectious agents, to facilitate the rapid development and validation of assays for infectious agents, particularly those with pandemic potential. Alternatively, consideration should be given to establish alternative validation strategies.

FDA should ensure that policies and requirements are consistently applied and that the scientific evidence and rationale for decisions are communicated effectively to diagnostic test manufacturers.

We recognize the difficulties regulatory agencies face and the context of the rapidly changing landscape of diagnostic devices and technology and appreciate the transparent process the FDA is undertaking now to improve the review process for medical devices. We offer our expertise to assist the member departments and agencies to develop a more consistent, evidence-based, and transparent policy for regulating diagnostic devices.

Thank you.


(No response.)

DR. SACKNER-BERNSTEIN: Daniel [sic] Crais?

(No response.)

DR. SACKNER-BERNSTEIN: Cynthia Rice? Don't worry. The time to come up doesn't count against your two minutes.


MS. RICE: I guess I could have stood a little closer, but I didn't expect all those other folks to not be here.

So hi, everyone. My name is Cynthia Rice. I'm with the Juvenile Diabetes Research Foundation. As many of you know, JDRF was founded 40 years ago by parents of children with diabetes to fund a cure and better treatments for the disease. JDRF has worked with many of the agencies here today to improve diabetes care and accelerate an artificial pancreas.

I want to start out by saying that -- and I expressed this to Jonathan and to Jeff Shuren when I first came in -- that we really do appreciate the focus of today's meeting. There is a real need for innovation in diabetes devices. You know, diabetes is obviously a huge public health need, as Dr. Klonoff discussed earlier, 24 million Americans. It's a leading cause of heart attacks, blindness, all sorts of amputations, coma, death. All of these are public health problems that could be prevented with better diabetes devices.

I would also add that people who live every day with Type I diabetes are using their devices year in -- you know, hour to hour, day to day, year to year, and so they're spending a lifetime with these devices, and so they're critically important to their daily life.

I want to briefly in this short time highlight three areas in regulatory, in research, and in reimbursement in which federal agencies can help improve and overcome barriers to innovation. We've spent a lot of time in the last couple of years working on artificial pancreas technologies and working with many agencies and outside organizations, such as private PAIRs, on issues related to continuous glucose monitors and the artificial pancreas.

Quickly, three points, one on research: Proof of concept trials are essential and are often underfunded. This has come up earlier today, that proof of concept trials in devices is an important area that the government hasn't funded as much of as they may have in bio. And, you know, we encourage agencies to fund more of this type of research. JDRF, as a private non-profit funder, is funding in this area. We have several successful areas that are already going on, though. NIH is funding in the diabetes device space with the artificial pancreas project, and we really appreciate that.

The Department of Defense, through the technologies for metabolic monitoring program, has done some important work that could not only help people with diabetes but is really focused on helping anyone with high glucose in the intensive care unit. And that's very helpful.

So we applaud those. But, in general, that kind of funding for proof of concept research is something we'd encourage more of.

On regulatory, we've been, you know, having a sort of five or six-year-long dialogue with the FDA, which we really appreciate. There is a lot of important work that the FDA is doing in this area, and we appreciate their help. I think the big picture comment is just that it's critical to the proof of concept trials that the regulatory system encourages them. They shouldn't be held to the same standard as trials conducted for pre-market approval, and figuring out, you know, what counts for proof of concept trials and what doesn't can be a time-consuming process. So to the extent that there is regulatory flexibility and clarity around that, I think that will help advance proof of concept trials and encourage more groups to fund them.

And, finally, on reimbursement, for devices to improve the health system, patients need to have access to them. Several speakers have spoken about that. I want to comment just briefly on billing codes, otherwise known as HCPCS codes. And CMS obviously runs that process with involvement from private payers. It's critically important.

One thing that I never knew until I got involved in this area is that they only make decisions once a year. And with continuous glucose monitors, the first company got approval in mid-2005. And by the time the applications were submitted and the first one was denied and the second one was submitted, the billing codes were actually not available to the healthcare system until January of 2008. And without the billing codes, it's very difficult for health plans who want to cover it to actually systematically process claims.

So I don't know all the complications involved in doing it, but I think it would be helpful if the process reviewed applications more often than once a year. I think that would help create the opportunity for access to patients of some new devices.

Thank you very much.

DR. SACKNER-BERNSTEIN: Thank you. Zen Chu?

MR. CHU: So my name is Zen Chu, and I come here representing the Wyss Institute at Harvard Medical School. Most of you will have never heard of that because we're a relatively new institute at Harvard. And it's an institute for biologically inspired engineering. It's a result of the largest donation in Harvard's history to focus specifically on a mission of translating innovation in partnership with clinicians and corporations as the enlightened wisdom to how to get new innovations to make a real impact. So that's the perspective I bring as well as having been an entrepreneur and a venture investor.

But our institute is focused squarely in this innovation gap, in this valley of death that many have mentioned today. And our institute is trying to change or add to a culture of innovation, and specifically, at the right place in the innovation life cycle, we feel, where kind of truth and efficacy are aligned for everybody because companies or venture investors are focused on testing and failing fast so that you don't invest on unneeded resources.

And so one of the most common challenges that we see over and over again as we try to either take a new technology and then match that technology to an unmet clinical need is that there are separate domains. This is interdisciplinary research that's needed with insights from both engineers and clinicians and payers for what should be the right endpoints. And, often, these engineers look where to apply, and their success of the technology is determined on where they pick in terms of the first clinical foothold, and oftentimes they don't pick correctly because it's not their domain.

And so this is also the challenge of every investor and entrepreneur and -- that we all struggle in that space of matching the right technology to the right clinical need.

And so I would propose that -- and I think there is some interest from the epidemiologists and public health folks -- to create a Wikidatabase, an open, transparent Wikidatabase of clinical needs. And you can imagine kind of a master table, if you will, that would be very unwieldy, but you would break this down into views of different indications, different clinical needs on one axis, and then the prevalence and the incidence, whether it's in the U.S. or the rest of the world, but you match that also with the comorbidities, with the right patient sub-segments, because so often, finding the right clinical need is really a challenge of sub-segmenting the right patients and comorbidities and surgical approach and how that fits within the current standard of care.

So it would be a public good that you could build and essentially codify knowledge, where communities of interest would come and build out some of this data and add in. And there will be inaccuracies and there will raise other issues. Many companies maybe wouldn't want to participate, but others would to highlight some of these clinical needs. And there would be kind of debates flushed out just as there are in Wikipedia.

But I think that would be a good forum and a good debate to try to highlight some of this, because then and only then, once a database like this existed -- and in contrast to what the CDC has and some others that are the start of this and could seed a database like this, but it isn't there yet -- would be to -- it'd be a worthy experiment to try to create a resource that these folks could access, could test some ideas on within a community of interest and get immediate feedback on.

So we'd be very interested in helping try to create this both across Harvard Medical School and the School of Public Health, but we would need to partner with other institutes, and I'd be very open to meeting anyone that's interested in that.

Thank you.

DR. SACKNER-BERNSTEIN: Thank you. Jose Bolanos?

(No response.)

DR. SACKNER-BERNSTEIN: Konstantin Goranov?

MR. GORANOV: Thank you for the opportunity to express my opinion. I'm co-founder of SALUTARIS. This is a startup company. We're involved in the development of new technologies with -- to cellular effects.

We have several products already sold out in Asia, Canada, Europe. Unfortunately, we had a hard time to approve these products here in United States. The 510(k) application we put about a couple years ago ended up in withdrawal, and I think the system definitely needs some improvements.

In my opinion, a 510(k) application is definitely a great vehicle to approve products, so-called me too products; innovative products had to go through a different route. There are a lot of standards we have to comply with, and as an inventor and person intimately involved with the 510(k) application, the standards will kill innovation. It's not possible for any new device or new technology to be proven with the effectiveness and safety with these standards.

We need to have a slightly different opinion -- very much are we aware of the importance of safety and effectiveness. However, we need a little bit more collaboration and more understanding from the regulatory body.

With these two minutes, I don't think we'll be able to express too much of the right structure of the approach. I'll say three important issues may be suggested, then, to consider to improve the process. First, the whole process of selecting new innovations; several opinions already expressed today. We can select a series of important products, important developments, through a special selecting process. And we probably need the collaboration between all of the stakeholders in this meeting here and define maybe somehow a staged review process, something that would put these innovations in a coherent manner to the next level of review. It's very difficult to find all the benefits of a new device or a new technology by simple review the whole package of evidences in one big set. And, again, I think the collaboration is the key for success.

Thank you for my opinion.

DR. SACKNER-BERNSTEIN: Thank you. I'm wondering if there are questions that any of the Council members have for the presentations so far?

(No response.)

DR. SACKNER-BERNSTEIN: Okay. Then we'll move on. Richard Eaton. Sorry. I should have announced --

MR. EATON: Good afternoon. First, I want to thank the Council on Innovation and FDA for holding this meeting on innovation today. It is an absolutely critical ingredient to new medical products and to bringing lifesaving healthcare benefits to American patients.

I'd like to talk about two things, the lack of regulatory predictability and how can barriers be removed.

First, I am from Medical Imaging and Technology Alliance, MITA. We're in Arlington, Virginia. We're the largest trade association which represents medical imaging products. We are also the leader in the development, deployment, and testing of the digital imaging and communications in medicine, the DICOM standard, which is universally recognized for its interoperability capabilities and communicating images and imaging information.

While there are many barriers to innovation, I think a lot of them can be subsumed under the rubric of lack of regulatory predictability. And why is this important? Well, first, what does predictability mean? It means being able to know what the regulatory requirements are, and that is before an applicant files an application, how long will it take to meet these requirements to bring my product to market, FDA status updates on the progress of a company's application, predictable cycle times, and how and when will I know when requirements change? We do know one thing, that the lack of predictability stops innovation dead in its tracks, so it is probably the most significant barrier.

So how can we remove this barrier? First thing is to establish predictability, and how do we do that? Ongoing, clear FDA industry communication on what the regulatory requirements are. Again, an applicant should know before an application is filed what the requirements are. Second, creating communication performance objectives and providing ongoing, regular status updates by FDA on pending applications. Now, there are FDA processing performance goals, but we think there should be an analogous creation of communication performance objectives so that applicants know on a regular, frequent, and ongoing basis what the status of their applications are. There should be adequate notification of changes in regulatory requirements so companies can prepare applications in a timely way, make adjustments, and marshal resources.

Again, MITA is willing and able to work with FDA and the Council on Innovation to help remove these barriers. And we thank you very much.

DR. SACKNER-BERNSTEIN: Thank you. Dan Cohen?

MR. COHEN: Thank you. I've been involved in the medical device community for many years, particularly with large-scale devices, but now to follow up on the comments of John Lasersohn and the venture capitalist community, I'm with an early-stage medical device company. And we do live in the world of living and dying on the venture capitalist funding, and the challenges to innovation come in the regulatory process. Within that process, it's clear that systemic change in the regulatory process would do the most to benefit the device approval process and to spur innovation.

The control that the FDA has under its congressional mandate over the regulatory approvals of a device has led to significant costs, particularly in the area of device lag, our ability to get devices to the market in a timely manner, and device loss, promising devices that are never commercialized due to the regulatory cost.

The challenge that we have is to suggest multiple options to bring balance back to the approval process for devices that more effectively includes physicians, patients, payers, and manufacturers with the regulatory approval agencies.

In an effort to comply with congressional mandates, FDA has defaulted to an approval system that results in unnecessary and unreasonable delays and denials because of a hyper-focus, primarily on efficacy, that harms patients' ability to reach -- have the devices they need to reach the marketplace.

We would suggest that it would be more desirable to move towards a system in which private firms, organizations, and other governments, along with a voluntary FDA approach, assured consumers of medical device safety and efficacy. Consumers want quality and safety in their device and, moreover, prior to purchase and use, they want the assurance of quality and safety.

There are three broad approaches that overlay any regulatory scheme that would provide quality and assurance that need to work in tandem: voluntary practices at institutions assure quality and safety because it's profitable to satisfy the consumer and to maintain corporate brand; tort remedy to protect consumers from fraud, false representation, breach of warranty and the like; and governmental imposed restrictions, where governments attempt to determine the quality and safety of a device and prohibit sale until it's been given permission.

We can't underestimate reputation in the process. A seller's trading partners and customers develop their opinions of their worthiness in the process. But historians of private professional regulation have shown that professional men and women built institutions to research, test, certify, and monitor device safety and efficacy even before the FDA existed. Such efforts often led rather than followed government efforts to assure safety and efficacy.

So what can we do? Third-party certification and review. Under the European Union, they have many strict control standards for medical devices, including extensive reporting. But they do so with very little government involvement. Low-risk products have a self-certification system. Higher risk products can be marketed after notified bodies have conducted tests and approval. Those notified bodies are indeed regulated by the EU standards.

Congress opened up opportunities for limited use of third-party certification, but since 1997 and the Modernization Act, the FDA has limited that to the relatively simple Class I devices. An expanded program along the European model would help spur innovation.

In that same line, international reciprocity in the device approval process would be useful. The U.S. government should establish reciprocity with countries that have proven records of approving safe devices, including most Western European nations, Canada, Japan, and Australia. Such an arrangement could reduce delays and eliminate duplication and wasted resources.

International reciprocity would eliminate a monopoly that the FDA has on device approvals and under which a system could be devised where an IDE equivalent would be submitted to the U.S. authorities that would provide safety and quality assurance, thereby gaining U.S. approval. But FDA would have to compete for its business with other approval agencies -- a concept.

A focus on proof of performance. A splendid reform for the medical device product market would be to adopt the CE mark standard for an efficacy requirement for device approval. The principal standard for CE approval requires a proof of safety and a proof that the device will perform as it is designed to perform. That would leave the cost/benefit analysis of efficacy to compare the effectives where it should rest, with the physician, the consumer, and the payer decision-makers.

Another option would be voluntary FDA "approval." And I put quotes around the word approval. To consider a voluntary certification by the FDA if doctors and patients wish to be assured of quality and safety apart from the FDA certification, they would be free to purchase devices that were clearly labeled "not FDA approved," somewhat akin and analogous to the off-label usage in the drug market.

Lastly -- thank you -- why would this work? An example could be found in other industries. In the electronics industry, Underwriters Laboratory provides the assurance of quality and safety. Many people in this room and including many of the Panel members would consult Consumer Reports before making a significant purchase.

The challenge to the U.S. healthcare system is how quickly and how efficiently it can bring safe and high quality medical devices to the market in a cost effective regulatory model that offers hope for treatment for patients, disease, and disability. The global reach of healthcare has offered opportunities beyond our borders. The U.S. should rethink its current approval process to take advantage of existing systems and market forces for the betterment of patient care.

Thank you.


(No response.)


(No response.)


MR. ROSNER: Thank you for the opportunity to highlight one of our nation's major unmet problems, namely the transformation of cardiology through early detection of cardiac dysfunction other than the unreliable and over 100-year-old EKG. And here I'd like to suggest a real solution that is ready but urgently needs government and industry support to reduce barriers for its use and achieve cost avoidance of billions of dollars.

Population growth, aging and accelerating insurance costs have pushed medical costs to an unsustainable 18% of GDP. If allowed to continue escalating and unchecked, every sector of our country's economy will suffer enormously.

At this meeting, as this meeting recognizes, it's critical that hospitals, medical professional groups, regulatory agencies, FDA, and the insurance and commercial enterprises actively focus upon and support a range of innovative, imaginative, and effective new measure and devices that will cap and hopefully reduce these unsustainable levels of expenditures without dramatic impact on a high quality healthcare system.

Such approaches will urgently require major shifts from costly later-stage, invasive emergency medicine, demonstrably effective, to begin adopting large-scale emphasis on preventive medicine that is not very profitable yet, as well as early detection of major illnesses, suffered by almost everyone aging in the United States.

The largest example and most costly in lives and expenses is heart disease that claims tens of millions of victims annually and takes over $400 billion per year from the overall $2 trillion in U.S. healthcare expenditures. A large fraction of this is avoidable through illness prevention and early detection without radiation.

While exercise and dietary discipline as well as pharmacological approaches seem to be beneficial in reducing heart attacks, we need to make it possible to adopt even more effective, new, viable technological approaches, such as CardioMax device, that can safely and reliably confirm early detection of cardiac disease or, most importantly, also its absence.

We must find ways to overcome debilitating existing time and cost barriers, mainly regulatory and financial, that only large corporations can afford and surmount.

A key solution is magnetocardiography. Having been directly involved in the tremendously successful growth of MRI, I see many similarities of technology that involves testing for early stages of heart disease through utilization of magnetocardiography. Like MRIs, the transition from research to commercialization as well as the multi-billion dollar business of public health opportunities are possible now through cardiac screening of the general public, as proposed by the SHAPE task force of Texas. The MCG technology can be reliable, rapid, safe, and relatively inexpensive cardiac functional results, thereby avoiding many unnecessary, costly procedures.

In conclusion, I should repeat, to limit runaway expenses, U.S. medical landscape must migrate to prevention and early-stage diagnostics that can also be achieved through monitoring of heart disease progression. I trust this CDRH meeting will find and recognize such innovative ways that will both improve public health and avoid further healthcare cost escalation.

Thank you.

DR. SACKNER-BERNSTEIN: Thank you. Lisa Salberg?

MS. SALBERG:  -- Hypertrophic Cardiomyopathy Association, and I'm going to change up things for just one second. Is there anybody else in the room who has an implantable medical device? Me? So I come from a slightly different perspective here.

So I'm going to talk about genetically mediated diseases. Hypertrophic cardiomyopathy affects 1 in 500 people. We can't do anything to stop the disease. There is no prevention. It's genetic. And there are a lot of people with genetic disorders. And where we differ than some other populations is that our population is identified earlier, we're treated younger, and we're going to have a lifestyle that is very different than other people with other disease because we are going to have interventions for much longer periods of time.

So our unmet needs in this population are the fact that we need devices that last longer, that are going to be far more durable in a young, active, healthy person because we're going to need more devices over life. I am 42 years old. I'm on my fourth implantable device. I will have many more in my future. My daughter was implanted at age ten. This is just a segment of our population. It's not just HCM. It's other diseases as well. So we need leadless technology because leads offer more complications than generators, and we need these devices to be brought to market in a very sensible, rational, logical way.

Those of us within the cardiac industry know that the 510(k) process works in some cases and it doesn't work in others. There have just been the recent harm of a lot of patients because of a rush to market for some products. I might agree or disagree with that.

We need longer battery lives of generators. Each generator replacement has complications. That's where your problems are going to happen, not with the device itself. In most cases, it's at the replacement time. So if we had devices with longer batteries, we would do better.

We need trials that include our population and that we can be recognized in the medical literature with smaller trial numbers. We are never going to get 10,000 patients in a trial for a rare genetic disorder. We need the FDA to recognize these smaller populations and come up with mechanisms that recognize us in the literature.

Lastly, I have these four different areas that I wanted to talk about. But I think we might benefit in the area of innovation by looking at what's been done elsewhere in our country. The victim compensation fund that was created for vaccines might be a good model to be looked at for the implantable device market. What happens often in a device alert recall or advisory is patients are going to rush to their doctor and they're also going to rush to their lawyers. We don't want patients to have to go through that process to seek restitution for damages. So how we encourage innovation and discourage litigation is to create a system in which people could be made whole. And that would be one option.

Another issue that I would like to discuss -- and we will submit our full thoughts in writing because a lot of you have made me think a lot today, and you've got some really good ideas out there, but from the patient perspective, we're going to twist it a little bit and look at it from our point of view. But I do think it's important that the FDA and industry continue to work together but make sure that the lines are very clear, because in the past, we've seen some blurring, and it's unclear as to who is actually leading and who is actually following. And as patients, we'd like to see that there is very clear oversight while we maintain innovation.

Thank you.

DR. SACKNER-BERNSTEIN: Thank you. Peg Graham?

MS. GRAHAM: Good afternoon. I'm an inventor and an entrepreneur, and I've been bootstrapping the commercialization of my invention for the past five or six years. And for those of you that have ever bootstrapped, you know that I'm pretty close to impoverishing myself trying to get this device seriously taken by the industry.

The concept for the device arose from my family's experience caring for my mother, who aged at home with a heart condition. For the Council, I would ask you to please add DME to your list of things to consider as you think about further collaboration between and among the agencies. I think it plays a unique, important role in supporting the decision to age at home, and if it's left out of the thought process, I think we're hurting the national interest in having people age at home.

I think it also plays a unique role in providing an opportunity to integrate health information technology, data capture, recording, and transmission into an EHR. These devices are -- I know they fill catalogues, now people's eyes roll in their heads, I know that DME is the stepchild of medical device innovation, but don't forget, it's everywhere. It's existing. There are channels for distributing, testing, training. It's all there. And if we could just figure out a way to get the HI tech folks talking to the DME folks, then I think that we could advance innovation in quite unique ways.

The other thing I would ask the group to consider, and this is particularly for the CMS, is the least costly alternative policy. Dumber is always cheaper than smarter, and if you're not looking at the externalities created by smart DME, fewer patients falls. Think about that. $19,000 for each episode of additional -- $19,000 of additional healthcare costs for each patient fall that's not avoided. Institutionalization is avoided because now somebody can engage in an activity of daily living independently. They don't need to go to a nursing home. They don't need to have the personal care assistant there, or if they do, they don't need that person to be there on demand. You can schedule certain DME -- certain ADL activities, I believe.

The other thing I would ask the group to do is to integrate the NIH into the pre-IDE process. Currently talking to Arizona State University about doing a clinical trial of this particular device out there in the Phoenix area. And I know we're talking about statistical power with Dr. Lyndon Joseph and the NIA, but that is going to impact whether or not CMS and FDA believe that what we did with the NIH study meets their needs. So I think that, you know, it can't just be the two agencies.

Having worked in the research community, think about the IRBs. I can tell you, having sat in on some of those meetings, sometimes what happens is IRB review falls into the game of who is the smartest person in the room and whose statistical knowledge can kill an idea. So please think about that.

And, finally, in addition to the U.S. patent that was granted, I have foreign applications, so again, think about DME innovation in that regard.

Thank you.

DR. SACKNER-BERNSTEIN: Thank you. Roland Winger?

(No response.)


(No response.)


(No response.)

DR. SACKNER-BERNSTEIN: Dennis Dannemiller?

(No response.)

DR. SACKNER-BERNSTEIN: Richard Longland?

(No response.)


MR. DRUCKER: Thank you for conducting this Panel on innovation in healthcare and what we think is a critical aspect of the future of healthcare and innovation for healthcare.

My name is Mark Drucker. I'm with a company in Boston called NewCare Medical. Our focus is on innovating for underserved communities, not only underserved patient communities but also underserved caregivers.

So our comment to the Panel would be that we hope you will increase your support from all these community members for innovations that touch the millions of people in the U.S. and worldwide who require care and who are non-physician caregivers. This includes health disparate communities, but also all those people along the continuum of care that includes nurses, med techs, family members, and spouses.

We hope that the Panel continues to identify and increase its support for these significant opportunities to innovate outside the traditional box.

Thank you.

DR. SACKNER-BERNSTEIN: Thank you. Mark Adlinger.

MR. ADLINGER: I thank the FDA and this Council for hosting this unique workshop and forum and hope that the dialogue continues.

I'm a former entrepreneur. I'd like to say that I am a recovering entrepreneur and presently principal engineer at a medical device business incubator and contract manufacturing and engineering firm in the Boston area.

I propose the promotion of non-profit companies and incubators for the development and distribution of devices that would not otherwise meet the full qualification of venture or for-profit funding but certainly fulfill healthcare needs.

There are a small number of non-profits existing for this purpose in the device and pharma areas. I believe that they are undervalued and underutilized. NIH, DoD, DARPA, and other agencies can greatly help in the funding of new initiatives and supplement and be supplemented by healthcare-related foundations.

This is no slight to the venture-capitalized companies. Certainly, I have been and my company have been the beneficiaries of such funding. But a device or modification of an existing device with a small market application or ROI and the profit motive model never allows for the initiation of the development. If the public good and public health and well-being is our intention, then we should take a more dynamic view of the business models that enable new or old innovative devices and therapies.

Considering the hundreds of billions of dollars invested in novel and creative financial instruments in our country, perhaps we can apply some novel and creative means of creating new medical instruments.

Thank you.

DR. SACKNER-BERNSTEIN: Thank you. Are there any questions from the Council to any of these presentations, or presenters I should say?

(No response.)

DR. SACKNER-BERNSTEIN: Okay. Is there anyone else in the audience who would like to speak?

(No response.)

DR. SACKNER-BERNSTEIN: I think what we're going to go ahead and do is move on to the presentations that are scheduled to start at 2:45 because it's going to take me a couple of minutes to get this ready. The first speaker is going to be participating remotely, so I just need a minute or two.


MS. NORMAN:  -- my apologies for not being there in person, but I think this will somewhat illustrate some of the barriers we're dealing with in delivering healthcare via non-traditional methods.

So, Jonathan, if you'll advance to slide 2. Why the focus on home health? Why are we talking about that separately today? Intel uses what we call the shift left philosophy, which applies to both technology in the home or mobile technology, different ways of delivering healthcare.

The further we get from the home, the lower the quality of life. So when we move from health and wellness to aging at home to moving to residential care, independent living, nursing home, and finally acute care, our quality of life goes down and the cost of care goes up.

We're focused, as is a large number of people in the industry, on moving healthcare to the point where we can keep quality of life higher, we can age in place gracefully, and we can keep the costs down so that the quality of care that we receive also is higher.

So, Jon, next slide. Three current barriers to innovation in the home health market, and remember, when I say home health, we're talking about not just technology in the home, but technology, basically, that's being used by lay users, things that were traditionally in clinicians' offices, delivered by clinicians.

We have three areas I want to talk about today, the first one being the market viability. We're asking for some changes in business models, and I'm challenging Food and Drug to give us a hand with opening up the acceptance of home health. Lack of standardized guidance for the industry, talk about that more in depth. And then changes that we see as essential to the current reimbursement policy to enable innovation.

So let's talk about market viability. This one really came home to me about two weeks ago, and I had tried to get some of the video released so we could include it in this, and I wasn't able to, but I can still tell you the story. Intel has -- one of our products is a remote telehealth monitoring system that connects clinicians and patients. We had about 300 units out for a business pilot with a very large payer. The pilot concluded excellent results, studies are going to be published, everybody loved it, but the payer didn't have the money in their budget for this type of capital investment this year. And because it's not being reimbursed, they just didn't see a way to go forward until some point in the future.

The flip side is a man named Jerry, diabetic, confined to a wheelchair; he loved the health guide. He had used it and watched his vital signs improve, things were back on track, and we had to take it out of the house. And it was heart wrenching. He offered -- his family members offered to pay for it directly, but because we don't have it set up for an over-the-counter use, only with use with a clinician present, there was no way to leave it with him. And it was devastating for a lot of Intel employees to see what happens when the business model at our potential customers doesn't support home health. And that's the very real face of it for us.

So right now, care in the home and mobile care, there are new models of care and there are also new business models. So as I was saying, companies don't necessarily have budgets set aside for this, for the capital investment, and what we found is we spend a great deal of time educating on the benefits of. Problem is, is that the benefits of telehealth, things like there are hundreds of studies published, for instance, on things like reduced readmissions to the hospital. There are a number of claims, telehealth, that have been proven in general, not specific to any one device.

Current process for getting products cleared for the market includes having all of your claims reviewed and cleared and having adequate data to support. There is some data that is true for any general telehealth device. This is good news that FDA could be sharing with the industry. Implementing a complete, end-to-end telehealth system, for instance, will result in these outcomes according to these studies. These are the types of features you should look for, just as an example.

And so what I'm asking here, the ask on this page, one of the two, is that if we did a study and pulled out what telehealth brings -- Food and Drug drives that -- that we could make some general claims and help the market itself start to understand the benefit not only of improving quality of life, lowering their costs as we bring healthcare closer to the home. And that's something that needs to be driven through the Agency.

Then the other thing that can help that goes along with this is do more public sessions for clinicians to adopt home-based technologies, using focus groups, publishing the research, convening workshops on new work flow models, et cetera.

So market viability is a huge issue, and until we solve that problem, home healthcare will not advance much further than it currently is. It'll continue at a very slow pace.

Next slide, please, Jon. Talking about standardized guidance. Just to back up a little, when we design a product for the home, we have an increased responsibility. So these are lay users. They're not used to using medical devices. Many are technology averse. Many don't have a PC in the home. The remote control for their television is a challenge. They're also now dealing potentially with a new health condition. They're stressed. They have this coming in the home. User interfaces is extremely important.

Also, what we deal with in these different environments, they're not predictable. I can tell you what I'll normally find in a hospital. I can't tell you if the wiring has been redone in somebody's home by their son-in-law, who really didn't know what he was doing, if they have two-prong-plug outlets or three-prong-plug outlets for grounding, if there's other devices around that are emitting -- we don't know. So that's another variable we have to bring into the equation.

And, lastly, the important point is, is why we have a special responsibility. There is no clinician available to intervene if an adverse event occurs.

So what we need to be able to do is look at what typical actions are. Do they believe that this home health device is a 911 system? What would be the adverse event that could occur from that? It would be pretty big if they were having a heart attack trying to use it to call for help. So have we adequately addressed informing the user of intended use, what to expect, what to do in different situations, and have we done it in a way that they can understand and remember.

So next slide, please? Safety, efficiency, consistency. We need standardized guidance. We need, number one, always to protect the user in the home. It's not enough to say we've checked the manual and everything in it is correct. What we need to know is, is the font large enough that an 89-year-old woman with limited eyesight is able to read the manual? Can she understand it? If it's her first time using technology, the word menu, select from the menu, is not going to make sense. So have we designed the manual in such a way and the user interface that it's intuitive and it's a tool rather than a new barrier and challenge for users in the home?

This is great that we're saying this, but there's a lot of companies that can't do that level of work. We need to define what are the different user groups, how do you test against those, what are the variable environments that, at a minimum, you should test that the device can handle. There is always going to be a one-off situation, but depending upon the risk for the device that we're dealing with, are we adequately addressing the environment where it's going to be placed.

All of this leads to how does this drive innovation. It becomes a more predictable process for the manufacturer. We now know what types of tests for what types of users we need to run. This can be updated as more is learned about the market, but it means we're all playing from the same-level playing field at that point. I know there's a number of activities starting, looking at how we provide guidance in the home. I am asking FDA to encourage that, participate, and do whatever they can to help drive this, provide information. And I do know personally that FDA has been involved in this area. Please continue.

Next slide. Reimbursement. Better define the market players. We have -- there is so much going on. Every time I turn around, there is a new iPhone application, there is a new place to store electronic records online where I can upload directly from a blood pressure cuff, and there are new telehealth devices. And they all have a slightly different flavor to them. Some are regulated. Some are not. If it's data capture, where it is a regulated blood pressure that merely uploads data to the PC, the PC is not controlling the interaction. It's only receiving data. That's a data capture device, and the PC is not part of a regulated system at that point -- versus a complete solution, where we have customized care protocols that direct medications, alert a clinician when vital signs are drifting out of bounds. We need to look at who the market players are and then use the 510(k) or the PMA processes to qualify devices for reimbursement. So we need clear definition in this area.

Next slide, continuing this. So we're asking for one-stop shopping. If FDA and CMS can sit down at the table together and talk about how a manufacturer can apply for reimbursement for their device, get a code, et cetera, without going through numerous processes -- we finish one, we start another. And it seems that we should be able to combine these into one review package, putting the right people at the table and making it a smoother process for everybody. If we are determining products are substantially equivalent, it's a good starting point. Probably be just a little bit more in the 510(k) package itself.

Next slide. Reimburse for broadband. This isn't Food and Drug's call. I understand that, but we're talking about where the barriers -- the National Broadband Plan, we need to support that and support it loudly and actively. We need to reimburse for broadband. The people that need it the most are the people least likely to have it. Seniors and disabled populations, we have found, are least likely to have broadband connectivity. We've gone in the home to connect home health devices, and we find rabbit ears -- well, this is in the past. It's updated now, but people do not have the connections. And there is no way to connect the devices. So plain old telephone lines work, but a lot of the media-rich content, which helps patients who are not used to technology, helps them to navigate the user interface, is lost. Broadband brings us the media-rich interface. It brings us a way to upload data quickly and accurately. It's easier to maintain the integrity of the data. So what we're asking is that installation costs and the monthly broadband fee be added to reimbursement packages when used as part of a home health system.

The last slide. We do have an opportunity. Home health is where we need to be pushing this industry. There will always be a need for hospitals. There will always be a need for the clinicians' offices, but the longer we can age in place, the higher our quality of life will be, and now is the time to get that started. We need help from Food and Drug, one, to help with the market viability, to help the industry understand the benefits of telehealth and health in the home. We need help with reimbursement to give us a level playing -- and we need standardized guidance for a level playing field.

So the last slide is my contact information. Please, any time, I am passionate about how we bring healthcare in the home in a way that it is a tool and not a barrier and have a user interface that works for Jerry, the diabetic, and my mother who is 89, and she is always the user I think about. And that's it.

Thank you.


DR. SACKNER-BERNSTEIN: Even though she can't hear us clapping, I'm sure she would appreciate that.

We're not going to really be able to ask any questions, but if there's comments in response by the Council?

MS. SLUTSKY: I would have liked to have asked her this question. On one of her slides, she talked about addressing the reluctance, you know, of clinicians to accept proven technology. And, you know, she described, you know, a series of activities, including focus groups and training. And I was curious as to whose responsibility she thought that might be. As a funding agency, we've actually funded grants and contracts in this area to increase adoption. But in this context, she was, you know, talking about, you know, almost a detailing of their innovation. So it would be nice if at another time we could follow up with her about what she was thinking about that.

DR. CROSS: Jonathan, I wish you could put that last slide up. I would love to have her phone number. But she's asking for information on effectiveness related to home monitoring, and I think we've got a boatload of it, and we're very interested in that, and I think we'd be very happy to share that with her.

DR. SACKNER-BERNSTEIN: I have the contact information. I'll forward it to you.

So with that, we can save up all of our questions and cross-examination skills for the next couple speakers.

Joe Smith?

DR. SMITH: I want to thank Dr. Sharfstein and the good Mr. Shuren. Thank you so much for allowing me to come here today. I am Joe Smith. I'm the Chief Medical Officer for the West Wireless Health Institute. I could have easily titled this slide "Cost as an Imperative for Innovation," and I think I tried to do that once, but clearly didn't get it done. And so that'll be about the flavor of my discussion.

I'm going to take you through a brief introduction to what we're doing against the backdrop of what is clearly a crisis of cost in U.S. healthcare, the imperative for change that that provides, the notion that we have a ubiquitous wireless technology distributed around the world -- it's estimated that there are on the verge of being 5 billion cell phones. And if you imagine that for a moment, you might say there are more people who have access to a cellular service than have access to flushed toilets. And so why wouldn't we take that broadly distributed utility and link it closely with healthcare?

And then, finally, what will it take to make an infrastructure-independent care model work for healthcare? And what I mean by that is we are currently engaged in a model of healthcare where the sickest among us find their ways to the centers of care, being hospitals and doctors and clinics, and isn't that a bit backwards? Wouldn't we like to have the sickest and most infirm among us get care where they need it, when they need it?

So the West Wireless Health Institute was started a year ago by two visionary and successful entrepreneurs, Gary and Mary West. They have so far funded it with $100 million, with the specific purpose, a very simple and specific purpose, and that is to lower healthcare costs. I think it's a novel organization in that regard, and it's one of the things that attracted me to it.

There are core functions of the institute, and that is to, certainly to innovate, not just on our own, but partnering externally, creating partners and incubation of novel technology through proof of concept, all against the backdrop of lowering healthcare costs.

We have to validate those technologies, either our own or others, by running the appropriate trials to validate that there is, in fact, cost savings, and not just cost savings to some intermediary in the healthcare chain but aggregate cost savings. And then also to be present for encouraging an enlightened public policy that will allow this kind of behavior to take place. And then finally, to invest, as we heard from, I think, Zen Chu earlier. There is certainly a valley of death that's been well described between great ideas and their ultimate impact, as risk capital, venture capital finds it difficult to engage in technologies when they see regulatory and reimbursement disclarity. They get a little concerned. And so we try to fill some of that gap, although obviously the gap is bigger than anyone's finite resources.

So I want to spend a minute getting us all familiar with what's happening in costs. We as a population are getting grayer. And those of us who get gray actually get clear up here after a little while in terms of balding. So we are as a population graying. And if you look at the growth rates of the aging population, you see that they outstrip the growth rates of our economy. And so we are getting older a lot faster than we are getting richer, and that matters because as we get older, we consume more healthcare, a disproportionate amount of healthcare as we age. And so as the population ages, you would like to have the dollars available for the incremental burden of healthcare costs in that aging population. And yet what's happening to all of us is that the number of workers available per person who has now reached an age of either retirement or decreased productivity or infirmity, that ratio is declining. There are fewer of us working to support the healthcare needs of those who are now walking into those timeframes in their life when they are going to need more and more care.

You know, we talk about the fact that we're spending 18% of GDP on healthcare. We argue that in the very near term, it's going to be $1 of every 5. But if you ask the Congressional Budget Office to project forward, their most optimistic estimate is that healthcare costs as a fraction of GDP are going to grow continuously, unabatedly until 2082. Now, I'm not going to be here in 2082. Congratulations to those of you who are. But you will be inheriting a world that I think is -- it is already at the moment difficult to imagine sustaining this for a few more years. And yet if we do as we're doing, we will walk into a system that is completely unrecognizable to any of us. And I point that this is one of the most optimistic of the projections. There was one, I think, that said that in 2082, we would be spending over 99% of GDP on healthcare, and obviously, that's fallacious. We wouldn't have a GDP if that's in fact what we're doing.

So one would like to believe that there is a nice relationship between dollars spent in healthcare and the quality that one achieves. It's been difficult to demonstrate that. I think in a prior life, if I presented this as a chart showing a relationship between two variables to my colleagues at the FDA, they would look at me with some enhanced skepticism. So I don't think we have evidence of a clear relationship between the money we spend and the quality of healthcare we get. And the money that we spend is growing remarkably, monotonically, and we already exceed any rational estimate of healthcare spent by any other country on the planet.

So I believe what has heretofore been difficult to describe in a regulatory setting about considering costs, you know, I believe that -- I have been frequently told, at any rate, that regulatory activities often are prohibited from imagining that cost is a variable for approval or for reimbursement. But I think it is clearly an imperative for innovation. We owe it to ourselves, to our children, to our patients, to our systems to innovate around cost as much as we innovate around other unmet needs. Cost is an unmet need. It's the largest one in front of us at the moment.

If you didn't believe that cost was an issue, I think you would have to believe that getting to a doctor is an issue. When one looks at the fact that we have just swept into access for healthcare another 20 or 30 million Americans into a system where we have not added one additional caregiver and that, at the same time, we are struggling to take our brightest and best and have them get enamored with primary care, we're realizing that we are facing a doctor shortage and that that doctor shortage is only going to grow going forward. That aging population that needs more healthcare, as it turns out, some of our doctors are aging as well, and they would like to work fewer hours or in less urgent care settings. And so what we're left with is a burgeoning amount of need for healthcare that is nowhere near met by our resources, either our dollar resources or our facility resources. And all of this, I think, points to changing the way we deliver care.

When one looks at where the dollars are spent, it is clearly in chronic disease. Adding these up, we spend about 75% of every healthcare dollar in chronic disease, mostly around the acute exacerbations, those costly collisions of people with chronic disease with our infrastructure. And so that, in fact, is the opportunity for us, is to manage or diminish those costly collisions and replace them instead with modest course corrections with technology that can reside where people are when they need it so that we can turn the need for those costly exacerbations into prevention and prediction.

Oops. I had a really good slide there. There, it's showing up. Look at that. It's magic.

So what we're talking about, really, is infrastructure-independent care, moving from where we are at the moment, which is relatively low frequency events, where you as a patient can be seen by a clinical caregiver, and they get a snapshot at that moment in time of how you are and what you're doing -- look at that. I know just who helped me with this slide, too. This is exciting. Okay. I'm not going to be able to show you that at any --

Okay. Real quick. Imagine it. The acute care-focused, appointment-driven, location-centric, and high cost. We're going to try to move to one which is high touch. Not high touch with your doctor, but high touch in the sense of being continuously monitored, the right treatment -- isn't that funny? God, Jen is going to kill herself -- when they need it, where they need it, and at lower costs.

So imagine for the moment that what that is, is you are where you are, and there are sensors in, on, or near you that are tracking activity of your disease, both your condition, your activity, how you're doing, how you're getting along, but also actionable metrics of your specific disease activity. And then you get provided information about how you're doing, you, your family, the clinical caregiver who is most responsible to you. Oftentimes, that's an alpha daughter, one of those 40 million women in the United States who step up and help to take care of an aging parent or in-law while they're also trying to manage their spouse and their children. But giving them the tools so that you know that, oh, by the way, you're doing a little less well today than you did before. Let's try to make some small course correction -- as opposed to sending you off on one day, seeing you back six weeks later not because your disease has some six-week time constant but just because my office calendar has a six-week opening. And imagine that you're just going to be on cruise control and make it the whole way. It doesn't usually work, and instead what we wind up are those expensive exacerbations of chronic disease.

Wouldn't it be better to have actionable, continuous diagnostics, smart systems that can keep track of how you're doing, exception-based reporting to your caregiver so that they needn't look at every piece of data because that's not useful information, but instead, look at the actionable bits of information, when you are clearly falling off some cliff slowly, to rescue you at that moment from home as opposed to waiting until you actually collide and then see you sick in the hospital.

And so the path to this, we believe, is the appropriate development and deployment of relatively smart sensors of specific disease activity, and then you integrate that with titratable therapies in a home setting or under clinical care direction so that we can do intermittent course correction.

There are historical precedents for this. A long time ago I was a cardiac electrophysiologist, and pacemakers and defibrillators were developed not because they could be developed but instead because they needed to be developed, because this was a circumstance where the underlying disease, when it presents, doesn't afford you the opportunity to get to a hospital or a doctor. If your heart stops, you need it started right away. You can't make an appointment for that. And so you need therapy which can continuously monitor you and engage when you need that care.

And so you can imagine taking that model and extending it to other forms of chronic disease with acute exacerbations. We're on that way. You know, we heard from the Juvenile Diabetes Research Foundation. We're on that way with the artificial pancreas, and here we have a continuous sensor linked to titratable therapy, and that is -- you know, has been just around the corner for 30 years, and we're on the verge of having it.

And so this model does work. It's a great way of taking care of chronic disease, and it can broadly extend our limited infrastructure so that we can take care of people outside of that infrastructure.

I was at a meeting the other day and said, hey, do you know why people go to a hospital? And everyone was quiet. No, why do they -- because they have to. That's the reason you go to a hospital. And so wouldn't it be great to avoid that need because that's, in fact, where great costs and complexity of care get created.

And so in this industry, we do have issues. At the moment, we are, as a group, more focused on the cute technologies that we have rather than the integrated solutions, and we heard issues earlier today about adoption. Once you come up with these cute things, how do you get them used? Well, if it winds up being a cute technology that doesn't fully contemplate the solution and the business of medicine, then it's difficult to get it adopted. If it doesn't seamlessly integrate into workflow for physicians and nurses, then it is very difficult to get it adopted.

I had a colleague of mine at Partners in Boston, and he said, well, if there is zero incremental reimbursement, then there needs to be zero incremental time because I'm already very busy. And so solutions that are going to be broadly deployed and leverage physician and clinician time really have to fit nicely into the existing workflow. And that was the point I was trying to make.

So I believe the solution here is device-based, closed-loop, feedback-controlled therapeutics. It aligns with our drive for personalized medicine. We would like not to all think that we're all just people with COPD or all people with asthma or all people with diabetes, that everyone is a little bit different. So we would like to have machines or therapies that learn from how we're doing and titrate and learn from us and do better.

There is certainly a societal trend for consumer-friendly care. Nobody loves going to a hospital. There is a growing need for infrastructure-independent care. We simply can't build hospitals everywhere we need to be. In fact, early this morning, I was testifying at the VA subcommittee, and that notion of our veterans living remotely doesn't afford an opportunity to put infrastructure where they all are. We have to go to them.

We do have the notion of chronic and constantly changing disease activity. If you have inflammatory bowel disease, you have a lot of it sometimes and you have very little of it other times. Why should you take a drug that's the same every day? Why wouldn't you take a modified amount of drug based on your disease activity that day? I don't put gas in my car every Tuesday. I put gas in my car when it runs out of gas. And so why should we do a better job of understanding how our cars work than how our bodies work?

We certainly have a pressure to decrease aggregate costs, and that I've given a fair bit of talk about.

We are clearly in the early stage of this particular industry. It's essential that we move forward because we are up against the backdrop of a essentially broken healthcare system to which we have just dramatically increased access.

There is an increased recognition with the ubiquitous cell phones and with everyone getting broadband at home, or almost everyone getting broadband at home. We can do better with that. But there is certainly a recognition of the opportunity, and we have been talking about it. You know, the economists in Business Week and Newsweek and everyone has been talking about when it's going to get here, and it isn't here yet and why isn't it here yet. There are important barriers, and those are things we can talk about.

We must solve reimbursement issues for this. It is not the case at the moment that despite some of these solutions being rather attractive, that they just magically get adopted. It has to be in the financial interest of the people who can determine their adoption in order for this to move forward. The VA CCHT study that was done was dramatic in the fact that by just applying relatively old telemedicine equipment, they were able to, in treating 43,000 people, decrease hospitalizations by 20 and at times as high as 50% in patients with chronic disease in rural areas. And so we can dramatically save our resources and save money, but we have to do it in a way where it is incentivized.

In the VA, because it's an encapsulated system, they can manage and control and envision the cost savings associated with it. In many of our healthcare systems, the silos are so non-communicative that it requires an investment on one side to save money on another and that makes sense not to the people who need to make the investment.

We do have a problem with risk capital because the money that it takes to get some of these novel technologies off the ground that can be part of these integrated solutions doesn't arrive, because venture capitalists, if there were some in this room, they would be sitting next to the door looking for an early exit. They're interested in making money and disclarity, this notion that I'm not sure what the regulatory path is, I'm not sure that this device is a cell phone or part of a medical device, and maybe it needs to be regulated.

When you talk to people at Qualcomm or at Intel or at any of the handset makers or the telecommunication industry, they are not terribly excited about that vagary of who will regulate me and for what and what are those hurdles, and I need to know them now if I'm going to make this investment now.

There are outstanding legal concerns. Privacy has been raised as an issue. I think if we can do banking over our cell phones -- I'm in the process of buying a house. If I can buy a house over my cell phone, I think I can transmit my EKG over a cell phone. I'm not so concerned about that issue.

Liability is an issue because if you've got wireless communications, somewhere there's large companies with very deep pockets, and as we've seen the plaintiff's bar, when they are interested in trying to reclaim for individuals, they will go to those companies with the deepest pockets. And so there is great hesitancy by the largest and most powerful and most broadly distributed companies to be so engaged because this business represents something that's important for Americans but a small part of the business of these large companies, and why would they extend their reach into a small niche area only to bear enormous liability?

Also, there is an issue about physicians. When they are the recipients of information that is chronically monitored and exception-reported, what happens if they get an alarm at 2:00 in the morning and they don't see their e-mail until 6:00 in the morning? What happened in those four hours to that patient, and are they liable then? Even if they perceive they're liable, it will blunt their adoption. And so we need to be quite clear about what these issues are, and I can imagine some solutions.

There is also the notion of interstate medical practice. You know, our medical practice in the United States is deeply balkanized. Every state determines whether or not you can practice. But if you're interested in doing this, as the VA has done, the VA physicians can practice across state lines, and that's one of the reasons they were able to do such remarkable cost savings. The average physician in the United States is not so fortunate. And so if your patients are traveling and you think you're doing a good job of taking care of them, you may enjoin some special liability if you make a decision about their care when they're not in state. So that needs to be addressed.

And we do have an entrenched healthcare system. If we are going to save money, that means some element of our healthcare system is going to get less. And anyone who perceives that they're in that segment will not rush to adopt this particular strategy. And so we need to have the difficult discussions of what part of our healthcare system are going to get smaller so that we can all do a better job?

So I joke with Jonathan at times about regulatory disclarity. He is quick to point out that that's not a word. And so we'll call it the absence of clarity at times. And this is not so much a bang as it is just the recognition that when technology moves, it is the case that it takes a while for our regulatory systems to be tailored to emerging technologies, and this is one of those cases.

There are some promising developments. This meeting, I think, is certainly one of them, and I'm delighted to be here, and I'm actually quite pleased that it's happening at all. I think it's a wonderful thing.

Coming up next month, we do have a joint meeting between the FCC and the FDA, and this really matters for this industry because when you think about sensors that are on the body measuring things, that sounds like medical devices. When you think about systems that make therapeutic decisions about patients either with or without a doctor involved, that sounds like a medical device. When you talk about a cell phone or a telecommunications link, to my way of thinking, that doesn't sound like a medical device, but those participants in this ecosystem need to hear that it isn't because in the absence, they imagine that it could be, and it blunts their enthusiasm. And so that'll be an excellent meeting to take place.

I did mention that the VA care coordination study was done, and hats off to a wonderful study. Adam Darkins did a great service, I think, to the industry and to patients. And there are other examples that I have provided here. You know, integrated care networks, accountable homes, those are strategies where I believe these technologies will be readily adopted. As we're going around and explaining what we're trying to do, we're getting a lot of uptake from payers and large healthcare systems. They are very interested in trying to drive down costs. We just need to be careful that we're not driving down the cost of some intermediate player which turns into augmenting someone's profit in the process. We need to drive down aggregate costs. And so we're really quite interested in trying to get that done.

So the critical success factors for us looking forward. Cost reduction while preserving and enhancing outcomes. We owe it, as we do this, to have outcome measures which are completely objective, rigorous, and that we can describe clearly.

Solutions as opposed to technologies. It won't be enough to have a cute thing if nobody will adopt it. And so you have to think it through all the way about who is going to use it and how and when and why.

Transparent and predictable regulation and reimbursement. It just couldn't be more clear that that's a vital part of this that will bring in the risk capital off the sidelines to innovate along these lines to make this vision more of a reality.

And then, finally, some notion of safe harbor. I think tort reform is one of the things we talk about but agree never to do, largely because so many of our lawmakers are attorneys and this just doesn't fit well with the current thinking. But I think the notion of safe harbor, perhaps, for those who adopt these technologies, so that we realize that we can go forward with something that's better than what we have, admitting that it isn't perfect, and then not torture ourselves over the residual imperfections but instead have a process of continuous improvement, realizing that we need to go forward.

That's all I had. Thank you very much.


DR. SMITH: Want to do Q and A?

DR. CROSS: First of all, thank you for your observations about the VA. I really appreciate those, and Adam Darkins and the work that he's done on home monitoring, home telehealth, and so forth. And using my wireless device, I have determined that disclarity is, in fact, in a dictionary.

DR. SMITH: If it is, it's a new entrant. Thank you very much. That's not Urban Dictionary, is it? That's a real one?

DR. CROSS: I'm afraid it is.

DR. SMITH: Oh, yeah. I might have added it to Urban Dictionary. All right. That's --

MS. SETO: So speaking as an agency that support -- and I personally endures [sic] home monitoring and sensor technology, I find it curious to compare, you know, automobile, a car, to our body. I think that clearly our body is a lot more complicated. As we move forward in understanding disease pathophysiology, we realize it's not a car engine. It's not. And so I want to make sure that that's not really an appropriate analogy.

DR. SMITH: Thank you. I don't mean to imagine that the human body can be approximated by even the most fancy German sports cars. Instead, I would only say that we need to make the analogies palpable to the average person so that they understand that what the opportunities are here. You know, perhaps draw on your patience -- I put air in my tires up to a certain level. When we give chemotherapy for malignancies, we don't give enough to kill the tumor. We give as much as the patient can stand without dying, you know? So we are driven more by maximizing therapies up to a point where we avoid complications than tailoring them specifically to the disease. And so there is a great opportunity, I think, in front of us, if we've got actionable sensors of disease activity.

DR. SACKNER-BERNSTEIN: So the comment you made about comparing these different countries and what's going to happen was underscored by another independent report that was released today based on 2007 data that puts us at a handful of western countries, seventh out of seven, in access to care for chronic diseases, and our per capita cost is twice as high. So we're last there also.

DR. SMITH: I don't think the alarm bells can ring any louder. It's really a question of are we listening. And then finally, if we are listening, it's time to do something. It's well past time to do something. It's just difficult to get us all on the same page.

DR. SACKNER-BERNSTEIN: Thank you. Michael Morton will give our next presentation.

MR. MORTON: Thank you. I have no automotive metaphors to share this afternoon.


MR. MORTON: Good afternoon. My name is Michael Morton, and I am a Director of Regulatory Affairs at Medtronic. I'm pleased to address this Council on behalf of AdvaMed. AdvaMed is an organization that represents manufacturers of medical devices. I'd like to thank Dr. Shuren, Dr. Sackner-Bernstein for organizing this workshop. I'd like to thank the members of the Panel for the time and the effort that you've put into this.

As I begin, I would like to recognize two current activities which are examples of efforts to address unmet public need and to facilitate innovation. One is the Pediatric Heart Valve Workshop: A New Paradigm for Obtaining Market Approval for Pediatric-Sized Prosthetic Heart Valves. That's quite a long title. But the key words there, new paradigm and pediatric heart valves. That workshop was co-sponsored by the FDA and AdvaMed and was held in January of this year. And follow-on work continues today. The other is the FDA workshop on computer modeling methods for cardiovascular device development, and that was held earlier this month.

Both these activities have drawn upon technical expertise within the FDA, the medical community, academia, and industry, and they are looking for new regulatory models and technical tools to bring devices to patients.

Today, I would like to discuss 11 items which AdvaMed member companies believe would help to provide medical devices for unmet public health needs and facilitate innovation in the medical device development. We've developed a number of detailed proposals.

First, we suggest a need for guidance for the humanitarian device exemption process. The standard for FDA approval for a device under an HDE is safety and probable benefit. This is a different standard than that for a PMA, which is reasonable assurance of safety and effectiveness. And this typically requires a full-scale prospective, randomized clinical trial.

At this time, there is no FDA general guidance to reviewers or to industry regarding the type or level of evidence that must be developed to demonstrate that an HDE does meet the probable benefit standard, so lack of guidance is my point here. This ultimately hinders the use of the HDE program. Without clear FDA guidance, demands for evidence can drift upward until, eventually, the demands seem to approach those which would be required for PMA approval.

So for this reason, AdvaMed recommends FDA develop general guidance on appropriate types and levels of data necessary for HDE approval, including examples of what FDA believes are the data needed to demonstrate probable benefit. That guidance also should state that RCT trials generally won't be needed to demonstrate probable benefit. Further, FDA should consider non-clinical data, published literature, historical data, surrogate endpoints, and statistical methods, and evidence from experience with similar devices.

On a related point, during FDA-sponsored pediatric stakeholder meetings, and this was in 2004, numerous participants pointed out that private insurers frequently refused to reimburse for pediatric HDEs. These private insurers reason that because an HDE can only be administered in an institution that has a functioning IRB, that therefore the devices must be investigational and therefore not reimbursable, would not be reimbursed. So we'd suggest that in a guidance that FDA make clear that the devices are -- that the HDE does have approval, and this would assist facilities and physicians in seeking reimbursement.

So our second point would be that we'd like to propose the use of more flexible regulatory models. And I mentioned earlier the pediatric heart valve workshop. FDA could develop regulatory tools and adaptive clinical trial designs that take in consideration the reduced sample size associated with orphan and pediatric diseases and conditions. For instance, FDA could approve certain devices based upon smaller confirmatory trials in conjunction with a long-term registry requirement either for an individual device or for certain types of devices. So related to this, FDA could post on its website examples of adaptive clinical trial designs that have already been successfully used to obtain on-label orphan or pediatric indications.

Third, we propose the use of alternative valid scientific evidence. The FD&C Act gives FDA the authority to utilize valid scientific evidence other than that from well-controlled trials. Importantly, the standard of reasonable assurance of safety and effectiveness is going to be the same no matter what type of scientific evidence is required. And while FDA does rely on many types of valid scientific evidence, in other areas, FDA may have an opportunity here to take advantage of its statutory authority in the case of pediatric devices.

FDA could make better use of all forms of valid scientific evidence to address the problems associated with the extremely small number of orphan and pediatric patients. For example, what may have evolved as the pediatric standard of care may be off-label. Doctors would therefore be reluctant to randomize pediatric patients to a surgical control if the minimally invasive procedure had become the standard of care. Also, parents would be very reluctant to have their children participate in such a trial. So in this instance, an FDA requirement to randomize pediatric patients to the surgical procedure creates a barrier that prevents the off-label use of the device from ever becoming on-label. And so where numerous articles document the effectiveness of a particular off-label use of a device that's become the standard of care, FDA should encourage the use of these data.

Fourth, we would like to reiterate the need for good communication and collaboration between FDA and sponsor. Too frequently, device development can be frustrating not because of some legal impediment but simply because it's unclear how the requirements apply to a specific technology. Frequently, a full and effective dialogue between the developers and the Agency would allow for a resolution of any issues that stand in the way.

As a general matter, therefore, we would urge the FDA to be more deliberately collaborative with developers of these products. And for example, FDA could offer to facilitate access to expertise within the Agency or outside the Agency. Further -- and this is an important point -- further, early agreement on innovative approaches to achieve full PMA approval or pre-market approval of these devices for orphan and pediatric populations would certainly spur development.

Fifth, we'd like to suggest that FDA develop a custom device guidance document. The FD&C Act, again, provides for the manufacture of custom devices intended for use by an individual patient in response to a clinician's order. In the ongoing dialogue concerning pediatric needs, clinicians have repeatedly reported that they feel compelled to jerry-rig or modify existing devices to treat their pediatric patients. And while FDA has a custom device program intended to address this problem, manufacturers have often been reluctant to participate because the rules are unclear and custom devices are limited to just one or a few patients.

AdvaMed recommends FDA develop guidance for custom devices that expands and clarifies the number of devices that manufacturers may customize for the orphan and pediatric populations. And this recommendation was echoed by Dr. David Feigel in 2008 at an NIH pediatric device workshop.

Sixth, we would propose the creation of a new compassionate use orphan and pediatric device provision. AdvaMed proposes a well-regulated mechanism to provide device access for super small pediatric populations that are not likely to be served under the FDAAA 2007 pediatric HDE program. AdvaMed recommends that FDA develop regulations that would allow manufacturers to distribute no more than 100 unapproved devices annually for patients afflicted with diseases or conditions that affect too few patients annually to justify an HUD device.

Seventh, we recommend a tiered triage approach for in vitro diagnostic tests. Many of the orphan diseases require new, more sophisticated diagnostic tests, and in order for those new tests to reach the market, FDA will need to focus its oversight primarily on the risk of harm associated, how the test result is used to treat patients. AdvaMed believes FDA should clear or approve higher risk tests through an approach where data requirements are commensurate with the level of the risk of the test. In particular, regulatory requirements should be determined on the basis of risks associated primarily with the clinical intended use of the test along with the consideration of the novelty of the analyte, the technology, or test platform, and the site of service. These changes would allow more rapid patient access to the tests and would promote innovation.

Eighth, we propose additional guidance on the use of general and specific device claims. FDA requirements for specific claims and their associated data can be an important barrier to device development for these small and dispersed orphan and pediatric populations. FDA could consider allowing for more general claims to enable device approval, and then conditions of approval, such as requirements for a registry, could be used to ascertain whether there are particular issues associated with, for instance, a particular age range.

And then, ninth, AdvaMed recommends the creation of an orphan and pediatric ombudsman within CDRH. Right now, there is no single individual within the center that has the responsibility or the expertise to assist and counsel manufacturers or other stakeholders on the regulatory pathways to achieve on-label indications for these orphan and pediatric diseases and conditions. So an ombudsman could also serve as the liaison with other federal officials interested in orphan and pediatric diseases and conditions.

Tenth, we recommend improvements in Medicare's coding, coverage, and payment processes. Reimbursement remains critical. We've certainly heard that repeated several times today, that reimbursement is critical to fostering device innovation. Today's reimbursement system is complex. It weaves together concepts of coverage, coding, and payment, and at each step in the process, access to important new discoveries can be delayed or denied.

Barriers to coverage can impede innovation in at least two ways. First, startups and other manufacturers concerned about the risk of coverage denial may prefer to invest in a more secure next-generation technology rather than a breakthrough technology.

Second, innovation does not stop when a device is put on the market. Devices can be refined over time and then applied to new indications for unexpected uses. So an early coverage denial not only holds back a technology for its original use in refinement, but it delays or eliminates opportunities for identifying beneficial indications.

And beyond coverage, the process of coding and securing adequate payment for use of medical devices likewise can produce barriers. A new device often needs a new CPT code to identify the service involving the product so that patients may submit accurate claims to Medicare, Medicaid, and private payers.

Under a best-case scenario, it takes about 14 months to receive a new CPT code. Once assigned a code, Medicare and other payers decide if patients will have access to the new technology by determining if it is covered by the public program or the private health plan. So timely Medicare coverage decisions as well as coverage of Medicare beneficiaries, routine costs associated with clinical trials further benefits patients.

Finally, Medicare payments must be sufficient to cover the cost of the procedure and the technology it embodies. Failure to do so can create significant disincentives for hospitals, physicians, and other providers to adopt and provide access to innovative technologies.

Finally, AdvaMed has recommendations to improve the orphan and pediatric device development in innovation that would require statutory changes. Many of these programs would help to offset the costs of orphan and pediatric device research and development and address small market size and commercialization risks. And these include a strong orphan and pediatric device research and development tax credit program, a tax credit for orphan and pediatric HDEs, expedited FDA clearance or approval on orphan or pediatric device applications, and clear pathways for reimbursement when such products are cleared or approved.

So, ultimately, FDA's regulatory program must be based on predictable criteria and timeframes to avoid stunting the development of innovated technologies. In the medical device industry, we are particularly dependent upon the investment by venture capital organizations willing to take the substantial risks associated with new product development, something we've heard several times today. Lack of regulatory clarity compounds the technical and clinical risks associated with product development and can discourage needed investment.

So, in closing, AdvaMed appreciates the opportunity to present our recommendations to meet unmet needs. Thank you very much.


DR. JACQUES: Let me maybe start with a couple of general comments about not necessarily about your talk in particular but just sort of reacting to some of the things that I've heard this afternoon.

There has been some conversation that, you know, both among Panelists and among speakers, during the day that there are sort of examples that one might see in other countries, in terms of the way they regulate things or the way they pay for things and things like that. And I think it is important when one is looking at another country to be wary of sort of cherry-picking one aspect without being willing to accept all of it. Arguably, if you don't have any cars that can drive any faster than ten miles an hour, you don't need speed limits because there is really no incentive to speed in some realistic sense.

So maybe the incentives in other countries are such that regulations can be, you know, tighter or looser, depending on how one wants to look at it, because maybe the risks of looser regulation are a little bit different.

I do want to point out that no is as predictable an answer as yes. And I do think it's important for people to acknowledge that although we can have conversations about predictability, if people only see predictability as I want a yes and I want it a little bit quicker and I want it -- I want yes more predictable, but if you want to say no, let's sort of leave that kind of murky, then it'll be hard to really sort of move forward here.

I do wonder sometimes, you know, where -- what happened to the promise of medical innovation. Certainly, there has been a lot of stuff that has been developed, and I realize that that trajectory may be not necessarily going upward with the same slope as it did before, but I do get a sense that there is sort of a, yeah and now what, kind of sobering thought.

There have been some editorials recently, now that we've had the tenth anniversary of the sequencing of the human genome, that have essentially been, gee, we thought we'd have a heck of a lot more to show for it by now. So now we know Francis Collins' genome and maybe somebody else's, but where -- what happened to the promise that somehow uniquely targeted therapies would solve a lot of issues. And I think what we've found is that what we have is a lot more questions.

You know, is there a ceiling to the practical benefits of new technologies? Let's look at airplanes because people have talked about cars and various other things. You know, when we were kids, we thought we'd be flying around like the Jetsons, you know, George and Judy and Elroy, and all the rest of them, and Astro. You know, we can't even get a supersonic commercial airliner, okay? That failed not for technological reasons but simply, apparently, because society felt that it didn't really add anything that was terribly worth, you know, continuing to invest in. And I do sometimes wonder as we sort of focus narrow technologies on, well, let's fix this little thing, let's fix this little thing, does that approach sort of doom us to sort of missing the big picture on this?

And then my last thought is on methodologic rigor. There has been some comment on, you know, why do we need randomized clinical trials? Why can't we use registries, why can't we use other things? I think there is a reasonable place for many different types of clinical study methodologies.

Methodology, essentially, is something that gives you confidence that the reported results actually are in the real world what is reported. And I think what that leads to is that the required methodologic rigor ought to be commensurate with what are the consequences of being wrong. If you're claiming to be the next great thing and people are going to stop doing other things or people are going to spend a whole lot of money, then arguably the cost of being wrong is significant.

On the other hand, if you're willing to say that at best you're an incremental change in a non-lethal condition, sort of the better Kleenex, so to speak, then arguably you don't need anything terribly rigorous. So are people willing to sort of come to the table, saying, "I'm going to acknowledge up front that frankly I don't have much to add. But in return for that, I'm going to ask you to hold me to a different standard"?

MR. MORTON: I actually would like to follow one of your points, which I think is excellent. And we were discussing at lunch that even though we, the group of AdvaMed representatives who are here today, are very much in favor of device innovation and we think that that offers great promise, but when you ask the question of unmet need, it always does not have a device solution to that problem. And I know that Katie O'Callahan is here. She has done a lot of work examining gender differences in the cardiovascular -- in treatment of cardiovascular disease, the difference in access of treatment of women compared to men. Men get more treatment. And in examining that, it doesn't always tend to be a device reason that that's the case. It has to do with who is doing the referrals; are the women themselves educated and focusing on cardiovascular disease that may be affecting them? So we use that as an example that we need to keep our eyes very broadly open.

MS. SETO: So I want to just respond to the ten-year post-genomic era and where has that gotten us. So in looking back, clearly, sequencing the human genome was -- and it's been characterized by Phil Sharp, Nobel laureate at MIT, as the third revolution in biomedical research. But we recognize that that might even have been the easier part. The phenotyping is tremendously challenging, from multiple dimensions, having the animal models and so on. And I'm still very optimistic even though we're ten years from now and we actually don't have really specific personalized medicine that's based on a genotypic makeup. But I'm still very optimistic that phenotypes are going to get us much more in-depth knowledge about targets. And I agree that not all of the unmet needs will be addressed or need to be addressed from a technological standpoint. There will be pharmacological and drug targets that we know that could not have -- we could not have achieved that knowledge without the phenotypes.

And so I think the targets -- the use of that phenotypic information to predict more the subtypes of patients, the ability to predict treatment outcome, so all of that is to say I am still very optimistic, even though we're now ten years post-genomic, that we are at a different level of medicine, what I call molecular medicine, than we were before the human genome was sequenced.

Lt Col WATTENDORF: Yes. So I want to echo the comments, too. I don't want to be the apologist of the human genome project or genomics, but there was a bead of the rise of the Luddites that I might have been hearing in that. And I tend to hear that quite a bit, I suppose. I'm one of the younger Panel members and maybe a little more on the sense of awe with technology than some others that may have more wisdom. However, you know, when we look at -- I would consider this a pretty transformational device, but it's based on radio technology, which has been around for an awful long time, and there is a lot in here that is very complicated. There is at least two Nobel Prize-winning capabilities in here and a transistor and the SIMO sensor.

And, you know, I certainly think when you're only ten years out from the genome project, declaring whether it's successful or not, in terms of -- and in the discussion of a medical device in the regulatory space, I think it's still too early. I think the problems have not been in the knowledge and the technology, but that it hasn't been simple enough for us to use.

And so going to the earlier presentation right before yours, I think what is it in the regulatory space that can be done to make some of these devices usable in a simple way, and how do we incentivize the simplicity, which doesn't mean the device is necessarily simple, but the utility of it is simple. And that seems to be the challenge that's the fundamental underpinning of many of the discussions today. And those are the pieces that I'm hoping that we can find ways from the federal side to create a environment, whether it's on the payers or on the reimbursement structure or on the regulatory space, where we would make the capabilities of this technology much simpler than it has been.

DR. JACQUES: Yeah. And I'm not suggesting that genomic medicine isn't worthwhile. And, in fact, there have been plenty of commentaries in the lay press and some in the medical press during the last couple of weeks or so.

After lunch, I had a short conversation with Jack Lasersohn, the person from the venture capitalists, and we both sort of separately started the conversation about the airplane. What would have happened to the airplane if, essentially, the question had been called in 1903 when the Wright Brothers flew. And arguably what kept the airplane technology alive long enough so that it could finally, frankly, be practically useful for someone was probably the federal government contracts to carry air mail, okay? Arguably, the plane wasn't necessarily good for a whole lot of things.

So I think it is an important question. How do these technologies that are immature be sustained at a reasonable level so that then you can decide long-term whether they're worth it or not to do long-term?

It is a bit of a challenge, though, for a health insurance payer, which is charged with paying for things that have a high likelihood of benefit, to be asked to sort of support the startup or adolescent-stage costs of technologies where frankly the question is still open. So it may well be that especially for our colleagues, you know, around us who are more into sort of that part of the technology infrastructure, it really may be in a more appropriate task for them, at least in the short-term.

DR. CROSS: I want to follow on to the comment about phenotypes. Where can you find phenotypes? And I would suggest that there is one possibility that you might want to consider. Electronic health records. And our organization has 25 to 29 million records, as I recall at the moment, vast amounts of clinical, descriptive, and objective information. And, in essence, putting all that together, a phenotype that, when matched up against a DNA pattern of some sort, might become very quickly productive.

MS. SETO: I agree. I think that's a huge resource. It's really a gold mine. We at the NIH, not just my institute, has supported several projects: one to develop the data mining tool; second is actually go and mine the data. And so one project, for example, at the Harvard Partner System, when they took I don't know how many number of patients' records, but collectively among all the Partners hospitals, a huge number of cases they see weekly. And through the IRB approval, they were able to create a subset of data from the patient record that they called Crimson, which is a research database. And it's been a tremendous tool in terms of looking at past trend. And there was a coincidence, just as an example, of Vioxx use and cardiovascular cases. So that's the kind of data and trends that you can use, and I bet you, you can also then do phenotypes with them. But I absolutely agree that the electronic health record is going to be a tremendous tool and resource.

DR. SACKNER-BERNSTEIN: Any other comments? All right. Are there any folks who want a last opportunity to make a comment from the audience? I mean, ideally, people who haven't had a voice?

(No response.)

DR. SACKNER-BERNSTEIN: I think that what we should --

UNIDENTIFIED SPEAKER: I just want to make several comments that --

DR. SACKNER-BERNSTEIN: You know, because it's being transcribed, though, maybe you could come up and --


DR. SACKNER-BERNSTEIN: Thank you. So what I'd like to do is just take a moment to thank all of the members of the Council not just for today but for the participation thus far. Also, folks from within each of the agencies who have been contributing along the way, including the CDC, who is not here. Mostly, I want to thank you for coming here today, those of you who have stuck to the end, the speakers throughout the day, because the purpose of the meeting was for us to get ideas, and a critical next step is to look at the submissions that come into the docket over the next month, which we really would encourage people to submit all the ideas. There is no idea that wouldn't be worth us knowing about and being able to share amongst this Council.

Once that information is in place, the next step really rests upon us to delineate for the public, to the public, how we're going to move forward with this two-prong approach, that is, identifying the unmet public health needs as well as addressing barriers that should be addressed, not just any barrier that exists, but a barrier that represents a level of difficulty that's out of proportion to that clinical need.

So we'll be doing that over the course of the summer. There will be a transcription from today's event posted on the website, as there will be even before the transcription is available of the different presentations so that people can access them and look at the information.

So, once again, thank you all for coming and for participating, and we look forward to continued interactions.


(Whereupon, at 3:30 p.m., the meeting was adjourned.)


This is to certify that the attached proceedings in the matter of:



June 24, 2010

Gaithersburg , Maryland

were held as herein appears, and that this is the original transcription thereof for the files of the Food and Drug Administration, Center for Devices and Radiological Health, Medical Devices Advisory Committee.



Official Reporter

Page Last Updated: 04/15/2014
Note: If you need help accessing information in different file formats, see Instructions for Downloading Viewers and Players.
Language Assistance Available: Español | 繁體中文 | Tiếng Việt | 한국어 | Tagalog | Русский | العربية | Kreyòl Ayisyen | Français | Polski | Português | Italiano | Deutsch | 日本語 | فارسی | English