Innovation and Invention in Medical Devices: Workshop Summary (2001)

The objective of the workshop that is the subject of this summary report was to present the challenges and opportunities for medical devices as perceived by the key stakeholders in the field. The agenda, and hence the summaries of the presentations that were made in the workshop and which are presented in this summary report, was organized to first examine the nature of innovation in the field and the social and economic infrastructure that supports such innovation. The next objective was to identify and discuss the greatest unmet clinical needs, with a futuristic view of technologies that might meet those needs. And finally, consideration was given to the barriers to the application of new technologies to meet clinical needs.

Harry M.Jansen Kraemer, Jr.

Chairman and Chief Executive Officer

Baxter International, Inc.

Medical devices have extended the ability of physicians to diagnose and treat diseases, making great contributions to health and quality of life. Beyond the considerable attention that such big-ticket technologies as computer-assisted tomography and magnetic resonance imaging have received, there is no doubt that these technologies have changed the mainstream practice of medicine. However, diagnostic technology is not limited to capital equipment imagers. It also includes analytical techniques using high-resolution chromatography,

polymerase chain reaction (PCR), and monoclonal antibodies, providing physicians with new, accurate, and rapid information.

On the therapeutic side, devices save lives and improve quality of life. Dialysis therapy extends lives for end-stage renal disease patients, orthopedic implants enable patients to walk again, and minimally invasive technologies allow surgeries that are safer, with less pain and trauma, requiring significantly shorter hospital stays.

The United States is clearly the leader in medical device innovation. In 1998, the United States medical device and diagnostics industry was responsible for nearly $70 billion in production, which is almost 50% of the total world consumption of medical technology. The United States exports significantly more devices than it imports, netting a trade surplus of almost $10 billion. Further, United States medical device patents outnumber foreign patents by more than three to one.

What makes the United States system so special? Kraemer believes that a fundamental part of the success of the United States is that it is a society that promotes and rewards innovation. The United States has a strong science base and an entrepreneurial culture of sophisticated and efficient financial markets, intellectual property protection, and a health care system that for the most part has been willing to pay for technological advances.

Despite being the best system in the world, however, the United States faces major challenges that can undermine the viability of innovation and access to better health care. One striking characteristic is that medical device innovation often requires the contributions of a diverse array of scientific and engineering expertise. Something as seemingly basic as the materials used in medical products, for example, have made lasting contributions to health care in ways people often take for granted. For example, the first plastic blood collection container enhanced the safety and quality of stored blood and made possible the separation of individual blood components. Thus, modern blood component transfusion therapy was made possible by plastics. Miniaturized circuitry for pacemakers, mathematical algorithms used in MRI, and greater understanding of fluid mechanics for heart valves were all developed outside the conventional areas of medical research.

Obviously researchers cannot script innovation. At times it occurs in a great stroke of luck or insight. At other times it is a gradual process with new devices piggy-backing off earlier ones. Innovation extends well beyond laboratories, including the many instances of clinicians finding new indications for already-launched technology, as well as breakthroughs that have occurred at the intersection of multiple scientific and technical streams of progress.

An example of an early innovation that continues to find new applications in medicine is the laser. Invented in 1958, the laser was first applied in health care as a non-contact scalpel. New applications of lasers include reshaping cor-

neas, photodynamic therapy for cancer, and transmyocardial revascularization (TMR) for severe angina. TMR for severe angina exemplifies the tremendous uncertainty at the leading edge of innovation. Despite its apparent early success, researchers are still not sure if it really works and, if it does, how. It also highlights an important aspect of innovation: the timing and methodological rigor of clinical trials to test innovations and the relation of these to regulation and payment for new technology.

Innovation requires time, insight, and sometimes luck. It also requires significant financial resources. What do researchers currently do to create an environment that is conducive to innovation?

First, the United States has a strong commitment to basic science, which forms the basis for future innovations. The United States funds more basic research than the next six countries combined. The National Institutes of Health (NIH) and the National Science Foundation (NSF) granted extramural funds of more than $13 billion in 1998 alone to support the sciences. Advances in basic sciences, such as understanding of genetics, biology, physics, and chemistry, enable the rapid pace of innovation.

This interaction between scientist, engineer, and clinician can be further promoted to speed the process of innovation. Although it is the single most prolific and important biomedical research entity in the world, NIH is organized primarily into disease- and organ-specific institutes in a way that is less than optimally supportive of technological disciplines that cut across institute priorities. While some institutes do invest in engineering tools and techniques, these are specific to institute priorities and do not sufficiently support the underlying science and engineering that spawn future advances in health care technology.

Support for biomedical engineering in the private sector is considerably greater than support by the government, and nearly all of the private sector investment is devoted to applied research. Shoring up the federal investment, particularly for basic research, would strengthen our national capacity for technological innovation.

Researchers increasingly understand that innovation is more than producing the latest widget. The downstream hurdles of regulation and payment affect not only whether a device gets on the market and becomes accessible to patients, but they also send feedback to the process of innovation itself. If the regulatory process is perceived as being slow and expensive for innovative devices, then incentives shift to produce more “me-too,” or derivative, devices that may have a less risky road to market.

For reasons both fair and unfair, the Food and Drug Administration (FDA) has been cited over the years as a major barrier to innovation. Kraemer is among the first to state that FDA, and specifically the Center for Devices and Radiological Health, have taken many important measures to streamline regulatory processes, making them more transparent and predictable. This is not to say that

interactions with FDA cannot be significantly improved. In fact, it may be somewhat ironic that the pendulum may have swung too far toward transparency, with the potential for confidential information to be released by the agency over the objections of industry. Still, rather than view FDA as an adversary, Kraemer would like to point out that researchers all share the goal of improving patients’ lives at the same time that they minimize the risks.

The most important issue, the one that researchers have all grappled with is that of over-regulation versus adequate regulation. The agency has acknowledged that devices are different than drugs. In general, devices have faster cycle times and tend to be characterized by incremental improvements, leading to longer-run significant advances. As such, earlier devices can provide considerable bases of information on the safety and efficacy of the next generation. It is difficult to have placebo controls in clinical trials of most devices or to double-blind physicians and patients as to who is getting which technology. Moreover, given the sizes of the target populations for many devices, it can be impractical to conduct randomized clinical trials with devices as they are done for drugs. Thus, clinical trial requirements and designs must reflect the technology at hand. And, whether for drugs or devices, there are always practical and ethical challenges of informed consent.

Clearly, devices need to be regulated differently than drugs, and FDA is increasingly aware of this. Even so, researchers have far to go in pursuit of the least burdensome provision of the FDA Modernization Act of 1997. There remain inefficiencies due to unnecessary regulatory impediments.

A fascinating and important issue arises out of the rapid evolution of technology itself, that is, the fit between new forms of technology and the FDA jurisdiction over technology. For regulatory purposes, how should xenotransplants—in which animal organs are genetically engineered to express human surface proteins, thereby suppressing rejection—be classified? Are they devices or are they biologics? Should they be regulated by the Center for Devices and Radiological Health or the Center for Biologics Evaluation and Research?

Under what model of regulation should these hybrids fall, and how prepared is the agency for handling this technology? This jurisdictional issue involves not only FDA but also the Centers for Disease Control and Prevention (CDC) and institutional review boards. In any case, a technology company that is weighing whether to push forward in innovative areas must consider the time and risk involved as well as the hurdles and other procedural matters.

Payment issues are of increasing concern to the health care industry. Everybody talks about the rising costs of health care and the need to find ways of managing costs more effectively. In fact, some continue to point to technology as a major culprit in increasing costs. To the extent that technology continues to improve health and quality of life for more people, it is not clear to Kraemer that

people are spending too much on health care. Nevertheless, the emphasis on cost and the role of technology in costs have placed part of the onus on the industry.

It is part of corporate consciousness that innovation is not just creating a better widget. It includes creating more cost-effective alternatives to current therapies. Of course, it is not quite that simple. There is increasing emphasis on cost-effectiveness as a yardstick for payment. While payers here in the United States and in other industrialized nations say that they are interested in cost-effectiveness, too often it appears that their interest is just plain cost cutting.

Cost is in the equation, but where are the benefits to health in that equation? In this environment, providers are getting pressured to use cost-saving technology as opposed to cost-effective technology. Manufacturers are under pressure to consolidate, to make technologies into commodities, and to compete on the basis of price alone. A major factor contributing to United States leadership is the willingness to pay for proven innovation. Price fixing and rationing, which are used in other nations, fail to account for the effectiveness side of the equation.

What are the incentives for innovators to start new companies and for manufacturers to invest in new technologies? What is the trade-off of risk for expected return on investment? It may be one thing for producing those commodity widgets and their costs. It is another for taking on the risk often required to create technologies that may have higher price tags but may yield significant leaps in health outcomes.

Part of what makes this business so complicated as well as personally gratifying is the obligation to help people who are making decisions about their parents, spouses, or children. People’s perspectives change significantly depending on whether they are dealing with abstract statistical populations or with their family, when cost does not matter and arguments that a technology is not cost-saving or cost-effective often fall on deaf ears. This affects the policy debate and the likelihood of success of policy initiatives regarding health care technology. It is a complicated issue that must be addressed.

There are, however, payment issues that can be addressed in the short term. Kraemer refers to the daunting task of navigating the maze of public- and private-sector payers within health care. Although FDA can be a tough customer, at least there is only one FDA. There are many parties that affect payment—the multiple Medicare carriers, state Medicaid programs, the Blues, Aetna, United Healthcare, Kaiser, Cigna—all with varying requirements and different approvals for payment. Kraemer in no way advocates the adoption of a single-payer system, but from the perspective of technology companies, the prospect of getting over the FDA hurdle just to face a multitude of payers is not a pretty sight. It does figure significantly into the risk equation.

One payer that continues to be highly influential is Medicare. Medicare is not necessarily the first to make a payment decision, but when it does decide to cover a new technology, it puts a lot of pressure on all the other payers. When Medicare does not cover a new technology it makes it easier for other payers not to make a positive coverage decision. The Medicare coverage process is a challenge to many companies. After FDA approval, the biggest concern for a manu-

facturer of an innovative device is denial of coverage. New devices that are improvements on earlier versions do not necessarily have this problem. As long as they are not great departures from their predecessors, the next stent or the next pacemaker has a much clearer road for achieving reimbursement.

Barriers to coverage can affect innovation in at least two ways. First, startups and manufacturers concerned about the risk of coverage denial may prefer to invest in safer next-generation technologies instead of breakthrough technologies. Second, innovation does not stop when a device hits the market. Devices can be refined over time, and applied to different indications to find unexpected uses. An early coverage denial not only holds back a technology for its original use and refinement, but it delays or eliminates opportunities for identifying other beneficial indications.

The situation has brightened for payment during clinical trials. Prior to 1995, it was Medicare’s policy to deny payment for virtually all non-FDA-approved devices and off-label uses, even those with predicates with proven safety and efficacy. This posed a significant disincentive for innovation by raising costs for manufacturers and making providers reluctant to participate in clinical trials. In 1995, the Health Care Financing Administration (HCFA)¹ and FDA agreed to an approach for classifying new devices into those that are truly novel and those that are basically next-generation versions, and making the latter categories eligible for Medicare reimbursement. This 1995 interagency agreement is an example of a strong collaborative step in the right direction for improving technology payment as well as access.

Payment problems remain, however, including the process of coding and securing adequate payment for use of medical devices. A new device needs a new Current Procedural Technology (CPT) code to make sure that doctors can be reimbursed. Even with a proper code, the Medicare payment level for that code may not be sufficient to cover the cost of the procedure and the technology it embodies, thereby constituting a significant disincentive for doctors to use the device. This entire coverage and coding process can take years from the time a product is launched (unless the new device already has a code in place), but the payment level is so low as to be a significant disincentive for doctors to use it. These payment hurdles are significant for large technology companies, but they are especially daunting to smaller companies that do not have the dedicated staff, experience, and other resources to handle these issues effectively.

What can be done in the short term to address these issues? HCFA recently revamped its national Medicare coverage process and is still working out some problems. However, HCFA also needs to reform its processes and requirements regarding coding and payment levels to make these much more timely, transparent, and fair. FDA and HCFA can cooperate to reduce duplication of effort significantly and streamline processes.

The need to demonstrate clinical and economic value is a central issue in regulation and payment, and it is clear that the trend is toward requiring more evidence rather than less for innovative devices. However, building evidence of value can be costly and time-consuming, and at some point reaches diminishing returns with respect to the availability and benefits of the technology.

In an ideal world, a decision maker has definitive proof that the device will improve health outcomes, be less costly, or improve outcomes at an acceptable cost, but that certainty comes with lost or delayed opportunities to improve people’s health and the quality of life. These trade-offs look very different depending on who a person is. If a person is a government agency charged with protecting against potential health hazards of something new, or charged with minding the public purse, it may make sense to err on the side of delay, but from the perspective of some patients and their doctors, that same delay can be a life and death proposition.

Even when researchers have enough evidence to understand the safety, efficacy, and cost of a device, the interpretation of value can vary depending on perspective, including those of patients, hospitals, managed care organizations, technology assessment agencies, the government, and society at large. For many of these perspectives, it is hard to argue against a device that works, reduces complications, saves lives, and costs less than the alternatives, but most of the time it is not quite that easy.

Left ventricular assist devices (LVADs) are a great example. Each year there are at least 60,000 patients with severe heart failure unresponsive to medical therapy who need a heart transplant, but who are unable to get one, due to either their age or other complications. Many do not get transplants simply because there are so few donor hearts available. Only about 2,500 patients, or less than 5%, get heart transplants each year. LVADs were developed as a bridge to cardiac transplantation for patients with severe heart failure. The need for LVADs is clear. They are proven, and they were approved recently by FDA. These devices clearly are not cheap. They can range from $45,000 to $65,000 per unit and, as is the case with so many medical devices, people tend to focus on the price tag. End-stage heart disease is not cheap, but its price tag is harder to discern and, unfortunately, end-stage means exactly that, end-stage. The potential of LVADs as bridges to transplant was limited by the number of available hearts. Now, given the continued shortage of heart donors, LVADs are in clinical trials as alternatives rather than bridges to cardiac replacement. This changes the size of the potential patient target population and changes the potential costs for payers. It will be interesting to see how this all plays out.

Society places a high premium on innovation. Patients expect and demand that researchers will continue to develop cures for an aging population. The pub-

lic wants the latest breakthroughs—not in years or in months, but today. Increasingly, researchers want and expect patients to be more interested, more informed, and more active about making health care choices, but the counterpoint to patient activism is patience with the time it takes to achieve that new breakthrough.

Can the current system continue to support these expectations? The promise of new developments in biotechnology, genetics, tissue engineering and, of course, computers and the Internet are transforming the industry, but researchers must continue to survey the innovation landscape and manage those hurdles, potholes, and inclement weather.

Chief among these issues is society’s definition of value and the evidence researchers will accept to prove that value. Researchers recognize that the process and products of innovation will continue to be tested. The bar is set high. Researchers need ongoing dialogue from all perspectives to define and redefine as necessary what they must achieve in clinical benefits and how they are willing to pay for these benefits. Solutions may not be easy and no one has all the answers, but it is at gatherings like this that researchers can put the issues on the table, provide practical and constructive review, and promote action to achieve their shared purpose of improving the health and quality of life for every American.