Click for next page ( 165


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 164
Summarizing Reflections WILLIAM W. LOWRANCE Two general impressions pervade the papers in this volume and my recollections of the symposium on which it is based.* The first is an overwhelming admiration of the medical armamentar~um that has become available. What a range of devices now can be drawn upon to measure physiological parameters, to peer right through the body, to deliver drugs precisely, to make surgical and other repairs, to replace tissues and bones and organs, to compensate for sensory and mobility losses, to bolster recovery! So although this volume was conceived and assembled in the interest of encouraging and guiding innovation, we should not feel too bad about the accomplishments so far. The second impression is a sense that the rubric "medical devices" covers an almost unencompassable range of technologiesfrom rather simple classical aids, such as crutches and eyeglasses, to novel high- technology instruments and implantable organs, and from inexpensive devices used intimately by individuals, to capital hardware used in large institutions for the benefit of many thousands. This makes the topic exceedingly difficult to analyze as a category, and frustrating to deal with as a policy and legal problem. Prank Samuel, paraphrasing Gertrude Stein on Oakland, has said of the medical device territory, "There is no there there...." * "New Medical Devices: Factors Influencing Invention, Development, and Use." Symposium sponsored by the National Academy of Engineering and the Institute of Medicine, March ~10, 1987, Washington, D.C. 164

OCR for page 164
SUMMARIZING REFLECTIONS 165 But Frank's own employment as an industry leader is evidence that within the territory, despite its unruliness, addressable issues exist. SALIENT TRENDS IN HEALTH CARE The many health care trends covered in the preceding papers need not be reviewed here, but a few with special relevance to the device enterprise are worth noting: ambulatory care, home care, self-care, noninvasiveness, long-term care, and rehabilitation. The overarching concerns of everyone in the enterprise are to ensure quality and to preserve the ethical complexion of care, even as costs are being subjected to vigorous campaigns of containment. Seymour Perry and others make it clear that health care is being monitored and evaluated more systematically than ever before. This is a crucial development, one long overdue. Finally we may learn what the paybacks really are from our personal and social medical invest- ments. But for innovators and providers, having people "looking over their shoulders" and reviewing their billingsinstitutionalizing caveat emptor, so to speak is unsettling. THE FLOW OF INNOVATION The evolution of a technology from conception to full use can be schematized as shown in Figure 1. A new medical device is conceived in a marriage between technical opportunity and medical need. Perhaps more than for some other kinds of technologies, the elements of the medical innovative partnership may be quite distinct, as was implied by the joint sponsorship of the symposium from which this book derives. Engineers, materials scientists, inventors, systems specialists, and others on the supply side seek beneficial uses for their technologies. Physicians, health care experts, and patient advocates on the demand side seek technologies to meet health needs. The problem is how to explore potential matches between the two sides. In some cases one person competently bridges between these universes. But more often, nowadays, the task requires organized teams. Following the almost magical step of innovation which may result from a stroke of genius, inspired tinkering, modest improvement of a conventional device, or strategically pursued fancy-technology R&~ a prototype is moved into development. Development may make variations on the initial invention, then subject prototypes to testing, evaluation, and improvement. Consid-

OCR for page 164
166 HOW TRENDS WILL INTERACT THE FLOW OF INNOVATION MEDICAL NEED ~ \ ~ ; l ll / \ \ _ ~ i ,, INNOVATION I R & D OPPORTUNITY ' DEVELOPMENT x~` I! `\" i ~ \ / 1 \ PREMARKET / I EVALUATION 1 / 1 . , / 1 / ~ " ~ MARKET INTRODUCTION DIFFUSION INTO ULTIMATE - USE - FUGUE 1 The flow of innovation. / / / oration is given to the pragmatics of manufacture. Consideration also is given to the vagaries of real-world application and use. If the invention looks promising, it moves on into premarket evaluation, being put through carefully staged testing, perhaps on animals first, then on people, to gain realistic assessment of its medical potential. In part because of the enormous diversity of these products, in part because of the intensity of their medical effects, and in part because of the pace and complexity of their design evolution, the choice of criteria by which medical devices should be evaluated is not always clear or fully anticipatable. If, after all this evaluation, the device still looks medically promising,

OCR for page 164
SUMMARIZING REFLEGTIONS 167 and looks financially promising to the vendor, and can meet the various government regulatory and other "filtering" standards of efficacy and safety, it is moved into marketing. If the innovation continues to meet all these criteria and its market grows, it undergoes wider diffusion into ultimate use. Several substages may have to be passed through before the device becomes fully established. Figure 1, though an idealized representation, helps to show the flow of innovation and the factors influencing it. For the moment, merely notice the dotted retro-connections indicated by dotted lines. These are feedback loops, a notion that both physicians and engineers are accustomed to. There is feedback between development and initial innovation, and between various later stages and development and innovation. And in the largest picture, there is feedback between both present and potential use and initial innovation. HEALTH OF THE DEVICE ENTERPRISE ITSELF A question expressed or implied by all the authors in this collection of papers is: Is the medical-device delivery enterprise itself, as a system, healthy? For the various reasons cited, it is difficult to generalize. But the external evidence is fairly heartening. Manufacturers are still turning relatively solid profits. Few of them are getting out of the business. Unlike Chrysler, the steel manufacturers, and the railroads, the medical manufacturers with the exception of a few producers of vaccines or intrauterine devices (IUDs~have not begged for federal bailout or special treatment. Entrepreneurial investors are still lining up. Thus for those who invent, develop, and make devices, the outlook seems far from bleak. For all of us who are the ultimate consumers of devices, surely, despite a variety of problems and costs, we have never been better served. The test questions to this broad issue are: Are any major, promisingly beneficial medical devices being denied to the world because of lack of support for, or impediments to, R&D? Or because of inadequate protection of proprietary rights? Or because of repressive regulation? Or because of the threat from unjustified legal liability? Or because of market failures? Tentatively since the authors do not review cases in economic detail- the answer seems to be: no; few, if any, lines of medical technology development have been stifled (with the possible exception of vaccines), although possibly a few have been slowed. Robert Mann refers to deceleration of R&D. John Moxley says

OCR for page 164
168 HOW TRENDS WILL INTERACT hospitals are slowing their purchasing of big instruments. From their broad perspective, Samuel Thier and Stuart Altman observe that cost- containment has not substantially retarded innovation, though it may have slowed sales and retarded increase in use. Edwin Whitehead and Alan Kahn, speaking as innovators, agree that there has been little damping of important invention. Walter Robb notes that, at least for some large manufacturers, innovation is shifting more to devices that can reduce per-unit care costs, and away from those that offer truly novel kinds of care. It appears that even for such controversial examples as diagnostic imaging, the industry continues to burgeon. New principles and models keep being introduced, and more and more citizens enjoy access to . . . Imaging services. One clear example of near-extinction of both innovation and use is IUDs. This volume does not discuss the Dalkon Shield lawsuits and related issues, although Susan Bartlett Foote refers to them. Peter Carpenter describes how his firm has gone about marketing its IUD, taking elaborate precautions to inform potential users of benefits and risks and secure informed consent. If any proposed devices are being orphaned because the potential market for them is small, despite serious medical neediness, their orphan status may be recognized and special support sought for them. Similarly, if any proposed technologies are languishing because of undue legal liability or other impediments, special subsidy or indem- nification may be sought; the National Childhood Vaccine Injury Act of 1986 attempts to provide such support (the act was passed but, as of mid-1988, has not been implemented). Beyond invention, obviously there are barriers to widespread prac- tical adoption. For instance, Seymour Perry and Stuart Altman point out difficulties the Prospective Payment System encounters in accom- modating new devices when they become available. Vibes of angst over uncertainty radiate from some of the foregoing papers. The system (such as it is a system) for developing devices seems encumbered, maybe even harrassed, and lacks predictability. Given the buffeting to which devices are subjected in regulation, the markets, and the courts, those who invent, develop, and sell these technologies understandably feel uneasy. Gratifyingly little whining comes through in this volume, but occasionally some does emanate from the industry. Chronic whining has been very debilitating to the pharmaceutical industry, and I hope it will be avoided by those who make devices. Kristine Johnson, speaking from her (unwhining) industry perspec- tive during the symposium, raised concern about five issues: uncertainty

OCR for page 164
SUMMARIZING REFLECTIONS 169 in the regulatory criteria; unpredictability of the market; confusion around the bounding provisions for reimbursement; overstandardiza- tion of design, to satisfy imposed rules rather than to make devices that would be the most beneficial; and overexpectations of technology assessment for newly devised technologies. All of these issues deserve attention. INSTITUTIONAL AND INFRASTRUCTURAL ISSUES No outstanding structural shortcomings are raised in this volume. Several authors wish for a National Science Foundation or other federally supported center of excellence in biomedical engineering or biomedical materials science. At the symposium John T. Watson of the National Heart, Lung, and Blood Institute proposed forming a committee to advise a consortium of federal agencies in identifying opportunities and reducing barriers to innovation. It is hard to evaluate either proposal; their merits might be discussed in future forums. Apart from a little of the customary grumping about the Food and Drug Administration (FDA), the authors do not generate focused criticisms of regulation. Perhaps the most vexing problem raised is the qualitatively different testing and evaluation problems that medical devices present for regulation, compared with pharmaceuticals. De- vices may, for example, be moved into at least a limited experimental clinical market when fewer than a hundred of them even exist. One does not have to be a statistician to recognize the limits this puts on statistical power in evaluating efficacy and safety. Moreover, some device risks reside in potential material failures or design flaws of a sort that only become evident during clinical use; these are extremely difficult to anticipate. Accordingly, design and performance criteria for each kind, or even each model, of device may need to be negotiated between the vendor and the FDA. When the symposium turned for reassurance to John Villforth from the FDA, he said, in effect, "Don't count on us for too much. A full- dress review of a single device can cost many thousands of dollars and a lot of expertise, and the FDA does not always have the capacity." He said he thought the FDA mainly should serve to pick up sentinel signs and provide feedback to the industry. Commendably, Frank Samuel argues that what the industry and all of us need is a "competent, efficient, swift, and credible FDA." Surely he is right, and we should hope that the Institute of Medicine (IOM) and other professional organizations will help nourish just those attributes in the agency.

OCR for page 164
170 HOW TRENDS WILL INTERACT REGULATORY CONTROL VERSUS TORT LIABILITY Susan Bartlett Foote reminds us that mere compliance with FDA regulations does not insulate a vendor from liability suits. She points out that very different values and criteria underlie the systems of regulation and tort (which are among the feedback loops midway down my scheme of innovation; Figure 1~. Regulation is notoriously cum- bersome, slow, and costly. Tort cases tend to be capricious, unaccom- modating to scientific evidence, and high in transaction costs; more- over, they fail to decouple compensation of those who are harmed from deterrence or punishment of those who cause the harms. Susan Foote's paper outlines a reform scheme that proposes to correct some of these deficiencies in the hybrid regulation-tort system. Harvey Fineberg offers the chastening observation that, until we accumulate a more robust background record of facts about efficacy and safety of medical devices, tort liability is likely to remain capricious and resist reform. ECONOMICS What is the relationship of technologies to health care costs? Understandably, this issue was brought up from the beginning of the symposium to the end. Many individual medical devices carry high price tags, and devices in aggregate add up to a substantial societal burden. Those costs are far from easy to appraise. What is clear is that simple appraisals can be fallacious. In some ways technologies increase costs; in some ways they decrease costs. Viewed in the micro, they are always expenses. The only way to evaluate them in the larger scheme of things without blowing a mental (or policy) fuse is to view them like all health care expensesas investments that return benefits to individuals and society. The question then transmutes into: What are the costs of these investments, to whom, and what are the paybacks, to whom? Stuart Altman gives us a solid economics lesson, based on two precepts: (a) There is no free lunch; and (b) economics matter. Further, he asserts two propositions that might be susceptible to empirical examination. First, market elasticity is not simple, and the Econ-1 version of supply-and-demand may well not hold; that is, raising medical costs may not necessarily result in people buying less. Second, because in general in recent years hospitals have experienced rising income from public funds, the income stream has swamped the substitution stream; overall demand simply has increased.

OCR for page 164
SUMMARIZING REFLECTIONS 171 Several authors deal with the economic dynamicsdiagnosis-related group (DRG) prospective payment and all that that shape the market both for standard devices and new ones. No need to go through those arguments here. They are complicated, situation-specific, and period- ically need to be examined with respect to given device applications. Of course, the distributive economic-ethical problem is that whereas medical costs mostly are borne by society's various collective health care financing pools, health benefits from those expenditures accrue to individuals and, in a more diffuse sense, to their families, associates, and society in general. DIFFUSION The spread of medical devices throughout the market depends on perceived need, economic considerations, ethical constraints, and a host of other factors covered in the various papers in this volume. A few problems might be pointed out here. Several authors, starting with Sam Thier, lament the inadequate preparation of physicians and other care providers to evaluate and properly use medical devices, especially new ones. This problem is exacerbated by the rapid rate not only of introduction of new devices but modification of existing ones. Peter Carpenter and Frank Samuel urge devotion of much more attention to development of "software" such as educational material and training seminars for users. Doubtless this need will increase as care shifts from hospital into outpatient center and the home (which, incidentally, as Susan Foote has remarked, also is likely to bring new rounds of lawsuits, legitimate and otherwise). Proper maintenance of devices in the clinic, nursing home, and home is another problem. So is prevention of misuse, misapplication, and outright abuse. Much of the responsibility for these matters will reside with vendors. Both postmarket surveillance and technology assessment are keys to stable diffusion. The first gathers the essential facts, the second conducts a structured evaluation of the evolving health care prospects. The IOM is involved in several major ways with assessment of medical technology. Contributions, not mentioned in this symposium, that I think the National Academy of Engineering (NAE) and IOM could make through joint effort have to do with applying decision analysis and related evaluations that grow out of operations research. Engineers already have assisted in this area, such as in optimizing regional blood collection and distribution systems. Although a small and growing band of thinkers

OCR for page 164
172 HOW TRENDS WILL INTERACT is working on these matters, there may well be need for projects here to critique the techniques, develop case analyses, or apply the methods to such problems as systems for collecting data on postmarket failure rates or side effects. Little can be added here about technology assessment. The endeavor is outlined in the IOM survey, Assessing Medical Technologies (Washington, D.C.: National Academy Press, 19851. Although this effort was not discussed in the symposium, everyone seemed to endorse it. A central issue is, how is technology assessment to be paid for? Surely we should channel some sidestream monies from the various reimbursement systems into clinical trials and assessments. How to accomplish this pragmatically is not clear, but it urgently deserves to be discussed. Again, analyzing policies for funding tech- nology assessment may hold a role for IOM, as may nominating devices for high-priority assessment. How to translate assessment into prescription? More effective ways need to be worked out for applying the feedback from technology assessment to medical strategy and tactics, to actual conditions of device use, to reimbursement schemes, and to the flow of innovation. CURRENT RESEARCH DIRECTIONS Leo J. Thomas reviews the exciting variety of bioengineering research possibilities explored in the 1987 National Research Council report, Directions in Engineering Research (Washington, D.C.: Na- tional Academy Press). These include systems physiology and mod- eling, neural prostheses, biomechanics, biomaterials, biosensors, met- abolic imaging, minimally invasive procedures, and artificial organs. At the symposium, John Watson urged more research on generic technical questions, to strengthen the whole endeavor. Also he said we need much more work on product reliability. Robert Mann, speaking in part from his experience with natural and artificial hips, emphasized the need to study the fundamental interplay between biochemical factors and biomechanical ones in the body (such as interfacing surfaces, rapidly pressurizing and Repressurizing tissues, electrome- chanical systems, and the like). Some of this basic and general research is actively being pursued, but some, according to our authors, is not getting the support it needs. No doubt there are openings for further IOM/NAE surveys of research needs, perhaps in quite specialized areas. Also deserving of Academy attention might be aspects of medical device reliability criteria, reliability testing regimens, and quality assurance. We repeatedly hear assertions that, despite so much general

OCR for page 164
SUMMARIZI1!iG REFLECTIONS 173 attention to these themes in recent years, with some devices special problems arise that deserve scrutiny and systematization, both to facilitate regulatory evaluation and to ensure quality for ultimate use. IDENTIFYING NEEDS AND OPPORTUNITIES In some ways it appears that the most neglected step in the innovation scheme is that last long feedback loop: The one from the ultimate user community back to the start of the whole process. How, for instance, do practicing physicians, or nursing home operators, or, for that matter, hockey trainers or just folks, who perceive a health problem send the request back to the device enterprise: Please develop such-and-such a gizmo to relieve our problem? Of course, such feedback may be sensed by that abstract entity we call "the market," or it may be conveyed by various health care providers or advocates, or it may be anticipated in behalf of users by those in a position to innovate. Here again, there may be roles for the Institute of Medicine and the National Academy of Engineering. From time to time panels might take stock and brainstorm either over health needs in search of technological solution, or over emerging technologies in search of application. One suspects that initiatives might be defined that would aid our shift toward more effective ambulatory care, home care, and self-care, for example. The congressional Office of Technology Assessment has, by the way, conducted some excellent reviews, such as the recent survey, Life-Sustaining Technologies and the Elderly (Washington, D.C.: Office of Technology Assessmnent, U.S. Congress, July 1987~. Sam Thier has remarked that we have not given enough attention to screening techniques and kits, or to rehabilitative devices. Massive- scale screening is getting more attention now, in defending against AIDS and other infectious threats, and in providing employment- related screening. Could IOM/NAE critiquing help? ENVOI The only way to make sense of the enormous range of existing devices, variations on those devices, emerging devices, and contem- plated future devices is to collect appropriate data and make compar- ~sons. Once descriptive comparisons are established, the key to broad

OCR for page 164
174 HOW TRENDS WILL INTERACT evaluation is to construe the applications of devices, or indeed, any medical measures, as personal and societal investments. Always, the ultimate question is, Medical devices for what? At the outset of the symposium, no doubt anticipating the frustrations we would experience as we wrestled with this amorphous issue, Sam Thier affirmed straightforwardly: "The end, of course, is prevention of disease, correction of disease, and rehabilitation from disease."