D I S C U S S I O N P A P E R N O . 1 4 Influences on the “Health Care Technology Cost-Driver” by Steve Morgan, Ph.D. University of British Columbia Jeremiah Hurley, Ph.D. McMaster University August 2002 Catalogue No. CP32-79/14-2002E-IN ISBN 0-662-32784-5 Although the views expressed in the paper are those of the author(s), each of the papers was subjected to an independent peer-review process. The Commission would like to thank the Institute of Health Services and Policy Research (IHSPR) of the Canadian Institute of Health Research for their oversight and administration of the peer-review process for these papers. The work of the authors, the reviewers and IHSPR will serve to make these papers an important contribution to the Commission’s work and its legacy. Contents Highlights iv Executive Summary v Introduction 1 Future Technological Change in Health Care 2 Technology and the Baby Boomers 2 Consumer-Oriented Means of Promoting Technological Change 3 Health Technology in the Genomic Era 4 Genetic Testing 5 Pharmaceuticals 5 Pharmacogenomics to Tailor Drug Therapy 6 Pharmacogenomics to Expand Markets 7 Surgical Procedures 7 Diagnostic Imaging 8 Technology in Summary 8 Pricing Health Care Goods and Services 9 Technology Assessment and Economic Evaluation 11 Allocation Decisions 11 International Experiences: The Case of Pharmaceuticals 13 Moving Forward 15 Bibliography 16 Highlights • Technological change in the health care sector will be influenced by many factors in the coming decades; most important among the determinants of the utilization of technology will be genetic sciences (specifically genetic testing), demographic changes, and consumer-directed marketing. • Progress made in health care technologies over the coming decade will be important, but incremental, when measured in terms of improved population health. • Technological changes in health care will not be independent of demographic changes, especially the aging of the baby boomer cohort. • Increased consumer-oriented promotion of new technologies will be a major determinant of changes in the utilization of both new and old technologies in the coming years. • Predictive genetic testing may become a major cost-driver in the health care system in the coming decades. Products such as pharmaceuticals, and services such as diagnostic imaging, that are sold on the basis of genetic test results may account for a majority of the associated costs. • Continued upward trends in both the cost and utilization of drug treatments will make pharmaceuticals one of the fastest growing components of health care in the next decade. • To ensure that technologies are adopted prudently, health care policy should aim not only to evaluate the efficacy and cost-effectiveness of technologies, but also to influence the decisions of thousands of providers in millions of clinical encounters. • If Canadian policy makers wish to better control the inflationary pressures created by health care technologies, substantial reform in the organization and funding of health care services is needed. iv Executive Summary Technological change will consist not only of innovation – the arrival of new products and techniques – but also of changes in the utilization of both new and old technologies in the coming decades. Progress made in terms of innovative health care technologies will be important but incremental in terms of improvements in population health. Major “breakthrough” innovations may occur, but they are impossible to predict. Both the direction of technological innovation and changes in the utilization of technologies will be driven in part by three overarching influences: demographic change, genetic sciences, and consumer-directed marketing of health technologies. The baby boomers’ expected demand for health care goods and services changes the financial incentives that guide research efforts in bio-medical sciences; consequently, technological changes in health care will not be independent of demographic changes. The cost impact of this relationship may be significant given the investment in technologies to address the needs of the elderly, the number of individuals in the baby boomer cohort, and the expectations that baby boomers will carry into their senior years. Although unquestionably a revolution in thinking about the mechanisms of illness, the impact of genomic sciences on population health will be incremental over the next decade. The ability to identify individuals at elevated risk of illness will (necessarily) outpace the ability to “treat” the related genetic anomalies. Moreover, the complexity of biologic systems dictate that the logical search for “cures” to genetic disorders will be a long and laborious one. In the coming decade, the cost of genetic testing services and the goods and services sold on testing results may become a major cost-driver in the health care system. The bulk of the related expenses may not be the price of the tests themselves, but rather the cost of the products and services that accompany the testing process and that are sold on the basis of test results. Such complementary services include the consultations with health professionals that may proceed and follow testing, as well as ongoing consumption of products and services for the purposes of disease surveillance or prevention. Expanded use of diagnostic imaging will be particularly focused on individuals at high risk of illness, for which diagnostic and surveillance costs can be justified. Continued upward trends in both the cost and utilization of drug treatments in major therapeutic classes such as cardiovascular medicines will cause pharmaceuticals to be one of the fastest growing components of health care costs in the coming decade. Important determinants of pharmaceutical expenditure will come from increased “consumer oriented” marketing activities, and on the promotion of pharmaceutical disease management for risks of late-onset illnesses detected by genetic testing. To reward innovators in a manner that provides incentive for socially valuable research, the price of health care technologies protected by patent should reflect their therapeutic value relative to the existing arsenal of health technologies – both new and old. Moreover, efficient incentives for future innovation also require health care technologies to be used only when appropriate and cost effective. Unlike the case for ordinary goods and services, relying exclusively on consumers’ willingness to pay for health care technologies is neither a reasonable v means of establishing a fair price nor an acceptable means of allocating technologies amongst individuals. Health care technologies require expert evaluation and critical assessment, balanced and accurate information for both practitioners and patients, and tailored institutional structures to provide incentives for efficient and equitable allocation decisions. Through continued, if not more rigorous, technology assessment and evaluation, Canada will be able to negotiate prices to some extent, but the agreed upon prices will likely have to fall within a narrow “price corridor” established by international pricing precedents. Once prices are set, the appropriate and cost-effective allocation of technologies will depend, in part, on how technologies are chosen for individual patients. To ensure that technologies are adopted prudently, policy should aim to influence the decisions of thousands of providers in millions of clinical encounters. Ultimately, efficient allocation of health care technologies will depend on providing proper information and incentives at the point of the clinical encounter. The essential utilization decisions take place at that time. Consequently, if Canadian policy makers wish to better control the inflationary pressures created by health care technologies, substantial reform in the organization and funding of health care services is needed. vi Influences on the “Health Care Technology Cost-Driver” Introduction “Prediction is very difficult, especially if it's about the future.” Nils Bohr When asked about the importance of technological change to the future of the health care sector, economists might turn to the markets for answers. The message found there is clear, summarized in a recent headline from the business section of the The New York Times: “this decade belongs to health care” (Munger Kahn 2002). Average price earnings ratios in high-tech health sectors are currently about twice as high as those in other industries, indicating that investors believe that health technologies are poised to pay big dividends in the near future. These beliefs are no doubt fueled by the highly publicized enthusiasm over medical technology st in the 21 century. “Health Technology” is a regular feature in the daily news, particularly in the financial press, where a steady stream of featured breakthroughs promises to change the medical landscape, the health of the population, the financial fortunes of one firm or another, and the future cost of health care. The recent burst of the e-commerce “bubble” should give one pause when considering economic forecasts based on stock market activity. There are, however, good reasons to believe that health expenditures will be significantly influenced by “technological change” in the coming decades. This technological change will consist of innovation and utilization. Innovation is the arrival of new products and techniques; it is the flow of new ideas into the stock of available technology. Utilization concerns how both new and old technologies are used in our health care system; it is what is done with the stock of available technology. In this paper, we argue that both technological innovation and utilization will be driven by the following three overarching influences in the coming ten to twenty years: demographic change, genetic sciences, and consumer-directed marketing. We review expert opinions regarding likely innovation in selected components of the health care sector and relate these trends to the overarching influences of aging, genetics, and marketing. Following this, we discuss the “pricing” of health care technologies, highlighting apparent tensions between the incentives created by patents and the desire to regulate health care prices. We then examine issues in the evaluation and allocation of health technologies that are central to achieving efficient utilization of both new and old technologies. We conclude with recommendations as to how various levels of government can work to create the institutional structures necessary to achieve that end. -1- Influences on the “Health Care Technology Cost-Driver” Future Technological Change in Health Care “I have seen the future and it is very much like the present, only longer.” Kehlog Albran In past eras of profound change in basic scientific understanding, revolutions in the health of populations and the practice of medicine have not materialized as rapidly as experts predicted (Porter 2000). Consequently, the safest prediction for health care technology in the coming decade is that progress will also be incremental – important but incremental nonetheless if progress is measured in terms of improvements in population health. This is because technological progress generally moves in a logical sequence. Most new ideas, products, and techniques tend to build incrementally and somewhat predictably on existing ones. Breakthroughs will occur, but their nature and magnitude are virtually impossible to forecast. Indeed, anyone who could foretell a specific technology that would dramatically improve population health in the near future would be very rich. While dramatic shocks are almost impossible to predict, changes in the tides of technological progress may be foreseen if forces that influence the direction of scientific inquiry are known to be shifting. Moreover, changes in the rate at which existing technologies are used or new technologies adopted may also be predicted if related and identifiable trends could also be foreseen. In this regard, three major factors that will influence trends in technological innovation and utilization over the coming decade are demographic change, genetic sciences, and consumer-directed marketing of health technologies. Health care planning can (and should) adapt to these reasonably predicable influences – while remaining prepared for unexpected shocks in health care technology. Technology and the Baby Boomers Born between 1946 and 1965 in North America (and other developed countries), members of the baby boomer generation are just now entering their years of high health care utilization. The impact that these baby boomers will have on health care costs will not be independent of the availability, cost, and utilization of health care technology. Nor will the availability, cost, and utilization of health care technology be independent of the baby boomers’ aging. The total cost impact of population aging is a function of both changes in average health care needs as the population ages, and changes in the quantity, type, and cost of technologies used to meet those needs. When the latter factors are held constant, the needs-related impact of population aging has been a modest cost-driver in recent decades (Fuchs 1984; Barer, Evans et al. 1987; Barer, Evans et al. 1995; Evans, McGrail et al. 2001). It is likely to continue to be a modest contributor to health care costs in the coming decades. However, historical experience suggests changes in the availability, utilization, and price of technologies to meet the needs of the elderly population will be important determinants of the cost-impact of population aging (Evans, McGrail et al. 2001). Health technology innovation and utilization are related to demographic change because the health care “market” is changing dramatically as the baby boomer generation ages. The changes in the age-related needs of baby boomers increase the financial incentives that guide -2- Influences on the “Health Care Technology Cost-Driver” efforts for both the development and the promotion of health care technologies. Bringing to market products that can be expected to be in demand by aging baby boomers is good business. “That’s where the money is.” Baby boomers are not only larger in number than other generations, they are also more affluent and independent than previous cohorts of elderly health care consumers, and they bring to the health care marketplace relatively high expectations about healthy aging and consumer-oriented health care (Clark 1998; Dychtwald 1999). Significant investment has already been made to develop forms of technology that are likely to be in high demand by the baby boomer generation. Almost two thirds of drugs currently in development by American pharmaceutical manufacturers are intended to “lengthen and improve the quality of life for seniors” (PhRMA 2001a). Therapeutic “markets” currently experiencing rapid growth due to the boomer generation include treatments for hypertension, type II diabetes, high cholesterol, and arthritis pain (Scott-Levin 2001). Consumer-Oriented Means of Promoting Technological Change Some believe that increases in the consumer-orientation of health care may be one of the most significant challenges of health care in the coming years (Porter 2000). This shift is driven in part by the advent of new health technologies addressing needs that go beyond the conventional definitions of “health care” and “disease” (Moynihan, Heath et al. 2002; Smith 2002), and in part by a combination of increased affluence and increased access to information that has engendered higher expectations among health care consumers. It is also driven by changes in health care marketing practices, aimed at capitalizing on notions of “consumer empowerment” and “patient centered” health care (Mintzes 2002; Mintzes, Barer et al. 2002). Health care technologies have traditionally been promoted through marketing activities aimed at health professionals who make allocation decisions on behalf of, and in consultation with, patients. However, recent changes in the marketplace have caused companies – pharmaceutical manufacturers, in particular – to seek new means of promoting sales by marketing directly to consumers, in addition to conventional professional-directed marketing activities. Perhaps most notably, increased emphasis on expenditure controls by managed care organizations (in the US) and governments (elsewhere) appears to have forced manufacturers to seek audience with patients in order to promote the sale of particular brands (Pinto, Pinto et al. 1998; Morgan 2002a). A recent paper commissioned by Pfizer Inc. appears to confirm this by portraying consumer directed advertising as a necessary means of countering unduly restrictive policies imposed by insurance providers (Rubin and Schrag 1999). While most forms of direct-to-consumer advertising for prescription – drugs are illegal in all but two countries – the United States and New Zealand – the trend toward consumer-oriented marketing is unmistakable (Mintzes 2002; Morgan 2002a; Rosenthal, Berndt et al. 2002). A watershed in this trend came in 1997, when the US Food and Drug Administration relaxed restrictions on television and radio advertisements. This opened the door to an explosion in promotional activities aimed at US consumers that now approaches $3 billion per year – much of which is viewed by Canadians, though these marketing practices remain illegal here (Mintzes, Barer et al. 2002). Pharmaceutical manufacturers are lobbying for the opportunity to advertise to consumers in other countries, including Canada. Companies holding patents on predictive genetic tests will not be far behind. -3- Influences on the “Health Care Technology Cost-Driver” The consumer-orientation of health care marketing will be a significant factor in determining changes in the utilization of health care technologies in the coming decades. There will undoubtedly be an increase in the use of those technologies most heavily promoted directly to consumers: if this were not the case, firms would not engage in such marketing practices. Health Technology in the Genomic Era While human genetics have been studied for some time, the complete mapping of the human genome is said to have launched “the era of post-genomic science” in which many claim that virtually all aspects of medicine will change (Baltimore 2001; Collins and Guttmacher 2001; Collins and McKusick 2001). The promise of this new era is far reaching. When leading scientists were recently asked to report on the prospects of medical research in the twenty-first century, genetics was cited as central to developments in the treatment of most diseases – including cancer (Livingston and Shivdasani 2001), cardiovascular illnesses (Lefkowitz and Willerson 2001), neurological and psychiatric illnesses (Cowan and Kandel 2001), diabetes (Olefsky 2001), and autoimmune diseases (Koopman 2001). Genetic information was also cited as a major impetus for advances in diagnostic imaging (Tempany and McNeil 2001), and pharmacologic and biologic therapy (Bumol and Watanabe 2001; Kaji and Leiden 2001). The Holy Grail in the genetic era of health care are products that will make it possible for “Physicians […] not only to use therapies to help patients live better with their genetic constitutions, but also [to] use novel therapies to alter the genetic makeup of the patient” (Kaji and Leiden 2001). Despite the high expectations of many involved (or invested) in the race for such discoveries, radical genomic therapies will not appear on our markets, or in our health care systems, within the next decade. Innovations will occur, but they are likely to come gradually. As the genes and proteins related to more illnesses are identified, the amount of genetic “data” is increasing dramatically; the task of turning that data into useful information, and ultimately useful treatments, will be one that takes decades. For example, the list of potential targets for drug therapy is expected to balloon from the 500 biological receptors currently targeted by conventional pharmacological therapy to as many as 30,000 targets for bio-pharmaceuticals and genetic therapies (Drews 2000; Horrobin 2000; Bumol and Watanabe 2001; Lemonick 2001). It has been questioned whether this impressive gain in “data” has been a case of finding more needles or bigger haystacks (Horrobin 2000). Even when the genetic origins of illnesses are known, “cures” are hard to find. The genes that cause single-gene disorders such as sickle cell disease, cystic fibrosis and muscular dystrophy have long been known; yet, no cures have been found for them (Baird 2002). As we look to detect the genetic causes of and find cures for common illnesses, the complexity of the problem increases substantially. Most common illnesses are caused by such a complex combination of environmental and multiple genetic factors that few (if any) could be fairly regarded as “genetic disorders” (Baird 2002). The number of genes with links to illness, and the complexity of the biological systems in which they play out, create analytic challenges so immense that “bio-informatics” has emerged as a new sub-discipline of computer science to aid in the analysis of related data (Pennisi 2001). In January 2000, for example, IBM launched a 5- year, $100 million dollar project to develop supercomputers to be dedicated to bimolecular simulation (IBM 1999). These developments portend the kind of obstacles that must be -4