The Costs and Benefits of Medical Innovation
An era of health care cost containment has come to an end. Once again concerns are expressed about rising health care costs and medical innovation is seen as an important driver of health care costs. Two presentations at the conference directly addressed the extent to which medical innovation drives up health care costs. One presentation looked at increases in total spending over the past 50 years, and the other looked at recent increases in aggregate prescription drug spending over the last few years.
There appears to be much greater emphasis in the public debate on the costs than the benefits of health care. An example of this perspective is a Washington Post editorial (Washington Post, 2001) “Back to Health Care Costs” published shortly before the conference. It stated:
Higher health care costs are like a tax increase or an increase in energy costs. They leave people, businesses and government—government in its role as major payer of medical bills—with less money to spend for other purposes.
The editorial focused on costs as if all health care expenditures are investments that evaporate and do not bring any value to the population. To redress the balance, several presentations sought to place a value on the benefits of medical innovation.
Finally, four speakers examined some recent developments in treating cardiovascular disease and metastatic melanoma and the cost implications of these developments.
ACCORDING TO CMS RESEARCHERS, TECHNOLOGY IS PRIMARY DRIVER OF HEALTH CARE COSTS
Researchers at CMS find that technological change has been the largest single driver of growth in health care spending over the past 50 years.
An estimated $4,660 per person was spent on health care in 2000—an increase of 838 percent from the $497 spent in 1950, assuming constant dollars. Sheila Smith, an economist with CMS, sought to evaluate the contribution of technological change, aging, insurance coverage, and other factors to historical growth in real per capita health care costs. Her approach was to identify the nontechnological factors contributing to growth in health spending and then to estimate their contribution, given a constant state of medical technology. The residual growth is then attributed to medical technology. The nontechnological factors taken into account included demographic factors (population growth and aging), relative medical price inflation, rising insurance coverage, increasing disposable income effects, supplier-induced demand and avoidable administrative costs. The last factor is defined to be the unnecessary costs associated with institutional structures within the health care sector.
After taking account of all these nontechnological factors, the residual implies that approximately 2.2 percent annual growth in real per capita health spending can be attributed to technology—the estimation range is 1.9 to 2.9 percent. Expressed another way, technological change has accounted for about half the real growth in health care spending over the period 1950-2000. This estimate of the impact of technological innovation is in line with earlier studies (Newhouse, 1992; Cutler, 1995).
INCREASED USE OF EXISTING DRUGS IS MOST IMPORTANT DRIVER OF AGGREGATE DRUG COSTS
Much attention in recent years has been devoted to increases in aggregate prescription pharmaceutical cost, even though prescription pharmaceuticals constitute only a small share of total health costs, less than 10 percent. Alison Keith, until recently an economist with Pfizer, Inc., examined the 13.6 percent increase in pharmaceutical spending in 2000 over 1999 (IMS Health, 2001):
Unit price increases of existing products accounted for 3.9 percentage points—not very different from the increase in the overall Consumer Price Index for all goods and services.
The biggest component, 7.5 percentage points, is attributable to increases in utilization of existing products.
The remaining component, 2.2 percentage points, is attributable to the cost of new products.
Keith explained that other recent years show a similar pattern and that this increased volume of pharmaceutical utilization arises primarily from treating more patients and applying new science (Dubois et al., 2000):
More patients—reflecting an aging population with more chronic conditions and co-morbidities, and a narrowing of the gap between prevalence rates and treatment rates for many diseases, in part due to a more widespread awareness of specific conditions and better detection and diagnosis.
New science—encompassing new understanding of disease processes and the importance of specific treatments, and new best practices in the clinic.
Keith suggested that this increased utilization can be understood to reflect a greater recognition of the value of prescription pharmaceuticals, where their direct costs are viewed in the light of both health and economic contributions, which often include offsets in other health care costs (Lichtenberg, 1996, 2001) and improvements in workplace productivity (Kessler et al., 2001).
THE VALUE OF INCREASED LIFE EXPECTANCY OVER 1970–1990 IS ENORMOUS
University of Chicago economists Kevin Murphy and Robert Topel sought to evaluate the social benefits of medical research by placing a value on aggregate improvements in longevity (Viscusi, 1993; Tolley et al., 1994; Cutler et al., 1998; Cutler and Richardson, 1999; Lasker Foundation, 2000; Topel and Murphy, Forthcoming). The first task was to estimate what an average American would agree to pay for a reduction in mortality risk that would add a year to his/her life. Murphy and Topel used data on what workers are paid in occupations with differing risks of job-related death to estimate the value of an additional life-year to be about $150,000, a figure that varies with age.
Over the period 1970–1990 increases in the life span of an average American have been significant. For example, the increase in the life span of a typical 40-year-old person is more than three years. Using age-dependent values of an additional life-year and the increases in life expectancy over this period, Murphy and Topel attribute a value of roughly $57 trillion or about $2.8 trillion per year to the increased life expectancy, indicating the public values improvements in health very highly. To put these figures
in perspective, improvements in life expectancy over the period 1970–1990 contributed about as much to overall welfare as did improvements in material wealth.
Kevin Murphy pointed out that investment in medical research has brought significant returns. In 1995, according to NSF calculations there were about $35 billion in investments in medical research. The gain in health, as measured by the value of added longevity, is about 50-100 times what we spend on research, even taking into account the fact that health improvements are due to a variety of factors.
Looking forward, Murphy said that potential future gains will also be very large. For example, eliminating cancer is worth roughly $47 trillion. Further, the economic value of disease reduction is increasing significantly over time. The value of disease reduction rises as the wealth of the population increases. In addition, the value of progress against any one disease rises as we make progress against other diseases. For example, as we have made progress against heart disease and, hopefully, make progress against cancer, the value of curing/mitigating Alzheimer’s disease increases. The reverse is also true. Progress against Alzheimer’s disease makes further progress against cancer or heart diseases much more attractive because of a better life in those later years as well as more years to live.
MAJOR RETURNS ON INVESTMENT IN MEDICAL TECHNOLOGY FOR CARDIOVASCULAR DISEASE
David Cutler, a Harvard University economist, explained that life expectancy has increased 9 years since 1950 with about half of this increase resulting from reduced mortality from cardiovascular disease. These successes in treating/preventing cardiovascular disease can be attributed to developments in the intensive treatment of heart attacks, new medications for chronic heart disease (hypertension, cholesterol, angina), and behavioral changes (less smoking, reduced fat intake, decline in heavy drinking). These developments, including the behavioral changes, are products of medical research.
To determine the return on medical care and basic research (Cutler, Forthcoming), Cutler attributed roughly one-third of the benefits to developments in intensive treatment, roughly one-third to new medications, and the remaining third to behavioral changes. For someone 45 years old the total increase in longevity is about 5 years since 1950, of which about 41/2 years is a result of reduced cardiovascular disease mortality, with 3 years from medical treatments and 11/2 years from behavioral changes. For someone 45 years old the average cost of medical treatment on cardiovascular disease is $30,000 in present value terms. The costs of providing behavioral advice are much less—David Cutler estimated about $1,000 to
cover the costs of research and consultation with health care professionals. For the purpose of estimating benefits, Cutler assumed an extra year of life to be worth $100,000.11
For the return on medical care, there is a cost of $30,000 in exchange for three extra years. These three extra years are not valued at $300,000 but at $120,000 because some of the benefits occur in the future and need to be discounted. Even so, the return for medical care is very large, on the order of 4 to 1. For the return on behavioral changes, there is a cost of $1,000 in exchange for just over an extra year. The discounted value of this extra time is $30,000. Thus, the return on behavioral changes (30:1) is much higher than the return for medical care.
SIGNIFICANT POTENTIAL BENEFITS FROM MELANOMA PREVENTION PROGRAMS
As Cutler pointed out in his presentation, life style changes have brought about significant reductions in cardiovascular deaths. Life style changes can also have an impact on the incidence of melanoma. Margaret Tucker of the National Cancer Institute said that although the incidence of melanoma is increasing, it is a disease that can be prevented by decreasing sun exposure. To achieve this, major cultural issues need to be addressed since having a tan is an important part of “looking healthy” in American culture. These cultural problems have been successfully addressed in Australia where considerable investment in a prevention program has resulted in melanoma incidence rates leveling off, possibly even decreasing. The Australian program taught the need for sunscreens and protective clothing and led governments to provide shade at nearly all outdoor pools and school playgrounds.
Tucker also said that secondary prevention/early detection is practicable. In Australia, it has been estimated that a family practitioner doing a 2-year screening for adults over 50 costs about $12,000 per male life-year saved, and $21,000 per female life year saved (Carter et al., 1999). In America, it has been estimated that a one-time screen by a dermatologist with treatment would cost $29,000 per life year saved (see below). These costs would decrease for targeted screening.
Robert Young of the Fox Chase Cancer Center said that screening for melanoma is still controversial primarily because it has not been fully assessed through a randomized control trial (RCT). Nevertheless screening is widely carried out. The American Academy of Dermatology, the American Cancer Society, and the NIH Consensus Conference all endorse regular
screening, while the Canadian task force on periodic health examination endorses screening for high-risk patients. On the other hand, the U.S. Preventive Services Task Force and the International Union Against Cancer do not endorse screening.
The issue of whether screening is cost-effective was addressed in a recent study by Freedberg et al (1999). This study examined whether no screening or a single one-time screen by a dermatologist could be cost effective for high-risk patients. The study found that it is cost-effective but highly dependent on the initial cost. If the screen costs $30 then the cost per life year saved is $29,170. However, if the screen costs $120 then the cost per life year saved is $110,000, a considerably higher figure whose acceptability is debatable.
In response to a comment that there is under investment in prevention research funding as compared to diagnostics/treatment research funding, Mark McClellan of the Council of Economic Advisers speculated whether the right reimbursement incentives were in place. Health care providers are generally paid more for doing more, for treating complications, and for treating the consequences of poor preventive care. John Ford, of the House Committee on Energy and Commerce minority staff speculated that the perceived under-funding of prevention research might move Congress to encourage more innovation in the area of prevention.
LOOKING TO THE FUTURE: COSTS OF TREATING CARDIOVASCULAR DISEASE LIKELY TO RISE
In a presentation on the future costs of treating cardiovascular disease, Dan Mark of Duke University observed that heart failure is a very important epidemic condition in the United States. As age-specific mortality is falling in cardiovascular disease the incidence of heart failure may be increasing. About 4.7 million people in the United States now have heart failure and a little over half a million new cases are added each year. The treatment options are palliating the symptoms, drugs that improve the prognosis, disease management (a low-tech collaborative approach), or attempting to reverse the heart failure state. The latter could involve giving the patient a new heart or inserting a Left Ventricular Assist Device (LVAD), a mechanical device similar to an artificial heart that is designed to increase the efficiency of the cardiovascular system.
Currently, about 2,300 heart transplants are carried out a year in the U.S. at a lifetime cost of about $300,000 per patient, resulting in an annual expenditure of $700 million. The number of heart transplants is limited by the total number of donated hearts. This has been stable for a number of years and is unlikely to increase anytime in the future. If an LVAD could be used instead of a heart transplant then potentially another 40,000–50,000 patients could benefit from such a device.
Currently, LVADs are approved as a bridge to a heart transplant to keep severe heart failure patients alive as they await for a transplant. LVADs cost in the range $50,000-75,000 excluding the costs of implantation and maintenance. If LVADs move from “bridge to transplant” use to standalone left ventricular support use not necessarily anticipating transplant then the device could have a huge economic impact. If the costs for LVADs were in the range $100,000-200,000 then that would add $5 billion-10 billion to annual health care costs. An ongoing clinical trial is testing this strategy. Its outcome is uncertain, but there is certainly the potential for explosive growth in the cost of the care of patients with heart failure.12
Mark concluded by saying that new technologies, most of which tend to be expensive (for example, LVADs for heart failure), and the aging of the U.S. population are going to drive up costs of cardiovascular care. There is always the potential for new technology to improve efficiency but, in Mark’s view, the U.S. system is too fragmented to take advantage of money-saving innovations.
NEW THERAPIES FOR METASTATIC MELANOMA ARE EXPENSIVE
Mike Atkins of the Beth Israel Deaconess Medical Center, Harvard Medical School, reported on developments in the treatment of metastatic melanoma. He said the disease had a bad prognosis—a median survival of 6-10 months and less than 5 percent of patients survive 5 years. Traditional approaches to treating cancer—surgery, radiotherapy, and chemotherapy— have not been successful. Although chemotherapy can produce tumor shrinkage in a small percentage of patients, these responses are usually of short duration. Overall it is unclear whether chemotherapy produces a survival advantage over simple observation. Clinical cost-effectiveness is very low.
Immunotherapy is currently the most promising therapy for metastatic melanoma. High dosage Interleukin-2 (HD IL-2) is very effective for a small subset of patients. Criteria for identifying this responsive subset are currently lacking. HD IL-2 is, however, costly and requires in-patient delivery. Typical costs per patient are $52,000, and CMS only reimburses up to $18,000. As a result, some major centers do not to treat metastatic melanoma patients, even those who can afford to pay for themselves. In addi-
tion, research efforts to improve IL-2 therapy have been hampered or even curtailed.
Bruce Hillner of the Medical College of Virginia reported on a study (Hillner et al., 2001) that confirmed the high cost of current treatments for metastatic melanoma. The study reported on an audit of the records of 100 consecutive new patients with metastatic melanoma at the University of Pittsburgh Cancer Institute (UPCI) after January 1997. An exceptionally high proportion (84 percent) of the group of patients participated in clinical trials—49 percent in Phase I trials, 10 percent in Phase II trials, and 25 percent in Phase III trials. In terms of therapies, 75 percent of the group received immunotherapy, 50 percent chemotherapy, 44 percent radiotherapy, and 23 percent surgery. Using assigned costs for the identified resources, the average cost per patient was $59,400. This figure represents a lower bound on the costs of treating the disease, since it omits the diagnostic costs prior to referral to UPCI and the costs of supportive care at the end of life. At the time of the analysis 82 percent of the patients were known to have died.
SOME RECENT DEVELOPMENTS IN IMMUNOTHERAPY FOR TREATING METASTATIC MELANOMA
Steven Rosenberg of the National Cancer Institute reported on his work to develop peptide vaccine strategies in combination with other therapies for the treatment of metastatic melanoma. He said that in the last decade we have seen the development of a fourth approach (after surgery, radiation therapy and chemotherapy) to cancer therapy—immunotherapy or biologic therapy, aimed at stimulating the body’s defenses to defeat cancer (Rosenberg, 2001). The use of IL-2 for in-patient treatment of metastatic melanoma and other cancers is the best example that immune stimulation can result in cancer regression. It is possible to incubate cancer cells in the highest achievable concentration of IL-2 and they will grow normally. All of the impact of IL-2 derives from its ability to stimulate the body’s immune system.
To develop immunotherapy further, a molecular understanding of the process is needed, particularly, an identification of the antigens involved in cancer regression. Using tumor-infiltrating lymphocytes, first identified in the late 1980s, tumor antigens in melanoma and other cancers have been discovered. Rosenberg said that these discoveries have opened up opportunities for new approaches for treating cancer patients by using their own immune systems. For example, peptides that mimic tumor antigens can be used to vaccinate patients and evoke an immune response. In pilot trials, the response rate to IL-2 has been doubled using peptides in conjunction with IL-2. A nationwide RCT is now evaluating IL-2 as compared with
IL-2 plus peptide. Regarding the economics of this peptide treatment, Rosenberg said that under GMP (Good Manufacturing Practices) conditions synthesizing enough peptide to treat 1,000 patients costs about $12,000.
Jonathan Lewis said that his company, Antigenics, is a relatively new biotechnology company specializing, among other things, in developing immunotherapy products.13 He said that building on the work of Pramod Srivastava and others, laboratory researchers had shown that treating animals with cancer with autologous tumor-derived heat shock protein molecules can generate an immune response leading to favorable results in terms of both survival and tumor regression. This had been demonstrated for a wide range of histologies and several different methods of inducing the cancer (see for example Tamura et al., 1997).
Lewis said that translating results in animals to humans is a big step. First, there is inadequate species molecular homology, in other words, humans are very different from mice. Second, laboratory experiments are carried out on very inbred strains of mice, whereas humans are very heterogeneous. These caveats are bypassed by heat shock protein biology. Researchers have demonstrated that it is possible to prepare heat shock protein vaccines for humans, that these vaccines are safe and tolerable for humans, and that the use of vaccines has elicited documented anti-tumor activity in humans. Regarding the latter point, in a recent uncontrolled trial carried out at the M.D. Anderson Cancer Clinic in which late stage and heavily pre-treated melanoma patients were given a heat shock protein vaccine, there was 95 percent survival at a median follow-up of 14 months in adjuvant patients, and 50 percent survival in residual disease patients.14 These were better results than had been previously seen at this center. A comparable study at the National Cancer Institute, Italy, showed similar findings.15 In addition, they saw some patients undergo a complete response, that is, all their cancer went away.
Lewis concluded by saying that studying and understanding melanoma will help gain a much better understanding of many other different types of cancer. For example, over the past year researchers at Antigenics had observed many important similarities between melanoma and renal and pancreatic cancer.
Carter, Rob, Robin Marks, and David Hill. 1999. Could a national skin cancer primary prevention campaign in Australia be worthwhile?: An economic perspective. Health Prom Int 14:73-82.
Cutler, D.M. 1995. Technology, Health Costs, and the NIH, Harvard University and the National Bureau of Economic Research. Paper prepared for the National Institutes of Health Economics Roundtable on Biomedical Research, September 1995.
Cutler, David M., Mark McClellan, Joseph P. Newhouse, and Dahlia Remler. 1998. Are medical prices falling? Quarterly Journal of Economics 113(4):991-1024
Cutler, David M., and Elizabeth Richardson. 1999. Your money and your life: The value of health and what affects it. In Alan Garber, ed., Frontiers in Health Policy Research, Volume 2, pp. 99-132.
Cutler, David M. Forthcoming. Better Medicine.
Dubois, Robert W., Anita J. Chawla, Cheryl A. Neslusan, Mark W. Smith, and Sally Wade. 2000. Explaining drug spending trends: Does perception match reality? Health Affairs 19(2): 231-239.
Freedberg, Kenneth A., Alan C. Geller, Donald R. Miller, Robert A. Lew, and Howard K. Koh. 1999. Screening for malignant melanoma: A cost-effectiveness analysis. J. Am. Acad. Dermatol. 41:738.
Hillner, Bruce E., John M. Kirkwood, and Sanjiv S. Agarwala. 2001. Burden of illness associated with metastatic melanoma: An audit of 100 consecutive referral center cases. Cancer 91(9):1814-1821.
IMS Health. 2001. Pharmaceutical Pricing Update. Vol. 8. No. 4. (June 2001). Plymouth Meeting, PA.
Kessler, R.C., P.E. Greenberg, K.D. Mickelson, L.M. Meneades, and P.S Wang. 2001. The effects of chronic medical conditions on work loss and work cutback. Journal of Occupational and Environmental Medicine. 43(Suppl. 3):218-225.
Lasker Foundation. 2000. Exceptional Returns: The Economic Value of America’s Investment in Medical Research. New York, NY.
Lichtenberg, Frank R. 1996. Do (more and better) drugs keep people out of hospital? American Economic Review (May):384-388.
Lichtenberg, Frank R. 2001. Are the benefits of newer drugs worth their cost? Evidence from the 1996 MEPS. Health Affairs 20(5):241-251.
Newhouse, J.P. 1992. Medical care costs: How much welfare loss? Journal of Economic Perspectives 6(3):3-21.
Rosenberg, Steven A. 2001. Progress in human tumour immunology and immunotherapy. Nature 411:380-384.
Tamura, Yasuaki, Ping Peng, Kang Liu, Maria Daou, and Pramod K. Srivastava. 1997. Immonotherapy of tumors with autologous tumor-derived heat shock protein preparations. Science 278(5335):117-120.
Tolley, George, Donald Kenkel, and Robert Fabian, eds. 1994. Valuing Health for Policy: An Economic Approach. Chicago: The University of Chicago Press.
Topel, Robert H., and Kevin M. Murphy. Forthcoming. The Economic Value of Medical Knowledge. In Exceptional Returns, Robert H. Topel and Kevin M. Murphy, eds. Chicago: University of Chicago Press.
Viscusi, W. Kip. 1993. The value of risks to life and health. Journal of Economic Literature, 31(4):1912-1946.
Washington Post editorial. June 9, 2001. Back to health care costs.