Barriers to Medical Innovation
The Conference also addressed the drivers of medical innovation and, at much greater length during several panels, the barriers to medical innovation.
Key drivers of innovation were considered to be:
a high level of public interest in health care issues;
strong public support for increasing NIH research funds;
substantial and increasing private investment in medical R&D, although private sector interest waxes and wanes depending on the attractiveness of other investment opportunities;
the aging population—a side effect of better health—moving the focus to other diseases (e.g. Alzheimer’s disease); and
public expectations in the United States that patients should receive the best quality science.
The conference discussed several barriers to medical innovation. These have been clustered under three headings—technical-level barriers, public policy barriers, and high-level political/economic barriers.
Regarding barriers to innovation, the focus of the conference was public policy and broader political barriers. Nevertheless, conference speakers
mentioned a number of technical barriers—two of these related specifically to cancer therapy and two were of a more general nature.
Inadequate understanding of the biology of cancer. Bruce Scharschmidt of Chiron made the point that through genome sequencing we have an abundance of targets. The key issue now is gaining a better understanding of the corresponding biology of these targets. We also need better information management techniques to handle data.
Poorly predictive pre-clinical models for cancer therapies.18 Scharschmidt also said that there are good animal models for developing drugs for Type 2 diabetes, but the same does not pertain to cancer products. Given the more sophisticated understanding of the molecular basis of cancer, there is not a comparatively sophisticated set of pre-clinical models. This is an opportunity for research investment.
Inadequate effort devoted to effectiveness analysis. Scharschmidt observed that cost-effectiveness is generally not rigorously assessed during the course of development. Drug and device development is an expensive and lengthy process. It is hard for companies to justify further dollars and time in cost-effectiveness studies, particularly as there are no agreed-upon set of measures by the industry and payers. As result, there is an opportunity to develop a stronger and commonly agreed upon scientific foundation for cost-effectiveness measures and studies.
Not enough patients entering RCTs (randomized controlled trials). Bruce Hillner of the Medical College of Virginia said that cancer patients are less likely than cardiovascular patients to be enrolled in RCTs. He identified several reasons for this. Patients are reluctant to accept the default arm of trials, often not considered an equivalent therapy. For many patients with cancer, treatment is a “one-shot chance” and they want to take the option recommended by their physicians. Further there is less reliance on evidence-based medicine in cancer treatment and too great a tendency for premature adoption of therapies based on presentations at major conferences. Finally, managed care plans generally refuse to encourage participation in trials.
PUBLIC POLICY BARRIERS
Reimbursement policies not friendly to innovation. David Lawrence reported on how the IOM Roundtable on Health Care Quality (Chassin et al., 1998) documented three types of quality problems—underuse, overuse,
and misuse. The underuse and overuse of medical technologies suggest that the right incentives may not be in place.
Both Lawrence and Laurel Sweeney of Philips Medical Systems pointed out that fee-for-service payment systems reward individual acts by individual people. They do not support very well integrated delivery capabilities increasingly necessary to treat a wide range of chronic conditions, such as congestive heart failure and diabetes. Sweeney reported, however, that there are some hopeful signs. The University of Maryland Medical Center is to carry out a study funded by CMS to test the cost-effectiveness of disease management services for congestive heart failure. The study will follow about six hundred patients. Half of the patients will receive traditional care focusing on the patient’s medication compliance and self-reported vital signs monitoring. The remaining patients will be split between the two test groups. Patients in one group will receive ongoing home visits from a nurse who will monitor their diet/nutrition and medications as well as keep track of their weight and blood pressure. Patients in the other group will use Philips Medical Systems’ in-home monitoring system to monitor their weight, blood pressure, and pulse, with the information being transmitted electronically to a computer in the doctor’s office.
Mike Atkins of Beth Israel Deaconess Medical Center and Harvard Medical School reported that Medicare does not fully reimburse the costs of HD IL-2 (high dose Interleukin-2) for the treatment of metastatic melanoma. As a result hospitals are not offering this treatment for metastatic melanoma and research into using HD IL-2 to treat metastatic melanoma is being curtailed. Several hospitals are not providing HD IL-2 treatment for metastatic melanoma even for those who can afford to pay, out of concern for equity issues, raising an ethical question on the right to the access of care.
Scharschmidt pointed out that currently there is no Medicare coverage for self-administered or injectable products unless there is an intravenous equivalent.19 This results in the anomalous situation of having two cancer therapies, herceptin and tamoxifen, for example, which are treated quite differently. Herceptin, which has to be given by physician/provider by intravenous injection, is reimbursable, whereas, tamoxifen, which is a tablet, is not reimbursable. There are currently about 30 oral anti-cancer products that are not reimbursable.
Rapidity of technological change threatens the ability of federal agencies to cope. John Ford of the House Committee on Energy and Commerce
minority staff emphasized the paradigm shift taking place in medical science/innovation. Investments in innovation are increasing—the budget of NIH is being doubled; and, in parallel, private sector R&D expenditures are increasing. Different types of innovation are occurring, for example, tissue engineering and genomics/proteomics. Further, individual elements of therapies are continually being improved requiring additional regulatory approval.
The implications of this are threefold. First, the flow of innovation is going to stress if not overwhelm the regulatory system. Second, the knowledge base of the regulatory agencies will need broadening. In particular, there is inadequate understanding among regulators of how multiple therapies are used in practice, leading to inappropriate regulatory practices for combination therapies. Third, new public policies may be needed to regulate highly targeted biologics with potentially different economic structures than those of standard drugs.
Ford said that a factor that could have an important bearing on the FDA’s ability to cope is the future of user fees for New Drug Applications. When demand at the FDA for approval of new drugs increases, the extra fees allow more resources to be made available. The Prescription Drug User Fee Act (PDUFA) of 1992 provided the FDA with increasing levels of resources for the review of human drug applications. The original act expired September 30, 1997, but the FDA Modernization Act of 1997 amended and extended PDUFA through September 30, 2002. The post-September 2002 arrangements for paying for New Drug Applications are currently under discussion.
Regulations inhibit innovation and are costly to implement. Lawrence in his keynote speech referred to excessive regulation of the health care industry. For example, CMS has 130,000 pages of rules, regulations, and guidelines whereas the IRS has only 10,000 pages. Kaiser Permanente did some very preliminary estimates of what it cost the HMO to deal with local, state, and national regulations in health care, and found that somewhere between 5 and 7.5 percent of the total annual revenue stream is devoted to meeting regulatory requirements. For Kaiser Permanente, that is a regulatory burden of close to a billion dollars each year.
Laurel Sweeney of Philips Medical Systems spoke about problems that might arise with the Health Insurance Portability and Accountability Act (HIPAA) of 1996 and the ensuing privacy standards released in April 2001. HIPAA’s privacy regulations pre-empt any state from enacting laws that are contrary to the Act. However, the privacy regulations do not prevent states from enacting more stringent requirements, which some states are already doing. Sweeney believed that the end result would be a patchwork of state
laws, which could, in some cases, affect the ability to conduct clinical trials, resulting in a negative impact on innovation.
Public policy changes a major uncertainty for venture capitalists. With regard to investing in medical innovation, Brandon Hull of Cardinal Partners said that although venture capitalists understand how to evaluate technology and development risks, it is difficult for them to assess the degree to which the public policy rules will change. Hull said that to foster innovation public policies must be consistent in order to have a salutary effect on prices, the investments required, and the timeframe from idea to marketplace. He asserted that price controls are disincentives to investment and speculated that a mandatory pharmacy benefit under Medicare would likely be accompanied by price controls and therefore could be a disincentive to investment and innovation. Shortening the life of patents is also a disincentive to investment. In addition, a consistent approach to coverage and reimbursement decisions is desirable.
In recent years investment costs have been increasing. It has been 10 years since the discovery of the cystic fibrosis gene and there is still no therapy derived from this discovery. The timeframes and the amounts of money needed to develop genomic therapies are still unknown. The requirements for clinical trials and regulatory review are important determinants of overall development timescales—up to 8 years out of a typical 15-year development cycle for a new drug. Hull said that recent public policy changes have helped with carrying out clinical trials, and over the past few years the time for regulatory review has also improved.
Older public policies perhaps no longer providing the right incentives. Susan Foote of the University of Minnesota commented that over the years Congress has been very active in establishing medical technology policy. Major initiatives in the 1990s included the establishment of the National Institute of Bioimaging and Bioengineering (1993–2000), the FDA Modernization Act of 1997, Biomaterials reform in 1996, ongoing Medicare coverage and coding reform and conversion of the Agency for Health Care Policy and Research (AHCPR) into the Agency for Healthcare Research and Quality (AHRQ). Sweeney alluded to the more recent Balanced Budget Refinement Act of 1999, which enacted fundamental reform of the hospital outpatient payment system including the transitional pass through for medical technologies. This was followed a year later by the Benefits Improvement and Protection Act which extended the pass through to hospital inpatient technologies.
Against this background of legislation and in the light of the accelerated pace and the broadening scope of medical innovation, Ford thought it might be appropriate to review some older elements of medical technology policy, specifically the Bayh-Dole Act of 1980, the Orphan Drug Act of 1983, and the Waxman-Hatch Act of 1984, to see whether these laws are
still achieving their desired public policy goals. Both the Orphan Drug Act and the Waxman-Hatch Act were enacted in the early days of the biotechnology industry, and Ford questioned whether the incentives written into these two laws were still economically relevant.
Policies for managing conflicts of interest may end up inhibiting innovation. Tom Fogarty of Stanford University suggested that too many rules devoted to managing conflicts of interest in the care giving setting at the interface of the patient’s needs and research objectives, may end up inhibiting innovation.20 He observed that academics and clinicians have quite different objectives with regard to innovation. Academics’ interests are focused on exploring new theories and carrying out experiments in laboratories while clinicians’ interests are focused on using science for clinical purposes and assessing clinical utility. Against this background, he said concerns about conflicts of interest had become pervasive, particularly in academic medical centers. Patients want to go to the physician who is utilizing the best technology to treat their disease. That physician is most likely involved in innovation, and as such, is often prevented from treating the patient because of the “perception” of monetary gain through the physician’s work in an innovative area. In such circumstances, Fogarty believes that the physician’s interest in providing the “best” treatment for his/her patient is considered suspect, when in fact his or her actions are predicated on treating the patient with technology that is in “the patient’s best interest.” In Fogarty’s view such conflicts of interest “are inherent to our very existence and represent a critical element in all relationships.” Attempts to legislate honesty and integrity or lack thereof will not work. Honesty and integrity should certainly be monitored, albeit at the local level.
HIGH-LEVEL POLITICAL/ECONOMIC BARRIERS
Congressional reluctance to address health care issues. Congressional reluctance to address health care issues is understandable given the highly technical nature and complexity of the issues and the fact that legislators and their staffs often lack the knowledge base to fully address the issues. Further, it is increasingly difficult for politicians to benefit from engaging
health care issues. Jamie Robinson of the University of California at Berkeley pointed out that government, like all the other stakeholders in the health care industry, is totally disenchanted with managed care (Robinson, 2001). He believed the key lesson of the 1990s for government, as for the other stakeholders, is not to get between the consumer and what the consumer wants to consume. Political capital cannot be gained by attempting to allocate scarce health care resources. From an electoral perspective it is better for politicians to criticize the industry from the periphery.
Responding to Lawrence’s comments about fragmentation on the health care delivery side, Susan Foote said that the political side was also fragmented. Institutional changes had occurred internally in Congress, and entrepreneurial politics had got in the way of developing comprehensive solutions to health care problems. Foote pointed out that this fragmentation is compounded by the piecemeal nature of medical technology policy. Despite the pervasiveness of medical technology policy, each major piece of legislation is responsive to different social and political pressures. Moreover, legislative responsibilities of the various House and Senate committees are jealously guarded in Congress. Sometimes these policies conflict, as for example, in the case of safety. The FDA defines product safety criteria, while in product liability legislation there are other definitions of product safety that emerge from court decisions at the state level.
Increasing scrutiny of prices could influence return on capital. Robinson pointed out that all stakeholders have become disenchanted with managed care and are increasingly reluctant to make choices on behalf of consumers/patients. So, by default, consumers have to make more of the choices (Robinson, 2001). As part of this move toward consumerism, employers have followed the pensions model and shifted from defined-benefit models to defined-contribution models for health care benefits. Employees will be offered a range of packages of health care benefits at different price points. At the same time consumers will be faced with higher deductibles and co-payments.
An important consequence of the trend toward consumerism in health care will be increased consumer scrutiny of costs. The American consumer values new medical technology, strongly supporting public investment in NIH research, for example, but is less enthusiastic about paying for high up-front R&D costs for medical products. Consumer concerns about costs could threaten the return on innovation through the emergence of differen-
tial pricing models, political pressure to cut costs, and pressure to shorten patent lives.
Lack of responsiveness to equity issues.21 Ford said that basic public support for federal funding of medical research programs rests to a certain extent on equitable access to the fruits of the research. Political support for high levels of NIH funding could be undermined by an increasing proportion of the population lacking health insurance or by the continued absence of a prescription drug benefit under Medicare.
Chassin, Mark R., Robert W. Galvin, and the National Roundtable on Health Care Quality. 1998. The urgent need to improve health care quality. JAMA 280(11):1000-1005.
Robinson, James C. 2001. The end of managed care. JAMA 285(20):2622–2628.