Keynote Speech: Encouraging High-Value Medical Innovation
Mark McClellan, Council of Economic Advisers
MEDICAL INNOVATION AT POLICY CROSSROADS
I have been nominated for the Council of Economic Advisers22 to coordinate economic issues in health care policy for the Bush Administration. The topic of medical innovation policy is an area of major interest and importance to the Administration. The President in his campaign emphasized the need for better mechanisms for assuring that people get access to valuable new medical technologies and continuing to improve access to existing technologies. It is very clear from research done by participants in this conference that medical technology has contributed enormously to the improvements we have seen in health in this country in recent years, especially improvements in the length of life and the quality of life for older Americans.
Medical innovation has also become a major part of U.S. economic growth. For example, in the last several years as investment in some of the Internet technologies has slowed down, there has been a continuing explosion in investment in biotechnology products. I think that is one of the major areas of venture capital investment today, which harbingers well for the future of this part of our economy in terms of continuing to contribute to economic growth.
Obviously an administration’s policy should be to encourage more of
this valuable medical innovation, but medical innovation has also been marked by rapid increases in health care costs. I do not need to tell you that changes in medical treatments, not rising prices or other factors, have been the major cause, at least up until now, of some of the increases in health care costs in this country, particularly Medicare costs.
So, we are at a policy crossroads. On the one hand we face very strong pressure to try to encourage innovation. On the other hand there are growing concerns about the fiscal and economic implications of the costs of medical innovation. This comes to the fore in the Medicare program. On the one hand, we are strongly advocating mechanisms that would help seniors get valuable new technology sooner. On the other hand we are equally strongly trying to make a case for the need for careful fiscal planning and economically sensible policies because of concerns about the longrun fiscal status of Medicare.
ENCOURAGING HIGH-VALUE MEDICAL INNOVATION
This comes down to the critical question of how do we encourage highvalue medical innovation? Not just innovation involving new technologies that may or may not have an impact on actual health outcomes, but how do we encourage more technologies that really are worth the cost? Ideally some of those would end up saving money as well. Again, we have not seen a tremendous amount of that in recent years. Perhaps through better policy making we could encourage more innovation that leads to lower health care costs as well as improved health. In terms of the incentives for innovation that policy might influence there are three major areas. I would like to hear what your views are on this, and whether there is an area that we are leaving out or not supporting properly.
Support for Biomedical Research
What many people focus on first when they think of innovation is support for biomedical research and fundamental research. This is an area where federal policy has seen a big infusion of dollars in recent years. The President just like the previous Administration has committed to a doubling of the NIH budget between 1995 and 2003. I think the original goal was to do it in a decade. According to our budget resolution, it will be done in just 8 years. So, at least in the biomedical community there has been a huge infusion of federal research dollars and that is going to continue over the next several years. Getting good products and valuable innovations into use requires more than direct federal support for medical innovation research, however.
A second part of the innovation process is coverage decisions. Once medical technology has been developed, what steps are needed to ensure insurance coverage and adoption in medical practice? In this area there have been concerns raised about delays. Partly this comes in the area of the FDA approval processes and the multiple steps required to bring a drug to market. Beyond that there are coverage decisions by the Center for Medicare and Medicaid Services (CMS, formerly the Health Care Financing Administration), and other payers.
The most interesting recent developments that we have been considering for application elsewhere have come from collaborative projects started by Richard Klausner at the National Cancer Institute between the NCI, the FDA, and CMS. This process currently involves primarily those technologies involved with imaging procedures in cancer and collaboration between all of the parties that need to reach agreement on the evidence available to bring a new product to market. This collaboration involves individuals from NCI, from CMS, and from FDA sitting down early in the innovation process, typically when a clinical trial is being designed. In one integrated process the drug or device manufacturer can pull together all of the evidence that is needed, not just for the FDA’s safety and efficacy evaluation but also for the coverage process which takes place at CMS and other private insurance companies. The information required for these different processes is somewhat different and historically the process has been much more of a sequential one. The clinical trial is first and that leads to a conclusion about the safety and efficacy of the treatment. Then further studies are carried out on costs and other issues that need to be addressed for coverage decisions.
Having an integrated process that brings all the parties together early on while a clinical trial is being designed can potentially speed up the process for moving valuable new technology into active practice by some years. This model may be applicable to other areas. It seems to me that it holds a lot of promise.
Diffusion of New Technologies
The coverage decision process is an important part of bringing innovations to market and into actual medical practice, but beyond that there is a third area that does not get enough attention in studies of medical innovation. This area is the process by which new technologies defuse into widespread practice. There have been some studies by a number of reputable experts on issues related to evidence-based medicine that show that some
medical technologies can take as long as 15 or 20 years to fully reach widespread adoption in medical practice.
In addition, studies show that many medical technologies are being used inappropriately. These technologies are widely used in some parts of the country and in some institutions. These variations seem much larger than can be explained for economic or clinical reasons, even though they may, in part, be due to differences in the preferences in the community and to differences in the skills of physicians and other health care providers in the community. It is these kinds of variations that account for much of the potential for improving the value of medical innovation in practice.
Some of you may have seen Secretary O’Neill’s testimony with the release of the Social Security and Medicare Trustees’ Annual Report in March. He talked about the potential for substantial savings from improving the way that medical care is delivered, for example, taking more steps at the local practice level to identify ways to deliver more cost-effective care.
He spoke of applying what he called the “Toyota” principles of management, whereby everybody involved in a complex organization like a hospital has the opportunity to think about the specific things that they can do to improve the quality and reduce the rate of errors in the services that they deliver. These are typically not very high-tech things, for example, ensuring that errors in transcribing of medications do not occur. They can lead to substantial savings and, at a minimum, substantially higher value in the delivery of care provided. This also extends to the use of medical technologies for indications that may not be very well justified or there may be evidence that the technologies are not being used appropriately. Yet, they still find their way into active practice. If there were some federal levers that we could use to help bring better practices into being, we could avoid delays in appropriate use of medical technologies, and move away from inefficient and wasteful styles of medical practice.
This is a very hard thing to do. As you all know, patients are different in important respects, and there are many patient preference factors and clinical factors that go into most medical decisions. Moreover, it is not something that is very easy to legislate or to regulate centrally. To move forward on this issue we have thought about whether there are ways to provide better incentives to organizations, to health plans, and to providers.
As Secretary O’Neill and Secretary Thompson have noted, it does not seem to make much sense in Medicare to pay more to hospitals that are delivering bypass surgery operations that result in complications and readmissions than to hospitals that can deliver a low complication rate surgical procedure with fewer re-admissions. This is a difficult area to make progress in because of the complexities of our current fee-for-service payment systems, and I do not want to make this sound like private insurance plans have all the solutions here either.
A big problem with capitated payments or payments that move away from paying on a fee-for-service basis is that they may not provide appropriate incentives to deliver costly care when it is truly valuable. We are struggling with ways to help policy move forward effectively. Once again some help from you on new directions for medical innovation policy would be very useful.
We look forward to the results of this conference and subsequent meetings of this National Academies working group with a lot of anticipation. We hope that they will lead to some ideas that we can translate into practice.
David Beier, Hogan & Hartson LLP: A question and an observation. As you think about the Human Genome Project being a merger of computer science and biology, increasingly you are going to have complicated problems if you only fund biomedical research and not the National Science Foundation and the other elements of the federal research portfolio. Could you comment on the President’s budget in this respect?
One of the problems of collaboration among the FDA, the NIH, and CMS is that you run the risk of creating a third approval hurdle of cost effectiveness for products, and of having federal bureaucrats essentially make that decision rather than the market or the medical practitioners. I would like to get a comment about how you avoid doing that if you go down that collaborative route.
Mark McClellan: With respect to the importance of innovation in areas that might not be thought of as directly related to biomedical science, I looked back over some of the recent funding decisions by the NIH and was impressed to see how much of their work is in areas that I think most people would consider computer science. There are major investment programs related to super-computing technology and collaborative projects across institutions to do next generation computing and information storage. The NIH knows that the future of biomedical research depends critically on innovations in such underlying technologies as medical information systems.
We are undertaking a review of whether there are better ways to fund some National Science Foundation research as well as to make sure it is coordinated with the expanded research enterprise at NIH. It has been interesting, though, to see the level of political support for NIH research, far and above some of the other areas of basic research that are clearly related to NIH. I think basic research is something that we want to make sure continues and that integration between basic research and applications at NIH also continues.
With respect to whether what we intend to be a streamlining of the technology approval process may become another bureaucratic hurdle in an already time-consuming process, I think that is a risk. I am not sure that the model that Dr. Klausner has developed so effectively in one part of NCI is a universally applicable model, although it does have support from manufacturers and is widely regarded as having shaved substantial time off the approval process. This is the kind of issue on which we could use some expert guidance to make sure that we are really taking time out of the approval process rather than just creating additional hurdles. These are not easy questions, and that is why I am glad the National Academies are focusing on them.
Gil Omenn, University of Michigan: Your third challenge about how to get the right incentives is directly relevant to a new study launched in April with a meeting here at the Institute of Medicine. It will be the followon to the To Err Is Human report and the Crossing the Quality Chasm report.
The specific mandate from Congress is to look at the quality oversight and the quality improvement activities of federal agencies including the Joint Commission on Accreditation of Healthcare Organizations (JCAHO), National Quality Forum (NQF) and other quality activities in the private sector and the not-for-profit sector. The present mandate, as interpreted by the IOM, excludes dealing with the reimbursement mechanism. In a few sentences, you made a powerful case that we need to find ways to tie the payment system together with incentives for support of what works and blocking things that do not—what is safe and what is not at the product level, the service level, and of course, the system level.
Mark McClellan: I think it is very important that reimbursement not be overlooked. It does have a big impact on medical practice and on the use of treatments. It is an area where bold thinking is needed. This is a controversial area. There are no obvious general solutions, but it is an area that we would like to incorporate in policy making, and so we look forward to any guidance you can provide.
Mary Jo Deering, Department of Health and Human Services:You observed that costly care is often valuable. You also noted that improvements in longevity are attributable to medical innovations. I think there is a large body of literature indicating that longevity is also due in part to public health and primary prevention efforts. Yet the percentage of health care expenditures devoted to prevention has been estimated to be as low as 1 percent and no higher than 5 percent. Is there anything that you see on the horizon that would help encourage innovation in less costly preventive interventions?
Mark McClellan: The question of innovation and less costly preventive treatments is a good one. It is related, in part, to the reimbursement
issue to the extent that providers and makers of medical technology and others involved in the health care system are paid more when they do more, paid more for complications, and paid more for treating the consequences of poor preventive care. It does not seem that the incentives are in the right place.
I noted you are from HHS and this is a concern that has made its way to the Secretary’s level. I think you will see some more announcements and initiatives coming out of your department in the months ahead. These are aimed at helping prevention regain a central place in our provision of health care and taking steps to try to make it easier for people in public programs, as well as in other insurance programs, to get the preventive care they need and to help them know when they need it.
I did not mean to say that the use of medical technologies when diseases actually occur were the only, or even perhaps, the main contributor to improvements in health and the quality of life that we have seen in recent years. Prevention is obviously an important part, and I hope that your work here on medical innovation can help us find ways to better support innovation that is geared towards prevention as well as treatment.
Philip Aspden, NRC: We have got a question from a webcast listener, Curt Shoemaker of Johnson & Johnson, and his question is brief. How are new technologies and new procedures selected for the NCI/FDA/CMS collaborations that you mentioned earlier?
Mark McClellan: I think they have focused on imaging procedures because the coverage issues that needed to be addressed are relatively well defined. If, however, this turns into a large and complex procedure, it can end up being quite time consuming and may not help achieve the desired goal of reducing the time to market.
Dan Mark, Duke University: As a practicing clinician, I see one of the biggest impediments to improving the efficiency of our care and to addressing the concerns about medical errors to be the lack of an appropriate informational infrastructure—an electronic medical record that spans across the entire system and allows us to integrate information efficiently.
The marketplace has been unable to respond to the deficiency perhaps because IT suppliers see doctors are still using many techniques we used in the last 100 years or so. William Osler, if he were reincarnated, would probably still be comfortable with our medical records. They have changed very little. If the marketplace is going to be very slow in responding to this, I wonder whether there is a role for the federal government. Is there some way to accelerate this process? I do not see that there is going to be the technological infrastructure to support some of the initiatives that are being talked about to improve quality if the information systems cannot be brought up to speed.
Mark McClellan: It has been very difficult to develop good informa-
tion systems in health care in contrast to many other industries. I think there are several reasons for this. One of them may be privacy concerns. As you know, there has recently been a controversial series of policy developments related to privacy issues. We are trying to go forward establishing some uniform privacy standards as soon as practicable over the next couple of years in a way that is not overly burdensome for the health care industry to implement.
Another impediment may be that the financial rewards for good information systems are not comparable to those in other industries. It is interesting to contrast the health care industry and the financial services industry. Very sensitive information is involved in both cases but I think no one would have any concerns about an ATM transaction passing across a shared network. It is handled privately. It is handled without error. It is handled with consistent standards across the board. In health care, we do not have in place either the privacy standards or the data standards or the infrastructure itself.
In terms of standards, I think the federal government is trying to move forward as quickly as we can with developing Health Insurance Portability and Accountability Act (HIPAA) and other coding terminology standards, and the establishment of the computer and data infrastructure needed to share data effectively. Again, privacy protection is an important part of this.
I still wonder whether some of these investments in computer technology are really all that promising in terms of reducing errors or delivering higher quality care. We hear a lot from hospitals and other institutions that just say, “Look, this new computer technology requires a substantial investment, and we just do not have the funds to do it.” I wonder if this is a function of reimbursement incentives that in many cases are not paying more and maybe even paying less to hospitals that deliver care without complications.
I hope that this or other groups can provide some guidance on how we can improve policies in this area. I know there are many proposals pending in Congress, ranging from grant programs to demonstration projects, to implement better computer systems. It is a critical part of health policy and innovation.