With total health care expenditures in the United States now projected to reach 21 percent of gross domestic product in 2020, it is vital that President Obama and Congress examine ways to reduce cost and enhance value in their health care systems. A federal center for comparative effectiveness is one such innovation. The Commonwealth Fund Commission on a High Performance Health System has endorsed the creation of a center for comparative effectiveness in the United States, estimating that it--with policies in place that encourage the use of better information to make better decisions--would save $634 billion over the next decade.

The United States has much to learn from comparative effectiveness review systems in Great Britain, France, Australia, and Germany. The National Institute for Clinical Excellence (NICE), the oft-cited U.K. example, studies the effectiveness of new drugs, devices, and diagnostic tools relative to existing technology and develops advice on clinical guidelines and management of individual medical conditions. While the relative price of similarly effective treatment is a factor in technology appraisals and clinical guidelines, it is not the focus: the primary goal of NICE is to improve the standard of care received by patients in the National Health Service. Review systems in France, Australia, and Germany share similar principles. All of these countries have found ways to do this in the context of their own health systems, and the United States needs to find a way to do so, also.

The American Recovery and Reinvestment Act includes much-needed federal support for comparative effectiveness research, and the Obama-Biden campaign proposal includes calls for the creation of a national institute to conduct such research. Several prominent groups have echoed these calls, including the Institute of Medicine, the Medicare Payment Advisory Commission, and the Congressional Budget Office. The establishment of a center to synthesize research, inform benefit design, and guide clinical practice was perceived as an effective or very effective way to reduce the growth in health care costs by 60 percent of health care opinion leaders in a recent Commonwealth Fund survey.

It is essential that the nation has an institute capable of generating information on the effectiveness of technologies and facilitating the use of safe, clinically proven care. We need to ensure that new technology yields value over and above existing technologies, commensurate with its incremental cost. The experience of our international peers demonstrates that investing in the knowledge needed to improve decision making, and incorporating information about relative clinical value into the design of insurance benefits, will both increase quality and yield substantial savings. Establishing and sufficiently funding a federal center for comparative effectiveness is a crucial step down the path to a high performance health system.

Collapse

Any conversation related to health care reform needs to begin with the question "What is the benefit to patients?" In the case of comparative effectiveness, we can say it is a fundamental component to providing the highest quality, most effective, safest health care for individual patients.

Physicians and patients need to have the best information about the effectiveness of diagnostics, medications, therapies, procedures and behaviors in order to make the best decisions at the point of care when it is needed most. It is critical to know not only what best treats disease for a group of patients, but also for individual patients.

Comparative effectiveness needs to evaluate the whole spectrum of treatment for a disease or condition -- not just head-to-head clinical trials. And it also needs to move beyond the diagnosis and treatment of disease to the way health care is delivered including the effectiveness of new technologies, processes and practice models. When we know what works best, we can eliminate the trial-and-error approach that creates waste and unsafe care in the health care system.

The first focus on comparative effectiveness should include diseases/conditions where little is known about what works best, diseases/conditions where data already exists but has yet to be interpreted for best practice, and chronic conditions for which care needs to be managed over time. We need to look at health IT to see what works best in communicating among providers and patients, and what works best in changing health care delivery -- getting the right care to patients at the right time in the right way.

We need to use comparative effectiveness to figure out what care will be best for patients, and then deliver it.

Collapse

There is widespread agreement among health experts that much of the health care delivered in America today lacks a basis in evidence, which is contributing to wide variations in practice patterns across the country. The end result is that patients may well be paying more for sub-optimal care.

Obtaining and making widely accessible objective information on best medical practices and protocols through comparative effectiveness research is imperative to improving the quality of health care and the affordability of insurance coverage. It can help better inform and educate providers and patients and produce better health outcomes.

The absence of good evidence can even result in harmful care. For example, in the mid to late 1980s, some studies provided limited evidence that autologous bone marrow transplant/high-dose chemotherapy (ABMT/HDC) was more effective than conventional chemotherapy for treating metastatic breast cancer.  However, after rigorous clinical trials were performed in the 1990s, it became apparent that conventional chemotherapy was superior to ABMT/HDC. Thus, approximately 30,000 women were unnecessarily subjected to ABMT/HDC.  ABMT/HDC is estimated to have caused nearly 600 premature deaths.

The ABMT/HDC story―with the approximately 10-year delay in producing credible comparative information―shows the negative impacts that can occur due to insufficient research being done in critical areas. 

Comparative effectiveness aims to assess how various procedures or interventions for a given ailment compare with each other. Comparative effectiveness is part of a broader movement to make sound science-based evidence the basis for medical practice.

Collapse

The proposal to fund comparative effectiveness research is inspired by studies showing substantial overuse of care by some patients while many others go without needed care. Researchers have documented extreme variation in medical practice and spending across the nation and have shown that there appears to be little or no relationship between spending levels and health outcomes. Fisher and Wennberg et al. have suggested that eliminating these variations in practice could reduce Medicare costs by up to 30 percent.

Funding comparative effectiveness research would produce evidence-based medical (EBM) guidelines that assist physicians in providing the best possible care. Unfortunately, the literature provides overwhelming evidence that existing EBM guidelines have had little effect on medical practice. One study of the Harvard Emergency Department showed that only about 55.9 percent of the care studied conformed to the relevant practice guidelines. Major reasons for non-adherence include lack of knowledge about the guidelines, lack of agreement, lack of confidence in the guidelines and the inertia of previous practice.

Studies have shown that merely publishing a guideline has little impact on medical practice. Commonly used approaches such as mass mailing and unsolicited conferences also have little effect. The most effective of the approaches studied used claims data to identify physicians who appear to be practicing outside of the guidelines and targets them for an interview with a trained physician in the physician’s office. But one study showed that this cost about $7,200 per physician intervention for each set of guidelines.

Generating new EBM data will have little to no impact on medical practice unless incentives are devised to encourage physicians to adhere to them. For example, government programs could deny payment for services that are not indicated under the guidelines. Also, increased patient co-pays could be imposed in cases where the service is considered inappropriate. But it is clear that developing the EBM alone will do little to control costs unless steps are taken to assure its use.

Collapse

Ilyse Schuman, Managing Director of the Medical Imaging & Technology Alliance has this to say about comparative effectiveness:

A national comparative effectiveness organization that is grounded in science has the potential to be beneficial to ensuring patient access to optimized health care which includes appropriate access to medical imaging technology. The question posed today, “What treatments should be among the first to be compared?” belies a sensitivity regarding comparative effectiveness research: Because diagnostics are different than therapies or treatments, they must be evaluated using different criteria than used to evaluate therapeutic treatments.

The medical imaging community strongly recommends separate methodologies to reflect the unique clinical roles of diagnostic interventions relative to the categorical treatment of pharmaceuticals or the categorical treatment of therapeutics and interventions with therapeutic outcome end-points.

Methods for comparing the effectiveness of alternative treatments correctly use therapeutic outcomes (eg, mortality, morbidity, quality of life) as the endpoint.  Diagnostics do not directly affect these therapeutic outcomes; rather, they provide information that informs the physician's choice of most effective treatment for the patient.  Consequently, comparative effectiveness methods for diagnostics  should use changes in treatment decisions (eg, surgery vs. medical therapy), as the endpoint, ie, diagnostic outcomes.

The authors of the “Comparative Effectiveness Research Act” insightfully included principles for ensuring that comparative effectiveness analyses are independent, transparent, scientifically rigorous, based on the most up-to-date data, and inclusive of all medical procedures and technology. However,  the medical imaging community also encourages that this legislation be enhanced to add specific language to ensure processes involving research development, evaluation of data, and dissemination of findings include appropriate diagnostic as well as therapeutic methodologies.

Collapse

The AMA supports comparative effectiveness research (CER) as a way to provide physicians with information on which treatment works best.  CER should be broadly defined to include a comparison of how to manage a specific health problem, condition, or disease.  CER should guide and support clinical decision-making, not dictate it.

National CER priorities should address the prevention, management and treatment of preventable disease, which is a major cost driver in today’s health care system. Key areas in need of further study and research include cardiovascular, endocrinology and metabolism disorders (including diabetes) and nutrition (including obesity).  We do not have CER findings in the area of wellness, prevention, nutrition and obesity, yet there is a wide range of available interventions.  Clarity from CER findings as to which intervention is the most effective will help physicians provide patients with the treatment most likely to have the best outcome.  

Of note, the inclusion of racial and ethnic health disparities—and health disparities more generally—should also be a CER priority area.

I recently testified to the Institute of Medicine (IOM) on this very issue, and the full testimony is available on the AMA Web site at http://www.ama-assn.org/ama1/pub/upload/mm/399/hsr-comparative-effectiveness-testimony-20mar2009.pdf.

Collapse

Comparative effectiveness research. Of course. But no research or improved information system is going to take away the human responsibility to decide what to provide and what to pay on the basis of at least some qualitative judgment. And if no one is empowered to make use of the improved information to increase effectiveness, even in an imperfect and uncertain world, then CER is not going to get at some of the cost drivers that are pushing more and more people out of private health insurance.

In order to know where to begin, we need to decide where we want to end.  We support increased funding for comparative effectiveness research and believe a key goal of CER should be to inform patients and providers so that they make good decisions.  Another important CER goal should be improving the quality of care.

To achieve these goals, CER must address a broad range of evidence gaps across the health system.  This includes prescription medicines and medical technology, as well as approaches to organization, management and delivery of care.  While much attention is appropriately paid to evidence-based care, our delivery system often fails to deliver care that is known to be evidence-based.  At recent meetings of the Institute of Medicine and Federal Coordinating Council on CER, this point was emphasized by many speakers from the patient, provider, public health and other health care sectors. 

CER also needs to start by asking research questions that reflect the real-world needs of patients, physicians and other providers.  This means providing a meaningful, ongoing role for patients, providers and other stakeholders through open, transparent processes.  We need sustained efforts to close gaps in evidence and health care quality.  We can't get there unless we start with a focus on patients. 

Collapse

There is nothing in principle wrong with comparative effectiveness research (CER). It goes on all the time, often with federal funding. The reason for the controversy is a proposal in Tom Daschle's book to use CER the way it is used by the British agency with the ironic acronym NICE. In Britain, NICE has decided that the National Health Service should not spend more than about $35,000 to save a year of life. Doctors, hospitals and area health authorities typically follow these guidelines. As a result, British cancer patients are typically denied access to drugs that are routinely available in the US and on the European continent.

Since Tom Daschle was chosen to be President Obama's Health Czar, we all have many reasons to be concerned. Although Daschle stepped down as the nominee, I think everyone views his book as the Obama road map.

Despite the many editorials and opinion pieces protesting otherwise, when it comes to addressing our nation’s health care issues, comparative effectiveness research (CER) is just “good science.” New investments in CER and related studies are critical to converting biomedical discoveries into more effective and more affordable health care services. These CER studies can extend beyond comparing two treatments to include knowledge translation from “laboratory bench to bedside,” different approaches to patient engagement, and new models of health care delivery.

But with so much at stake, and with so many possible medical conditions to be considered, how do we determine which CER studies should come first? As we make these decisions, it would seem that certain criteria should be paramount. In the immediate future, studies of comparative effectiveness should focus on situations that meet as many of the following four criteria as possible: 1) diseases and disabilities that impose the heaviest personal burden on patients and financial burden on society; 2) conditions for which there is a high degree of uncertainty in the medical community about the “right” thing to do (given the range of approaches and interventions available); 3) decisions that have especially significant consequences for patients (e.g., high-risk interventions); and 4) questions for which the data we need are largely available and can be quickly gathered and analyzed.

Certainly the problems of our health care system do not lend themselves to easy answers. Our investment in CER projects must be large enough to produce meaningful results, rather than simply engender more uncertainty. In addition, there must be sustained investment to expand the scientific workforce skilled in CER, as well as the continued build-out of the national CER infrastructure. Further, we must take care to implement CER through means that are synergistic with the ongoing discovery of even newer clinical innovations. By any measure, CER to better inform patients and clinicians at the point-of-care must be supported as a critically important component of any comprehensive effort at health care reform.

Collapse

Comparative effectiveness research (CER) is controversial, but it is an idea whose time has come for Americans. Developing the capability to help physicians and patients understand treatment options and make the best decision is fundamental to health care reform. The most politically astute approach is to involve health care stakeholders in establishing criteria that would guide and prioritize what should be studied. The topics for a CER agenda might be narrowly focused, for example, on the most prevalent conditions, or go broader to include “systems of care” such as the concept of medical homes. The agenda should also encompass developing mechanisms for comparative effectiveness research such as using claims data, registries and randomized controlled trials. Because CER is vulnerable to attack, it requires developing a uniquely American way to spend the $1.1 billion federal funding for CER appropriated for 2009. We may not do CER the way the National Institute for Clinical Excellence (NICE) in the U.K. reviews drugs and treatments, but we must do it.