Comparative effectiveness research is about information. And information is essential to good decision making. Those two statements don’t sound groundbreaking, but with all the recent debate about CER, I am starting to wonder.
CER is a small but significant piece of the larger effort to reform the nation’s health care system. The various reform bills in Congress all feature the creation of a new organization or activity to rationalize and strategically organize CER. This is an important step toward improving clinical quality and reducing cost in the United States. More and better-organized information will enhance clinical interactions and improve decision making for physicians.
The policy interest in CER is driven by its ability to “bend the cost curve” while improving quality. It is not a secret that there are large variations in health care practice; some areas of the country spend a lot more than others, and these variations are not associated with improved quality. The Dartmouth Atlas group has reported this data for some time now, and Atul Garande recently crystallized it into a compelling narrative in The New Yorker.
The argument for CER is that a lot of the variation in medical practice results from large evidence gaps and a failure to translate available evidence into clinical practice. Without this evidence, decision making may be more easily influenced by other factors—financial incentives, medico-legal worries, traditional practice, specialty biases, etc. The idea that doctors frequently make decisions based on financial incentives counter to a patient’s interest is absurd, but non-clinical considerations have influence when there is ambiguous evidence in a system designed to maximize margins. If we had more quality information on what works, we could eliminate the wasteful interventions that constitute up to one-third of health care spending by some estimates.
The lack of a comprehensive CER strategy leaves many gaps in our knowledge. Take low-risk prostate cancer. The Institute of Medicine named this cancer one of the initial 100 CER priority areas. Why? There are a number of treatment options available—active surveillance, surgery, and a variety of radiotherapies—and prostate cancer is the number one cancer diagnosed in men, but there are no good data on which treatment is best! Numerous men ask questions about the comparative benefits and side effects of these treatments every day ,and yet we have never compared them all to each other in randomized trials. Our current fragmented research enterprise has never created the incentive to do so, leaving training bias and dogma in place of actual data. When good research is available, it is usually slow to inform clinical practice. A study published in the New England journal showed that only 55 percent of people receive recommended care.
Much of the pushback against CER has been to say that it will allow “the government to get in the middle of the doctor-patient relationship.” How so? The doctor-patient relationship is about developing rapport based on mutual trust and respect. New investments in CER deal with the generation, compilation, and distribution of information. Having more information will not erode rapport, but enhance it. I’m sure the next patient that I see with low-risk prostate cancer would be much happier and more confident in our relationship if a focus on CER enabled me to accurately discuss the plusses and minuses of various treatment options. They would then be able to make more informed decisions based on their personal values. A comprehensive CER strategy that helps fill information gaps would not make government the decision maker. Instead, patients would be empowered—armed with better information and council from their physicians.
Another erroneous suggestion is that CER will lead to “one size fits all” medicine. On the contrary, a robust CER enterprise would aid the development of personalized medicine. As medicine becomes more individualized, larger data sets and new methodologies will be needed to determine the influence of personal factors—such as demographics or genetic signatures—on treatment and outcome. These studies cannot be done by individual researchers but require collaborations and an enhanced research infrastructure.
Current medical knowledge is limited by large gaps and difficulties translating new findings into practice. These limitations lead to variations in practice that don’t correlate with quality and open the door to other influences. A robust CER enterprise would seek to address these issues by compiling available information, identifying evidence gaps, generating new evidence, developing alternative research methodologies, and disseminating medical information for implementation into clinical practice. More and better information—who could argue with that?
Brian Alexander, MD, is a resident in the Harvard Radiation Oncology Program and a former White House Fellow.
For more information, see: