Laws requiring the use of evidence-based medicine (EBM) have been proposed in a number of states, including Colorado, and it already is required by law in New Jersey. Since the early 1990s, EBM has been promoted as essential to improving health care.
Its proponents claim that using epidemiological data, the study of group health outcomes, will let physicians make better clinical decisions for individual patients. This shows their naïve understanding of epidemiology. Group outcomes are averages of individual outcomes; however, the average may not apply to anyone in the group.
Think about it this way: Even if the average dress size is a size 12, not all women wear a size 12 and no sensible person would suggest that they should. If made law, EBM would greatly curtail the freedom of physicians to tailor treatments that can solve the individual problems of the patients they treat.
EBM confuses statistical averages with science. Rather than use evidence from all sources, it elevates randomized controlled trials, "mega-trials," and meta-analyses to the level of "gold standard," the most reliable evidence with which to judge the efficacy of medical treatment.
This misapplication of group data to the care of individual patients is so well known that it even has a name, the "ecological fallacy." The ecological fallacy is taught as a reminder that group averages tell us nothing about causal processes in the individuals who make up those groups. Top notch medical care uses group averages in concert with a physician's clinical judgment and experience, qualitative factors unique to each patient, and the wishes of the individual patients who want to choose treatments that fit their own priorities.
A careful examination of EBM reveals that its recommendations on medical treatment would more accurately be called "opinion-based medicine," as it typically prescribes treatments that are based on the opinions of academics and journal editors.
The language used by EBM proponents is an indication of their "we know what is best for you" attitude. They promote themselves as rational, objective, and altruistic. Any opposition is portrayed as illogical, self-indulgent, unscientific, and greedy. Along the way, a whole world of good practice - real evidence from real experience with real people - is ignored and subverted by rhetoric.
Story Continues Below
EBM transfers clinical decisions from those with the clinical training, experience, and extensive knowledge of health and disease, the physicians, to "systematic reviewers" from the disciplines of biostatistics, epidemiology, and health economics. In addition, it has been used to develop various tools, including standardized practice guidelines, in order to correct the "problem" of clinical practice variation.
Practice guidelines already are being used in the largest health-care system in the United States. The Veterans Affairs system uses them to measure "quality of care" at all of its sites.
Practice guidelines specify which treatments should be used by all physicians, but they may place the physicians' incentives in conflict with those of their patients.
For example, a recent guideline issued by the American Academy of Sleep Medicine specifies the proper treatment for patients with sleep-related breathing disorders such as obstructive sleep apnea. It specifies that physicians prescribe continuous positive airway pressure (CPAP).
However, this treatment is unwieldy and difficult to use, and two recent studies have found that patients prefer other treatments for the disorder. Yet if physicians don't prescribe CPAP for their patients they will be at variance with the practice guidelines, and could find themselves down-rated, just as teachers now are being rated under the federal program, "No Child Left Behind."
Eventually even their compensation could be tied to "performance," as now occurs under Medicare's "Pay For Performance" (P4P) program, a money-saving scheme that coerces physicians to follow practice guidelines.
Physicians who actually care for patients and who have been trained their entire professional lives to use the best medical information available are skeptical of EBM and have been slow to accept it. As a result, EBM and its resulting "practice guidelines" are not being used as much as proponents would wish.
They have resorted to various schemes to "encourage" its adoption. One organization, the American Academy of Family Physicians, has doubled the credit offered to physicians who take its own brand of evidence-based continuing medical education (CME). But, as the director of its Division of Continuing Medical Education, Nancy Davis, Ph.D., admits, "How do you know [evidence-based] CME is any more effective than traditional CME? We don't know yet."
One effect of this scheme surely will be that physicians will take less CME courses.
Proponents of EBM say that everyone's medical treatment should be the same as the group average, and they want to transform medicine into a top-down command and control system that defines, prescribes, monitors, and enforces average practices upon doctors. But there is no evidence that clinical practice can be reduced to the routine application of checklists and formulae.
Practice variations are not necessarily a bad thing, since for treating the individual patient there is never only one best treatment. In addition, individual patients who have multiple diseases and who must take a number of different drugs, where the physician must judge the likely risks and benefits of each, make the use of EBM problematic if not impossible.
Perhaps the most important problem with EBM is that of its ethics. EBM is not a physician driven agenda. It has bypassed the clinicians (those physicians with clinical training, experience, and an extensive knowledge of health and disease) in favor of an alliance of managers and their statistical technocrats, who are empowered to define "best practices."
Their paychecks depend on churning out these definitions. These non-clinicians thus have acquired substantive influence over millions of clinical consultations without sharing any of the responsibility for the clinical consequences.
Special thanks to Linda Gorman, Ph.D., for editorial assistance.