By Janis M. Miyasaki, MD, MEd, FRCPC
Few words generate as much controversy and confusion as "evidence-based medicine." For some, the term implies that medicine before the 1980s was based on belief and bias alone. But the true intention of the term is this: All practice is based on evidence. Evidence takes many forms including experience, what we learned during residency training, case series, and randomized controlled trials.
Evidence-based medicine (EBM) is simply the conscious effort to obtain the best evidence to make decisions, putting this evidence into practice, and evaluating the outcome. In this construct, "best" means having the least risk of bias. Therefore, case series without controls have the highest risk of bias while a study with a control arm, predetermined endpoints, and randomized assignment to treatment or control arms has the least.
In fact, EBM is just good applied science.
The great expansion of medical knowledge poses a challenge to clinicians. While improved performance is the goal when clinicians use a medical literature database to apply scientific results at the bedside, the technology is only as effective as the user's ability to capitalize on its benefits in real time. Although EBM in daily practice should be quick and readily available, the usefulness of our search results is limited by how much time and effort we have to devote to refining the scope of our search.
When seeking information about a specific problem, we should make our question as specific as possible. The more precisely the query is worded the more likely it is that the literature retrieved by the database search will be relevant to the problem. For instance, if you want to learn the best way to diagnose myasthenia gravis, searching PubMed for the keywords "myasthenia gravis" is less effective than searching for "myasthenia gravis" and "diagnosis" and "gold standard."
Paradoxically, frequently encountered clinical problems often require more refined searches than a unique situation in an unusual patient, because the potentially relevant medical literature that must be sifted through is much larger. Even limiting the scope of the search to reviews in general journals such as New England Journal of Medicine, Lancet, Neurology, Archives of Neurology, and Brain often generates a considerable reading challenge for a single practitioner. We need expert help in reviewing and assessing the literature on neurologic diseases and their treatments.
The American Academy of Neurology seeks to guide our specialty by developing evidence-based guidelines or parameters. These are not proscriptive. Each guideline includes the following language:
This statement is provided as an educational service of the American Academy of Neurology. It is based on an assessment of current scientific and clinical information. It is not intended to include all possible proper methods of care for a particular neurologic problem or all legitimate criteria for choosing to use a specific procedure. Neither is it intended to exclude any reasonable alternative methodologies. The AAN recognizes that specific patient care decisions are the prerogative of the patient and the physician caring for the patient, based on all of the circumstances involved. (emphasis by the author)
The guidelines are meant to provide the best available evidence to clinicians and patients to aid in decision making. Each guideline takes a minimum of two years to complete. Hundreds or thousands of abstracts are reviewed and many more full-length articles than the final manuscript lists are read and evaluated. A minimum of 12 subcommittee members review each guideline several times until it is ready for evaluation by volunteer and invited stakeholder reviewers, then the Practice Committee, and finally the editorial reviewers of Neurology®.
A typical number of reviews requiring a response by the guideline authors is 30; As such an author, I have personally responded to more than 75 reviewers for a single guideline. This means that the guideline is highly vetted before publication. AAN members see a final product that, among guidelines from other national medical organizations, is recognized as rigorous and of high quality.
AAN members also should realize that many of our colleagues in other areas of medicine and surgery do not have guideline development policies and are rushing to put them in place. Why? They feel that hospital administrators, policy makers, and insurers are viewing how they practice with a skeptical eye. A specialty without an evidence base for its diagnostic and treatment decisions is vulnerable to arbitrary cost containment strategies. The Academy's guidelines prove to decision makers and the public that neurologists want to provide the best possible care, based on the best available evidence. Furthermore, our guidelines highlight areas of research need. We want clinical research to respond to clinicians' information needs and to do so in a rigorous manner.
What can be more laudable than these goals: To educate, to improve patient outcomes and to challenge science to do better? These are altruistic—and attainable— goals of medicine, the AAN, and EBM.
Dr. Miyasaki is the co-chair of the Therapeutics and Technology Assessment Subcommittee of AAN. This subcommittee is charged with developing evidence-based guidelines for AAN. Within the last 24 months, she has received personal compensation from Teva and Boehringer Ingelheim for consulting work. In the same period, she served as an uncompensated member of the Movement Disorders Editorial Board. Within the past five years, Dr. Miyasaki received research support from Teva (clinical trial), Allergan (clinical trial), Solvay (clinical trial), National Parkinson Foundation (comprehensive care grant), Eisai (clinical trial safety monitoring committee), National Institutes of Health—NET PD studies (independent medical monitor), and Parkinson Society Canada (outreach program).