Evidence-based medicine (EBM) is a juggernaut that has taken practitioners by storm. Over the last 20 years, my colleagues at McMaster University, as well as at other academic institutions, have spearheaded the EBM movement to the point where journal articles with “evidence-based” in the title are ubiquitous. While EBM is not perfect and is a continuing work in progress—qualitative levels of evidence seem to change definition regularly—the fundamental principles of EBM are unquestionably an advance in the management of patients.

As a rule, physicians are dedicated to their patients and usually convinced that they deliver the best possible care. It is difficult to accept the possibility that our patients might not be getting optimal care because of our decision-making processes. For decades, the main teaching methods in medical schools consisted of non-interactive didactic teaching sessions and rounds with attending physicians and senior residents where each would try to “out anecdote” the other. These rounds usually created a competitive environment, in which the primary focus was placed on cases we had never seen before and would never see again. From these “canaries,” the words and experience of the senior attending, which were based on his “vast experience” of two or three patients, were gold nuggets. Nobody questioned, let alone challenged him. Over the last three or four decades, it has become clear that not everything in medicine is black or white and that disagreements among clinicians is quite common.

So how do we reconcile the “If I say it, it must be true” or the disagreements? In retrospect, the answer was so simple and yet until recently, the technology did not exist to make the transformation to evidence-based medicine. Since one of the main purposes of publishing papers is the dissemination of information, it seems only logical to use the Internet to perform a literature review and see not only the experience of others, but also if any well-designed studies had been done addressing the issue. Unfortunately, in the early to mid 1980s, computers were slow and the internet was still in its infancy, and so performing literature reviews for every clinical question was highly impractical.

Over the next 10 to 20 years remarkable advances have facilitated literature reviews and spawned a generation of practitioners who practice evidence-based medicine. The main advances included the vast expansion of the World Wide Web, the availability of portable and powerful computers to access the Internet, and the recognition that the uncritical acceptance of the professor’s experience, while somewhat useful, was very limited and often wrong. In parallel, the National Library of Medicine simplified the performance of literature reviews within an enormous database and made it widely available and easily accessible.

It is difficult to justify the “professor’s experience” approach to clinical decision making while ignoring the vast amount of relevant literature that is accessible in a handheld device or laptop computer. The fundamental principles of EBM include an approach to patients that identifies key issues, generating clinically important questions that address these issues, performing literature reviews that will identify papers that address the question, interpreting the quality, conclusions, and applicability of the literature, and finally applying the knowledge to patient management. Although many of us have considerable expertise at obtaining and refining “best available evidence,” where some of my colleagues and I often fall down is the knowledge translation. This incorporates patient preferences and values into decision making about interventions. I am not very good at this critical aspect of EBM because it is usually very time consuming and can be mundane; my time is valuable, and I am always busy.

EBM is here to stay. Despite passionate complaints about the arrogance of some proponents of EBM, continual debate about the relative strength of randomized trials versus observational studies and the lack of uniformity in grading levels of evidence in consensus conferences and guidelines, there is little doubt that EBM represents a significant advance in our approach to patients. EBM does not exclude the “professor’s opinion” as evidence, but simply puts it into perspective as the lowest level of evidence. It is also critical to understand that EBM is in its infancy and is constantly evolving and adapting as problems arise.

In order to make individual decisions about health care (eg, performing or withholding an intervention), it is important to incorporate the best available evidence with patients’ preferences. With few exceptions, patient preferences “trump” physician preferences. To a variable extent, the physician has several layers of responsibility. The first is to obtain the best available evidence in favor of (or against) an intervention (this will be further discussed later in the chapter). The second is communicating the evidence, in understandable terms, to the patient and/or their guardian. Finally, patient preference should be ascertained. In a perfect world, data providing evidence are clear and noncontroversial, patients understand the pros and cons of the intervention and the physician and patient have similar values and preferences. However, it is far more common than not that at least one of these criteria is not met and/or physicians fail to communicate effectively with their patients.