Sackett DL, Richardson WS, Rosenberg W, Haynes RB. 250 pages. New York: Churchill Livingstone; 1997. $24.99. ISBN 0443056862. Order phone 800-553-5426.

Evidence-based Healthcare: How to Make Health Policy and Management Decisions

Muir Gray JA. 270 pages. United Kingdom: Churchill Livingstone; 1997. $29.95. ISBN 0443057214. Order phone 800-553-5426.

The germ theory of disease, the introduction of bedside teaching, the move to problem-based learning, the clinical application of molecular biology-landmark concepts in medicine appear regularly and require physicians to use new skills. The discipline of evidence-based medicine may be such a concept. Provoking enthusiasm and energetic implementation from some and resistance and outright hostility from others, evidence-based medicine seems to be here to stay. At the moment, it is an evolving science, and its influence on practice and health policy is increasing.

Two new books make an important contribution to the growth, practice, and teaching of evidence-based medicine. Evidence-based Medicine: How to Practice and Teach EBM, edited by David L. Sackett and colleagues, offers perspectives from Canada, the United Kingdom, and the United States. A companion volume, Evidence-based Healthcare, was written by J.A. Muir Gray of Oxford, England.

The growth of evidence-based medicine is visible everywhere. Special sections devoted to evidence-based medicine are appearing in traditional scientific journals. Medical students in Manchester and Oxford have organized a conference and an Internet forum on evidence-based medicine. The University of Louisville recently advertised a meeting entitled "Evidence-based Practice in Oncology." The U.S. Agency for Health Care Policy and Research redefined its mission to include the dissemination of evidence-based information on clinical effectiveness in the management of important disorders; the agency will also choose a group of evidence-based practice centers to support. This year, a conference assessed the effects of 5 years of experience with evidence-based health care policy implemented by the National Health Service in England. In addition, Cochrane Centers are springing up worldwide. The Cochrane Collaboration is an international initiative organized to prepare, maintain, and disseminate the results of systematic reviews of existing health care interventions. The ultimate aim of the collaboration is to review all clinical trials and publish them in a form that is accessible to persons in the health professions. The Cochrane registry of controlled trials exceeds 110 000, and 258 Cochrane-sponsored systematic reviews have been completed or are in progress. This effort has recently been described as an "enterprise that rivals the Human Genome Project in its potential implications for modern medicine" [1].

The books by Sackett and colleagues and Muir Gray define the practice of evidence-based medicine as making medical decisions by combining the best external evidence with the physician's clinical expertise and the patient's desires. Evidence-based health care extends to managerial and policy decision making. Thoughtful physicians have long struggled with the gap between the best of what is known in medicine and what usual clinical care includes for many patients. Studies regularly show the disparity referred to by Weed [2] as the "voltage drop" between science and patient care. Evidence-based medicine is a method of problem solving that 1) recognizes that no individual person can know all that is needed to practice effectively across the spectrum of care and 2) acknowledges that not all therapies or health care decisions that are used have been validated. Adherents of evidence-based medicine propose to reduce the knowledge gap between research and practice by understanding conflicting results and by assessing the evidence: that is, determining how good it is, how strong it is, and whether it is applicable to patients and to health care policy.

In the 1980s, the staff of the Faculty of Health Sciences at McMaster University published a series of articles on critical appraisal of the literature, produced an influential textbook [3], and began offering annual 5-day workshops on practicing and teaching critical appraisal. Over the years, their focus shifted from how to read the medical literature to how to use the medical literature, with emphasis on the care of the individual patient. In 1992, this new approach was christened evidence-based medicine [4]. This group's efforts have resulted in a movement for more informative abstracts for journal articles [5] and the publication of two journals, ACP Journal Club and Evidence-Based Medicine. These journals provide structured abstracts of recent studies that are assessed for validity and presented in a format consistent with the standards of evidence-based medicine.

The two new books review and extend concepts that have been previously published. The book by Sackett and colleagues is small, uses simple and understandable language, and provides excellent teaching aids, such as helpful short bibliographies, usable tables, clinical examples, and pocket cards that simplify retrieving formulas and doing calculations. It provides URLs for Web sites that support evidence-based practice and sources of appropriate software. It addresses such topics as when to use a textbook, when and how to use available software, and when to do a full literature search to answer a clinical question. The book is remarkably free of jargon and is clearly organized. It demythologizes evidence-based medicine and confronts the controversies about it. Strongly anchored in clinical problem solving, the book is built around how to identify and frame an important clinical question that arises out of a clinical encounter; how to find (or at least search for) the best evidence; and how to determine the validity, importance, and clinical relevance of that evidence. The major areas addressed are diagnostic tests, prognostic markers, therapy, harm, practice guidelines, overviews, and economic evaluations. The authors promise to update the book regularly, and an electronic version is probably not far off.

Sackett and associates' book also introduces readers to such searchable evidence-based resources as Best Evidence (available on disk or CD-ROM), which contains ACP Journal Club and Evidence-Based Medicine, and the Cochrane Library (available on disk), a compendium of the full texts and structured abstracts of the systematic overviews prepared and updated by the Cochrane Collaboration. It also provides aids (such as modifiers and hedges) to help with searching MEDLINE for evidence-based articles (such as randomized clinical trials, which are the gold standard for evaluating therapy).

The chapter entitled "Searching for the Best Evidence" takes the reader from how to define an important searchable question to when to use a primary electronic source (MEDLINE) or a more efficient secondary source (Best Evidence, the Cochrane Collaboration). To answer the question, "What are the sensitivity and specificity of various noninvasive diagnostic tests for Helicobacter pylori?" the authors walk the reader through the entire bewildering array of resources available on the Internet and in various electronic media and help the reader identify sources that offer the quantitative data that can best inform clinical decision making.

The authors state that "our primary perspective in this book is the individual clinician." For someone who is starting from scratch, however, hard work is required to learn these skills. The vocabulary and abbreviations are challenging: NNT, CER, EER, ARR, CI, NNH, RRR, PEER, pre- and post-test probability, and odds ratios are just a few to be conquered. One cannot just open this book at any point and begin; using it requires either some serious experience with evidence-based medicine or a true commitment to learning about this discipline. This book will be most valuable to persons who have already started down this road and find it intellectually stimulating and empowering. For the busy physician prepared to master evidence-based medicine, this book is an excellent but demanding guide. It is certainly suitable for use after attending one of the increasing number of evidence-based medicine workshops that are being offered.

The greatest strength of Sackett and coworkers' book is its practical, extremely helpful section on how to incorporate teaching and role modeling of evidence-based medicine into the daily lives of students and trainees. One chapter offers an outstanding set of recommendations, almost all of which are patient-based, on how to incorporate evidence-based medicine skills into rounds, conferences, journal clubs, and the ambulatory arena. However, the authors point out that acting as a role model for using evidence-based medicine in clinical problem solving is the most powerful way to teach these skills. Evidence-based medicine is most rewarding when done not alone but with others who are equally committed to its practice and are familiar not only with its strengths but also its weaknesses-for example, areas for which no strong evidence exists to guide decision making. To help expand the numbers of clinician-teachers who are comfortable in this new world, this book adds an important section on how best to teach the teachers.

We can provide some insight into the effectiveness of this approach. Despite the presence of eight medical schools in the New York area, no organized support for the teaching and practice of evidence-based medicine had emerged before 1994. In that year, a coalition of medical school and hospital faculty and librarians collaborating with the New York Academy of Medicine (a training arm of the National Library of Medicine) initiated the New York Evidence-based Medicine Project to target physicians who train residents. Using the McMaster University evidence-based medicine course as a model, the New York Evidence-based Medicine Project produced a 4-day training course that modeled several of the above teaching techniques, created a World Wide Web page with evidence-based medicine links, published a newsletter, and made a commitment to disseminate information on evidence-based medicine through New York's internal medicine residency programs. The group is now planning a resource center and a Web discussion group. Sackett and colleagues' book will be an excellent resource for the continuation of this effort.

The greatest challenge to the application of evidence-based medicine skills clearly exists at the level of the practicing physician. Can the busy practitioner find the time to learn and apply these skills? Is it reasonable to expect that they can be used during a patient encounter? In our experience, application of these skills in practice is a work in progress. During a crowded clinic session, it is almost impossible to find the time to search MEDLINE for relevant articles, assess the validity of abstracts, retrieve an article, and determine the applicability of the article to the patient. Practitioner use of evidence-based medicine skills will increase only when high-quality evidence is more easily identified, all published recommendations (such as guidelines and consensus conferences) include explicit statements about the level of evidence, and systems exist for accessing evidence-based medicine resources in as close to real time as possible. Sackett and colleagues advise us that increasingly accessible electronic sources are being developed and will appear regularly; the latter is clearly becoming the prime requirement for the dissemination of evidence-based health care.

Major concerns have been raised about the use of evidence-based medicine to set health policy, or, as Muir Gray says, "doing the right thing right." The task of applying population data to individual persons is broadened (and complicated) by issues of cost. Such issues raise questions about whether evidence-based health care will be used to disenfranchise certain groups if cost-per-year-of-life-saved is used as the metric. Will policymakers use data from evidence-based medicine appropriately? Will the physician's freedom to offer state-of-the-art therapy be jeopardized if that therapy is expensive or its efficacy is not yet definitively proven? If no high-quality evidence shows that a treatment is efficacious, will that translate into "no evidence supports using this treatment" and, consequently, will nihilism prevail (or failure to reimburse ensue)?

Muir Gray's book is an ambitious effort to define, advocate, and teach how to make evidence-based management and policy decisions and how to develop these skills in individual persons and in organizations. The book has an accessible format and is clearly designed to avoid overwhelming the reader. The basic tenets of evidence-based medicine are reiterated, with expanded emphasis on outcomes, equity, effectiveness, cost-effectiveness, quality, and patient satisfaction. The book has valuable appendices that direct the reader to resources that support the author's recommendations. It describes a Web site that updates the book and regularly adds to the resources. The section on key components of an evidence-based organization states that "the culture of an evidence-based organization is an obsession with finding, appraising and using research-based knowledge in decision making." It plunges into the question of how an "evidence-based chief executive" can move an organization to provide evidence of validity for all change (evidence-based purchasing and evidence-based primary care, for example). He notes that the most "challenging item on the new management agenda" is to get clinicians and managers to work together to guide evaluation of clinical practice. Some of the themes parallel those that are produced in the transition to managed care, with its tension among quality, cost, and access.

For the past 5 years, the National Health Service in England has supported evidence-based decision making. Because Muir Gray's book presents a British perspective, exactly how some of his suggestions will be received in the United States remains unclear. Are there many evidence-based medicine chief executive officers out there who are prepared to lead their organizations to meet the goals and practices described here? A challenge indeed.

A hard-to-shake obstacle to implementing evidence-based health care remains the belief on the part of many physicians that its practice takes the art out of medicine. In the first edition of their book on clinical epidemiology in 1985, Sackett, Haynes, and Tugwell [3] addressed this concern by describing their efforts as developing a "science of the art of medicine" and emphasizing that the origins of their efforts were "in clinical practice, as its authors struggled with the diagnoses and management of their patients and fell slowly behind in their clinical reading." In the same preface, they caution against substituting a new tyranny of unachievable "methodological rigor" for the old tyranny of unteachable "clinical art" [3].

Whether evidence-based medicine teaching can change and improve practice and how that outcome can be measured are still largely uncertain. New avenues of evidence-based clinical research are attracting increasing numbers of young physicians. These two books lay the groundwork for understanding evidence-based medicine and its advocates and for deciding whether and how to be a part of this approach to clinical practice and health care policy. The next few years will tell us about its effects.

For those who want to keep up with what's out there, the McMaster University Web site (http://hiru.mcmaster.ca), the Web site of the Center for Evidence-based Medicine, Oxford (http://cebm.jr2.ox.ac.uk), and ACP Online (http://www.acponline.org) all offer the opportunity to learn about evidence-based medicine, to explore evidence-based medicine resources, and to link to other evidence-based medicine sites.

Eleanor Z. Wallace, MD

Long Island Jewish Medical Center; New Hyde Park, NY 11040

Rosanne M. Leipzig, MD, PhD

Mount Sinai School of Medicine; New York, NY 10029-6574

Dr. Leipzig: Department of Geriatrics and Adult Development, Mount Sinai Medical Center, One Gustave L. Levy Place, Box 1070, New York, NY 10029-6574.