The Medical Literature as a Resource for Evidence Based Care

The Medical Literature as a Resource for Evidence Based Care

Contents

• Abstract
• Introduction
• Medline Studies
• ACP Journal Club
• Systematic Overviews
• Cochrane Collaboration
• Evidence Based Guidelines
• On-line Resources
• References

Abstract

Evidence-based medicine (EBM) is an approach to health care that promotes the collection, interpretation, and integration of valid, important and applicable patient-reported, clinician-observed, and research-derived evidence. The best available evidence, moderated by patient circumstances and preferences, is applied to improve the quality of clinical judgments. The Health Information Research Unit of McMaster University Faculty of Health Sciences in conjunction with others at McMaster and elsewhere has developed, evaluated, and made more accessible information tools to facilitate the practice of EBM. These include users' guides to the medical literature, strategies for improving the yield of MEDLINE searches, standardized formats for abstracts of journal articles and guidelines, new journals, systematic reviews and meta-analyses, and software tools that bring high quality information to the point of clinical decision making. This article discusses these and other EBM resources for health care professionals.

Introduction

Traditionally clinicians have favored past experience, prevailing practice, professional training, and peer opinion as guides for day to day decisions about patient care. These sources have become less reliable as the volume and complexity of medical information has grown exponentially. In the 1970s, clinicians and researchers at McMaster University began to address this clinical information overload by developing tools to help clinicians to discern the clinically valid, important, and applicable details. "Critical appraisal" of the medical literature was introduced as an approach to reading that would help practitioners sift clinically relevant and methodologically sound studies from the countless articles appearing in medical journals. The original critical appraisal series was published in the Canadian Medical Association Journal, then translated into many other languages, and disseminated widely. A new series of "users guides to the medical literature", appears in the Journal of the American Medical Association (Guyatt, 1993a, 1993b, 1994; Oxman, 1993, 1994; Jaeschke, 1994a, 1994b; Levine, 1994; Laupacis, 1994; Hayward, 1995), and emphasizes the perspective of busy clinicians who need quick and simple ways to make sense of a burgeoning medical literature.

The "evidence-based medicine" (EBM) movement (Evidence-Based Medicine Working Group, 1992) asserts that potential advances in health care must be tested and proven to do more good than harm before they are incorporated into medical practice. EBM is an approach to patient care that promotes the collection, interpretation, and integration of valid, important and applicable patient-reported, clinician-observed, and research-derived evidence. The best available evidence, moderated by patient circumstances and preferences, is applied to improve the quality of clinical judgments. The EBM movement has grown in parallel with the introduction of problem-based curricula into health sciences training programs and has spread to many other institutions, including the University of New Mexico and Harvard in the United States, the Suez Canal University in Egypt, and Maastricht in the Netherlands.

The medical literature contains evidence that can be used to improve patient care, but only a small portion of the medical literature describes solid advances in the diagnosis, prevention, treatment, cause or prognosis of illness. Finding and understanding the important and valid studies is tough for busy clinicians, and an impediment to the wide adoption of EBM. Efficient, effective information retrieval tools, on the other hand, could ease the information overload of practitioners, and help them apply evidence to clinical practice. In this article, we review the rationale for EBM and describe information tools to facilitate the practice of EBM, emphasizing the research done in the Health Information Research Unit (HIRU) and the Department of Clinical Epidemiology and Biostatistics at McMaster University.

MEDLINE studies

Much of the medical literature, including editorials, letters, case reports, reviews, laboratory and animal studies, is interesting and informative. A relatively small fraction of the literature, however, reports specific, scientifically sound advances that can and should change how clinicians care for patients. Such information is most often found in reports of experiments in people, where groups of persons are systematically exposed to alternative interventions to evaluate the effectiveness of prevention, diagnosis, and treatment of specific health conditions. EBM practitioners need fast and efficient access to subsets of the medical literature enriched with this kind of clinical research. Fortunately, the most trustworthy reports of the effectiveness of health interventions bear common characteristics that can be used to identify them among the huge number of clinical studies and to enable their selective retrieval and synthesis.

The best evaluations of health care result from preplanned, time-limited trials in large numbers of humans, with criterion standard, sensitivity and specificity measurements for diagnostic studies; randomized controlled trials for therapy, prevention, quality of care, and economics; randomized controlled trials, cohort studies, or case-control studies for harm or etiology evaluations; cohort studies for prognosis; and meta-analyses or systematic overviews for reviews of the results of many primary studies. The results of the best evaluations of health care are first published as journal articles, then registered in bibliographic databases maintained at the National Library of Medicine (NLM). Indexers apply medical subject headings (MeSH) to each entry in NLM databases (MEDLARS), and some of these index terms describe how studies were designed and executed. The quality and consistency of methodological indexing was given a boost in 1990 and 1991 when the MeSH vocabulary was expanded to include specific tags for different types of studies (e.g., controlled trial) and publications (e.g. meta-analysis). By including methodologic terms in their search strategies, users of the NLM databases now can look for studies that avoid the most common threats to the validity and applicability of their results.

With the proliferation of personal computers, modems, CD-ROM products, wide-area networks and Internet, MEDLINE searches have become more accessible to clinicians, who are the fastest growing population of MEDLINE users. MEDLINE-derived databases, interfaces and search software packages have steadily improved, further easing the way for clinicians. Indeed, different systems no longer differ substantially for price and performance (Haynes, 1994c, 1995). Some interfaces to MEDLINE, however, may work better for particular subgroups of users. In one study librarians' strategies produced more rewarding MEDLINE searches using on-line systems and clinicians' strategies produced better results with CD-ROM systems (Haynes, 1994c). This suggests that software interfaces and search engines should be selected based on personal preferences, available support and training opportunities.

When MEDLINE access was provided on the wards and clinics of a hospital, experienced clinicians retrieved relevant citations, enjoyed searching, and felt that their searches enhanced patient care outcomes (Haynes, 1990b, 1993; McKibbon, 1990). After experiences with approximately 8 searches, clinicians also retrieved just as many relevant cita

[1] [2] [3] [4] 下一页