Evidence-based Practice in Radiology: An Introduction to the Series

Abstract

Dermot E. Malone, MD, FRCPI, FFRRCSI, FRCR, FRCPC Late one evening in 2004, my home telephone rang. It was Dr Harald Stolberg. In typically enthusiastic fashion, he told me, “Dr Tony Proto wants a series for the Evidence-based Practice section of Radiology (1) that will explain to practicing radiologists what evidence-based radiology can mean to them. He says, ‘Make it readable!’ Would you like to join me as a co-editor?” I accepted (of course) and we laid out the series outline together over the next few weeks. Unfortunately, in January 2005, Harald became acutely unwell and passed away after emergency abdominal surgery and a short illness. It is not my intent to repeat his obituary (2), but no introduction to this series would be complete without acknowledging the vision, drive, and commitment to excellence in radiology that made Harald a champion in the cause of integrating evidence-based practice (EBP) principles into radiology. He was an inspirational figure. After forming and coordinating the production of the article “Evidence-based Radiology: A New Approach to the Practice of Radiology” by the Evidence-Based Radiology Working Group (3) and submitting it for publication in November 2000, he met several of the authors for dinner at the Radiological Society of North America 2000 Annual Meeting—not just to celebrate the conclusion of that task (although we did that) but to ask, “What’s next?” When Harald and I worked on the outline of this series, we considered that “evidence-based” practice has become a catchphrase in recent years. In 2004, professor Paul Glasziou (director of the National Health Service Centre for Evidence-based Medicine [CEBM], University of Oxford, England) revived terminology originally used by David Sackett (4) that we adopted and that will be used throughout the series (Paul Glasziou, written communication, 2004). He noted that to some, EBP identifies a series of centers charged with producing evidence reports and technology assessments to support guideline development by other groups (3). To others, it means the centralized production of guidelines and their integration into practice. These definitions have in common the assumption that the ordinary practitioner is best served by a centralized process, external to his or her practice. These schools of thought can be termed top-down EBP (4). While centralized academic activity is important and the key to a good evidence base, there is another school of thought about EBP. To this group, EBP is “the integration of best research evidence with clinical expertise and patient values” (4,5). It is a process that was originally developed at McMaster University, Canada, and subsequently refined there and at the CEBM. This approach has roots in the fields of clinical medicine, epidemiology, and adult learning theory (problembased learning). It is concerned with information, individualized problems, and the use of the Internet and modern informatics to get the best research evidence into practice. The underlying assumption is that the practitioner is best served by a decentralized but exact five-step approach internal to his or her practice. He or she asks an answerable question, searches the literature to locate evidence produced by the top-down EBP groups, appraises it explicitly, searches primary literature (if necessary) to cover any relevant temporal gaps, and then proceeds to apply and evaluate the best current evidence with due regard for local circumstances (5). In radiology, the radiologic expertise is the central component; research evidence, clinical circumstances (including comorbidities), and, in particular, the patients’ and referring physicians’ preferences become part of the rational decisions made. This approach is derived from Published online 10.1148/radiol.2421060010