 Canadian Journal of Ophthalmology The official journal of the Canadian Ophthalmological Society |
||
|
Volume 39, no.1, February 2004 |
||
|
Editorial The evaluation of scientific information in medicine Je n'y ai décrit que ce que je crois connaître, laissant de côté ce que je ne connais pas. Et chaque jour me prouve que ce que je connais est bien peu de chose auprès de ce que j'ignore. - Pierre Masson, Tumeurs humaines1
There are many ways to obtain scientific information in ophthalmology, including textbooks, conferences and publications as well as informal exchanges of information between colleagues. Numerous myths have been circulating throughout the years regarding the evaluation of scientific information. For example, "basic science research is more scientifically relevant than clinical data." This is not necessarily accurate. Basic science data are usually obtained from well-controlled experiments; however, these data do not always reflect the actual clinical dilemmas we face with our patients. Another myth is that "only information obtained from peer-reviewed publications is valid." This is a common misconception, particularly in view of the fact that there is no uniform standard policy for the peer-review process. Scientific journals in particular tend to target the specific area of interest of the majority of their readership rather than providing a wider range of topics. Owing to this trend, it has become increasingly difficult for the individual ophthalmologist to determine what information is accurate as well as relevant to his or her clinical practice. Evidence-based medicine is a very important tool to help the clinician find and critically appraise information. The strategies that are usually used are in the form of clinical questions that are pertinent and answerable, which enables one to track down articles and decide which articles to read and which information is believable. Three basic questions should always be asked when a physician is seeking information in order to evaluate a particular clinical dilemma:
According to Guyatt and Rennie,2 "'yes' and 'no' are often not adequate answers to these questions…. Evidence comes in shades of grey. Frequently results may be valid, perhaps demonstrate an important effect, and they might improve the patient's condition." Considering that most physicians use a bibliographic database such as MEDLINE to find the answers to their questions, it is important to have criteria to critically evaluate the level of evidence of published studies. A recent publication in the Canadian Journal of Diabetes summarizes the criteria for assigning levels of evidence to published studies.3 As shown in Table 1, different levels are applicable to studies of diagnosis, treatment and prevention, and prognosis. Guyatt and Rennie2 propose a simpler approach (Table 2). They classify primary studies into categories such as therapy, diagnosis and prognosis, and suggest specific questions in each category. For example, for studies of therapy, questions include whether patients were randomly assigned to treatments and whether all the patients involved in the trial were properly followed. Concerning studies of diagnosis, an important question is whether there was an independent, masked comparison with a reference standard. This information is intended as a tool for acquiring scientific information. However, it is imperative that scientific journals provide a solid, well-balanced variety of papers dealing with clinical and basic science reports. The reader should keep in mind that common sense should always prevail, particularly in circumstances in which a patient does not fit in any of the categories described in clinical trials. Miguel N. Burnier, Jr., MD, PhD Editor-in-Chief References 1. Masson P. Tumeurs humaines : histologie, diagnostics et techniques. 2nd ed. Paris: Librairie Malonie; 1956. 2. Guyatt GH, Rennie D. Users' guides to the medical literature [editorial]. JAMA 1993;270(17):2096-7. 3. Canadian Diabetes Association 2003 Clinical Practice Guidelines for the Prevention and Management of Diabetes in Canada. Can J Diabetes 2003;27(Suppl 2):S5. |
||
|
|
||