Article Text

Download PDFPDF

Best Evidence 4
  1. Kenneth A Locke, MD
  1. Mount Sinai Hospital University of Toronto Toronto, Ontario, Canada

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Best Evidence 4 is a compilation on CD ROM of the entire contents of ACP Journal Club (ACPJC), Evidence-Based Medicine (EBM), and the 1999 edition of the textbook Diagnostic Strategies for Common Medical Problems (DS). It includes all published editorials from ACPJC and EBM and an archive of studies now judged by the editors and study authors to have become obsolete since original publication in the journals. The CD is targeted to readers of both ACPJC and EBM, although most of its references are most relevant to internal medicine (a reflection of the preponderance of ACPJC articles).

    ACPJC and EBM are both journals of secondary publication and include abstracts of selected articles on various topics, including diagnosis, aetiology, treatment, prognosis, economic analysis, and quality improvement. DS focuses on the diagnosis of such common medical problems as myocardial infarction, acute monoarthritis, and pulmonary embolism. The criteria for study selection are explicit and valid for both ACPJC and EBM, but the criteria used to select evidence for DS are less clear.

    Several options are available for searching, including browsing the table of contents for each of the component publications; using a text word search function across all publications, with filters that can be invoked with a single keystroke for diagnosis, prognosis, therapeutics, aetiology, economics, quality improvement, or clinical prediction guides; using an alphabetical index; using a “clinical content map” designed to group related material together; and using a “favourites” function with user definable folders. The internet can be accessed without leaving the software, and web pages can be inserted in the favourites folder.

    The screen interface works like a simple web browser. It uses 2 screens, 1 for searching and the other for displaying search results that can be either resized to appear on the same screen or toggled. The same interface is used when web pages are accessed, so most regular browser functions are not available. Search results are presented in a ranked order by number of hits of the search terms used. The source publication (ACPJC, EBM, or DS) is noted beside each item. It took me several tries to understand exactly what all the browser functions were and how to use them. Once mastered, however, it becomes relatively easy to navigate within the program, although the need to keep toggling back to another window to change search strategies remains a nuisance.

    I used this software to answer clinical questions that arose from my clinical teaching practice. While searching for the operating characteristics of the clinical examination for deep venous thrombosis, the search function (“deep venous thrombosis” and “clinical”) yielded 77 citations (not all relevant), 1 of which was a useful review of the topic.

    By using the clinical content map, I found the same article faster, but I had to guess that the relevant category would be “cardiovascular illness.” Exploring DS for the same question disclosed no mention of this article or of an earlier one by the same group. When I searched for evidence about the tests used in the diagnosis of temporal arteritis, there were several useful tables in DS. For the best method of diagnosing haemochromatosis, 1 citation in which the word was only mentioned in passing was retrieved.

    A search for evidence for the treatment of hepatitis C found 5 articles, 2 of which were not relevant, and only 1 of 2 recent large trials of combination interferon-ribavirin treatment was mentioned. Other diagnostic and therapeutic questions yielded similar types of results.

    Best Evidence 4 gives one the ability to rapidly search 2 of the premier evidence-based resources (ACPJC and EBM) currently available. It is most useful for quick searches for high quality studies. Its major problem is that studies not reviewed in its parent publications will not be found.


    Methods/Quality of information: ★★★★☆

    Clinical usefulness: ★★★☆☆


    • Best Evidence 4 may be purchased at subscriptions{at} for £70 (BMA Members £50 only and 10% off CD ROM for joint subscriptions to EBM and Best Evidence). Best Evidence 5 is expected early in 2001.