This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
—We agree that an overview should be motivated by and rooted in an understanding of the clinical problem and the underlying biology. However, an overview is not a review of the "overviews," as Dr McBride suggests—it is a review of relevant primary research. Overviews, like any other journal article, can be used as a platform for "editorializing" whether or not they are comprehensive. "Completely reviewing the subject area" often means covering a wide range of literature superficially and unsystematically and in no way provides protection against bias. To the contrary, it provides ample latitude for bias. McBride's last sentence refers to "guidelines" that use explicit methods and extensive literature reviews. If McBride means clinical practice guidelines, we agree entirely that the methods should be explicit and we address this in a subsequent Users' Guide. However, readers should be cautious about confusing "extensive" with systematic and valid. We would
Oxman AD, Cook DJ, Guyatt GH. Another View on Overviews-Reply. JAMA. 1995;274(3):218. doi:10.1001/jama.1995.03530030037029