Background: While systematic reviews should be of assistance in decision making, currently many reviews are unreliable, poorly reported or not clinically useful. It is therefore important that they are read critically.
Objectives: The objective is to provide a logical way to appraise systematic reviews. Particular objectives are to demonstrate how to use a validated general guide for appraising systematic reviews; amend the guide so that it can be used to appraise systematic reviews of controlled trials, prognostic studies and studies evaluating diagnostic tests; choose between discordant reviews and assess the usefulness of good systematic reviews.
Discussion: Guidelines for undertaking and reporting systematic reviews should eventually improve their quality and usefulness. In the meantime, reviews must be appraised critically, particularly those outside the Cochrane Library.