Systematic reviews of diagnostic tests in cancer: how much evidence is there to support clinical guidelines?
Susan Mallett and Doug Altman
Diagnostic accuracy is a cornerstone of therapeutic treatment. In cancer multiple tests are often needed to diagnose, stage and monitor treatment of disease. However, the accuracy of many tests used to diagnose cancer is not supported by systematic reviews which compile and assess the best available evidence.
A list of systematic reviews of diagnostic tests relevant to identifying cancer in symptomatic patients has been compiled. Systematic reviews provide the best available evidence on the accuracy of diagnostic studies. We explored the availability of such studies to inform the diagnostic test recommendations in Clinical Practice Guidelines for cancer.
We found that there are comparatively few systematic reviews of the accuracy of diagnostic tests in cancer. In addition current guideline recommendations for diagnostic tests are often unsupported by a review of research evidence. For example, only 14% (3/21) of the recommended tests in ovarian cancer could be linked to evidence from systematic reviews.
We hope to feed the information from this study into the planned diagnostic section of The Cochrane Library.
