Background: Statistical and methodological issues have, until relatively recently, dominated the test accuracy research landscape. Consideration of clinical context (the intended setting application and role of tests, the downstream consequences of test results and the use of test accuracy measures to convey contextual information) has only relatively recently received attention. Consideration of test accuracy in the absence of contextual information may mislead when making recommendations about test use.

Objectives: To investigate the extent to which a representative sample of diagnostic test accuracy (DTA) reviews represent the clinical context in which an index test is to be used when formulating a review question, deciding on synthesis methods, reporting results and making recommendations, and on the basis of results to provide recommendations for the reporting of contextual aspects of DTA reviews.

Methods: Published and unpublished reviews were sought by interrogation of the DARE database; the Cochrane Database of Systematic Reviews; the database of systematic reviews hosted by the Aggressive Research Intelligence Facility, University of Birmingham; the UK NHS National Research Register; and contact with the Cochrane DTA working group. A final random sample of 100 reviews was included from 271 reviews eligible on the basis of title and abstract.

Results: 100 included reviews represent 17 disease topic areas and include between 1 and 50 index tests. Preliminary scrutiny of included studies suggests: lack of clarity of question formulation, particularly with respect to index test role and current testing practice; a lack of consistency between question formulation, methods and reporting of results; and unsupported use of test accuracy outcome measures.

Conclusions: The findings of this review will have implications for the development of standards for reporting of DTA reviews and the development of the ummary of results table for Cochrane DTA reviews.