4.8.1      Study results

Results of the studies included in a review can be organised either by entering all data from a single study together even when it includes several tests (data tables by study), or by entering all data for a single test from several studies together (data tables by test).  Each test requires entry of the number of true positives, false positives, false negatives and true negatives.  Where results are available for a test at several thresholds these can either be added as separate results with a covariate (see below) specifying the threshold, or several tests defined each with a specific threshold.  When the review is published all of the data entered will be displayed in paired sensitivity-specificity forest plots, even if these are not included in the review. This flexible approach to organise and enter data differs from that for Cochrane reviews of interventions where the data structure (comparisons, outcomes, subgroups) defines the analyses which will be undertaken. In reviews of test accuracy all data are entered before the structure of the analyses is specified.