Although there is international consensus regarding the general principles of testing for lupus anticoagulants (LAs), no agreement exists as far as the analysis of the clotting time results is concerned. Twenty-nine laboratories participating in the Fifth International Survey of Lupus Anticoagulants (ISLA-5) reported the activated partial thromboplastin time (APTT)-based clotting times obtained on seven defined test samples and a normal plasma (NP) using the same two reagents with low and high phospholipid (PL) concentrations, respectively. These clotting times were used to analyse how various methods of calculating the results may influence the apparent sensitivity of LA tests. We found that the use of a separate screening test may lead to the exclusion of samples where the presence of LA would have been detected by a combined screening and confirmatory method. For instance, the dilute APTT (dAPTT) gave a sensitivity of 53.5% (screening test), while the calculation of a ratio between the clotting times obtained with two different PL concentrations gave a sensitivity of 68.1% (confirmatory test). The normalisation of results by dividing with the corresponding results of NP increased the apparent sensitivity. The screening test ratio between dAPTT results of test samples and NP gave a sensitivity of 84.7%. The normalised ratio between the clotting times obtained with the two reagents (lupus ratio, LR) gave a sensitivity of 95.1%. We conclude that when testing for LA, all samples should be tested with both low (screening procedure) and high (confirmatory procedure) PL concentrations. These two clotting times should be evaluated in relation to each other and to the corresponding results obtained with a reference plasma (normalisation).