Despite the severity of ankle syndesmosis injuries, it has been argued that they are relatively poorly detected. This study investigated the interrater reliability of four orthopaedic tests for ankle syndesmosis injury and assessed their ability to predict the extended recovery times that have been reported as a consequence of this injury. Nine physiotherapists, working in random pairings, examined 53 subjects, all receiving treatment for ankle injury at two private clinics. Each subject was tested by two physiotherapists who independently performed the palpation test, external rotation test, squeeze test, and dorsiflexion-compression test. Kappa coefficient testing indicated that the external rotation test had the best interrater reliability (kappa = 0.75). The squeeze test was found to have moderate reliability (kappa = 0.50), and the palpation and dorsiflexion-compression tests both had only fair reliability (kappa = 0.36). The degree of pair-wise association between the results of the four tests was low (phi < or = 0.30 for all test combinations), suggesting that if all four tests were performed on the same subject, it was not likely that they would achieve similar results. Follow-up interviews were conducted to determine the time taken for subjects to walk 10 m without pain and, for sports injuries, the time taken to return to training and then to competition. For each test, Mann-Whitney U values showed no significant difference between the recovery times of subjects with positive or negative test results, although subjects with recovery times markedly longer than normal were detected by three of the four tests. When an either/or combination of the external rotation and dorsiflexion-compression tests was considered, subjects with a positive test result took significantly longer to return to playing sports.