To determine the accuracy of subject and author indexes in five anaesthesia journals we examined all 1989 and 1991 volumes of Anesthesia and Analgesia, Anesthesiology, British Journal of Anaesthesia, Canadian Journal of Anaesthesia and European Journal of Anaesthesiology, and counted the number of entries and the number of errors in both indexes. The number of errors was expressed as a percentage of the number of entries, and the incidences of errors were compared by chi-square contingency table tests. The overall error rate in both indexes and for both years (1989 and 1991) differed significantly among the five journals (P < 0.001). Many of the individual differences were also significant. British Journal of Anaesthesia had the lowest error rate (0.9%), followed by Anesthesia and Analgesia (2.4%), Anesthesiology (3%), European Journal of Anaesthesiology (2.9%) and Canadian Journal of Anaesthesia (11.4%). The combined index accuracy was improved from 1989 to 1991 for Anesthesia and Analgesia, British Journal of Anaesthesia and Canadian Journal of Anaesthesia, did not change for European Journal of Anaesthesiology and was worse for Anesthesiology.
