The colour coding was wrong in the pie charts accompanying this news article by Susan Mayor (16 November, p 1129). The correct data from the National Confidential Enquiry into Perioperative Deaths are as follows. The percentage of cases of deaths after surgery that were discussed by anaesthetists in morbidity and mortality review meetings was 36%; 57% of cases were not discussed; and in the remaining 7% of cases it was not known whether they were discussed. The corresponding figures for discussions by surgeons were 77%, 19%, and 4%.
