Twenty years ago in the BMJ the Society of Cardiothoracic Surgeons of Great Britain and Ireland published data from its cardiac surgical register.1 Figures for 1977-82 showed wide variation in provision of coronary operations—from 14 per million of the population in south west England to 200 per million in the London area. The north west and Mersey were well down at 50 per million (see figure on bmj.com). Reporting was by hospital and the identities were concealed before collation. In this issue the north west surgeons, in response to a clamour that has been around for some time, provide survival figures for 25 named surgeons.2 Others have posted surgeon specific mortality on the web.3 These data are believed to be “robust, rigorous, and risk adjusted,” as required by Alan Milburn, former secretary of state for health, in parliament.4

Does disclosure of results serve patients' best interests? To put the question another way, what purpose is served by concealment? For an analysis, read Atul Gawande (who wrote Complications5) in the New Yorker.6 He describes the experience of …