Thirteen newly appointed senior house officers (SHOs) at two hospitals were assessed in the recording of two variables: Glasgow Coma Scale (GCS) and diagnostic coding at Hospital A, and GCS at Hospital B. At Hospitals A and B baseline recordings of these variables were established. Active feedback (in the form of data presentation and discussion) was given at 6 and 11 weeks and a final 4-week audit was performed on GCS recordings at 20 weeks. Analysis of GCS recordings of head injuries at Hospitals A and B showed an increase from baseline values of 162/401 (40%) to 336/420 (80%) after feedback 1, and after feedback 2 recordings increased to 379/429 (88%). This was maintained in the last 4-week audit which showed recordings of 220/244 (90%). Further analysis of Hospital A's data showed the SHOs recorded diagnostic coding in 1335/4406 (30%) of cases. After feedback 1, recordings increased to 2550/3327 (77%). After feedback 2 there was no change in recording 2712/3530 (77%). Active feedback improves data recording. The first feedback has a greater impact than the second. Two feedbacks produce a sustained effect in 6 months.