The aim of our study was to assess the accuracy of the theatreman system for data retrieval and to identify possible causes of the inaccuracies found.

A retrospective analysis was undertaken in our orthopaedic and trauma theatres at Morriston Hospital, Swan-sea. We reviewed 110 operations carried out in the department over a six-week period.

The following sources were assessed: case notes, theatre logbooks, theatre coding sheets and data from the-atreman.

Our study identifies inaccuracies and problems in data collection and its retrieval. This problem has been already highlighted by other sources.

As the codes are such a source of inaccuracy, with modern high processing capacity computers, we believe accuracy could be vastly improved by using plain language data entry. This avoids manual conversion to codes and eliminates inte-operator discrepancies and reluctance for some complex code entry. For only 25% of the cases to be retrievable from the theatreman system shows that the whole system and not only the input of data are at fault. As a unit, if we used this information to represent our workload, we would significantly be underestimating our workload.

There is a need for a computer system that recognises words, has ease of data input, generates operation notes and perhaps linked into the patient’s notes. We acknowledge that electronic case notes may help to correct some of these problems but worry that a system introduced with accuracy of data retrieval similar to the theatreman system, is worthless.