Abstract
Background Hospital Episode Statistics (HES) is often used by hospital managers and politicians as a reflection of departmental workload. The accuracy of this data is often questioned. We aimed to ascertain the reliability of this database for trauma admissions.
Method Between August 2002 and July 2003, all admissions were recorded by doctors using a separate departmental database. Data was collected during the daily trauma meetings. This data was compared with the HES return for the same period.
Results 2496 patients were recorded in the trauma admissions database. Overall, 36.4% of the patients were either not recorded by the HES database or wrongly coded in terms of type of admissions or diagnosis.
HES data for all 2496 records was analysed by type of admissions and speciality.
4.2% of trauma patients were incorrectly classified as elective or day cases.
2.9% of trauma patients admitted to hospital were not recorded in the HES data as orthopaedics admission.
The accuracy of HES diagnosis coding was tested on 300 records randomly selected by a statistical package.
HES recorded the wrong diagnosis in 29.3% of cases.
Conclusion A significant number of trauma cases were not counted in the HES data. This may have significant implications for trauma funding.
HES data does not accurately record diagnoses and therefore can not be used as a research tool for specific injuries.
Data recording practice should be changed to improve HES data accuracy.
Theses abstracts were prepared by Professor Roger Lemaire. Correspondence should be addressed to EFORT Central Office, Freihofstrasse 22, CH-8700 Küsnacht, Switzerland.
