An accident and emergency based child accident surveillance system: is it possible?

OBJECTIVE To evaluate the possibility of setting up a database on childhood accidents within an accident and emergency (A&E) department. DESIGN A proforma detailing epidemiological details about the child and details of the injury was designed and tested for accuracy. It was completed in parallel with the existing case documentation. A retrospective sample was analysed after completion of the study to determine times when data collection was poorest. SETTING Regional paediatric hospital (total catchment population c. 500,000). RESULTS Of 13,958 patients in whom full information was available, 65% had all the information available on the forms. Areas of discrepancy in the remainder included (1) transposition of date and time of injury with date and time of attendance (73%), (2) wrong coding (11%), (3) illegible digits (6%), (4) inaccurate data entry to computer (6%). Form completion was worst between 1800 hours and midnight. CONCLUSIONS Data collection is feasible in the A&E department and is a necessary step in effective child accident prevention. It should be done using real time entry onto computer systems. Additional audit staff must be employed to ensure data collection is as complete as possible as close to the time of the initial attendance as possible. Regular analysis of the findings is essential.