Incidence, nature and impact of error in surgery.

BACKGROUND: Adverse events occur in 3·8-17 per cent of hospital admissions. The purpose of this study was to analyse the incidence of medical errors and assess the feasibility of an error registry for quality improvement programmes.
METHODS: Errors were recorded prospectively in a complication registry between 1 June 2005 and 31 December 2007. Events were coded according to the Trauma Registry of the American College of Surgeons; the nature of events was recorded and the severity graded using the 1992 Clavien system. Recorded events were discussed by the medical staff on a daily basis and, if by consensus judged to be errors, were saved to the registry database.
RESULTS: Of 12,121 patients admitted to the surgical ward during the study interval, 2033 (16·8 per cent) had a complication and 735 (6·1 per cent) had an error documented in the registry. Of 873 recorded errors, 607 (69·5 per cent) were of little or no consequence (Clavien grade I) and 220 (25·2 per cent) required therapeutic intervention (Clavien grade IIa and IIb). Errors leading to permanent injury (Clavien grade III) occurred in 41 instances (4·7 per cent) and five patients (0·6 per cent) died (Clavien grade IV).
CONCLUSION: This study shows that errors are common in surgery, and that near misses are more frequent than errors with serious consequences. It is hypothesized that registration of near misses might prevent errors with serious consequences and thus improve quality of care.