Use of mortality rate after aortic surgery as a performance indicator.

Abstract

Hospital episode statistics (HES) regarding death after aortic surgery were analysed to evaluate their potential value as a performance indicator. HES data for all acute hospitals in England and Wales from 1996 to 2001 were analysed retrospectively. In-hospital mortality was calculated for all patients over 40 years of age who underwent abdominal aortic procedures or died in hospital with a primary diagnosis of aortic aneurysm. Some 38 319 cases were identified, of which 8.9 per cent were complex, 46.8 per cent were elective and 44.4 per cent were emergencies. The elective mortality rate was 6.4 per cent overall; that after emergency operation was 35.0 per cent, rising to 41.6 per cent if urgent procedures were excluded and 63.1 per cent if unoperated cases were included. Over the 5-year interval a median of 68 (interquartile range 30-108) elective procedures were carried out by individual hospitals. Considerable variation was identified in the proportion of elderly patients, tertiary referrals and the proportion of emergency admissions that had surgery. The use of mortality data may be misleading owing to identifiable differences in case mix and selection. Some redefinition of groups may help to provide more valid data, but ultimately only high-quality data with clinical information are likely to allow meaningful comparisons of performance.