Mortality following aortic surgery as a possible indicator of clinical performance

Abstract

Abstract Background The mortality rate following aortic surgery has been suggested as a possible indicator of clinical performance. Methods Routine hospital data on mortality following aortic surgery were collected for a 2-year interval in the Trent region. Procedural and diagnostic codes, mode of admission, district of residence, treatment and specialty data were analysed and compared with audit data and the Operating Theatre Information System. Results Inaccuracies in coding using the HRG (Health Resource Group) codes meant that 21·4 per cent of elective aortic cases (HRG Q01) were probably emergencies and 26 per cent of probable emergency ruptures were not coded as vascular emergencies. The HRG system does not identify complex procedures carried out in specialist centres, thereby distorting the overall mortality figures. Case mix and patient selection introduced a bias, apparent between tertiary and district general hospitals, with higher-risk elective and lower-risk emergency cases at tertiary referral centres. For patients aged over 80 years, two districts undertook no elective aortic surgery, and the rates for emergency surgery varied between 16 and 25 per cent in the districts and 77 per cent in the tertiary centre. After correction for these factors, there remains significant variation in mortality rates between tertiary referral centres and the district hospitals (Table). Conclusion Some of the difficulties in using crude mortality rates as an indicator of performance have been highlighted. They are subject to bias and distortions owing to the collection of incorrect information, patient selection differences between hospitals, case-mix differences and tertiary transfer of complicated cases. Other case-mix factors can be accounted for only with more detailed collection of clinical data such as POSSUM (Physiological and Operative Severity Score for the enUmeration of Mortality and morbidity) scores. However, there remains considerable and worrying variation in selection and outcomes of patients undergoing aortic surgery.