Target setting for elective infra-renal AAA surgery: A single mortality figure?

Area of Science:

• Vascular Surgery
• Health Services Research
• Biostatistics

Background:

• National standards for elective abdominal aortic aneurysm (AAA) repair mortality are set at <6% by the UK NAAASP.
• There is an ongoing aim to reduce elective AAA repair mortality to 3.5% by 2013.
• The statistical validity of these mortality targets requires investigation.

Purpose of the Study:

• To investigate the statistical validity of mortality targets for elective abdominal aortic aneurysm (AAA) repair.
• To analyze data missingness in the National Vascular Database (NVD) for AAA procedures.
• To model the impact of unit size and case-mix on observed mortality rates.

Main Methods:

• Interrogation of the National Vascular Database (NVD) to assess AAA data completeness and geographical variation.
• Utilisation of published data (2006-2008) and funnel plots to illustrate NHS Trust mortality rates.
• Application of a binomial distribution model to calculate mortality variation based on patient volume and a "true" mortality rate of 3.5%.

Main Results:

• The NVD AAA dataset exhibits significant data missingness (median 22%), with higher rates in non-preferred variables.
• Simulated funnel plots show greater 3-year mortality variability in smaller surgical units compared to larger ones.
• Data missingness for preferred clinical variables ranges from 5% to 50%, impacting risk modeling.

Conclusions:

• Sole reliance on a single mortality figure for performance assessment is problematic due to statistical variability and unaddressed case-mix factors.
• Further research into mortality target setting is crucial.
• Developing national strategies and facilitating international comparisons are recommended for improved performance evaluation.