Getting operating theatre metrics right to underpin quality improvement: understanding limitations of NHS Model Hospital calculations.

Abstract

The Model Hospital is an NHS online resource summarising performance data for, amongst other things, operating theatres categorised by NHS Trust and specialty. As an official source of information, it might be assumed that metrics, such as 'average late start time', 'average early finish time', and 'average late finish time', are calculated in a way to reflect performance in these domains, but this is not the case. These values are, respectively, only for those lists that start late, finish early, and finish late, with the number of lists in each category unreported. The Model Hospital reports utilisation appropriately as 'touch time' (the time delivering anaesthesia and surgery) but prefers a 'capped' measure, in which any touch time occurring in late finishes is ignored. The Model Hospital aggregates utilisations across lists in a mathematically invalid way, which leads to the assumption that small aliquots of unused time on lists can be combined to create larger time blocks, in which to complete more operations. We present alternative, more intuitive, and mathematically conventional methods to derive performance metrics using the same data. The results have implications for hospitals developing their own dashboards and international organisations seeking to create national databases for operating theatre performance.