It may also come as a surprise, then, that the trendiest quality measures for hospitals relate to mortality: the Hospital Standardised Mortality Ratio (HSMR) and the Summary Hospital Mortality Index (SHMI, or ‘shimmy’). The theory goes that hospital quality and mortality rates are inversely related: the higher the quality, the lower the mortality rate.
To be clear, HSMR/SHMI relate to particular cohorts over particular time periods, and they provide a number, which shows whether there are more or fewer deaths than expected. This, ‘expected’ bit is crucial; it relates to historic data and is the sum of the estimated risks of death. A HSMR/SHMI of 100 is ‘average’, meaning that 50% of hospitals will have lower than ‘expected’ and 50% will have higher than ‘expected’. Unfortunately, statistical niceties are lost on most of the public and journalists, so phrases like, ‘worse than it should be’, or ‘2,000 excess deaths’, are commonly heard.
There is huge debate and argument amongst the medical community about the validity and reliability of HSMR/SHMI, but the upshot is that they are here to stay and hospitals are judged on them. Mortality ratios are to hospital bosses what the share price is to the CEO’s of listed companies; they live or die by them!
I have attached a couple of documents that explain the ratios.