A few weeks ago I was informed I had an “excess mortality”, and premonitions of doom flooded over me. Over the preceding five months I was reported to have had 10 excess deaths, “indicating poor performance”, compared with what would be expected on the basis of age, sex, primary diagnosis and route of admission, according to the Dr Foster database. When I actually looked at records of the patients concerned, it was apparent that “primary diagnosis” was an irrelevance for many of them: most had multiple, and often end-stage, co-morbidities, and some were simply incorrectly coded – one with a total anterior cerebral infarction had gone down as “syncope and collapse”.

So why tell this sorry tale to the membership of the BGS? Firstly, in these days of performance monitoring, it’s wise to ensure the right things go down on the KMR1 front sheet – and under “Payment by Results” Trusts will have an interest in all relevant diagnoses going down as well. Secondly, and more seriously, is the potential threat to those of us who choose to look after frail old people, of comparisons with systems such as the Dr Foster database, relying on arbitrary principal diagnoses, rather than on the total patient. Just as there’s anecdotal evidence to suggest that NCEPOD and mortality monitoring may discourage surgeons from operating on frail people, will there now be pressure on us to avoid the care of those who need us most, or to try and discharge dying people as soon as possible to nursing homes?

Edmund Dunstan