COVID as a nosocomial phenomenon
Squeezing the juice from NHS hospital activity data
The problem with updating your database with the latest official figures from NHS Digital is that you want to go and play with them and invariably end up being distracted. This is precisely what I have gone and done!
In my last newsletter I plotted out COVID admissions proper for all service providers across NHS England, ‘proper’ in this sense meaning people coming in to hospital through the front door (though they may or may not be suffering symptoms and may or may not be infectious and may or may not be carrying SAR-COV-2). Strictly speaking these are not necessarily genuine COVID cases, being people who have tested positive with a controversial diagnostic test, though it is convenient to refer to them as such. On the same plot I plonked COVID admissions improper, these being inpatients who tested positive at some point during their stay. By definition these people did not attend hospital because of COVID and either caught the genuine virus whilst waiting for an operation or caught a false positive test result. For some obscure reason the NHS counts these people as COVID admissions from the date of the test result.
OK, so we have two data series representing outpatient admissions and inpatient admissions and I simply could not resist deriving the ratio to see how this curve panned out over time…
…for those not au fait with the use of the ratio I’ll just remind us that a ratio >1.00 means more inpatient COVID than outpatient COVID, with a ratio of =1.00 representing parity. There are very few occasions when the red line dips below a ratio of 1.00 and a fair few occasions when the red line rockets past a ratio of 2.00. Back in Sep ‘20 we see the ratio nudge 5.00, which means hospitals were handling five times as many COVID inpatients as outpatients. In plain English, for every one COVID case coming in through the front door five cases were detected amongst inpatients already lying in bed.
Big picture-wise, over the period 1 Apr ‘20 to 31 Jan ‘22 some 186,536 COVID cases came in through the front doors of hospitals across NHS England whereas 393,108 COVID cases were spotted in their beds, this giving a grand mean ratio of 2.11.
The rise and fall of this curve is fascinating because it is not really marking out pandemic peaks of Dec - Jan ‘21 and Dec - Jan ‘22. That substantial hump centred on Sep ‘20 after a glorious summer smacks more of hospitals going swab crazy more so than a virus running wild so I shall be taking a look at nose poking rates in my next newsletter. More noses means more cases!