Emergency Department Admissions: Analysis of CDS Dataset (part 9)

I analyse an anonymised data dump of 1.9 million admissions records to the emergency departments of an undisclosed NHS Trust for the period June 2017 – September 2021

Back in part 6 I baked a slide that attracted some interesting comments, this being a dual time series plot of admissions to the emergency department (ED) against in-hospital deaths. We noted that when daily admissions fell through the floor then daily deaths went through the roof. Though alarming it’s tricky to fathom just what is going on here so I decided to refine this analysis by plotting weekly admissions to the ED alongside weekly deaths in the ED. Try this…

OK, so we don’t have the same situation. There’s no peak in departmental deaths when the admissions count drops during national lockdown. We may conclude that it was other people doing the dying elsewhere in the hospital: they weren’t coming in through the front door as critically ill cases.

To get a better handle on the matter I thought we ought to squint at a scatterplot of weekly ED departmental deaths against weekly in-hospital deaths for the period January 2017 – December 2020 and colourise the data cloud into pre-pandem…