COVID uncovered (part 8)

Bed occupancy and NHS staffing levels: back to basics.

In part 7 of this series I produced a slide of case severity index (CSI) against percentage of staff absences due to COVID that I found perplexing. I had expected the two series to complement each other on grounds that a greater percentage of staff would be going absent during the pandemic peak, which would also be when we’d see elevated levels of case severity¹. Against expectation CSI due to COVID and staff illness due to COVID were antagonistic series, causing some readers to speculate on staff absence forcing bed closure. Whilst this is true it misses the point made, so I thought I’d roll out some basic slides that deal with volume and not ratio…

All beds and all absences…

Here is our starting point with everything in the pot. What should be obvious is the strong correspondence between lack of available beds and staff absence back at the beginning of the pandemic. There are two seasonal dips in bed occupancy that correspond to two peaks in staff absence. I wouldn’t pay too much attention to lags because of delays in reporting procedures.

All beds and staff absences due to COVID…

We expect COVID bed occupancy to mirror staff absence for COVID, these being indicators of disease prevalence amongst the population as a whole. It is worth noting the flipped peaks - more staff went absent in the first wave than the third (seasonal) wave, whereas more beds were occupied in the third (seasonal) wave than the first. Beyond this initial ‘big flip’ it is worth observing just how closely the two series mimic each other.

MV Beds and staff absence due to COVID…

Another slide that makes sense. We observe an initial ‘big flip’ once more but something curious happens during the 2021/22 winter season: MV bed use is in decline at a time when staff absence peaks. This is a reversal of what went before and may be due to an influx of milder cases or a change in treatment protocol (or both).

Case Severity Index and staff absence due to COVID…

This is similar to the previous slide but I’ve replaced total MV bed use with the ratio of MV bed to general bed use for COVID cases (a.k.a Case Severity Index). The presumption here is that increased reliance on MV beds in preference to general beds is an indicator of pandemic ‘hot periods’. If this is a reasonable assumption to hang on to then we should note that - apart from the first wave - staff absences are less during such periods. This is the conundrum faced in part 7 of this series and is not a staffing levels leading-to-less-beds issue but something else entirely.

Why would staff absences due to COVID be lower during periods when reliance on MV beds for more severe COVID cases be higher? This is not an easy paradox to unpick. One idea that is running through my mind is that staffing requirements for critical care beds are much more stringent than general beds, so if you have a handful of nurses turning up for duty then you’ll be able to run more general beds than MV. Low levels of absence thus give the Clinical Nurse Managers more leeway to open intensive beds. Oddly enough this means the data series for MV beds may be more of an indicator of managerial options than it is a proxy for case severity!

We find ourselves in murky waters once more, so I shall suggest we go get that…

Kettle on!

1. Defined as the ratio of mechanical ventilation (MV) to non-MV general bed occupancy for COVID cases.