The retired former government scientist I am acquainted with has been doing some crunching of numbers about how full our hospitals were and are during the Covid ‘pandemic’, and this is what he found.
Most of us hold an image of the pandemic in the UK as being thousands of stricken souls being ferried to overcrowded hospitals on a daily basis. We have the mainstream media to thank for this, along with the rawness of image pouring into our social media feeds from China and Italy.
Some dared to question the narrative after discovering empty hospitals and plodding A&E departments, to be beaten down by a friend of a friend who had been working non-stop around the clock. Working non-stop around the clock, patients waiting several hours on trolleys and empty wards due to lack of nursing staff was pretty much normal in my old inner-city teaching hospital, so the curious among us never really got to understand what was actually going on above the level of the anecdote.
Here I reveal what actually happened across all 315 service providers that make up the NHS England bed pool using official data; to say I am surprised by what I am finding is a serious understatement.
I shall start with a slide of daily admissions of confirmed COVID-19 cases onto which I have also plotted daily diagnoses of COVID-19 arising from the testing of inpatients. Daily figures are defined as activity in the preceding 24 hours as at 8am each morning.
Hopefully this slide speaks for itself but I realise a few words may be needed. When I first saw this chart, my response was that I had made a serious number crunching error, and had probably got the two series mixed up; either that or my spreadsheet had mangled the figures somehow. After triple-checking the data I can report this chart is accurate.
In a nutshell we may say that the COVID-19 pandemic in the UK has primarily been an in-hospital disease. At the peak of the outbreak between 1st April and 10th April just over five times as many existing inpatients were diagnosed with COVID-19 as were coming in through the front door with a diagnosis of COVID-19 (x5.35). This is not the image that had been painted by the mainstream media.
We may now also see the reason for the confusion across social media. At a peak intake rate of 531 admitted cases across all 315 service providers on 9th April, NHS England’s emergency services never got truly stretched and many A&Es would indeed have idled as some have reported, though all depends on catchment area (my own A&E was always bursting to the seams even on a sunny day in May). In contrast, some poor souls had the task of managing beds of the 2,582 inpatients who were discovered to be lying there with COVID-19 on 3rd April. Hospitals were certainly busy but not with an influx of the public from the streets.
A final but important point to make is that, by definition, inpatients were in hospital for reasons other than COVID-19 (otherwise they’d be classed as COVID-19 admissions). This tip of a co-morbidity iceberg that is seen in terms of bed occupancy also reveals itself in mortality where we find that, of a total of 29,531 COVID-19 deaths (as at 26th August), only 1,390 were for those presenting with no pre-existing medical condition. WHO directives to medical authorities state that death certification must show COVID-19 as cause regardless of co-morbidity, an extraordinary fact that prompted me to quip “you can no longer die of cancer”. (Emphasis added)
In terms of bed occupancy, the serious end of the pandemic concerns mechanical ventilation (MV) bed usage. I shall set aside arguments for and against ventilator use in the treatment of COVID-19 since this is a rather thorny and complicated issue. When it comes to the crunch intensivists working within the NHS would have had no option but to place patients on ventilation if ARDS set in and their condition was deteriorating. Mechanical ventilation will have killed some patients but then again they would have died without it, for at this stage no prophylactic such as HCQ or Ivermectin will have afforded a safe and sensible clinical strategy. In many cases death will have been inevitable and unavoidable.
In the attached slide we get a feel for the size of the sharp end of the problem, with service providers within NHS England providing around 2,800 MV beds per day during the peak in early April. This equates to an average of roughly 9 beds per provider, though tertiary centres will have mostly taken the strain. Ventilation, like death, is a pretty robust indicator of what the disease has been doing and is now doing regardless of what the press and government may claim. Effectively we are looking at a disease that has come and gone. (Emphasis added)
The green curve for MV beds for other patients tells another side of the story. In early August we see this bobbing around 2,000 – 2,500 beds per day; we may assume this is business as usual for summer. If 2,000 – 2,500 MV beds is business as usual we may well ask what happened to the 1,500 patients per day that must have been denied a MV bed back in April in preparation for COVID-19. Did this bed shunting induce excess death or is a degree of MV bed occupancy unnecessary? This is one of those rather unpleasant questions we need to ask; I shall put it bluntly: did NHS MV bed management kill people? (Emphasis added)
Another take home message from this slide is that ‘business as usual’ runs at 2,000 – 2,500 MV beds per day during the lean summer months whereas the peak pandemic only reached ~2,800 MV beds per day. This is a difference of just ~300 MV beds per day across 315 service providers. Can this really be called a pandemic? It will be interesting to see what happens when seasonal ‘flu hits this coming winter (assuming the PCR test fiasco permits seasonal ‘flu).
Length of Stay
Since we know the daily bed days occupied by COVID-19 cases together with the daily discharges what we can do is calculate accumulated bed days and divide this by accumulated discharges to arrive at an estimate of the average length of stay over time. If this is a bit of an early morning head-banger let us try a worked example…
Up to 31st March of this year NHS England reported 65,972 bed days that had been allocated to COVID-19 patients. During this period some 3,414 COVID-19 cases had been discharged. Thus, on average, their length of stay equates to 65,972 / 3,414 = 19.32 days. Obviously, some will have recovered much quicker than this and some will have not but nevertheless we may say that, on average, the length of stay was 19.3 days. If we repeat this for 1st April we find 78,031 beds in use up to this point and 4,264 discharges giving an average length of stay of 18.3 days. Repeat this process and you end up with the attached chart.
Length of stay is a complex variable that acts as a proxy for both the condition of the patient and the effectiveness of treatment. During the last two weeks of March we observe elevated stays of around 20 days, and this will largely reflect the generally poor condition of the first COVID-19 patients. We noted above that the majority of cases during the early stages of the pandemic were inpatients rather than admissions.
This chart also reflects the learning curve of medical teams, and in this regard, we see length of stay plunging to just 14 days within the space of a fortnight. By mid-April presenting condition and medical care had settled down into a routine which has slightly improved over time.
The Hidden Cost
In the wake of reports of deaths on waiting lists – something I warned about at the outset, what we are looking at in this slide is bed occupancy for all 315 service providers that go to make up the NHS England bed pool.
As we can see from occupancy levels for the relatively ‘normal’ week beginning 7th August our hospitals normally expect to provide around 100,000 beds on a daily basis. Back in April this was curtailed to 40,000 daily bed spaces per day – that’s a lot of missing beds and this would have translated into a great deal of pain and misery for the folk who would have used them.
If we take 100,000 beds per day as the reference for healthy service levels, then deduct occupancy for both COVID-19 and other diseases, then we see that 1,666,772 bed-days were lost across the 315 service providers since 1st April because of Covid policies.
I had noted elsewhere (in my previous article https://principia-scientific.com/uk-govt-play-fast-loose-on-covid-data-to-justify-more-restrictions/) that the dramatic rise in new cases – now being called a new wave – can be largely attributed to the dramatic rise in testing by calculating the Positive Swab Rate (PSR). When this is done, we find a very modest rise and a very definite spike that cannot be considered to be robust evidence of a ‘new wave’; spikes are not waves last time I looked. However, we also noted a rise in hospital admissions; though these are not admissions as such since the count includes inpatients who have been discovered to be positive during their stay. As already noted, a positive case does not necessarily mean an infectious case and an admission with COVID-19 doesn’t necessarily mean an admission because of COVID-19; there’s that old chestnut ‘of’ and ‘with’ again!
At this point we may consider something may well be going on beneath the numbers despite all the additional testing and so I devised an indicator I am calling Bed Occupancy Rate (BOR). This is simply the accumulated number of daily bed days attributed to COVID-19 cases divided by the accumulated number of daily reported new cases. If reported new cases are translating into bed use then we should see this indicator rise over time or at least remain static. Except it doesn’t, it is decline. (Emphasis added)
Attached is a plot of BOR for the period 1st June to 17th September. What we are looking at is a nose-dive that is indicative of a pandemic in decline. This is a splendidly graphic way of saying bed use is in decline whilst reported cases are rising; a bold and brazen clinical fact that is not commensurate with the government’s claims of a ‘new wave’. A new wave of just what, exactly? Irresponsible mass testing, perhaps? (Emphasis added)
If I were to put an alarmist spin on this, I’d claim that this is showing bed stays for COVID-19 cases are reducing because the stratum of the population the virus is now hitting is healthier, with all the vulnerable cases being hit back in April. There’s no evidence for this and similar claims that attempt to explain away lack of deaths, of course, but that doesn’t stop people waving their hands about and failing to provide evidence in support of their argument.
The good news here is that I can analyse length of stay (LOS) to see if this is diminishing, thus bolstering the case for alarmists. I calculated this at a steady decline in average stay from 14.01 days at 1st June to 13.16 days on 2nd September for service providers making up NHS England, this representing a 6.1% drop in LOS. During this exact same period BOR dropped from 3.40 beds per case on 1st June to 3.07 beds per case on 2nd September, this representing a 9.7% drop in BOR. We now see that shorter length of stay is indeed contributing to the situation but the net result is that BOR is still diminishing, which is indicative of a pandemic in general decline. (Emphasis added)
The data I’m obtaining for my bed occupancy project (BOP) from NHS England is valuable in that it reveals just who was admitted to hospital as a positive case and who was an inpatient who tested positive. As we have recently discovered the gov.uk coronavirus dashboard calls them both ‘admissions’.
The NHS England daily hospital activity file also usefully reveals bed use for COVID-19 and other patients as well as discharge counts for COVID-19 by age band but this file is only updated monthly with the last release dated 10th September. This means I am unable to verify or refute headlines as they happen, which is rather frustrating since the data are already sitting there on NHS England’s servers.
This morning I have noted that NHS England are now producing another activity file – the COVID-19 daily situation report. This sounds like it is going to be useful but it only provides data on admissions; and there’s a rather cheeky twist. Five tables are provided and I am going to list their headings to see if you can spot the cheeky new game…
1). Total reported admissions to hospital and diagnoses in hospital
2). Estimated new hospital cases
3). Estimated new admissions to hospital from the community
4). Estimated new hospital admissions from the community with 3-7 day lagging
5). Total reported hospital admissions and diagnoses from a care home
That’s right, three of the five tables are populated with estimates rather than what is actually happening on our wards. These estimates are based on… reported new cases! (Emphasis added)
Yes indeedy, they are converting those rather suspicious new case counts and generating a hypothesised admission scenario. I prefer to call it fudge. I think we can guess what tables are going to be used for broadcast to the nation.
Can anyone spot the second game? Yep, NHS England, who used to distinguish between genuine admissions and inpatients testing positive up to 10th September, have now lumped them together in table 1 so the public are not going to know who is being admitted to hospital and was already in hospital suffering from some other condition. (Emphasis added)
I alerted folk to a change in how COVID-19 admissions data are now coded and presented to the public. In a nutshell an ‘admission’ is now both an admission proper and an inpatient who has been found positive during their stay. The date of an inpatient test is now taken as the date of their COVID-19 ‘admission’ even though they’re already been admitted! The astute among you will realise this improvement in data capture will lead to double-counting of bed occupancy. (Emphasis added)
To illustrate the situation as colourfully and clearly as possible I’ve plotted out three pertinent time series for the period June – September. The time series marked with a red line represents the new NHS England daily activity admissions indicator (table 1 in the new file), for which admissions and inpatients are lumped together.
The time series marked with an orange line represents a variable I calculated from the old data file that I called ‘COVID-19 bed load’, this being derived from admission and inpatient case counts. As we may see there is a period of overlap with excellent agreement, which gives us faith in comparing old and new data files. With this in mind let us now consider the time series for genuine admissions only, this being represented by the green line. We now see that genuine admissions accounts for a fraction of the total COVID-19 NHS England bed load. As I have stated before, COVID-19, if anything, is primarily an institutional pandemic. (Emphasis added)
With the admission and inpatient count now starting to rise in late September (red line) what we still do not know is whether this is the result of a rise in COVID-19 amongst admissions proper or a rise in COVID-19 amongst inpatients. This distinction is critical in our understanding of whether we are at the beginning of genuine second wave or simply looking at an artefact of testing strategy. For example, back in the week beginning 26th March pillar 1 testing averaged out at one test between two people in clinical care (Pt=0.5). The test rate then proceeded to climb in a steady manner such that during the week beginning 27th August we may observe just over 2.5 tests per patient in clinical care. My trusty hand-held calculator reveals this represents a five-fold increase in the testing of those already known to be sick. (Emphasis added)
This may well represent an improvement in care but it is going to skew results, and all this lather is without considering the wrinkle of the PCR test detecting the remains of an earlier infection! If an earlier SARS-COV-2 infection left patients with various ailments e.g. kidney damage then we are going to see a rise in hospital admissions for deteriorating renal conditions for people who are going to test positive but not be infectious. (Emphasis added)
The critical issue here is that we’ve abandoned traditional methods of diagnosis by an admitting physician (what we call differential diagnosis in the trade) and replaced this by an automated numbers generator with the controversial RT-PCR test in the driving seat – and all because the WHO say so. This is no longer clinical diagnosis as I used to know it. (Emphasis added)
The bottom line of this work shows that while in the real world the pandemic is declining, governments and the media tell us the exact opposite, and that more restrictions are needed to save us all from dying.
One of my social media friends wrote this recently, which sums up the situation pretty well:
Lots of kids in a class room – fine.
Working with a full workplace – fine.
Trains and buses full of people – fine.
Being in a pub with drunk people – fine.
Packed supermarket – fine.
On an airplane with hundreds breathing the same recycled air – fine.
Eating out at any restaurant – fine.
Having more than 6 family members or friends in the same house – too dangerous.
And they wonder why people don’t follow the guidelines…..
I rest my case M’Lud.
About the author: Andy Rowlands is a British Principia Scientific International researcher, writer and editor who co-edited the new climate science book, ‘The Sky Dragon Slayers: Victory Lap‘