Emergency Department Admissions: Analysis of CDS Dataset (part 3)

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 2 of this series we took a look at the weekly counts and percentages of what I’ve been calling Probable ILI/COVID. There haven’t been many of these cases, TBH, so after a think I decided to relax the criteria to bump numbers. What if a probable ILI/COVID case is now an admission with a respiratory diagnosis that also gets treated for a respiratory condition? Sounds good? Feels crunchy but with a mellow centre? I think so, and so here is a pair of revised slides for ogling:

Note how similar these two curves are despite one being for raw counts and the other being the percentage of total admissions. This points to stability in that these cases represent a constant percentage of the workload week in, week out. Not very pandemic-like I admit, but it is what it is: the pandemic that wasn’t (at least in terms of front door medicine). Ask a Martian (or a Venusian if you prefer) to point to when they think a pandemic might have occurred and I’d bet good money they’ll go for December 2017.

Cocked-Eyed Coding

But what if coding is a wee bit cocked-up? What if staff couldn’t make up their mind between coding for a respiratory condition and coding for an infectious disease when it comes to the symptomatic mystery that is the COVID thing? An excellent suggestion made by a keen subscriber in this regard is to look at anybody with a respiratory diagnosis OR an infection diagnosis rather than my rather narrow approach of looking at anybody with a respiratory diagnosis that was treated for such. Here’s what the raw count and percentage of all admissions looks like for these people:

Here we have another strike for a pathogen passing through the population in 2019, with a peak count of 1,648 admissions for 2019/w52 (week ending 27 December), this representing a peak of 19.7% of all admissions. The percentage chart offers a secondary peak of 15.1% that sits within the pandemic period, being 2020/w13 (week ending 27 March). There’s no way of knowing what the infectious disease was or the respiratory condition and in this regard we should note similar-sized peaks for previous years.

If you are going to withhold antibiotics for PCR-driven COVID ‘cases’, as was done during this period, then there’s no telling what will transpire, though we can be certain that something nasty will come about as a result of such unethical and misguided practice. Just what the fudge happened to our medics back then? Then there are those experimental drug trials to consider…

What Drug Trials Would They Be?

Back on 25 March 2021 I penned a worry-piece for an 11k strong Facebook group I ran at the time. Herewith an extract from that piece:

On 18th March 2020 Tedros Adhanom Ghebreyesus announced the SOLIDARITY study to explore the most effective drug combinations in the battle against COVID-19. The primary contenders for this global trial were remdesivir, lopinavir/ritonavir (Kaletra), and interferon-β in combination with Kaletra and hydroxychloroquine and chloroquine. These drugs offer a small toxic to therapeutic margin and more than one senior voice has expressed concern about WHO’s recommended dosage. In addition both hydroxychloroquine and chloroquine are difficult for the body to breakdown, having an effective half-life of 1 – 2 months, meaning that it is easy to overdose through accumulation. Among the countries joining the trial were Belgium, Spain, France and Switzerland - all of whom exhibited unusually high case mortality rates in the early stages of the outbreak. But that’s not all.

On 22nd March 2020 the French biomedical agency INSERM announced an additional European-wide trial called DISCOVERY that was designed to test the same drugs with the exception of chloroquine. By 8th April the University hospital in Nice (CHU) had to stop experimental treatment with high dose hydroxychloroquine due to risk of cardiac arrhythmia. In contrast, come 8th June 2020 Belgium had settled on high dose hydroxychloroquine in the treatment of COVID-19… and was leading Europe in COVID fatalities per capita. But that’s not all.

On 3rd April 2020 the UK government announced that around 1,000 patients from 132 hospitals had been rapidly recruited for a trial called RECOVERY. With thousands more participants promised, the UK government aimed to conduct the largest randomized controlled trial of COVID-19 drug therapies in the world. This study focused on lopinavir/ritonavir, dexamethasone, hydroxychloroquine, azithromycin, tocilizumab and a monoclonal antibody preparation called REGN-COV2. The hydroxychloroquine arm of the study was abruptly discontinued on 5th June 2020 with claims that the drug offered little “benefit”, yet Engelbrecht et al in their book ‘Viral Mania’ state that 25.7% of treated patients had died. Shortly after, on 17th June 2020, the hydroxychloroquine arm of WHO’s SOLIDARITY study also abruptly ended. But that’s not all.

A global trial called REMAP was instigated to explore treatment for patients on ventilators or in shock owing to a rapidly deteriorating condition. High dose hydroxychloroquine was again the focus but in combination with lopinavir/ritonavir. Trials took place at 200 sites in 14 countries including Belgium, Spain, Netherlands, UK and USA – again all countries that were notching up significant death counts.

I don’t know about you folks but I’m getting an uneasy feeling that all these exploratory trials designed to test a raft of drug therapies to combat COVID-19 may have generated iatrogenic death on an unprecedented scale via high dose and cumulative toxicity leading to liver failure, sudden cardiac death and multi-organ failure. Quite where we go from here I have no idea because attributing causality is near impossible at the best of times, and especially if placebo arms are ruled out on grounds that it is unethical to withhold treatment, and especially if autopsy and post mortem is ruled out under emergency Laws.

Thought provoking is it not? I haven’t the time to dig deep into this and verify all claims made but maybe somebody out there has more time and better info. Whatever the truth of the matter what we most certainly do know for a fact is that several experimental drug trials took place back then, thus whenever we spot a COVID-like peak of something untoward, as we have done for the respiratory/infectious cases, then such trials should be added to the list of suspects, especially if they interfered with normal protocols and/or the new drugs turned out to be unsafe.

Slicing The Cake

What I’d like to do now is slice my Probable ILI/COVID indicator cake - the one based in respiratory Dx with Tx - three ways…

1. Pre-pandemic (2017/w1 – 2020/w10);

2. Pandemic (2020/w11 –2020/w49);

3. Post-vaccine (2021/50 – 2021/37);

…and see what drops out when I cook up an error bar plot for mean percentage of probable ILI/COVID admissions:

Well it can’t be much clearer!

If we use respiratory diagnosis requiring treatment with a respiratory procedure as our proxy for ILI/COVID within the CDS dataset then the pandemic era washes up as a flop. Odd for a novel and deadly respiratory virus don’t you think?

Right, now then [pauses for an early morning slurp of refreshing hot lemon tea], I’m going to cut this article short because I want to introduce readers to some basic demographics so that they get a feel for what goes in and what pops out of a typical emergency department before we get into the nitty-gritty of wading through cases of interest.

Kettle On!