The Puzzle of Australia’s Respiratory Mortality Season 2020

Written by Martin Neil, Jonathan Engler and Norman Fenton on March 6, 2024. Posted in Current News

An article by Arkmedic published in October 2023 provided an excellent and comprehensive analysis of multiple aspects of the COVID pandemic which supports many of the arguments we have been making in the last three years on this substack.

However, we challenge one of Arkmedic’s claims – that PCR testing was reliable enough to prove that there was a novel virus that caused deaths. Specifically, Arkmedic uses official statistics in Australia that purport to show that influenza had disappeared in 2020 and that, despite an initial outbreak in Victoria state, Covid-19 largely disappeared too. Arkmedic claimed that the absence of influenza, coupled with the increased surge in influenza PCR testing, proves that the PCR test for SARS-CoV-2 was reliable.

PCR specificity and supposedly novel signs and symptoms

Arkmedic claims that PCR is ‘really’ specific, and that the specificity rate published for some PCR manufacturers can be trusted (Roche is quoted at 99.8%). However, if you read it closely you will find that nowhere in the Roche documentation is there any description of the cross-reactivity tests undertaken that might support this low value. In a recent article we showed that failure to account for cross-reactivity systemically overestimated the specificity of PCR testing.

Arkmedic makes the correct point that the false positive rate is partly driven by prevalence, but then splits the population of those tested into those with symptomatic covid and those without (i.e. asymptomatic) and assigns to each of these a different prevalence rate. Strictly speaking, this is a mistake because prevalence means the proportion of people in the whole community, not prevalence in particular subsets of individuals with or without symptoms.

Logically a patient known to have an infection has a prevalence for the infection with probability value one (certainty), and a patient known not to have it has a prevalence with probability value zero (impossibility). But we are dealing with uncertainty about the infection in the population before we take account of the evidence we have about a particular patient. Hence, it is wrong to twist the definition of prevalence and define it based on symptoms.

We do not believe that an asymptomatic PCR test positive provides strong evidential support for a SARS-CoV-2 diagnosis (see, for example, this article). Objective confirmation cannot be obtained by PCR test because of the inability of swabs and PCR to reliably collect and identify causative agents (as reported by the CDC EPIC study in two 2015 NEJM articles – one done on adults and one on children). Hence, a positive result gained from a sample taken from the upper throat or the nose does not automatically mean an infection in the lung or elsewhere is caused by the detected pathogen. Therefore, it is not currently possible to differentiate between

• someone who has viral fragments in their upper respiratory system, detected by PCR, and who is and will remain ‘uninfected’ and will never show symptoms; and
• someone who might go on to develop an infection in the lower respiratory tract, or elsewhere, and show symptoms.

Hence, we cannot neatly separate a given population into asymptomatic and symptomatic groups and estimate the prevalence of a virus in each. All we can do is estimate and diagnose using probabilities over a single population.

To deal with symptoms properly we should therefore make them conditional on population infection prevalence in the same way we did with the PCR test. Both ‘signs and symptoms’ and the PCR test result are observable consequence of these various competing latent pathogens. Hence, the direction of causality is: infection causes the PCR test result and also causes the signs and symptoms, as shown here:

Critically, the signs and symptoms for ‘covid’ are no different, or hugely overlapping, from the signs and symptoms for other Influenza-like-illnesses (ILIs). Here is a quote on differentiating between RSV (respiratory syncytial virus) and Covid-19 from the advisory committee on immunization practices (ACIP) at the Center for Disease Control (CDC):

“…RSV is a well-recognized respiratory pathogen of infants and a common cause of respiratory disease in older adults. In adults, RSV can be difficult to differentiate from COVID-19 and influenza based on symptoms alone — and is frequently overlooked as a diagnosis for a viral respiratory disease in adults.”

Likewise, here is the UK NHS advice on the similarity of covid-19 symptoms with colds and flu.

Hence, observing these in a person does not discriminate between these competing causes but does make both more likely than ‘no infection’. Likewise, if a person is asymptomatic (no signs and symptoms) then the likelihood for either ILI or SARS-CoV-2 decreases in tandem to the same degree.

Let’s look at an example where both ILIs and SARS-CoV-2 have the same 10% prevalence. The PCR test’s sensitivity and specificity are assumed to be respectively 99% and 75%, as reported here. In our illustration, if a person has an ILI there is 50% probability of experiencing signs and symptoms and with SARS-CoV-2 infection this is the same (half of the infected population will show symptoms and conversely, half will not). And absence of signs or symptoms means there is a higher probability of being infected with neither.

Model a) shows the diagnosis for a patient when they have no signs or symptoms and no SARS-CoV-2 PCR test result, and model b) where they show signs and symptoms and have no SARS-CoV-2 PCR test result. Note that in neither case does the absence or presence of signs or symptoms differentiate between ILIs or SARS-CoV-2. Without signs and symptoms, the probability of SARS-Cov-2 and ILI infections are the same at 6% for each, and similarly in the presence of signs or symptoms the probability is 50% for each.

There is no differential diagnosis based on symptoms, except where it differentiates between no infection and any infection (logical but useless from a decision-making perspective).

Australia 2020 PCR testing

Arkmedic goes onto claim that the maximum false positive rate was 0.2% for New South Wales in Australia during the ‘covid lull’, the period from September 2020 to July 2021, and that because this matched the Roche false positive rate, one confirms the other as an accurate measure of specificity. Hence, there was no SARS-CoV-2 circulating at that time and no competing pathogens that might trigger false positives. This later point is backed up (according to Arkmedic) by the fact there was a high rate of influenza PCR testing done and that the test positive rate dropped to 0.1%.

So, if we accept these arguments we are left with two possible explanations:

• That the flu and SARS-CoV-2, as well as all other competing pathogens, genuinely (almost) disappeared from Australia, or
• The PCR kits being used were producing results that were manipulated, but manipulated in ways that differed from the way PCR was used elsewhere globally.

The first explanation is not credible given that Victoria state in Australia suffered a covid peak in July – August 2020, yet the state next door, New South Wales, did not. For this to make sense we would have to believe viruses stop at borders or that the lockdowns and border restrictions are dramatically effective. Evidence to date is overwhelming that neither of these can be true. Where did this highly infectious virus go after ‘landing’ in Victoria in 2020?

On the second of the possible explanations Arkmedic thinks Australia is the exception that proves the rule about PCR testing, but there are several points worth bearing in mind:

• There were no transparent, public, independent evaluations of PCR testing throughout the pandemic, except for those discussed here and it is abundantly clear that the false positive rate is high in the presence of competing pathogens.
• New multiplex array test kits were being introduced to test for both flu and SARS-CoV2 and we know of at least one test which exhibited ‘interference’ between flu and SARS-CoV-2. Likewise, the rate of test positives for all Biofire test kits dropped and did so for all competing pathogens in 2020 – it defies belief that this was a purely biological phenomenon.
• Some hospitals were not culturing flu viruses in the same way as before, so the gold standard had changed just as the pandemic began, with a switch over to the multiplex array PCR. It is conceivable that this change may have occurred globally.
• It is possible that, during the ‘covid lull’ period, confirmatory PCR testing was happening. As explained in this article this has the effect of drastically – but artificially – reducing the false positive rate.

There is ample evidence that the influenza surveillance systems had been altered. Here is a report from an Australian GP in a previous comment thread:

This first-hand report does not match the official influenza sentinel surveillance data:

However, looking closely at official mortality for Australia in 2020, we find something quite puzzling. The data for the ICD-10 influenza and pneumonia (J09-J18) deaths were similar in 2019 and 2020, as reported by the Australian Bureau of Statistics:

2019:

2020:

Influenza and pneumonia deaths in 2020 were 60% of the 2019 deaths (it varies considerably year by year; in 2018 there were 3102 deaths).

These mortality rates are similar to those reported for the USA and the UK in 2020 as described here. This shows there was no significant change in respiratory mortality and hence no pandemic.

But notice that 70% of the deaths in 2019 are categorised as pneumonia ‘organism unspecified’ (J18), compared to 94% in 2020. What can we conclude from this? That, despite the supposed absence of influenza and absence of SARS-CoV-2, the pneumonia death rates remained largely unchanged, while coincidentally the rate of ‘unknown’ causes of pneumonia death rose from an already significant proportion, 70%, to 94%!

Also, at this time people in Australia were searching the internet for the flu, as we reported here:

To a jaundiced, untutored, eye Australia had a normal respiratory mortality season in 2020 and that the only thing that varied was PCR testing rates, technology, and results. The facts on the ground remained the same and we can only assume the PCR testing done was highly suspect.

Source: Substack

Please Donate Below To Support Our Ongoing Work To Defend The Scientific Method

PRINCIPIA SCIENTIFIC INTERNATIONAL, legally registered in the UK as a company incorporated for charitable purposes. Head Office: 27 Old Gloucester Street, London WC1N 3AX. 

Trackback from your site.