The Growing Gap Between Reality and Pop Science

The gap between the mainstream narrative about the success of the COVID-19 vaccination campaign and empirically verifiable outcomes continues to grow.

The early trial results for the mRNA vaccines were hailed for showing ‘95% effectiveness,’ with experts quoted by NBC calling it ‘the grand slam:’ ‘”We do not have good examples of vaccines with this level of efficacy across age, gender, race and comorbidities,” (Dr. Greg) Poland said. “I’ve never seen something like this.’”

The trials did not show efficacy against transmission or mortality, even when data from multiple trials for both mRNA vaccines were aggregated to increase the power, but this did not stop governments rolling the vaccines out for all age groups and risk categories and mandating them for wide sections of the working age population, for a disease where the post-working age population was by far the most at risk.

How does that 95 percent protection against infection look in retrospect? 

According to the recent preprint by experts from Harvard, Yale and Stanford (three of the top ten most highly rated universities in the world), 94 percent of the US population had been infected at least once by November 9, 2022, only two years later. 

So, the supposed 95 percent protection reportedly resulted in 94 percent infection.

The mass vaccination campaign completely failed to prevent almost the whole population from becoming infected. And yet protection against infection was the one claim that appeared to be validated by the randomized controlled clinical trials (RCTs), which are thought to be the highest level of medical evidence. How can this be?

Bear in mind that by November 9 (according to the OWiD Data Explorer), 80 percent of people in the US had received one or two (69 percent) doses of the vaccine, so coverage was very broad, but not universal.

We also have to acknowledge that the claim from the Harvard, Yale, and Stanford team is an estimate based on modeling, and modeling does not count as high-grade evidence within the same hierarchy of medical evidence. Their model is a black box – they do not reveal any details about how it was constructed or what the key data assumptions were that drive the model. See also the devastating critique of COVID-19 modeling in general by Ioannidis et al.

The sudden formation of what I am calling the ‘grand strategy,’ which set out to provide interim protection through lockdowns until an effective vaccine became available, was mostly based on modeling, both of the magnitude of the possible loss of life and of the effectiveness of the countermeasures (in averting the imagined additional loss of life). If such modeling cannot be relied on, then the grand strategy (which was in fact neither grand nor strategic) falls to the ground anyway.

Fortunately, there is a relatively firm basis for estimating the cumulative prevalence of SARS-CoV-2 infection. The CDC’s nationwide commercial laboratory surveillance system estimated 57.7 percent seroprevalence over the period January – February 2022.

Given that infections accelerated over the course of 2022, and that ‘estimates of infection based on antigen testing results are likely underestimated,’ it does seem plausible that a very high proportion of the population had been exposed by the beginning of November. Moreover, the same report estimated that 91.5 percent of people had antibodies to either SARS-CoV-2 or the vaccines. Any further gains are likely to be marginal.

There is some support in some observational studies for a protective effect against COVID-related mortality for up to 6 months, particularly. But there is little or no evidence for a reduction in all-cause mortality, which is the acid test as it avoids selective criteria about the cause of death.

The problem to be solved is excess mortality, so the primary function of a vaccine should be to reduce all-cause mortality, not just mortality from a particular cause. The Zhengzhou University team’s meta-analysis shows high levels of protection only against COVID-related death over unspecified time periods. 

We need straightforward comparisons between a never-vaccinated group and a comparable group from the time of the first vaccination dose – no exclusions, no categorizing of the partially vaccinated as ‘unvaccinated.’ We want to see total outcomes over a meaningful period of time. Most of these studies are only showing partial and short-term effects.

The recent study coming out of Indiana by Tu et al. compares mortality outcomes for matched pairs of unvaccinated but infected individuals and vaccinated people and finds a 37 percent advantage for the vaccinated. 

This is a carefully designed study, but you need to look at the fine print: ‘Matched pairs were censored when an infected participant received a vaccination or a vaccine recipient became infected.’ So, if vaccine recipients died after then becoming also infected, this was excluded from the analysis? Writing in Medscape, Perry Wilson commented: ‘I am concerned that this would bias the results in favor of vaccination.’

Compare Chemaitelly et al., who found that: ‘Effectiveness of primary infection against severe, critical, or fatal COVID-19 reinfection was 97.3% (95% CI: 94.9-98.6%), irrespective of the variant of primary infection or reinfection, and with no evidence for waning.’ This was based on cohort studies from the national database covering the entire population of Qatar. So, prior infection is the best defense available against future infection, and nearly everyone has had it.

Observational studies are prone to being affected by extraneous factors, which is why they rank below RCTs in the evidence-based medicine hierarchy. Different choices for inclusion, exclusion and timing can lead to different results. Research groups should undertake sensitivity analysis more often, to find out how changing each of the key parameters would change the results. Are the findings robust in all scenarios? 

The studies showing vaccine effectiveness may have internal validity but lack external validity for a population as a whole over the two years of the vaccination campaign. If this is the case for the studies claiming protection against infection, it is likely to be equally true of the studies claiming protection against death, as they have the same limitations and are equally unable to determine the holistic outcomes. Deferring some deaths for a few months would not be sufficient.

Another example of these anomalies of measurement is worth mentioning. In my last contribution, I remarked on the fact that the US V-Safe data showed that 7.7 percent of people reported seeking medical attention after vaccination, whereas the comparable Australian figure was less than 1 percent. But having now read the fine print, I find that the AusVaxSafety data is based on a survey sent out on Day 3 after vaccination, whereas the V-Safe check-ins run for 12 months after the last dose. So, the Australian active surveillance data is very short-term. The US system is more thorough, but was not transparent as the data was only made public by court order after legal action. 

Researchers analyse the data that they or the government agencies choose to measure or reveal, which can be very selective, and indeed misleading. Short-term results are extrapolated to project long-term outcomes that do not eventuate. The research gives us only snapshots – the micro, not the macro perspective.

The public expects vaccination to protect them against getting infected. Yet recent studies show that the vaccinated are actually more likely to get infected, such as the Cleveland Clinic study by Shestha et al. Indeed, the Cleveland Clinic study shows a dose-response type of correlation, with the number of infections progressively increasing with the number of doses, and the authors discuss two other studies which had similar findings. They deserve credit for publishing their findings, which they describe as ‘unexpected.’ 

But they would not be unexpected to those of us who were paying attention to the vaccine surveillance reports from Public Health England, which showed vaccinated people having higher rates of infections compared with the unvaccinated (for example, see Table 14 in the report for week 13, published on 31 March 2022). PHE greyed these out hoping that we wouldn’t notice the data that doesn’t fit the narrative. Their successors at the Health Security Agency solved the problem by discontinuing the reports altogether.

In that previous contribution I pointed out that the European mortality curves have been flatter in the last two years of vaccination, which is consistent with some mortality being at least deferred by mounting hybrid immunity. But deferred for how long? And what are the relative contributions of infection and vaccination? No one knows. 

Grand claims about millions of lives being saved by vaccination are not falsifiable as they rest again on averting hypothetical counterfactual scenarios in which many more deaths would supposedly have occurred without the vaccination campaign. But these deaths may occur only in the virtual world of computer modeling and may be averted only for a short time. Policy should be based on factual information and the big picture.

Government programs need to be rigorously evaluated, particularly when they affect public health and individual rights. The objectives should be clear, whereas in this case they were vague and constantly shifting. And the outcomes data should be straightforward, whereas in this case they depend on complex and variable statistical processing of small samples.

Policymakers and politicians have been making big calls on the basis of uncertain data. They need to know for sure that the pandemic is being substantially moderated by policy settings and not prolonged.

The primary goal of government strategies should have been to prevent excess mortality, yet excess mortality remained high through 2022, peaking at just over 23 percent (UK) and over 10 percent (US) (see OWiD again). There is no hard evidence that excess mortality was reduced over the past two or three years overall.

How can continuing the mass vaccination campaign be justified if the population already has equivalent immunity, vaccination increases the risk of infection (and adverse effects), and other benefits are uncertain? 

The WHO, government agencies and scientists started in 2020 with the proclaimed goal of ‘controlling the pandemic’ which evolved towards the hope that COVID-19 vaccination could ‘end the pandemic.’ It didn’t.

They soon had to concede that vaccinations would not provide full protection against transmission or infection, but maintained they were ‘substantially effective’ against infection. 

And yet everyone got infected, many times over in some cases. 

Failure is being spun as a triumph – but is it a triumph of misinformation? Is it a grand illusion?