The Great Barringtons & the Threat Posed by Covid-19
As the Covid-19 pandemic moves, with the help of vaccines, into an endemic phase in some parts of the world, while continuing to rage in others, there is much debate over the most appropriate response. At the same time inquiries are beginning in some countries, including the UK, into the public health strategies pursued in the early stages of the pandemic. Given that significant outbreaks of infection are still occurring even in countries with a relative abundance of vaccines, discussions of current and past policies are closely linked.
The Great Barrington Declaration was named after the American town in which it was launched in October 2020, by three well-established academics: Martin Kulldorff of Harvard; Sunetra Gupta of Oxford; and Jay Bhattacharya of Stanford. [1] The ‘Great Barringtons’ have become the leading scientific voices in challenging the public health response of seeking to contain Covid-19 by restrictions on social mobility, including ‘lockdowns’, that have been pursued in different forms by governments around the world. The Great Barringtons favour ‘focused protection’ from Covid of the elderly and clinically vulnerable, while allowing the infection to spread among the young and fit, who are at relatively low risk of serious disease or death. Now that vaccines are available, the Great Barringtons favour using them to protect the vulnerable at a global level, while, as before, allowing immunity acquired through infection among the relatively healthy to facilitate the transition to an endemic state.
The Great Barrington Declaration provoked a hostile response from scientific, medical and health authorities. [2] At the same time, the Great Barringtons attracted support from a wide range of critics of the mainstream response to the pandemic, particularly among those concerned about the collateral damage caused by lockdowns to the economy, civil liberties, education, mental health and other causes of illness and death, such as cancer and cardiovascular disease. The scientific case made by the Great Barringtons provided the basis for the demands made by diverse ‘lockdown sceptics’ for the relaxation or abandonment of lockdown regulations. [3]
In this article I focus on the evidence marshalled by the Great Barringtons in relation to the threat posed by Covid-19. In a follow-up article I will turn to the policy discussion, including the claim by the Great Barringtons that lockdowns and other society-wide restrictions caused greater harm than Covid itself, or at the very least that lockdowns caused more harms than they avoided.
Sunetra Gupta on the threat posed by Covid
All three Great Barringtons, individually and collectively, have made statements on how dangerous they estimate Covid to be, at the population level and to particular groups. However, Sunetra Gupta has written the most and has the strongest background in disease modelling. Indeed, according to Kulldorff: ‘In the UK, you have the world’s preeminent infectious disease epidemiologist in professor Sunetra Gupta.’ [4]
In a recent article for the online magazine UnHerd, looking back on the estimates of the seriousness of Covid she made over the past 18 months, Gupta asked: ‘have my Covid hypotheses held up?’ In answer to her own question, she said: ‘I now think those estimates were somewhat overly optimistic, but not outrageously so.’ [5]
The most basic estimate she is referring to is one she gave in an interview with UnHerd 15 Months earlier, in May 2020. In that interview, as she recounts in the recent article, she suggested that the Infection Fatality Rate (IFR) for Covid was likely around 0.05%. This is another way of saying that out of every 10,000 people who caught Covid, she expected around five of them to die. Related to this, she also hypothesised that many more people had caught Covid than the mainstream scientific community tended to believe, and that large cities such as London may have been approaching ‘herd immunity’ by end May 2020: ‘In May 2020 in the UK, particularly in London and the South East, it was indeed probable, and in my opinion remains so today, that around half the population had been exposed to Sars-CoV-2 and the Herd Immunity Threshold had been reached for that environment, which is why deaths and infections were coming down so rapidly.’ [5]
It is hard to over-emphasise just how surprising many scientists found and continue to find these claims. In May 2020, Gupta was claiming, and today she is still claiming (with the ‘somewhat overly optimistic, but not outrageously so’ caveat), that the IFR for Covid is ten, perhaps twenty, times lower than most others believe.
It is of course widely known that IFR varies hugely with age, and therefore also with location for this reason as well as others. But as is clear from her application of these numbers to London and the UK in general, Gupta believed her estimate applied in countries such as the UK, which was the shocking point — Neil Ferguson’s team at Imperial College had predicted an IFR of around 0.9%, or 18 times higher than her 0.05% figure, in the UK. Clearly, with an 18 fold difference, it is not possible for Gupta’s estimates to be ‘somewhat overly optimistic, but not outrageously so’, without Ferguson’s and others’ being wildly wrong. So, which is it?
The simple truth is that Ferguson and others were broadly correct, and Gupta was and is mistaken, by a huge amount. In April 2021, towards the end of the most recent winter wave of Covid infections and deaths, the UK’s Medical Research Council estimated that just under 15 million people in England, or around 25% of the population, had caught Covid since the start of the pandemic. [6] At that moment in time it is thought that somewhere between 116,000 and 129,000 had died from Covid in England. [7] Taking the lower figure, to be as generous as possible to Gupta, this leads to an average IFR over the pandemic to that point, the point when mass vaccination was starting to change the dynamics, of 0.8% in England. This is close to Ferguson’s early (March 2020) estimate, and 16 times greater than Gupta’s.
Surely I must have made some mistake? How could the ‘world’s preeminent infectious disease epidemiologist’ not only make such a basic error in May 2020 but continue to defend it in August 2021? Gupta’s calculation is 35,000 deaths by mid-May 2020 out of half the UK population infected. That would be around 0.1% for IFR. At the time (May 2020), she thought the 35,000 figure an over-count. She no longer thinks this, it seems — hence the recent ‘somewhat overly optimistic, but not outrageously so’ self-assessment / slight correction (She doesn’t spell this out, but this would double the IFR from 0.05% to 0.1%). However, 0.1% is still vastly different to Ferguson’s March 2020 prediction of 0.9%, and the MRC’s April 2021 assessment of around 0.8% averaged over the pandemic.
The difference in her May 2020 assessment as compared with Ferguson’s predictions, and others’ analyses at the time, is explained by her related claim that many more people than the mainstream authorities recognised had caught Covid early in the pandemic. Hence, in the IFR calculation, her denominator is much larger than others’, giving a much smaller final answer. Indeed, it was this claim, made two months before her May 2020 Unherd interview, that first caused a stir in the press. The Financial Times reported that Gupta believed that up to half the UK population had been infected with Covid by late March 2020, at a time when the pandemic was only beginning to cause significant disease and death. [8] Gupta presented this analysis in a scientific paper [9], which was updated in December 2020 (though without any change in the basic parameters). [10]
While this argument may seem plausible, even if it is at odds with the mainstream consensus (and passing over the inconvenient fact that half infected by March 2020 must mean more than half by May 2020), it fails for a number of reasons. Firstly, according to analysis by leading virologists, no more than 10% of the UK population had caught Covid by the end of the first wave in June 2020 [11] and around 25% by April 2021 (this is what informs the MRC’s assessment). Gupta claims that the tests used are unreliable, and that markers of infection fade with time. However, authoritative voices with relevant expertise insist that the tests can provide accurate results many months after exposure. [12]
But there is more than this, something so basic that it is tempting once again to wonder how someone with her expertise could make such an error. In a recent article entitled ‘The Beauty of Vaccines and Natural Immunity’, Gupta, writing with the other two Great Barringtons, undermines her own case, by acknowledging that previous infection continues to provide protection against serious outcomes even if it does not always protect against infection: ‘it makes little sense to ignore the scientific fact that infection does confer long lasting future protection for the millions of people who have had Covid’. [13] She presses this (uncontroversial) point further, insisting that ‘while natural infection may not offer permanent infection-blocking immunity, it offers anti-disease immunity against severe disease and death that is likely permanent.’ This is the crux of the matter: why, if so many were already immune following recovery from infection (Gupta’s assumption), did so many more people die from Covid in England and the UK as a whole after May 2020? According to Gupta, by May 2020, half the population had contracted Covid and 35,000 had died, but those who survived had acquired protection against serious outcomes. [5] Thus, following her logic and numbers, even if the entire remaining 50% of the population were to contract Covid, we should expect around 35 000 further deaths, because the half who had been infected and recovered (Gupta’s figure) were protected from death even if not always from re-infection. But in reality, a further 90,000–100,000 people died across the UK between May 2020 and Spring 2021. Neither Gupta nor the other Great Barringtons have questioned the figures for deaths after May 2020. Indeed, they have cited the official figures themselves. [14] As several critics have pointed out, if Gupta’s maths were correct, the number who caught Covid must have far exceeded the total population.
In Conclusion
Gupta’s analysis of the threat posed by Covid does not stand up to scrutiny, and her continuing defence of it in the face of overwhelming evidence raises another set of questions. How this relates to the Great Barringtons’ policy proposals, and, to separate out the issues to some extent, what were and are the merits and demerits of their proposals on their own terms, will be discussed in the next article.
[5] https://unherd.com/thepost/sunetra-gupta-how-have-my-covid-hypotheses-held-up
[6] https://www.mrc-bsu.cam.ac.uk/now-casting/report-on-nowcasting-and-forecasting-1st-april-2021
[7] For a discussion of the different methodologies used, see: https://blog.ons.gov.uk/2020/03/31/counting-deaths-involving-the-coronavirus-covid-19/
The numbers in the text are derived from the cumulative figures given by the Office for National Statistics up to and including week 13 of 2021. I have taken their numbers for England and Wales combined and multiplied by the fraction England population size / England & Wales population size. The lower figure is what the ONS calls ‘Deaths due to Covid-19’; the higher what they call ‘Deaths involving Covid-19’:
[8] https://www.ft.com/content/5ff6469a-6dd8-11ea-89df-41bea055720b
[9] https://www.medrxiv.org/content/10.1101/2020.03.24.20042291v1
[10] https://www.medrxiv.org/content/10.1101/2020.03.24.20042291v2
[12] A test developed at Oxford University is able to detect antibodies in 88% of people more than six months after a PCR-confirmed infection. For a discussion of this see:
https://www.ukbiobank.ac.uk/media/x0nd5sul/ukb_serologystudy_report_revised_6months_jan21.pdf
That there is a known rate of reversion allows an adjustment to be made and thus an improvement on the already excellent 88% figure for the purposes of population estimates. The Oxford anti-body test is also used by the ONS in its ongoing studies:
[13] https://www.smerconish.com/exclusive-content/the-beauty-of-vaccines-and-natural-immunity
[14] https://www.spiked-online.com/2021/08/02/the-smear-campaign-against-the-great-barrington-declaration
