The scientist whose calculations about the potentially devastating impact of the coronavirus directly led to the countrywide lockdown has been criticised in the past for flawed research.
Professor Neil Ferguson, of the MRC Centre for Global Infectious Disease Analysis at Imperial College in London, produced a paper predicting that Britain was on course to lose 250,000 people during the coronavirus epidemic unless stringent measures were taken. His research is said to have convinced Prime Minister Boris Johnson and his advisors to introduce the lockdown.
However, it has now emerged that Ferguson has been criticised in the past for making predictions based on allegedly faulty assumptions which nevertheless shaped government strategies and impacted the UK economy.
He was behind disputed research that sparked the mass culling of farm animals during the 2001 epidemic of foot and mouth disease, a crisis which cost the country billions of pounds.
And separately he also predicted that up to 150,000 people could die from bovine spongiform encephalopathy (BSE, or ‘mad cow disease’) and its equivalent in sheep if it made the leap to humans. To date there have been fewer than 200 deaths from the human form of BSE and none resulting from sheep to human transmission.
Foot and mouth disease cost the UK billions of pounds
Credit: DAVID CHESKIN/PA
Mr Ferguson’s foot and mouth disease (FMD) research has been the focus of two highly critical academic papers which identified allegedly problematic assumptions in his mathematical modelling.
The scientist has robustly defended his work, saying that he had worked with limited data and limited time so the models weren’t 100 per cent right – but that the conclusions it reached were valid.
Michael Thrusfield, professor of veterinary epidemiology at Edinburgh University, who co-authored both of the critical reports, said that they had been intended as a “cautionary tale” about how mathematical models are sometimes used to predict the spread of disease.
He described his sense of “déjà vu” when he read Mr Ferguson’s Imperial College paper on coronavirus, which was published earlier this month.
That paper – Impact of non-pharmaceutical interventions (NPIs) to reduce COVID19 mortality and healthcare demand – warned that if no action were taken to control the coronavirus, around 510,000 people in Britain would lose their lives.
It also predicted that approximately 250,000 people could die if the Government’s conservative approach at the time was not changed. The research, which was based on mathematical models, was key in convincing the Prime Minister that “suppression” – and subsequently a lockdown – was the only viable option to avoid huge loss of life and an NHS meltdown.
This week, a second paper authored by Mr Ferguson and the Imperial team further predicted that 40 million people worldwide could die if the coronavirus outbreak was left unchecked.
But scientists warned last night about the dangers in making sweeping political judgments based on mathematical modelling which may be flawed.
In 2001, as foot and mouth disease (FMD) broke out in parts of Britain, Ferguson and his team at Imperial College produced predictive modelling – which was later criticised as “not fit for purpose.”
At the time, however, it proved highly influential and helped to persuade Tony Blair’s government to carry out a widespread pre-emptive culling which ultimately led to the deaths of more than six million cattle, sheep and pigs. The cost to the economy was later estimated at £10 billion.
The model produced in 2001 by Professor Ferguson and his colleagues at Imperial suggested that the culling of animals include not only those found to be infected with the virus but also those on adjacent farms even if there was no physical evidence of infection.
“Extensive culling is sadly the only option for controlling the current British epidemic, and it is essential that the control measures now in place be maintained as case numbers decline to ensure eradication,” said their report, published after the cull began.
The strategy of mass slaughter – known as contiguous culling – sparked revulsion in the British public and prompted analyses of the methodology which has led to it.
A 2011 paper, Destructive Tension: mathematics versus experience – the progress and control of the 2001 foot and mouth epidemic in Great Britain, found that the government ordered the destruction of millions of animals because of “severely flawed” modelling.
According to one of its authors – the former head of the Pirbright Laboratory at the Institute for Animal Health, Dr Alex Donaldson – Ferguson’s models made a “serious error” by “ignoring the species composition of farms,” and the fact that the disease spread faster between some species than others.
The report stated: “The mathematical models were, at best, crude estimations that could not differentiate risk between farms and, at worst, inaccurate representations of the epidemiology of FMD.”
It also described a febrile atmosphere – reminiscent of recent weeks – and claimed that this allowed mathematical modellers to shape government policy.
“The general impatience that met the wait for the full extent of infections to become apparent, accompanied by an ever increasing number of outbreaks and piles of carcasses awaiting disposal, was perceived as a lack of success of the traditional control measures and provided the opportunity for self-styled ‘experts’, including some veterinarians, biologists and mathematicians, to publicise unproven novel options,” the researchers said.
An earlier report, in 2006, Use and abuse of mathematical models: an illustration from the 2001 foot and mouth epidemic in the United Kingdom, identified Professor Ferguson’s modelling as having been the biggest driver of government policy.
The paper said that “the models were not fit for the purpose of predicting the course of the epidemic and the effects of control measures. The models also remain unvalidated. Their use in predicting the effects of control strategies was therefore imprudent.”
On Friday, Professor Thrusfield said: “When we wrote those two review papers, we thought it would be a cautionary tale for the future if foot and mouth disease struck again. We didn’t think it would be a cautionary tale for a new plague in the human population – but of course the cautionary tale is fully valid.
“This is déjà vu. During the [FMD] epidemic there was quite vocal opposition from members of the vet profession – especially those who had their hands soaked in blood, killing perfectly healthy cattle.
“There was also a major economic and emotional impact on those involved, [because] the slaughter of these animals that were perfectly healthy. This was serious stuff. This was farmers losing their livelihoods. They need not have been slaughtered but they were because the predictions were wrong.”
Neil Ferguson self-isolated after presenting coronavirus symptoms
Last night, Dr Paul Kitching – lead author of Use and abuse of mathematical models, and the former chief veterinarian of Canada’s British Columbia province – raised fears over the modelling being done on coronavirus.
“The basic principles on modelling described in our paper apply to this Covid-19 crisis as much as they did to the FMD outbreak.
“In view of the low numbers of Covid-19 tests being reported as carried out in affected countries, it is difficult to understand what informs the current models. In particular the transmission rate. How many mild and subclinical infections are occurring?”
“The model driven policy of FMD control resulted in tragedy. Vast numbers of animals were slaughtered without reason. Untold human and animal suffering was the result – not to mention the financial consequences.”
However, Sir David King, who was the Chief Scientific Advisor to the government in 2001, said that criticism of the epidemiological modelling was “misplaced.”
He said: “I would agree there was some unnecessary culling taking place, but this is simply because there wasn’t a unity in the way the thing was being handled.”
Professor Ferguson said of his modelling for FMD: “A number of factors going into deciding policy, of which science – particularly modelling – is only one. It is ludicrous to say now that our model changed government policy. A number of factors did.
“We were doing modelling in real time as the other groups were in 2001 – certainly the models weren’t 100% right, certainly with limited data and limited time to do the work. But I think the broad conclusions reached were still valid.”
Of his work on BSE, in which he predicted human death toll of between 50 and 150,000, Professor Ferguson said: “Yes, the range is wide, but it didn’t actually lead to any change in government policy.”
Others have directly criticised the methodology employed by Ferguson and his team in their coronavirus study.
John Ioannidis, professor in disease prevention at Stanford University, said: “The Imperial College study has been done by a highly competent team of modellers. However, some of the major assumptions and estimates that are built in the calculations seem to be substantially inflated.”
Professor Ferguson said anyone who thought the coronavirus was akin to seasonal flu was “living in cloud cuckoo land.”
He defended the conclusions reached “in terms of the overwhelming demand on healthcare systems imposed by this virus.”
“It is ludicrous, frankly, to suggest that the severity of this virus is comparable to seasonal flu – ludicrous and dangerous. People who are doing so have not analysed the data in any level of detail.”