The expert behind the Covid-19 shutdown was wrong on CJD and foot-and-mouth, and is probably wrong again now. But he’s not alone

It may beggar belief that the UK government took the advice of a professor who’s messed up before. But let’s not forget that it’s our leaders who chose to implement the measures causing long-term harm to our economy and society.

In the UK government's daily press conference on Covid-19 on Sunday, the deputy chief medical officer, Jenny Harries, explained the government's approach to the unprecedented measures put in place to tackle the pandemic. Her manner was calm, but her message depressing: “Three weeks for review, two or three months to see if we’ve really squashed it, three to six months ideally, there’s lots of uncertainty in that, to see at what point we can get back to normal, it is plausible it could go further than that.”

The measures are astonishing by the standards of wartime, never mind peacetime. Essentially, the UK is now under a form of house arrest. Healthy citizens under the age of 70 are allowed out only for exercise, to buy essentials, to go to work if they really can't work from home, and to help with the needs of others. For everyone over 70 or who has a health condition that leaves them particularly vulnerable to Covid-19, the strong advice is not to leave their homes at all till mid-June at the earliest.

While these measures do not, as yet, go quite as far as those in Spain, Italy and France, they nonetheless mean an enormous loss of freedom and will cause substantial damage to the economy and living standards. Far from these measures merely being “advice”, police forces across the UK are using their new powers to enforce these edicts.

Are such measures justified? The government has been heavily influenced by the group of modellers at Imperial College London led by Professor Neil Ferguson. A major turning point in the UK's battle with Covid-19 was a report by this group, released on 16 March.

Up to that point, the government was intent on keeping schools open for as long as possible, and allowing pubs and restaurants to continue trading while advising people to avoid them. The aim was, as the Imperial report described it, “mitigation” — to slow the spread of disease, allowing some healthy people to get it while shielding the most vulnerable. Eventually, enough people in the general population would have had the disease to make it safe for more vulnerable people to return to normal life.

The scary numbers of deaths projected in that report — 500,000 in a “do-nothing” scenario — seems to have persuaded the government to change course. That figure was always highly unlikely as some measures were always going to be implemented. But the Imperial report suggested that even the “mitigation” strategy could still be disastrous, leading to around 250,000 deaths. While the peak of cases would be significantly smaller than under “do nothing”, hospitals would be overwhelmed nonetheless.

Instead, the authors of the report offered an alternative: “suppression”. This involved an even more stringent set of measures to stop the spread of the virus, which is now the policy of most European countries, including the UK.

But the report contains some important caveats. The more successful the strategy in the short term, the larger any “second wave” epidemic would be, because fewer people would have immunity. Thus, until a vaccine becomes available, the report argues that these rules would have to be in force at least two-thirds of the time in order to keep demand on healthcare services under control.

Does the modelling justify these measures? Any attempt to model reality has to address three issues, all of which mean that models are just making educated guesses about reality.

First, understanding the particular nature of this virus — for example, how quickly it spreads, at what stage in the disease it is contagious, and how many people will be contagious but asymptomatic.

Second, the limitations of data. Some countries have excellent data thanks to widespread testing, but others (like the UK) have not expanded their testing as much as they would like. So models can only guess at how much the disease has already spread, the infection fatality rate (IFR) and reproduction rate — how many other people each carrier will infect.

Third, the models themselves are only a simplified version of reality. They cannot reflect the complications of different situations in which people get infected or infect others. Capturing the specific characteristics of a society, and of sub-sections of that society, such as church-goers or football fans, is always going to be difficult.

The modellers at Imperial have made some best guesses at all of these factors. Other research teams have made different assumptions to guess at how the virus might progress. For example, one group at the University of Oxford speculates, on the basis of their own model, that half the population of the UK may already have the virus. The lead author, Professor Suneptra Gupta, said: “I am surprised that there has been such unqualified acceptance of the Imperial model.” Forecasts for how many people might die could swing wildly depending on relatively small variations in these assumptions.

Although these difficulties afflict all modelling, there has been particular criticism in recent days of Professor Ferguson's track record. He worked on initial estimates for the possible death toll of variant-CJD, aka the “human form of mad cow disease,” in 1996, estimating with others a range from 50 deaths to 50,000. But these were far from the most lurid estimates for a disease that has actually killed just 176 people. During the outbreak of foot-and-mouth disease (FMD) in 2001, his estimates led to a far wider cull of animals than previously thought necessary, ultimately costing the economy an estimated £10 billion.

One critic, Michael Thrusfield, professor of veterinary epidemiology at Edinburgh University, has since written two papers that are highly critical of the research led by Ferguson on FMD. One of the papers, from 2006, argues 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.”

Another academic critic, Dr Paul Kitching, told The Telegraph: “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?”

Ferguson has responded to the FMD criticism, saying: “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.”

The real problem is the way that models are latched on to at a time when information is very limited. They take on a life of their own, rather than being treated as preliminary estimates, feeding into a sense that we should always prepare for the worst-case scenario rather than the most realistic one. Interestingly, the government has delayed introducing tests for the disease until they are proven to be accurate. So why is it apparently relying on models that haven't been tested?

But while criticism of Ferguson may be valid, he is right to point out that his work is just one line of expertise and argument feeding into policy. There are plenty of others, too, from behavioral scientists to economists and, of course, political voices. It wasn't Ferguson who made the decision to focus on short-term deaths rather than long-term harms to society from suppression measures. On what basis was that decision made? Has the government made any attempt to model the harm that could be caused — to the economy, to health, to education, to the general social fabric — of an extended lockdown? If so, where is this research?

Governments are also faced with a moral dilemma: how to balance different kinds of harm. For example, it is clear that Covid-19 is particularly lethal to the elderly and those with serious health problems already. Even Ferguson has admitted that two-thirds of those who would die without suppression policies could be dead by the end of the year anyway. Are we hobbling society for years to come in order to buy some of its members a few more weeks and months?

From the outside, it looks like the government had a plausible mitigation strategy, but then panicked under pressure from the media and members of its own party to adopt a more draconian policy. If Ferguson is right, senior politicians will congratulate themselves on listening to the experts. If Ferguson is wrong, he may end up being the fall guy.

Think your friends would be interested? Share this story!