No, British Epidemiologist Neil Ferguson Has Not 'Drastically Downgraded' His Worst-Case Projection of COVID-19 Deaths

But he has raised his estimate of the virus's reproduction number, which implies a lower fatality rate than his research group initially assumed.


Contrary to what you may have read or heard, British epidemiologist Neil Ferguson has not suddenly reduced his worst-case projection of COVID-19 deaths in the U.K. by a factor of 28. To the contrary, he says the policies adopted by the British government, which are in line with the aggressive control measures recommended by a highly influential March 16 paper that Ferguson and other researchers at Imperial College wrote, should keep the number of deaths below 20,000.

"We assessed in that report…that fatalities would probably be unlikely to exceed about 20,000 with effectively a lockdown, a social distancing strategy," Ferguson, who is himself recovering from COVID-19, told a parliamentary committee on Wednesday. "But it could be substantially lower than that."

In other words, it is not true that Ferguson "is presenting drastically downgraded estimates," as The Daily Wire claimed, or that he "just walked back the apocalyptic predictions," as The Federalist asserted. But Ferguson did revise one of his key estimates in a way that suggests a lower case fatality rate (CFR) than his group assumed in their modeling.

"What we've been seeing in Europe in the last week or two is a rate of growth of the epidemic which is faster than we expected from early data in China," said Ferguson, who testified from his home via video link. "So we are revising our central, best estimate of the reproduction number [i.e., the number of people the average carrier can be expected to infect] to something on the order of 3 or a little bit above rather than about a 2.5 level." In his view, that revision "actually adds more evidence to support the more intensive social distancing measures applied this week, because the higher the reproduction number is, the more intensive the controls need to be to achieve suppression of the epidemic."

A substantially higher reproduction number implies that the COVID-19 virus can be expected to spread more quickly than the Imperial College group imagined. But it also means that many more people in the U.K. already have been infected, which implies a bigger gap between known cases and the actual number of infections. That, in turn, implies that the true CFR is lower than the 0.9 percent rate that Ferguson and his colleagues used in their projections.

The Imperial College CFR estimate is far lower than the crude CFR for the U.K., which is currently about 5 percent. The difference reflects the understanding that the true number of infections is bound to be much larger than the official numbers reflect, because many people with mild or nonexistent symptoms (as is typical of COVID-19) will not seek medical treatment or testing. The size of that group is a crucial question in estimating the true CFR.

Ferguson believes the number of undocumented infections is not nearly as high as a recent estimate by researchers at Oxford University, who suggested that half of the British population is already infected. If that were true, the CFR for COVID-19 in the U.K. would be something like 0.002 percent, making the disease much less deadly than the seasonal flu, which has an estimated CFR of 0.1 percent.

"I don't think it's consistent with the observed data," Ferguson said of the Oxford estimate, citing the results from comprehensive testing of Italian villages and the Diamond Princess cruise ship's passengers and crew. Raising the reproduction number from 2.5 to 3 or more nevertheless implies that the number of undocumented infections is higher than Ferguson's group originally thought.

In last week's paper, Ferguson and his co-authors, writing on behalf of the Imperial College COVID-19 Response Team, projected COVID-19 deaths in the U.K. and the United States based on a range of policies and a range of reproduction numbers. In their worst-case scenario, which assumed a reproduction number of 2.6 and "the (unlikely) absence of any control measures or spontaneous changes in individual behaviour," they projected 550,000 deaths in the U.K. and 2.2 million in the United States. Although those horrifying numbers got a lot of attention, they were never plausible, as the paper itself said, because they were based on the clearly unrealistic premise that "nothing" is done to contain, suppress, or mitigate the epidemic.

In Ferguson et al.'s best-case scenario—based on a reproduction number of 2 and isolation of people with symptoms, home quarantine of everyone else in their households, and early implementation of school closures, coupled with "social distancing of [the] entire population"—they projected just 5,600 deaths in the U.K. But when they raised the reproduction number from 2 to 2.6, the number of deaths more than doubled. They projected 12,000 deaths in that scenario.

Ferguson now says the reproduction number is probably "a bit above" 3, nearly as big a change as the shift in assumptions described in the paper. Yet he thinks the number of deaths is "unlikely to exceed" 20,000 and "could be substantially lower than that." Although it certainly seems that Ferguson has become more optimistic about the fatality rate, he denies that. "Our lethality estimates remain unchanged," he said on Twitter yesterday.

Another difference between last week's projections and Ferguson's testimony this week is the expected peak of COVID-19 cases in intensive care units. The report projected that those cases would peak in late November or early December if the U.K. adopted the combination of policies that the authors deemed most effective. But on Wednesday, Ferguson said he expects that the peak will be reached by mid-April.

"If the current measures work as we expect them to," Ferguson said, "we will see intensive care demand peak in approximately two and a half to three weeks." Because of those policies and increased ICU capacity, Ferguson said, he is "reasonably confident" that British hospitals will not be overwhelmed by COVID-19 cases, although their capacity will be strained in some parts of the country. He emphasized that the main goal of the suppression strategy favored by his group is to avoid a hospital crisis that "will have unintended consequences on health for the entire nation."

Ferguson also seemed to take a different stance this week on the question of how long aggressive COVID-19 control measures should remain in place. The Imperial College projections assumed that enforced social distancing of the entire population, which the British government is trying to achieve by ordering everyone to stay home except for essential purposes, would be maintained "for 5 months or longer," although it could be turned off and on based on trends in ICU cases. The authors added, rather alarmingly, that "to avoid a rebound in transmission, these policies will need to be maintained until large stocks of vaccine are available to immunise the population—which could be 18 months or more."

On Wednesday, Ferguson acknowledged that such a policy is not feasible. "We clearly cannot lock down the country for a year," he said. Even with a lockdown of relatively short duration, he observed, "we'll be paying for this year for decades to come."

The question, Ferguson said, is how the government should "allow the economy to restart," a process that "is likely to rely on very large-scale testing [and] contact tracing," along the lines of what South Korea has done. "We are looking at that as a model," he said. "The U.K. does not have the testing capability to replicate South Korea right now, but I think it's likely in the next few weeks we will." Widespread testing would also help clarify the lethality of COVID-19.

The Imperial College paper acknowledged that "the social and economic effects of the measures which are needed to achieve this policy goal [of suppressing the epidemic] will be profound." But the authors expressly did not "consider the ethical or economic implications" of choosing an aggressive "suppression" strategy rather than milder measures aimed at "mitigation."

In his testimony, Ferguson conceded that the economic impact of a nationwide lockdown is "a very important consideration"—"one that the government and scientists are grappling with." And he noted an important aspect of the epidemic that should inform any attempt to weigh costs and benefits.

"We don't know what the level of excess deaths will be in this epidemic," Ferguson said. In other words, we don't know the extent to which COVID-19 will increase annual deaths above the level that otherwise would have been expected. "By the end of the year, what proportion of those people who've died from COVID-19 would have died anyhow?" Ferguson asked. "It might be as much as half to two-thirds of the deaths we're seeing from COVID-19, because it's affecting people who are either at the end of their lives or in poor health conditions. So I think these considerations are very valid."

Although a cost-benefit analysis that considers not just deaths but years of life lost "sounds very utilitarian," Ferguson said, the issue is obviously relevant. An epidemic that primarily kills healthy children, teenagers, young adults, and middle-aged people will result in a much bigger loss than an epidemic that primarily affects the elderly and people with serious pre-existing medical conditions.

Pace New York Gov. Andrew Cuomo, that observation does not mean the lives of the people who are most vulnerable to COVID-19 are "expendable" or that "we'll just sacrifice old people." But it does, or at least should, figure in policy decisions that may be economically ruinous, imposing severe burdens on millions of innocent people who are vulnerable for different reasons.