Are lockdowns working?

By Christopher Monckton of Brenchley

In recent weeks, behind the scenes, a battle royal has been raging among the epidemiologists advising governments. On one side are the activists, who argue that the Chinese virus is both more infectious and likely to prove more fatal than influenza, a deadly combination.

The activists’ strongest arguments are that in the early stages of a pandemic the daily growth rate is exponential; that in the absence of determined control measures a quarter of the global population would be infected by the end of May; and that continued exponential growth at the daily compound rate of almost 20% (entailing a doubling every 3.8 days) that prevailed until mid-March would rapidly overwhelm not only the hospitals but also the morgues, as has already happened in Spain and northern Italy.

On the other side are the passivists, who argue that after a few weeks in lockdown people will cease to observe the restrictions, introducing a second wave of infection. They hold that the best thing to do is let everyone become infected, let the old and the sick die, let the health services collapse, and leave the population to acquire what the lamentable Chief Officer of Health in London described at a press conference some weeks ago as “herd immunity”. The international outcry at this crass remark led the British government to backtrack at once.

I declare an interest. When it comes to preventing pandemics, I am an activist. The earlier one interferes with the exponential growth of a pathogen as infectious as the Chinese virus, the less the cost in lives and treasure. When HIV first emerged, I minuted the Cabinet to the effect that there should be universal testing, followed by immediate, compulsory and permanent isolation of carriers. No such action was taken, unfortunately. The result is that some 50 million have died of HIV, another 500,000 a year die of it, and the cost of treating those who are HIV-positive is heavy. Nearly all those deaths were preventable.

The Chinese virus is considerably more infectious and more fatal than HIV. Realizing this, the British Prime Minister, after weeks of listening to the internal wranglings between the activist and passivist public-health scientists, who were unable to agree among themselves, took a command decision to lock down the United Kingdom firmly, completely and for as long as might be necessary. He was persuaded by modeling from Imperial College, London, showing just how rapidly the National Health Service would be overwhelmed if things went on as the passivists wished. It was clear to the Prime Minister that patients suffering from diseases other than the Chinese virus would be placed at risk as the health system collapsed.

Mr Trump, who, like Mr Johnson (and me) was by instinct reluctant to subject the entire population to house arrest and to cause dislocation and damage to the economy, eventually came to a similar view. The situation is more complicated in the United States, where the individual states rather than the Federal administration are chiefly responsible for public-health measures. But in many states, as in many nations round the world, lockdowns of varying severity have been introduced. The activists have thus far prevailed.

But are the lockdowns working? A simple performance indicator, clear enough to show people whether or not the house arrest and related measures to which they are being subjected should be persisted in, is necessary. Remarkably, however, no such benchmark test is yet available. Therefore, I have been researching the statistics and propose the following test. The reference period for the test is the three weeks from January 22 to 14 March 2020, the date on which Mr Trump declared a national emergency. During the reference period, the mean compound daily growth rate in confirmed cases was 19.8%. Confirmed cases were thus doubling worldwide every 3.8 days.

To demonstrate the extent to which mitigation measures are or are not working, the benchmark test calculates the mean daily compound growth rate in confirmed cases of infection for successive seven-day periods ending on every day from March 14 to the present. Here is the test for the world excluding China and occupied Tibet (whose Communist regime cannot be trusted to tell the truth about case numbers, or about anything else much); for the United States, and for the two worst-affected European nations, Italy and Spain:

All four nations show an inexorable reduction in the daily rate of growth (though it remains dangerously high). The most impressive results are those for Italy, the first country in Europe to impose a strictish lockdown. During the reference period, the Italian growth rate was more than 30% per day, and cases were doubling every 2.6 days. But the lockdown is beginning to work. In the week to April 2, the daily growth rate in Italy was down to 5.2%. Even that is an alarming value: it would lead to a doubling of cases every two weeks. But the trend in the daily growth rate is firmly downward, and it will probably continue that way – provided, that is, that people can see, as they can from this test, that the lockdown is indeed working. In the world outside China, as more and more countries introduce lockdowns, the daily growth rate has declined from 19.8% in the reference period to 11% in the week to April 2. In the United States, the daily growth rate has declined a little, from 23.1% in the reference period to 16.2% in the week to April 1.

Here is the benchmark test for four more countries: three in Europe and one for South Korea. All four countries show declines in the daily growth rate of confirmed cases. But in South Korea the pandemic is almost under control:

The reason for the success in South Korea is that, following the SARS epidemic, the public health authorities fully understood the paramount importance of very widespread testing, immediate isolation of carriers and vigorous contact-tracing, including use of the cellphone network to identify where the carriers had been and whom they had met. The EU has picked up this idea, though the UK – in this as in much else – lags behind.

In particular, ever since the SARS epidemic the Korean public health authorities have maintained a very large testing capacity. They activated it as soon as they realized that the director of the World Health Organization, who has close links to the Peking regime and had as recently as January been parroting Chinese propaganda to the effect that the virus could not be transmitted from human to human, could not be relied upon.

Britain will be calling for an independent investigation of the WHO’s gross misconduct in this affair as soon as the pandemic is under control.

South Korea also adopted national lockdown. The public health authorities also recommend use of personal protective equipment (notably face-masks) not only by health professionals but also by the general public when outdoors. In this respect, too, the South Korean public health authorities disagree with the WHO, which has today announced it is reconsidering its notion that masks are valueless. The director of Korea’s public health authority bluntly says that the evidence that masks work is overwhelming.

Following his advice rather than that of the useless WHO, I wear a full-face motorcycle helmet and gauntlets whenever I leave our own grounds. Full-face protection is useful, according to the South Koreans, because the Chinese virus can enter the body not only through the nose and mouth but also through the mucous membranes of the eyes. Even wearing spectacles provides some measure of additional protection. As South Korea’s expert made clear in an excellent recent interview, it is necessary to obtain every advantage one can, because each additional barrier to transmission helps to bring the pandemic under control.

It is South Korea, then, that provides the clearest evidence that prompt, determined and vigorous control measures work, and work well.

Both Germany and France have done quite well in beginning to control the pandemic. Their mean daily growth rates were down from more than 30% in the benchmark period to around 10% in the week to April 2. The United Kingdom, however, had a daily growth rate of 16.4% in that week: a value scarcely better than the global 19.8% during the reference period from January 22 to March 14. The UK is the worst-performing of the 12 territories tracked here.

Germany and France both took advantage of the EU’s system for supplying both testing kits and personal protective equipment for health professionals. The UK, however, failed to respond to the EU’s email in time. Worse, British civil servants are so used to acting simply as passive agents for the Brussels tyranny-by-clerk that they were more or less completely unprepared for a pandemic, and the flapping-around is saddening to watch.

The former director of “Public Health England”, a grim but useless bureaucracy, was asked four times yesterday why it was that Germany had tested more than 500,000 of its citizens in all, while Britain had not yet managed to test 10,000 in any one day. He could not answer.

Here are benchmark tests for four more countries: Canada, Australia, Sweden and Ireland. Note that for Ireland the benchmark period is the two weeks to March 14 rather than three weeks, because Ireland began to report cases later than other countries.

From the point of view of the passivists, Sweden is the most interesting result. For its public health authorities are passivists: they have not introduced a lockdown. Yet their daily growth rate has fallen to 10%, among the lowest anywhere. Nevertheless, there is growing concern among health professionals in Sweden that the do-little option may yet prove fatal. It is possible, then, that Sweden will follow other European countries in imposing a strict lockdown in the near future. In the past ten days, other countries have seen a decline in the daily growth rate of confirmed cases, but Sweden, uniquely, has not.

Overall, the benchmark test show – at this early stage – that the lockdowns are beginning to work. The daily growth rate in confirmed cases is falling in those countries that have been locked down, and is tending to fall fastest in countries with the most determined control measures.

The next few weeks will be particularly interesting, because it is in the nature of exponential growth curves that, just as the growth is very rapid if control measures are not tough enough, the slowing of growth is just as rapid when the measures really begin to bite.

Over the next few weeks, the extent of the lockdowns’ success or failure will become evident. For this reason, I propose to update the benchmark tables daily until further notice.

It should be made clear that the benchmark test is not policy-prescriptive. It merely shows, in a dispassionate fashion based on the available data (warts and all) the extent to which control measures are or are not working, territory by territory and for the world excluding China.

Finally, the question arises whether the official data on which I have relied are trustworthy. The answer is that they are not, for the lack of widespread testing has entailed a very substantial understatement of the numbers infected.

Take the United States as an illustration. On average the Chinese virus takes five days to incubate and a further 16 days to kill those to whom it proves fatal. The least unreliable of the official statistics are those for deaths caused by the virus. On February 29 the United States reported its first death from the virus. The World Health Organization, which had originally estimated a death rate of 2% (as it had with SARS, whose death rate was actually 9.6%), now estimates it at 3.4%. In that event, 21 days previously, on February 8, there must have been 1 / 3.4%, or 29 cases. However, only five cases were reported. But if there were 29 cases on February 8, and if the growth rate for unreported cases is the same as for reported cases, the true number of cases by February 29 was not 5, as reported, but more than 2300.

Performing a similar calculation for each day until April 2 would lead us to conclude that there were not 26,500 cases of infection in total by that day, as reported by the U.S. administration, but 36 million. Curiously, if this were true it would not be all bad news. For the death rate would then be less than 0.02%, rather than the WHO’s 3.4%.

What is more, since only 6000 deaths have been reported in the U.S., the vast majority of those infected would have suffered symptoms little worse than those of the common cold and have recovered, in which event the “herd immunity” of which the British public health commissar spoke is being built up at a rapid rate.

If the death rate is only 1%, it is possible that 123 million people – more than one-third of the U.S. population – are already infected. If, however, it is 10%, as for SARS, then about 12 million U.S. citizens are infected.

What, then, is the true death rate? This early in the pandemic, the answer is that nobody really knows, even to within an order of magnitude. The standard method of obtaining a preliminary assessment of the death rate in the early stages of a pandemic is to consider the closed cases – those who, having been infected, have either recovered or died. Until April 2, 135,447 people outside China and occupied Tibet were reported as recovered from the infection, while 49,845 had died. Therefore, 185,252 had either recovered or died, and the deaths represented not 2% nor 3.4% but almost 27% of all these closed cases. I have not seen that figure reported anywhere, but that is the figure.

If the death rate is indeed 27%, then only 4.6 million U.S. citizens are infected, compared with the reported. However, the 27% figure should be regarded with some caution, since it takes no account of the under-reporting of cases, many of which will have been recoveries or asymptomatic. But it does suggest that of the currently-active 748,153 confirmed cases outside China more than 200,000 will be likely to die worldwide.

The Chinese virus, then, will be a biggish killer, either because far more are infected than are being reported or because the death rate is higher than the WHO imagines, or both. At this stage, we do not know: but no responsible government, seeing figures such as these, would consider itself as acting responsibly if it were to fail to ensure that energetic control measures were put in place.

In all this mishmash of competing statistics, the one certainty is the daily mean rate at which reported cases have been increasing. That is why I have chosen this measure as the basis for the benchmark test.

My hypothesis is that, thanks to the decisive measures taken by most governments, the daily growth rate of total confirmed cases will continue to fall, and that about 1-2 weeks from now the fall will become quite rapid, perhaps buying enough time for health services to increase their capacity to handle intensive-care patients on ventilators, and to perform antigen tests for the presence of the virus and, no less importantly, antibody tests to demonstrate that those who have recovered are immune.

If the daily growth rates do not fall very quickly to South Korean values, then the capacity of health services will be overwhelmed. As of yesterday, the hospital ship sent by President Trump to New York had just three patients on board. Expect the ship to be filled to capacity within days.

Keep safe. And come back here daily for the updated benchmark test.