Why do the forecasts for the development of the coronavirus pandemic have a very serious estimating error? What is the problem with the current tests for SARS-CoV-2, and whether BCG vaccination can really affect the incidence of COVID-19? Read about this in an interview with Professor Raul Gainetdinov, Director of the Institute of Translational Biomedicine at St Petersburg University, Academic Supervisor of the Pirogov Clinic of High Medical Technologies of St Petersburg University and one of the most highly cited researchers in the world according to the Web of Science.
Currently, many research organisations, both in Russia and overseas, are trying to forecast the development of the COVID-19 pandemic locally and globally. Have you made any COVID-19 forecasts?
I am a medical practitioner engaged in science. I am used to processing statistical data and I am familiar with the diagnostic methods used for detecting COVID-19 today. I’ll say that predicting the future of the COVID-19 pandemic with data available today is pure guesswork. Polymerase chain reaction testing (PCR), currently used to confirm the infection, calculates the quantity of RNA at the peak of the disease, when the virus population is multiplying in the host’s body. These tests cannot detect the infection at very early or late stages of the disease. Besides, only a small percentage of the population is tested: those who had to be hospitalised, their relatives, or those who were sitting next to them on an airplane. While a vast majority of the human population, people in the street, who do not show obvious symptoms, remain untested.
In your opinion, how many people in a population may be silent virus carriers with no obvious symptoms?
Different assumptions have been made regarding the proportion of the population that does not exhibit symptomatic illness. According to various estimates, between 40 and 80% of individuals are asymptomatic virus carriers. The most radical opinion has been expressed by German epidemiologists. They believe that the proportion of coronavirus patients in hospitals to those who were not hospitalised is ten thousand to a million. We can determine how many people have already been infected only by counting coronavirus antibodies. An infection triggers an immune response. This reaction enables detection of those who had been infected with COVID-19 virus and recovered. These tests, however, are just being introduced worldwide. In Russia, they have been used since Monday; that is, literally, for a couple of days. In several Moscow hospitals, doctors have been tested: up to 7% of medical staff showed the presence of antibodies. These are preliminary findings, indeed. Every day I check official information on testing more people, and not only in hospitals.
Once the necessary data have been collected, will it be possible to provide accurate forecasts?
Much will depend on the sensitivity and accuracy of the method. Thus, the PCR method, currently used to detect the infection, has an accuracy of up to 70%. That is, the remaining 30% are either false positives or false negatives. We use this method in our daily work, and I have to stress again the necessity for repeated testing to confirm the accuracy of the results. The accuracy of the antibody test is still unknown. I have not yet seen any data to form a judgement. However, the USA has been critical about the antibody tests developed in China. These tests have flooded the market; yet, their effectiveness has not been proven. Many of the tests turned out to be very non-selective: they detect coronaviruses that have been around for a long time but they cannot ‘catch’ the novel coronavirus (2019-nCoV). I can say nothing about the newly developed Russian antibody test’s effectiveness. I just hope that more accurate data will be available soon.
Therefore, all the forecasts and predictions at this stage are akin to playing darts blindfolded. When we do not know all the initial conditions accurately, we can write a beautiful but erroneous formula, and we will obtain false results that have nothing to do with reality. Some people think that the real number of coronavirus cases is 10 times higher than we see in the reports, because the majority of the population have not been tested. Other people believe that the real count is 100 times higher. In my opinion, the number of people infected with the new coronavirus in St Petersburg is much higher than the official statistics suggest. Many times higher. The question is how many times higher is it. But you need not be afraid of this, because the more people who have contracted and recovered from the virus, the higher the herd immunity.
Why are we so concerned?
Why are people scared of the coronavirus? The mortality rate for COVID-19 is higher than for influenza. Initially, they said that the death rate is 1%, then 4%, then 5%... However, if you do not know the real number of coronavirus cases, that doesn’t tell us much. Firstly, we don’t know the number of infected people, and, I repeat, most likely, the real number is much higher. Secondly, we do not know the real number of deaths associated with the novel coronavirus, which is most likely to be lower. Since COVID-19 appears to be lethal mainly for people with concomitant chronic conditions, it is not always possible to be sure that the cause of death was the coronavirus infection. Due to this disproportion, the figures simply do not add up or make sense, which scares everyone.
I’ll give you an example. A few weeks ago, US experts projected that between 2 and 3 million people could die in the United States as a result of this pandemic. A week later, Donald Trump perplexed the world when he said that 600 to 800 thousand Americans would die from COVID-19. Two weeks later, Trump mentioned 150 to 200 thousand deaths associated with the coronavirus pandemic. And three days ago, an American immunologist Anthony Fauci, Senior White House coronavirus adviser lowered the US coronavirus death forecast to 60 thousand people. In fact, influenza-associated deaths in the United States range between 35 and 60 thousand cases per year. Thus, Fauci admitted that the COVID-19 mortality is not much different from that of the flu. The logical question arises: why are we all scared?
Unfortunately, the World Health Organization (WHO) also contributed to growing concern and unease about COVID-19. Personally, as a scientist, I was angry with some of their statements. For instance, they stated that the mortality rate for COVID-19 is much higher than that for the flu. How can they say this when no one knows how many people really got infected? They said this as early as during the coronavirus outbreak in China, hitting the panic button. In other words, the panic was caused by some ‘finger in the air’ estimates. We need the real data to guide our judgement in this case.
How would you comment on the hypothesis that universal BCG vaccination can decrease susceptibility to the new coronavirus and, consequently, reduce its morbidity and mortality?
On this hypothesis, our country need not worry about COVID-19. I have to make a disclaimer – this is still just a supposition based on certain correlations. These correlations, however, make this supposition probable. Moreover, the supposition is supported by several studies conducted in different countries. Besides, some of these studies were done long before the current pandemic. Generally speaking, these studies explored the effects of BCG vaccination on viral infections and herd immunity. It is believed that early BCG vaccination can boost innate immune responses through a process termed ‘trained immunity’, increasing the body’s resistance to various infections. The results were rather optimistic; all the data is publicly available on the Web.
According to the WHO website, ‘the bacille Calmette-Guérin (BCG) vaccine has existed for 80 years and is one of the most widely used of all current vaccines, reading >80% of neonates and infants in countries where it is part of the national childhood immunisation programme. BCG vaccine has a documented protective effect against meningitis and disseminated tuberculosis in children. It does not prevent primary infection and, more importantly, does not prevent reactivation of latent pulmonary infection, the principal source of bacillary spread in the community. The impact of BCG vaccination on transmission of the ethiological agent of tuberculosis is therefore limited.’
As for the latest correlations regarding the current epidemic, in nations with mandatory policies to vaccinate against tuberculosis, the COVID-19 mortality and morbidity rates appear to be ten times lower compared to ‘leading’ countries. Countries with mandatory BCG vaccination include: Russia and the former Warsaw Pact countries; Southeast Asian countries, in particular China, Japan, and Korea; and most of the countries of Africa and Latin America. In most countries that are leading in mortality and morbidity today, BCG vaccination has never been universal or was cancelled years to decades ago. Among the countries that do not have universal BCG vaccine policies are Italy and the United States. The UK, the western regions of Germany, France and Spain, used to have BCG vaccine policies but ended them a long time ago.
I have checked some of the data recently again, just in case. Interestingly, in the German lands that used to be part of the German Democratic Republic (GDR) where BCG vaccination was mandatory, the mortality and morbidity rates for COVID-19 are much lower than in Western Germany, where BCG has not been vaccinated for a long time. The same applies to Portugal and Spain: the former has a general BCG vaccination policy, while in the latter the BCG vaccination programme was discontinued. And the numbers speak for themselves. Compared to Portugal, in Spain there are six to eight times as many infections.
If we compare the numbers of coronavirus deaths per million in Italy and Russia, there is a three hundred-fold difference. Theoretically, at present, Russia is at the peak of the epidemic. If we take the official statistics for today, the mortality rate in Russia is just over one case per million, while in Italy it is 358 cases per million, and in the USA – somewhere around 80 deaths per million. In the countries without universal BCG vaccination, exponential growth in case incidence has been observed. Whereas the countries with mandatory BCG vaccination, including Russia, exhibit a linear growth trend. Therefore, it would be premature at the current stage to say that all the countries are to expect an exponential growth of COVID-19. Even more so – it would be premature to make serious forecasts.
In other words, at the moment, any forecasts cannot be considered accurate?
Exactly. For instance, in the calculations recently published in St Petersburg, there is a second serious error: the calculations are based on the exponential growth model, which would work for Italy and other countries without mandatory BCG vaccination. Whereas in Russia, the growth of infection is linear, which fits the data accumulated in the countries with universal BCG vaccination.
A global scientific experiment on the epidemiology and transmission dynamics of the coronavirus is unfolding before our eyes. Soon everything will become clearer. But today it is obvious that there is no big difference in the transmission dynamics of COVID-19 between countries with drastic and moderate stay-at-home restrictions. Thus, the number of infections and deaths are now decreasing both in Italy and Spain, which have taken strict quarantine measures, as well as in neighbouring Switzerland, where severe measures were not implemented. The difference between the countries with a history of BCG vaccination is much more impressive.