The majority of Bengal fire sparks are safe, but some can easily start a fire – a similar story seems to be true for coronavirus carriers. The more cases of Covid-19 infection scientists analyze, the more they lean toward the version that the driving force of the pandemic was not random contact with strangers, but cases of mass infection within a particular social group. This means that by banning mass gatherings and maintaining social distance, it is possible to avoid a large-scale second wave of infection.
Imagine holding a Bengal fire in your hands. Most of the sparks flying from it are completely safe, and even if they fall on your hand, you are unlikely to feel a burn. But sometimes a spark falling on a pile of hay or a fluffy carpet is enough to start a fire. According to many scientists, the Covid-19 epidemic is spreading. Most carriers of the virus may not pose a significant threat to those around them, but sometimes just one infected person is enough to start a chain reaction and lead to a major outbreak of the disease. This type of spread is called clustering and is characteristic of many infectious diseases to a greater or lesser extent. The global “fire” was ignited by relatively harmless sparks from a Bengal fire that accidentally fell on a fluffy carpet.
South Korea, which experienced an outbreak of Middle East Respiratory Syndrome (MERS) in 2015, was better prepared for the epidemic of the new coronavirus than most other countries. Authorities were able to keep the infection under control for a long time: in the first four weeks after the first infection in the country, a total of only 30 cases were identified. The World Health Organization held up the Koreans as an example to the rest of the world – until one day, February 19, all the inspiring statistics went downhill. A resident of the city of Daegu, who went down in pandemic history as “Patient 31,” turned out to be a deeply religious woman. Even after she became infected, she continued to attend church meetings-even after she began coughing and her temperature rose. The name of the Taegeu resident known as “Patient 31” is kept secret by medical professionals and authorities. Over the next two weeks, the number of infected people in the country skyrocketed more than 100-fold to over 4,200. Nearly two-thirds of those infected were members of the same religious organization and contracted the virus either directly from a religious patient or from other infected members of the sect. This case is just one of more than two hundred examples of mass spread of the coronavirus collected by scientists at the London School of Hygiene and Tropical Medicine (LSHTM) from 25 countries. The countries are different, but the pattern of mass infection is the same: a person infected with the virus rapidly spreads the disease within a particular social group (cluster).
Here are just a few examples of similar stories from the same database: four music concerts in Osaka attended by infected individuals (resulting in 80 cases), dance classes in South Korea (65), a high school in New York (60), a shopping mall in Singapore (87), a business meeting in Boston (89), and so on. Cluster outbreaks of Covid-19 have been recorded in nursing homes, ski resorts, cruise ships, churches, restaurants, hospitals, and prisons – virtually any place where people congregate indoors.
It is not so easy to keep distance in the Moscow metro. More than 20 similar cases of mass Covid-19 infections have occurred in workers’ dormitories in Singapore. The largest of these was in the massive S11 dormitory, which was quarantined in early April. By May 17, there were more than 2,500 infected people there. Cluster infections are characteristic of many infectious diseases, especially those transmitted by airborne droplets. In the case of Covid-19, however, it appears that this is the primary mode of transmission: the pandemic is fueled not by random contact, but by the rapid and efficient spread of the virus within organized groups.
Keep your distance: inventions of the coronavirus era The discovery gives scientists hope. If the main driving force of the pandemic is indeed superspreaders who, for various reasons, infect their surroundings in large numbers, even accidentally, then to stop or at least greatly slow the spread of the disease, it is enough simply to restrict mass gatherings such as concerts or football matches. Other restrictive measures can be lifted or greatly eased if social distancing is maintained.
In different countries, the recommended length of social distance varies: in Russia – 1.5 meters, in the United Kingdom – 2 meters. Genetic analysis of viral samples collected from over 200 Israeli patients with a confirmed Covid-19 diagnosis showed that, on average, only 1-10% of all infected individuals infect approximately 80% of patients in the “next wave”. This study has not yet undergone the peer review process, but a similar result was found in Shenzhen, where 80% of those infected contracted the virus from only 8-9% of the “previous generation” of patients. Scientists in London have come to the same conclusion after analyzing all known cases of mass spread. “It appears that about 10% of infected individuals account for 80% of new infections,” says LSHTM lecturer and epidemiologist Adam Kukharski. All in all, the discovery is hardly sensational. Scientists knew 10 years ago that many infectious diseases (such as malaria) spread in a similar way and formulated the so-called “80-20 rule”. In the case of the new coronavirus, this proportion may be even higher. This means that the results of the latest studies may significantly change the epidemic control strategy. John Ioannidis is a professor of epidemiology at Stanford University. He teaches in several departments in the School of Medicine and in the Department of Statistics in the School of Natural Sciences and Humanities.
Anyone can be a potential superspreader, so it’s better to keep your distance. “Most of the clusters we know about seem to be related to events that took place indoors, with large numbers of people in close proximity. And if that’s the case, keeping people indoors doesn’t seem like the best strategy,” he muses. Most infectious disease transmission models are based on a parameter known as R0 – the reproductive or “infectiousness” coefficient. It determines how many healthy people, on average, are infected by each infected individual. Under normal conditions (i.e., in the absence of social distancing and other restrictive measures), the R0 of the current coronavirus is estimated to be between 2.2 and 3.0, depending on various calculations. However, in light of recent research, it seems that this indicator is not significantly different from the well-known saying about the average temperature in a hospital, where someone has a fever of 40 and another has already died and is lying in the morgue refrigerator, but “on average” the temperature is normal – 36.6.
The average temperature of all the patients in the hospital will not help determine what treatment each of them needs. We explain quickly, simply, and clearly what happened, why it matters, and what happens next. The number of episodes should stay the same: episodes End of story. Podcast advertising. In any case, it makes little sense to base calculations solely on R0, since it appears that the majority of infected individuals do not transmit the virus to anyone, i.e., their R0 is zero. Now, if it turns out that the cluster theory is true, then the first wave (in most countries, including Russia, it has already begun to decline) may become the last. In order to limit the explosive spread of the infection, it is necessary to limit mass gatherings of people in confined spaces and strictly require everyone to observe social distancing. This should be enough to prevent the influx of coronavirus patients from causing a collapse of the healthcare system. In order to more accurately calculate and predict the development of the pandemic, scientists are introducing a new indicator – the so-called “clustering coefficient” (k), which is calculated on a scale from 0 to 1. The smaller the k, the lower the proportion of carriers who are the main spreaders of the infection. And correspondingly, the higher the percentage of patients who pose no threat – like sparks of a Bengal fire falling in the snow. According to preliminary calculations, the clustering coefficient of the new coronavirus is about 0.1. By comparison, the clustering coefficient for SARS (severe acute respiratory syndrome) was slightly higher – 0.16; for Middle East respiratory syndrome (MERS) – 0.25; and for the “Spanish flu” that caused the 1918 global epidemic – about 1. At the same time, clustering estimates from Covid-19 can vary slightly. For example, the authors of a similar study conducted in Hong Kong concluded that approximately 20% of infected individuals contribute to the spread of the epidemic in the city – as a somewhat established rule.
Bars, arcades and parks in Hong Kong were temporarily closed, but markets continued to operate – although authorities officially advised people not to congregate in groups of more than four. Either way, in practice this means that the majority of virus carriers are unlikely to pose a threat. And that may explain some of the mysteries of the epidemic. For example, why didn’t the virus that originated in China spread around the world much faster? One version is that the first cases appeared there in September at the latest. Or why didn’t the first cases in Italy, France, or the U.S. in late December or early January cause an immediate outbreak: their carriers simply didn’t transmit the virus to enough people. According to Adam Kuharsky, if the Covid-19 clustering coefficient is indeed 0.1, then the majority of transmission chains die out on their own. In order for the virus to have a chance of reaching critical mass and triggering a chain reaction of a large-scale epidemic, the scientist calculates that it would have to enter the country unnoticed at least four times.
Identifying infected individuals at the border is virtually impossible, especially if they have no symptoms of the disease. However, it is estimated that the virus has been imported into the UK at least 1300 times, mostly from France, Spain and Italy. Professor Tom Jefferson of the Evidence-Based Medicine Center at the University of Oxford agrees that it is important and necessary to study cluster infections. But he cautions colleagues against jumping to conclusions, saying the same thing over and over again: scientists still know too little about the new coronavirus. “Our evidence base is literally growing by the day. Since the beginning of the year, more than 3000 scientific papers have been published on the coronavirus. However, such an abundance of information often leads to what we call ‘pandemic fog’ – where the results of one high-quality study sometimes directly contradict the conclusions of another,” says Professor Jefferson. Until recently, our understanding of respiratory viruses in general was based mainly on the well-studied influenza virus,” says the scientist. However, enough research has been done during the pandemic to say with confidence that there is little in common between influenza and Covid-19.
Scientists still have a poor understanding of what other factors influence the rate at which coronaviruses spread. “Covid-19 is a new, recently discovered pathogen, and therefore the results of the LSHTD research (as well as all other scientific studies of coronaviruses) cannot be considered definitive,” said Professor Jefferson. “This study was a sensible one to conduct, and its authors worked very efficiently. However, many more similar studies are needed to draw far-reaching conclusions. The conclusions reached by the authors potentially affect everyone who has been forced to self-isolate or go into long-term quarantine, the scientist emphasizes. Therefore, it is simply necessary to further understand the dynamics and methods of virus transmission in detail: “It is necessary to understand under what conditions the virus is most effectively transmitted: at what time, in what rooms, at what humidity and temperature, etc.”. So far, the most common type of cluster described is residential: the virus is known to spread easily between members of the same family. However, this almost never happens in schools – school clusters have rarely been encountered by scientists.
Scientists know that influenza spreads quickly in schools, but this has not been observed with the coronavirus. However, Jefferson says, it is possible that these seemingly paradoxical results are related to the fact that school closings were one of the first restrictive policies implemented, and clusters simply did not have enough time to form.
I generally agree with him, as does Kenneth Lin, a professor of family medicine at Georgetown University in Washington: “Any review of the scientific literature is as reliable as the evidence base is comprehensive. In this case, the sample [of the LSHTD literature review] seems incomplete and not reliable enough to make decisions about returning to schools, offices, gyms, and so on.”
In Moscow, quarantine measures began to be relaxed. But John Ioannidis of Stanford University is more optimistic. “Errors are inevitable in any study,” he agrees. “They are explained by both the selection of sources and the selectivity of our memory, and determining the extent to which they bias the results is always a challenge. However, I do not think they can negate the overall conclusions of this study.
Will every 100th person infected with the virus die, who is at risk, and who are “superspreaders”?