Is it possible to predict an end to the corona crisis?
Is the coronavirus really unpredictable? In Germany as in other countries, epidemiologists are working with mathematicians and geophysicists on computational models to predict the corona virus’ spread and the effect of containment measures. One thing is certain: The world needs to be prepared for a long haul if it is to successfully combat COVID-19.
#COVID2019 and #Flattenthecurve are probably the most visited and commented Twitter hashtags worldwide today. What helps to flatten the coronavirus disease’s exponentially growing curve – and when will the effects set in? The scientific community cannot and will not provide a clear answer yet – we still know too little about SARS-CoV-2. If there is one thing that can be said about the pandemic with certainty, it is that nothing is certain. Biologists and physicians in all parts of the world are working on a vaccine and effective drugs. But mathematicians, statisticians and geophysicists are also coming under the spotlight. On the one hand, they are involved in processing the data on the outbreak to date and drawing conclusions from it. On the other hand, they are developing models to help predict the virus’ further spread.
Data on the coronavirus’ spread in Germany
One of these research groups works at the Karlsruhe Institute of Technology’s (KIT) Center for Disaster Management and Risk Reduction Technology (CEDIM). Together with Risklayer GmbH, an analysis database for risk assessment, they have published maps of the spread of the virus in Germany and worldwide. The site http://risklayer-explorer.com/event/100/detail is updated continuously.
Amongst other things, this analysis makes it possible to track how the number of new infections per day has changed over the past two weeks. "Our goal is to provide an overview of the number of people infected with the novel coronavirus compared to the population", says geophysicist James Daniell, co-founder of Risklayer. The team uses official statistics from health ministries and local governments. So far, the scientists have analyzed more than 5,000 data sources using scraping – the process of combining information by collecting the required data from websites. They have also launched a crowdsourcing initiative, and with the help of volunteers in Germany they want to collect the latest data even faster. In addition to the number of COVID-19 cases, they are also evaluating demographic information such as the number of inhabitants or the age of the people affected.
One of the problems in Germany, as in many other industrialized countries, is that there is no open online portal that summarizes the current data from the federal states down to the district level to be evaluated in real time. This is different in China, in other Asian countries and in many developing countries: "This shows that countries that have been regularly affected by severe natural disasters such as tsunamis or earthquakes in recent years have made provisions for the next emergency with such open portals," explains James Daniell. The figures presented daily by the Robert Koch Institute (RKI) have always lagged at least 24 hours behind the actual figures. That's exactly how long it takes for the cases recorded by the municipalities to reach the federal level and thus the RKI.
This time span could increase with an increasing number of infected people. The Ministry of Health in Baden-Württemberg gives one reason for the delay: "The current high workload of the local health offices can lead to a delay in data input and case transmission, particularly in health offices with a high number of cases."
The question is what helps in this situation: Does a total lockdown help to curb new infections? Or is the partial isolation of particularly vulnerable people with previous diseases and in old age the better choice? Another approach: Should we simply let as many people as possible become ill with COVID-19 in order to achieve what is known as herd immunity and thus prevent the virus from getting a chance?
Mathematical models of the dynamics of the corona outbreak
The Research Centre Jülich is looking for answers to these questions in the fight against the coronavirus and is making its supercomputer capacities available for this purpose. "As a matter of urgency, we are now bringing together experts from our scientific network to obtain precise data on the virus’ spread and drug development as quickly as possible. At the moment, I expect that we will have initial results soon," says Prof. Thomas Lippert, Director of the Jülich Supercomputing Centre (JSC) at Forschungszentrum Jülich.
Together with other international research institutions and companies, the Research Center has joined the Canadian quantum computer manufacturer D-Wave Systems Inc.’s initiative to support researchers in developing solutions to combat the corona pandemic. D-Wave provides free access to quantum computer systems for users researching COVID-19. Partners from CINECA, Kyocera, NEC Solution Innovators Ltd., the QAR-Lab at LMU Munich, the University Tōhoku and Volkswagen, in addition to Forschungszentrum Jülich are helping to implement the project.
Together with the University of Heidelberg and the Frankfurt Institute for Advanced Studies (FIAS), the JSC is developing mathematical models of the dynamics of corona eruption in Germany, which can also be used to simulate the effect of containment measures. The goal is to predict when the outbreak will peak and how many people might become ill. In addition, the JSC and its partners at the Gauss Centre for Supercomputing (GCS) are providing computing time.
The EXSCALATE4CORONAVIRUS (E4C) project needs such computing time. The project will use the currently most powerful European computer resources in Barcelona (BSC), Bologna (CINECA) and Jülich to simulate computer-assisted active substances for the development of drugs against the novel corona virus (in-silico-drug design). This approach allows the fast analysis of simulation results and reduces the time required to discover new therapeutic agents. The Jülich Institute of Neuroscience and Medicine, Computational Biomedicine ist also involved in the project.
Could the principle of herd immunity work?
Another approach is to hope that so-called herd immunity will be established as soon as possible. This is the case when a sufficiently large number of people have experienced the disease and are subsequently immune. These people then no longer transmit the disease. Infected people no longer meet enough people to spread the virus. The number of infected people therefore decreases.
The mathematicians Wolfgang Bock from the Technical University of Kaiserslautern and Thomas Götz from the University of Koblenz-Landau, together with Polish researchers from the Medical University and the Technical University of Wroclaw, have shown that this leads to problems in their project "Modelling Corona Spread" (MOCOS): "We were able to see from our simulations that a strategy aimed at letting the disease pass through in a controlled manner without overburdening the health care system fails. The range of contacts that anyone outside the home could have is too small for this," said Bock and Götz. In other words: According to MOCOS calculations, herd immunity would not yet be established in a scenario in which the disease is allowed to pass through in a controlled manner. At most 15 percent of the population would be affected. That is too few to protect against a second corona wave. At the same time, the health care system would be severely burdened by the high number of patients suffering from the disease and intensive care patients in parallel. The scientists conclude that "fast, effective and hard measures are sensible until the disease has passed – the harder the measures, the faster one gets through the crisis.”
Sebastian Funk, an epidemiologist researching at the London School of Hygiene & Tropical Medicine, is also working on modelling the COVID-19 pandemic. He told SPIEGEL: "At the moment, it is difficult to imagine how it will be possible to eliminate the virus globally.” But if complete eradication is not successful, says Funk, the only remaining options are to make the lockdown of public life a permanent state until a vaccine is developed – "or otherwise to accept that a large part of the population must be infected."
A radical lockdown with strict curfews for many weeks or even months, however, has – that much is certain – not only a devastating effect on the economy, but also on the population’s mental health. In addition to the loss of many freedoms, there is the threat of financial worries if employers or clients cut back their operations or even go bankrupt.
However, if the restrictions were relaxed too soon in the event of a short-term or even supposed decline in the number of illnesses, a renewed strong wave of infection could occur. This in turn would have to be answered with severe restrictions in public life – a spiral is created.
In Europe and in the USA, which have now become the center of the coronavirus’ spread, it seems that at present there is only a choice between several evils to deal with corona. Very vividly, an article in the WDR science magazine "Quarks" (in German) summarizes the different options we have to fight corona. All of them come with unpleasant consequences for the population and show that the pandemic will accompany us for a long time to come. Research on SARS-CoV-2 is therefore more important than ever to make the virus manageable. It will probably not be easy to eradicate.
Research at Karlsruher Institut für Technologie (in German)