The New York Times
03/09/2020 - 17:07
- Clarín.com
- Society
Adam Kucharski studies how diseases spread, but does not operate with viruses in a laboratory or treat patients in a hospital. He is a mathematician, works at the London School of Hygiene and Tropical Medicine and uses mathematics to understand outbreaks of diseases such as Ebola, SARS, influenza and now the coronavirus (Covid-19). Its objective is to design better ways to control outbreaks .
By strange coincidence, before the current outbreak he wrote a book called The Rules of Contagion, which has been published in Britain and will be published in the United States in September. He talks about it that the mathematics of contagion not only implies physical illnesses, but also ideas, rumors and even economic crises .
For the recent centenary of the 1918 flu, Kucharski worked on an experiment with a mathematical colleague and BBC presenter, Hannah Fry, from University College London, and with collaborators from the University of Cambridge to make a documentary, Contagion: The BBC Four Pandemic (Contagion), using a cell phone application - BBC Four Pandemic - to track social contacts and map how an infection could spread.
The news of coronavirus epidemics around the world implies an avalanche of numbers that are a challenge for anyone who is not a scientist to say. I asked Dr. Kucharski to help us navigate through some of these numbers and tell us which ones we should pay attention to. We communicated by phone and wrote by email last week. What follows is an edited version of those exchanges.
The English mathematician Adam Kucharski
-We hear a lot about the percentage of sick people who are dying. Is that the death rate?
-The death rate measures the risk that someone who has the symptoms of the case dies at a certain time from the infection.
-And how is that index calculated?
-The ideal is to monitor a large group of people from the moment they develop symptoms until they later die or recover, and then calculate the proportion of those who have died within all cases.
A stockbroker works this Monday at the Wall Street Stock Exchange in New York (United States). There was a sharp crash due to the coronavirus crisis. (EFE)
- It reaches then with looking at the total number of deaths and the number of cases in force?
-The problem of dividing the total number of deaths and the total number of cases is that this does not take into account the cases not reported or the period from illness to death. That period, that delay is crucial: if a certain day 100 people with Covid-19 arrive at the hospital and today everyone is alive, that obviously does not mean that the mortality rate is 0 percent. We have to wait until we know what happens to them over time.
All deaths are from people who got sick two to three weeks ago, so it's not just about the deaths of the moment divided by the cases of the moment. In addition, some cases could go unnoticed: if in two cases there are two deaths, as happened in Iran last month, most likely several other cases have been overlooked.
-We have seen all kinds of figures in terms of death rates. Does the latest estimate of 3.4% worldwide make sense?
-In the beginning people took into account the total number of cases and deaths at present, which, as I said, is a miscalculation, and the conclusion was drawn that the fatality rate of the cases should be 2% based on the China data. If the same calculation is made with China's totals for yesterday, an apparent death rate per case of about 4% is obtained. People then speculate that something is happening with the virus, when in reality it is only this statistical illusion that we know from day one. Based on the best available data, I would say that, after considering the undeclared cases and the various delay periods involved, we are probably facing a fatality risk of between 0.5 and 2% for people with symptoms.
Two officers wait for passengers to do the temperature tests as part of the prevention of the coronavirus at Damascus airport. (Reuters)
-What happens to another number that we hear about all the time, R, the reproductive number, that is, how many people can a specific patient get? Why is it important and what is involved in its calculation?
-In its simplest form, R is the answer to the question: to what extent should we worry about contagion? If R is greater than one, each case, on average, is infecting at least one more person. Growth is observed.
If R is less than one, a group of infected people generates less contagion. From the point of view of health policy planning, this provides a very clear objective. For example, in the response to Ebola in 2014, it was a truly prominent part of the response. The goal was to get R to stay below one.
- It seems very simple and direct, but you comment in writing that it is more complicated than it seems. In your book you say that to calculate R you have to know the duration, the opportunity, the probability of transmission and the susceptibility (the "Points to connect"). Let's take them one at a time. What is the duration?
-The time during which someone turns out to be contagious. If someone is contagious twice as long, they are circulating and spreading the infection two more times.
-Do we know the duration of this coronavirus?
-On average, we are probably talking about a week or two . Of course, if patients are hospitalized, they do not roam the community spreading the infection in the same way.
-The second component is the opportunity. How is it determined?
-It is a measurement of the number of people with whom someone infected comes into contact per day. With something like the flu, you are not contagious for a long time, but many of your interactions could spread it. While with something like HIV, the duration is much longer but the number of sexual partners you have in relation to the number of conversations you have is obviously much shorter.
-And the probability of transmission?
-It is a measure of the probability that the infection is transmitted during a given interaction. For example, during a sexual encounter, the virus is not necessarily transmitted.
-Finally there is susceptibility. How is that determined?
- Susceptibility measures the possibility that the person at the other end of the interaction will contract the infection and get infected.
-Once you have the numbers of these four components, what is the equation to reach R?
-If you multiply them with each other, they give you the reproduction number. So if any of these things increases or decreases, it directly affects the value of R.
-How does this knowledge help public health planning?
-Generally, susceptibility is the easiest to reduce if we have elements such as vaccines. If we do not have them, we should consider pointing to the other aspects of transmission, such as reducing opportunities through social distancing, or the probability of transmission during issues such as handshakes, encouraging handwashing.
- What happens if one is not in public health but thinks about their own personal probabilities and what their behavior should be?
-If we imagine that the reproduction figure is two, each person is infecting two more, on average. But some situations are more likely to spread the infection than others. We have discovered that for cases like Covid-19, the most important seem to be the closest interactions .
What we have to think about - and what most of our data modelers are undoubtedly thinking about - is not just how many transmissions are occurring, but where those transmissions are occurring. If you are going to change your behavior, think about how to reduce those risk situations as much as possible.
-If you were an average person, what would you pay attention to in terms of news and numbers?
-A signal to pay attention to is whether the first case of an area is a death or a serious case, because that indicates that there has already been a lot of community transmission. As an impromptu calculation, suppose the case mortality rate is around 1%, which is likely. If there is a death, the victim probably got sick about three weeks before. That means there were probably about 100 cases three weeks ago actually. In the next three weeks, that number could well have doubled, doubled and doubled again. So you currently have 500 cases, maybe a thousand cases.
I think the other thing that people really should pay attention to is the risk of serious illness and death, particularly in the older age groups, those over 70 and 80 years old. In general we are seeing that perhaps 1% of symptomatic cases are fatal at all ages. There is still some uncertainty about it, but what is also important is that 1% is not evenly distributed. In the younger groups we are talking about 0.1%, which means that when you enter the older groups, you may be talking about 5%, with 10% being fatal cases.
When thinking about social behavior and our own interactions, the question should be: "How do we prevent transmission from reaching groups in which the impact could be really severe?"
The author of the note, James Gorman, is a scientific editor and presenter and screenwriter of the ScienceTake video series. He joined The New York Times in 1993 and has published several books, including How to Build a Dinosaur, written with paleontologist Jack Horner.
Translation: Román García Azcárate