In a popular YouTube video, physicist Brian Cox shows that a bowling ball and a spring fall at the same speed in a vacuum, that is, without air resistance. Although it was known in advance that this would be the result of the experiment, it is still impressive and not intuitive to see a light spring falling from a height of tens of meters at exactly the same speed as a heavy and dense bowling ball. The law of universal gravitation, discovered by Isaac Newton, accurately predicts this result, as it determines that the acceleration an object experiences due to the gravitational force of a planet depends only on the distance to the planet and not on other properties of the object. The air resistance, which makes the spring fall more slowly, in obvious contradiction to this law, disappears when experimenting in a vacuum.
In physics, it is possible to predict the results of such experiments very accurately, since it is possible to separate the effects of the various factors. The recent descent of the Perseverance probe on Mars, after nearly 500 million kilometers, occurred precisely on schedule because the contributions of the various factors resulting from the effects of the atmosphere and the gravitational pull of the sun and planets could be considered separately .
Unfortunately, this separation of factors is much more difficult or even impossible in other areas such as the social sciences, economics or medicine. Despite the efforts of millions of scientists, we still don’t have good models for the Covid-19 pandemic. It is true that in record time it was possible to determine the structure of the virus, the sequence of its genetic code, the mode of transmission, the mechanisms of infection and, with this knowledge, produce effective vaccines.
However, human-to-human transmission of the virus depends on so many factors that we still do not know how to explain many phenomena. For example, we don’t know the real reasons why the virus mortality rate is much lower than predicted in dozens of low-income countries per capita such as Bangladesh, Thailand, Vietnam, India or Nigeria. We don’t know if school closings are a major contributor to stopping the virus from spreading, as different studies have produced different or even conflicting results. We do not know the level of immunization that is required to achieve group immunity, as this depends crucially on the spreading patterns. We don’t even know if restrictions are particularly effective in the face of conflicting studies. We ignore how atmospheric conditions, humidity, ventilation and occupancy of houses, temperature and other factors affect the spread of the virus. We don’t know why, under certain circumstances, some people are super spreaders, infecting hundreds of people, while others don’t seem to spread the virus significantly. We also do not know exactly why we see a rapid increase in contagions at certain times and a rapid decline in these contagions at other times, as happened last month not only in Portugal but around the world. We do not even know which physiological, genetic and behavioral traits cause the highest risk of infection and, in particular, the highest risk of serious illness.
The reader will argue that we know of some conditions that increase the risk of death, such as age or the presence of certain diseases. But unlike in physics, where it is possible to separate the various factors involved, in a pandemic like this we couldn’t do one. Every person is different and the effects of the virus depend heavily on these differences. Our knowledge of the mechanisms by which the virus causes the disease is still insufficient, and it will be decades before controlled and randomized studies can determine with certainty the safety of several factors.
Regarding the spread of the virus, it could be argued that there are mathematical models that make it possible to model the spread of the virus and the evolution of the pandemic. This is true, but these mathematical models simplify reality and assume that it is possible to separate the various factors involved. The most common models, so-called compartment models, divide people into compartments depending on their condition. The simplest model, the SIRD, uses four categories or compartments: vulnerable (S), infected (I), recovered (R), and dead (D). Knowing that at the beginning of the epidemic, each person, on average, infected a certain number of other people (the now famous basic reproduction number R0) and the time that characterizes the different stages of the disease, it is possible to determine with great precision who Developing the number of people in each compartment and modeling the dynamics of the epidemic, including the first phase of exponential evolution, the achievement of group immunity and the subsequent decline in infections.
We have to use all available instruments. Above all, we must avoid relying on monolithic opinions and not base our actions on a very small number of specialists who sometimes place unjustified trust in their models.
However, these models are only accurate if the transmission conditions remain essentially unchanged as the epidemic evolves. This is not the case with covid-19, where legal regulations and changes in behavior change the parameters of the models, which then have to be estimated day by day. In practice, these models can only predict developments with reasonable certainty in a short period of one to two weeks. From then on, it all depends on people’s behavior, atmospheric conditions, acquired immunity and many other unpredictable factors. This means that our ability to predict how this pandemic will develop in the medium and long term is very limited. There are arguments in support of very different positions on issues as simple as the future need for masks, the level of vaccination that leads to some normalcy, the level of infection that leads to the immunity of the group, that Future vacancies appear, etc. This diversity of opinion is normal in a situation of high uncertainty like the one we are in.
Does this mean that we shouldn’t trust scientists, epidemiologists, mathematicians, doctors, virologists, and economists who are trying to use models to predict the future development of this crisis? No, on the contrary, we must use all available instruments. However, we must avoid relying on monolithic opinions and not base our actions on a very small number of specialists who sometimes place an unjustified reliance on their inherently fallible models. In his book Wisdom of the Crowds, James Surowiecki argues that, in general, many minds are better than one, and that taking different opinions into account generally leads to better decisions.
For this reason, it makes perfect sense to create a scientific council of experts from different areas and with different visions, which has the task of analyzing the situation of covid-19 in a structured and systematic manner and working out a series of recommendations that are possible from used by the government to make decisions. The existence of a scientific council with the responsibility to define and develop documents and criteria to support the decision by consensus would be the best guarantee that we will respond appropriately to the high uncertainty situation that this pandemic continues to create in the EU will be near future.