The asymptomatic, or the difficulty of controlling the virus without testing

A number of important questions about the coronavirus remain unanswered. One of them is the percentage of asymptomatic infected people it generates. The answer could explain why the epidemic has grown so fast in countries such as Italy or Spain (where reliable estimates from Imperial College gave several million actual infected at the end of March).

From my analysis of this question here, the conclusion seems to be again that it is easier to control the epidemic with a continuous and abundant supply of reliable tests (and I am referring here both to PCRs, which are very reliable but need more time and specialised staff, which limits the capacity to perform them, and to the different modalities of the so-called rapid tests).

By asymptomatic infected we mean a person who

1. has been infected by the SARS-CoV-2 virus;
2. can in turn pass it on to others;
3. but has not developed symptoms of Covid-19 disease.

For the purpose of this post, we will slightly relax the above condition: the asymptomatic has either not developed symptoms of Covid-19 disease, or the symptoms developed are so mild that they are not suspected of being infected, or even take longer than usual to develop more obvious symptoms of Covid-19 (weeks instead of days).

Knowing the percentage of infected people who are asymptomatic is, by its very idiosyncrasies, a very complicated question. Since they have no symptoms, what we might call the visible part of the disease, they themselves (and others) are unaware that they are infected and the only way to detect them is with reliable tests. But we know that in quite a few of the Covid-19 affected countries there has been or still is a shortage of tests, and so they are mainly used to confirm the disease in people who already show clear symptoms of the disease. For this reason, if one looks in the scientific literature for percentages of asymptomatic people for Covid-19, one will find huge variations; here are just three examples of many more possible (borrowed from the CEBM of Oxford University https://www.cebm.net/covid-19/covid-19-what-proportion-are-asymptomatic/)

1. On the Diamond Princess cruise ship, the estimated percentage of asymptomatic people was 16-20%.
2. Iceland’s insularity and small population allowed large-scale testing of the population. As a result, it was estimated that 50% of those who tested positive had no symptoms.
3. The World Health Organization has gone so far as to suggest that asymptomatic people may account for as many as 80% of those infected.

Statistics can also lend a hand here and help to estimate the percentage of the asymptomatic infected. For example, it is quite possible that the government’s forthcoming survey of the state of the epidemic, using a random sampling of more than 60,000 people, will provide reliable estimates for the percentage of asymptomatic people in Spain. In addition to the modelling and statistical calculations, such a study will require tens of thousands of tests, most of them in the form of rapid tests for antibody uptake, although given their sensitivity, it may be necessary to supplement them with a few tens of thousands more tests of higher reliability.

If we consider the estimates of those actually infected in Spain, and the percentages mentioned above, we see that the epidemic, as it spread during the month of March, has very likely generated from several hundred thousand asymptomatic people in Spain to perhaps more than a million. That tells us that there has been a high chance that quite a few of those asymptomatic people were attending services where, even in the containment period, they had frequent contact with other people. Think of health care workers, caregivers in nursing homes, supermarket checkout clerks, or those who serve us in a bakery or pharmacy. In the entry Beware of the baker and of polling station officials…, we explained how the evolution of an epidemic is greatly accelerated when the infected include what we called “popular” people: people who are in frequent contact with many others because of their job. Here again, the importance of the use of masks, gloves and appropriate protective equipment (something we have indeed not had enough of in Spain) can be concluded.

If reliable tests are the only way to detect asymptomatic cases, we must conclude that a continuous supply of such tests is essential to control the growth of the epidemic. Just the opposite of the situation in Spain in March. Perhaps this may shed some light on the fact that the confinement on 15 March slowed but did not stop the development of the epidemic in Spain.

From what has been said, it can be inferred that in order to cut off the virus’s ability to infect, all those people who, even during the containment period, will necessarily have to establish multiple daily contacts, should be tested preventively and without the need for them to present symptoms beforehand.

Containing this first outbreak of the coronavirus, and preventing future outbreaks, is therefore a matter of ensuring a continuous and abundant supply of reliable tests. In the current circumstances, with the epidemic affecting more and more countries, consideration should be given to manufacturing sufficient tests here in Spain; not only to ensure self-supply, but also to help alleviate the shortage of tests in so many countries where they are already needed.