Lower Covid-19 lethality raises doubts about isolation, says epidemiologist

There are still important unanswered questions about Covid-19, but we’ve made enough progress to avoid measures of social restriction as much as possible, says John Ioannidis, professor of epidemiology at Stanford University (USA).

He has just published a paper that calculates a lethality rate due to coronavirus infection much lower than the approximately 4% previously predicted. On average, 0.23% of those who contract Sars-Cov-2 die, according to Ioannidis calculations, with great variation in relation to age: for those under 70, the lethality is 0.05% of those who are infected .

If the potential number of deaths is less, skepticism about drastic restrictions increases, he says, “because there is less to be gained, compared to the social, economic and health costs of these measures”.

Specializing in the epidemiology of infectious diseases and a global reference on how research practices, designs and methods can optimize the quality, reliability and usefulness of scientific information, he has devoted much of his time to Covid-19 projects.

“It was the perfect combination of interests: a major emerging issue, requiring the best possible data to respond to it.”

From Athens, where his mother is visiting, Ioannidis spoke by phone with the report about what remains to be researched in this pandemic and what lessons she leaves for the future.

*

QUESTION – Since the first months of the pandemic, mr. warned that decisions were being made without reliable data. Could governments have acted differently?

JOHN IOANNIDIS – Without information to evaluate the best option, it is necessary to predict the most pessimistic scenario. The data we had on China, and some from Italy, looked very disheartening. We saw a very aggressive virus, spreading very quickly and with a very high fatality rate due to infection, because we saw few asymptomatic people.
Since the risk of dying seemed on average very high, it was fair to make draconian decisions, such as aggressive lockdowns. But it was also essential to get reliable information quickly, because feedlots have tremendous consequences for health, the economy and society.

Q – For which questions do we already have satisfactory answers?

JI – We have a good notion that this is a virus that spreads quickly and infects a very large segment of the population. And now we have better data on fatality. For people on average, the fatality rate for infection is very low compared to what was previously believed, 0.23%, with tremendous variability.

The difference in lethality between a child and someone aged 90 is about a thousand times. We know it is a complete disaster if someone gets infected in an asylum. Lethality can reach 25%, one in four infected die.

PO that remains to be answered?

JI – We still don’t know what proportion of people need to be infected in order for the pandemic to subside.

We saw that the infection rate can reach 55% in slums in Mumbai (India) and Argentina and probably in densely populated areas in Brazil. This comes very close to the theoretical calculation of 60% for the so-called herd immunity.

In some non-densely populated places, however, the epidemic appears to be subsiding in much smaller percentages. In Sweden the prevalence is nowhere near 60%, but there has been no strong increase in cases, and models suggest that immunity can happen in much lower percentages because people are not meeting anyone, they are not mixing with each other.

One issue that already has data, but it would be necessary to deepen it, is that of pre-existing conditions that affect immunity, such as exposure to other coronaviruses. There are different percentages of pre-existing immunity in different countries, and this should also contribute to the need for such a high number of collective immunity. But we’re still not sure.

Another frontier is that of restrictive measures: closing workplaces, closing schools, working from home. Some are more disruptive than others, and are being used in different combinations, being adopted based on different indicators – number of cases, hospitalizations, other aspects.

It is a priority to evaluate the key measures we are offering as solutions.

Q – Is the concern about the negative impact of these restrictions?

JI – We have to be honest and say that we don’t know if they can make a difference in a positive direction, as this pandemic probably won’t be gone in the next week.

I have no reason to test hygiene measures, wash my hands, avoid congested places, keep my distance, wear masks if I can’t stay away. We already know that it all works. But confinements, closing schools and companies, this is something that needs to be studied much better, compared to what we know today.

Q – Without data, what should be taken into account when making decisions?

JI – Uncertainty. We know a lot, a lot more than we did when the pandemic started. And much of what we know today is more optimistic, compared to what we thought it would be before. But it is still a serious problem. We must admit that we do not know if it is necessary to close everything, or for how long, and do studies to understand if the balance of this is positive.

Q – Is there already research on the impact of restrictions?

JI – There are some studies. In Norway, there was no significant difference in terms of the number of infected people between reopening gyms or keeping them closed. Schools have happily opened in many places, but there is research to understand how to keep them open and prevent further closings.

This requires a different mindset than people. Instead of waiting for the government or the scientists to have an answer for everything, we should be ready to accept that this is what we know and this is what we don’t know and want to know. For that, it is necessary to test.

P – Your work shows less lethality from infection than previously estimated. Is this another argument against drastic measures?

JI – A lower lethality rate is good news, but it also creates more skepticism about aggressive restrictions. There is less to gain.

Isn’t it necessary to look at absolute numbers as well? If transmission is very fast, can’t there be many deaths, even with a low lethality rate?

Yes, and that number of how many people will be infected until the epidemic loses strength is one of the information about which we still have uncertainty. Until we know more about that, not too disruptive measures are important.

There are other variables. We don’t know when there will be a viable vaccine, but the reasoning about what measures to change will be different if and when it is available. The same is true with better treatments for Covid-19.

Q – What does this pandemic teach scientists and decision makers?

JI – The main lesson is that we have to be prepared. We need more investment in health systems, especially in primary care, because this is a disease that can really be treated in primary care.

Tests are also important. Countries that were aggressive in early testing, such as Taiwan, Singapore and Iceland, smothered the epidemic before it became a big wave.

It is not possible to make specific plans, as we do not know when or if a new pandemic will come, what type of pathogen it will be, what is the lethality, the capacity for infection, the risk. But if we have the health system in place and also a plan to collect the information we know to be the most important right from the start, the information we talked about, we can avoid mistakes.

X-RAY

John Ioannidis, 55, is a professor of medicine, epidemiology and statistics at Stanford University and co-director of Metrics (research innovation center), at the same institution. Born in New York, grew up in Greece, graduated and doctorate in medicine at the National University of Athens and residency at Harvard Medical School hospital