The most fundamental truth about Covid-19 is that almost everything related to it is uncertain. Our knowledge of the virus and of the sickness that it causes remains very limited.
Unfortunately, our nature is to crave certainty. How long is this going to last? How many people are going to die? Where did this come from? What is the best way to manage the impact on our societies? When can we go back to work? How do I know if I have had Covid-19? If I had the virus am I immune from getting it again? We have no good answers.
The most important thing to keep in mind, in the midst of the numbers of what has happened and projections of the future, is to be modest and try to live with our ignorance. Many are groping around in the dark, convincing themselves that there is some way of understanding what is happening. We all feel an urgent need to know what the future will bring.
The many statements of “facts” are largely illusions. Numbers and models meant to make us feel better about the future are barely more than guesses. These remarks deal with all countries, including Bangladesh. Patience and courage are needed in the face of this uncertainty.
Do we know how many cases of Covid-19 infections there are?
We have very little idea in any country as to how many cases of Covid-19 there are. A person is infected by Covid-19 when this is revealed by a laboratory test. The number of such tests is likely to be far lesser than the number of cases; many persons are infected but show only very mild symptoms. Some scientists believe there are millions of cases, most of which are not recognized. Unless one does sample surveys of the population, administering a test that reveals if you have antibodies for Covid-19, there is no way to estimate how many persons have been infected.
There are now such sample studies of the population mostly in Europe. There are two conclusions that can be drawn from these studies. First, there were many more cases than the authorities had believed from the testing programs. Second the percentage of the population that had been infected was of the order of 5-10%. It is clear that we have little idea of how many people are infected. In every country, the case numbers have little meaning. We also know that the infection rates are far short of what is needed to establish herd immunity.
Do we know how many deaths are caused by Covid-19?
Again we do not have accurate estimates. We have the report of the attending doctor, if there is a doctor, as to the cause of death. But in most countries, such reports are notoriously unreliable. The only way to be sure of the cause of death is from an autopsy. In Bangladesh there are few such investigations, usually only in criminal cases. In the United States, there is a suspicion among disease experts that the deaths arising from the virus may be 50% to 150% higher than the “official” estimates. But in the absence of autopsies there is no way to know.
In the United States, autopsies carried out on persons that died in early February revealed deaths from Covid-19. This was much earlier than anyone expected and suggested that Covid-19 was already well established in the United States by December 2019 or January 2020. It adds to the mystery.
In Bangladesh, in 2019, there were approximately 900,000 deaths or 2,500 every day. It is often difficult for the attending physician to determine the cause of death and most people die outside the hospital, usually without a doctor. An increase of 100 deaths a day, spread around the country, would not be noticed. With the best will in the world, one cannot get an accurate count of the number of deaths arising from Covid-19 without widespread autopsies.
In Bangladesh, deaths are counted as due to Covid-19 if the person was tested positive and then died. In New York, they are virtually counting anyone who dies in the hospitals as a Covid-19 victim. Under such rules, accurate numbers are impossible to determine.
I emphasize that this problem is not peculiar to Bangladesh. All countries face the question of accurate assessments of virus-induced deaths. Only after the pandemic is over and medical researchers can carry out more extensive research do we have any chance of determining the actual numbers associated with Covid-19. For the Spanish Flu that raged around the world from 1918-1920, death numbers continue to be argued and researched.
How accurate are the tests being administered?
All tests of a condition such as “Do you have Covid-19?” produce a “yes” or “no” answer. Sometimes there is an “undetermined” outcome. When the test says “yes” it means that the answer is probably yes but not certainly yes. There will be a lot of “false positives” ie, you do not have the disease but the test says that you do.
There are also “false negatives” where the test says that you do not have the disease but in fact you do. If you want the test to identify most of the people that have the disease, you want a very low number of false negatives. But along with this low number of false negatives comes a much larger number of false positives. To develop a test that finds all the people with the disease will result in a lot of people who do not have the disease being told that they are infected.
Most of the tests being used for Covid-19 have had limited testing so the magnitude of false positives and false negatives is not yet established. All of this reinforces the difficulties in actually counting cases.
A recent test widely used in the United States is reported to have 15% false negatives, ie, about one in six of persons told they were free of Covid-19 actually were infected.
The lesson is that treatment of data must be very careful and one should draw conclusions too with reservations about the accuracy. Numbers often convey the idea of real knowledge even when there is little. Caution is necessary.
Forrest Cookson is an economist who has served as the first president of AmCham, and has been a consultant for Bangladesh Bureau of Statistics.
