For prevention, we need to know how a disease is transmitted and simply let people know of it so that they refrain from traversing the avenues which pave the way for transmission. This is what we know more or less. Epidemiologists learn in their basic course that for the prevention of the transmission of a disease, one does not need to identify the causative organism although it is all the more better for various reasons. It is, in fact, sufficient to know what is the vehicle or vector of transmission, which could be biological, physical, mechanical or environmental, singly or in combination. In this regard, how cholera was controlled in London based on the findings of John Snow, the anaesthetist of Queen Mary, on the water distribution system in the city of London in 1854, is cited as a classical epidemiological example. The cholera vibrio itself was isolated 30 years later in 1884 by Robert Koch.
For prevention, we do not need to know how many people have been infected and who they are. For prevention, barriers are created in the avenues of transmission, when we know about them. This should halt transmission. Prevention should largely be population-based action. But to sensitise people to an effective and efficient behaviour- and practice-based action for prevention, we need to develop a supportive and conducive attitude and belief among people. There are specific time-tested techniques to change people’s mind first. Preventive measures may be better fitted if the nature of a disease-causing agent and the disease itself are understood well. This is especially important if the cause is a virus and more so, if it changes its physical and genetic formations and structures fast, which is a reality in case of an RNA virus.
What estimates should we need to measure the profile of COVID-19? First of all, what should be the aim of making COVID-19-relevant estimations? Then we need to decide what the bits of information are that we need to make estimations? The aim obviously should be the prevention and control of the infection, ie COVID-19. Based on the aim, some bits of baseline information should be in order. They may be on the expanse of the problem, for example, to fit responses, ie we plan and gather resources accordingly by type and volume — match for match.
There may be several other aims of making estimations. The expanse of a disease may be gauged by the number of the people who have been infected, as is obvious; this should include all people irrespective of the type and degree of signs and symptoms that they have been inflicted with or irrespective of any other characteristics. For the answer to the question ‘how many’, we need to test everyone, specific to the location in question, eg everyone in a village, in an upazila, in a ward or city or the whole country. Some location-wise answers may be quite challenging while some would be improbable for various reasons, eg risk, time constraint and limitation of fund. For COVID-19, we cannot get the estimate of ‘how many’ in the country because of these constraints.
The second question will be: who are the people that have been infected? This is, in fact, a mirror image of the question ‘how many’ as if we know ‘who’ we would know ‘how many’ by simply counting the who’s. We need the answer to the question of ‘who’ to control transmission effectively through, for example, the isolation of the people who have been infected. But as may be understood, an answer to this question is as improbable as ‘how many’ as this would also require that everyone in a given location should be tested. A closer strategy may be based on the knowledge of who is suspected to have been infected. This also is as formidable a question as the first two. Because if we do not know all the who’s that are infected, we would not know who to suspect as infected through contact with the people who have been infected.
In the given context, the next best alternative is to test as many people as possible, which for the sake of efficiency, should begin with those who are evidently ill from the disease and their contacts. While the ailing will be easy to identify, the identification of their contacts will be quite complex contextually. This analysis takes us to the realisation that it will be impossible to find out all those people who are the risk of transmission of the disease. This is also complex because many infected with COVID-19 will not show any clinical features or will exhibit mild features for any reckoning. This is a pointer to the fact that the route of finding out who are at potential danger of transmitting the infection is to test as many people as possible irrespective of whether they exhibit symptoms or not. But as those who show signs and symptoms are more potent in the transmission and the more serious the signs and the symptoms, the more is the risk of transmission, it is imperative that we should identify such people to begin with and if the situation permits we expand our net of identification as much as possible, without any magic number.
The contacts, to be clear, should mean the people who mingled with the identified cases in the neighbourhood and the people who the cases mixed with in any other places. Clearly, one cannot give any rule of thumb figure for estimation here because it would be so variable from case to case. Also those who have greater potential of mingling all around because of their potency to move around are the ones who are without any post-infection symptoms or complications.
Pragmatic alternative
ALTHOUGH the total number or the identity of the infected may not be a pragmatic consideration, the expanse of any disease may be estimated in the form of rates, which will tell us ‘how many’ but without identifying ‘who’ beyond those who have been examined. This will not help in preventing or controlling the disease very efficiently, but will underscore the effect of the preventive and the control measures undertaken. This will also help in identifying at least those who are considerably inflicted by the disease and will help at least to control the spread of the disease from them. For this an, epidemiological survey will have to be done through a random sampling process. Sampling process decides what will be the sample size for an analytic study or a survey. In a survey, the bigger the sample size, the better should be the accuracy of the estimate. Some other parameters also are adopted to ensure better accuracy, eg the limit of accuracy sought, the correction terms to be used for the weaknesses which are due to the sampling type and process. Statistically speaking, for 95 per cent accuracy, when the population size is more than 100,000 to take the sample from, and when the expected incidence or prevalence is taken to be 50 per cent in case it is unknown, the sample size would be 384, the highest for any location. If the locations are districts, the required sample size would be 24,576 from the 64 districts of the country. But this will not be based on the COVID-19 suspected cases alone, which however, would include them. The other information that this type of survey provides is what percentage is silently infected and what percentage is not infected.
The current survey gives us the result of the tests done on considerably sick COVID-19 patients. This gives a trend of the expanse of the disease, ie if the cases are increasing or decreasing. If the number of tests is increased more, cases will be falsely positive if there are fewer actual cases among the tested and the test positivity will fall as a result, as this will fetch more of those who might be suffering from other diseases or are not infected. So it might not give a true estimate of the ground reality. False positivity, in fact, increases when an incidence or attack rate or a prevalence rate of a disease is less than 30 per cent in the population; on which right now, we have no information because of the apathy of those who are at the helm.
AM Zakir Hussain is a former director, Primary Health Care and Disease Control, former director of IEDCR, DGHS, former regional adviser of SEARO, WHO, former staff consultant, Asian Development Bank, Bangladesh and working group member of Bangladesh health watch (BHW).
No comments:
Post a Comment