By Anjan Roy
Humanity has faced outbreaks of many, many pandemics over historic times. Bubonic plague several centuries back to Spanish influenza just a hundred years back had wiped out large sections of human population.
But many of their lessons have not been learnt and those have been lost. This time it is different. We now have one tool which was not available earlier. It is not just medical aid, but it is statistics that can make a significant difference this time.
This viral infection – COVID-19 – has no specific medicine till date. Sometimes use of medicines can even cause harm.
However, in fighting the pandemic that coronavirus has become, data science will in all likelihood be the most useful tool. The common expression now is — flattening the curve. As we are witnessing, the doctors are now looking closely at the progression rate of the infection among population to predict the course of the spread.
These are data-driven and their close analysis will hold clue to determining what state we are in the spread of the disease and what protection or action agenda is best suited for the time being.
Throughout the world, the process of data studies are being stepped up and medical data analysis is taking up an important space. Thus, for example, the spread of the epidemic and their fatality are sought to be understood through data analysis.
The most stark comparison is the testing of patients. In South Korea, testing of suspect cases is being put at 3,692 tests per million people, while in Italy it is as low as 826 per million. Higher testing per population group captures more infected people in the initial stage thus limiting spread of the disease.
On the other hand, a connotation of that exercise is possibly the much lower fatality rate in South Korea compared with Italy. In South Korea there were only 66 fatalities, whereas in Italy the deaths have crossed a thousand-mark.
But then the demographic data is being overlaid on these primary figures to also pinpoint and explain the differences between South Korean and Italian results. Apparently, South Korea has many more young people than Italy and the disease fatality goes up with age factor. Thus, South Korean population is less vulnerable to the disease than Italian. Another factor that has been put out is the gender difference in the extent of severity. That is, males are more prone to the disease turning severe than women. These are all trends witnessed from the data and sure these will be further related or cross verified with various control data.
In the USA the process of testing is turning out to be even lower as reportedly there are fewer test kits available or facilities are limited as even sick people have been turned back from doctors chambers.
On the basis of large numbers, a formula has also been developed which is predicting deaths and spread of the disease. Dr K.K. Agarwal, past president of Indian medical Association and currently president of the Confederation of Medical Associations of Asia and Oceania, has put out various results, including estimated death rates, on the basis of the averages.
Dr Agarwal has given the following prediction of possible peaks of Indian cases:
1. 1st peak on 4th 22 cases
2. Expected second peak, wave, doubling time 7 days, 10th 15 cases
3. Expected third wave 17th 23 cases
4. Expected wave on 24th 20 cases
5. Small waves on 19th, 20, 21st with 10-15 cases each
(Dates are all for the month of March)
Whether it comes true to the dot or broadly matches, these efforts show how the pandemic and large scale incidence of disease is leading to use of data generated for the trend of the spread as well as some precautionary measures.
As far as testing is concerned, India appears to have given up the strategy of large scale testing as the dimensions of the exercise, given the massive population of the country, appear impractical, though its consequences could be equally fatal like Italy.
The Indian Council of Medical Research (ICMR) is currently engaged in a nationwide survey of reported cases and their origins to determine what stage currently Indian spread is. The surveys should indicate whether there have been community based infections or otherwise.
That is, the survey should reveal whether Indian infection cases have been picked up from vulnerable spots abroad by tracing in every case the travel history and contact points or there have been some cases where the infected person did not have any travel history but infection picked from domestic sources.
If there are significant number of the latter category then it is a sign of caution and indicator of community based infection. This could prompt a radical shift in the approach to tackling the spread than in the earlier scenario. If there are significant differences among the different regions of the country, then deferent approaches would have to be adopted for tackling the disease in different areas.
There are reportedly some modelling excises on for bringing out underlying factors which can be useful for formulating suitable strategies and approaches, which could give optimal results in fighting the disease. These will keep out much longer after the pandemic is over and gone. That should be useful in preventing such mishaps at a much earlier stage.