Role of Data Science in Managing COVID-19

In managing the COVID-19 pandemic, it is important to generate and focus on appropriate data more relevant to the existing context (stage) of the pandemic in the country.

In the early phase of the pandemic where there were only clusters (not a community spread), one to one testing for contact tracing was a strategy to identify all possible asymptomatically infected individuals for the purpose of prevention of community spread. The idea was to quarantine and isolate all-infected people to delay the community spread as much as possible till the other effective measures such as satisfactory vaccination level is achieved. But only few countries have been successful to prevent the community spread until the desired full vaccination coverage is achieved. Even they are also facing a surge of cases with the new Delta variant which is highly transmissible even among the fully vaccinated people. 

A few other countries are still trying hard to prevent a community spread till the desired vaccination coverage is achieved in the near future. However, the daily reporting of COVID-19 positive cases would be an important epidemiological parameter when the community spread is not yet established in a country as it is useful data to understand the trend or progression towards the community spread. Because, then it is possible to implement required strategies timely to delay the said progression. However, it would be not possible to completely prevent the community spread indefinitely, when we are dealing with a pandemic caused by a respiratory virus even though it can be delayed for a small period.

"Daily reporting of COVID-19 positive cases would be an important epidemiological parameter when the community spread is not yet established in a country as it is useful data to understand the trend or progression towards the community spread."

But, when a country is in a stage of community spread and also in the verge of exceeding the surge capacity of health care systemto cater services for the COVID-19 related complications (moderate to severe diseases), the reporting of daily COVID-19 positive number would not be a priority as it would be technically impossible to have an accurate figure for that (instead, there are other parameters to be focussed in that stage). Underestimation of daily positive cases would be the result of such attempt. That is because; in the stage of community spread it is not possible and also not necessary to test all asymptomatic cases.

 Further, it might not be the time to allocate all our resources and efforts to detect each and every COVID-19 case in the community to count the exact number of cases in the countrydisregarding the fact that there is a community 
spread anyway. 

Therefore, now it is the time to focus more on the numbers of COVID related complications presented to the health care system and to make sure that numbersare within the limit of the surge capacity of the health care system. Therefore, it is useful to scientifically countthe numbers of Oxygen dependant cases and the numbers of the patients needing inward care (for clinical indications), high dependency care and intensive care. If the number of cases with such complications are exceeding (or close to exceeding) the maximum capacity of the health care system, itwill be the bettermarker to early identify the pending disaster andthe inadequacy of the currentcontrol measures to prevent the speed of the virus transmission in the community.

"When the surge capacity is exceeded, it is not only the COVID-19 care, but also the other essential medical care services would be compromised leading to increase in fatalities both in COVID-19 and non-COVID-19 diseases"

In the stage of community spread, it is important to understand in advance when it is very close to exceed the surge capacity of health care system to act proactively to reduce the speed of virus transmission. When the surge capacity is exceeded, it is not only the COVID-19 care, but also the other essential medical care services would be compromised leading to increase in fatalities both in COVID-19 and non-COVID-19 diseases. Therefore, additional vigorous measures should be promptly implemented in advance to prevent the collapse of the health care system as it would take a week or two to achieve the control due to the already established surge of infections waiting in the long incubation period. After achieving the control and managing to get the numbers of COVID-19 complications under the surge capacity, next step is to implement new normal life style to prevent such surge appearing again. This might be the needed ongoing practice till the desiredvaccination coverage is achieved. That is also only if we are fortunate enough not to discover any emergence of a resistant, more virulent variant of SARS CoV-2 escaping the vaccine induced immunity by such time in future.

Therefore, the main objective at this stage of the pandemic would be to filter the patients needing inward hospital care for clinical indications and monitor the trend of the number of complications real-time in relation to the surge capacity of the health care system. Other patients with mild disease and people who are exposed should be managed in a system of home care with a focus on both infection control measures and early identification of warning signs for hospital admission. 

"The main objective at this stage of the pandemic would be to filter the patients needing inward hospital care for clinical indications and monitor the trend of the number of complications real-time in relation to the surge capacity of the health care system"

COVID-19 death rate is also a very useful parameter to decide on the level of movement restrictions needed in a given time. But it is very important to correctly identify the deaths due to COVID-19 separately from the deaths associated with COVID-19 (incidental finding of SARS CoV-2 infection in diseased people who died due to other causes). In the stage of community spread, it is possible for some dead people to be incidentally infected with SARS CoV-2, but not having a causal relationship to the death of the person. If such deaths are counted as COVID-19 deaths, figures might be exaggerated unnecessarily affecting the rational policy decision making.
Another set of data that should be generated at this stage of pandemic is about reinfections and break through infections after vaccination. Such data should be collected categorically for mild disease and moderate/severe disease after past infection or vaccination. Further, there should be an on-going review about the disease outcome of the new variants in children, as well. Similarly, we are in a stage where we need to plan to generate local scientific data on specifically identified new areas for ongoing review of the evolving situation for future policy decision making and it is well beyond the concept of discussing and arguing about daily COVID-19 positive numbers at this point. 

The author is a Consultant Medical Virologist attached to the Medical Research Institute