BENGALURU: Researchers from Indian Institute of Sciences (IISc) have developed an ensemble forecast with 972 scenarios (varying 7 control variables) for the emergence of third wave in the state.
The team was led by Professor Sashikumaar Ganesan, chairman of Department of Computational and Data Sciences along with Deepak Subramani and some PhD students and also Giridhar babu. In an exclusive interview with The New Indian Express, he explains the model as to how it forecasts social distancing, early emergence of new variants and vaccination are responsible for wave three. It also predicts that worst social distancing which is pre-COVID levels guarantees emergence of new waves even with two times vaccination rate.
Preparations are going on in full swing in Karnataka to mitigate the third wave. But there are a lot of uncertainties in the data. with this kind of a situation on hand. How was this mathematical model built for the third wave?
We are aware of it and like you rightly mentioned there are many uncertainties in terms of the covid spread and there are data that are missing in order to accurate model predictions and also fitting the parameters.
To account for all these things especially for the wave three forecast, what we have done is we've done an ensemble forecast. It means that we consider the several key parameters that are influencing the wave 3 and we vary all those parameters and try to build different scenarios.
Which are the parameters?
We have considered seven key parameters to build the model and we then varied it with different values and then we built 972 experiments or scenarios. We have accounted for all the uncertainties in the model.
Parameters are: Case to case infection ratio, the duration of antibody waning; antibody waning in 1/3rd,2/3rd, and full population; emergence of immune escape new viral variants (july, sept, nov); reinfection due to new variants (1/3rd, 2/3 and full); vaccination rate per day (2.8 lakhs, 4.2 l (+50%), 5.6 l (+100%) and finally Covid appropriate behaviour (good-lockdown like compliance, bad or relaxed situation; worse (like in precovid scenario)
Three key factor that we were interested in the study were 1. chances for wave 3, 2. date of peak and caseload (age wise at the peak).
Can you explain a few key parameters that were analysed?
Though there are seven key parameters amongst that we have observed three key parameters
Covid appropriate behaviour which is the most significant parameter and the second is the timing of the new variant and third one is the vaccination rate. Once we have all these three in control, if the state and its people handle it well then we may not even see a third wave.
Based on this model, when could we expect a third wave to hit Karnataka? And also when is it expected to see a peak?
What we have found is if very good social distancing is observed, follow a lockdown kind of scenario and if a new variant comes in Mid of August or end of august then the chances of increase in number of cases is 50:50.
If a new variant comes beyond sep, oct, nov months and if we follow the good social distance scenario like during lockdown we won't see a wave three at all. But this is very unlikely as we cannot expect a lockdown scenario presently.
So then we move on to the highly possible scenario which is the social distancing "bad" or relaxed SD.
It means that in this distribution of cases in case the vaccination rate is in the current phase the model predicts many cases. If a new variant comes in August itself interestingly, if vaccination is ramped up then what we are seeing is the maximum caseload is coming down.
Meanwhile, if there is any new variant seen by sep-oct we will have to bring back all covid appropriate behaviour and restrictions. But the good thing is if there is an increase in vaccination then there will be 10 times reduction in active case load. Otherwise if we continue with present scenario then it case load can go to closer to 10,000 to 7 to 8 lakhs also.
Date of peak then would be near December and go on till March 2022. Starting from November to next March there could be a cluster of cases that is what scenario is showing. If new variants are coming beyond November then u can say that there is cluster of no wave scenarios that means several cases are predicting that there won't be any new wave.
So we have to be watchful for the next three to six months as they are the key periods.
If we avoid the emergence of new variants till november then I don't really see a peak in Karnataka. Before that if variant emerges then definitely we will see a peak immediately five to six weeks after emergence of new variant.
What would be the case load and severity of this wave at it's peak?
In the worst case scenario if a new variant emerges in Aug then we could see 4.51 lakh cases in August and we may see about 20 lakh cases between 22 August to 10 September. But if the vaccination is at the same pace and we have a new variant then we could see with relaxed SD around 1.73 lakh cases between 1.44 lakh to 2.07 lakh cases.
However, if the social distance is good then it will be only around 36,000 cases by 15 oct and about 36 to 48k cases till the end of October. Also if a new variant comes in September or in November then in a good scenario no wave at all.
Is there any particular population which will be most affected?
We analysed the population by dividing into five subgroups of age. 0-11, 12-18 and 19-45, 46-60, 60plus. We have used the vaccination phases module and divided them accordingly. If new variant comes in August or end of August.. probability of wave 3 is 0.78.
Active COVID cases during peak 2 and wave 3 when you compare it would be five times more but if new variant comes in November age group that will be vulnerable is children. They will be more affected almost- 12.63 times more extra than during the second wave.
Physical classes is set to begin for class IX to XII from August 23. Is there any suggestion to the government about this, based on the prediction from the mathematical model?
This issue was discussed In the TAC meeting where I have suggested that social distancing rules in school has to be prioritised. Only 50 percent of children and staff should be allowed to come, CAB have to be monitored. Also very importantly if there is an outbreak in school then a mechanism to contact trace childrens’ primary and secondary contact has to be traced . Only after all this schools should be reopened
Can you summarise the key observations?
If there is a delayed variant then the impact will be less, the probability for new wave is less. second one is CAB, it is the general public and government's responsibility to monitor CAB.
Key elements from the study
New immune-escape variant was responsible for Wave-2 [diagnostic analysis]
Emergence of new variants & poor social distancing is most likely to be the main reasons for Wave-3
Emergence of new variants beyond Sep’21 will not induce new wave when the social distancing is good (lock-down like scenario)
Increase in vaccination rate reduces the peak active caseload in Wave-3 § consequently, hospitalizations, ICU and Oxygen requirements decrease
Vaccination rate should be doubled to reduce the probability of Wave-3
Wave-3 impact on children (Age 0 -17 yrs) could be up to 13 times (mean) more than Wave 2