07 May,2021 04:43 PM IST | Mumbai | Nascimento Pinto
Dr Sitabhra Sinha, a professor in the physics group of the Institute of Mathematical Sciences (IMSc). Photo: Sitabhra Sinha
While the rate of spread has seemingly slowed in a few states including Maharashtra, the pandemic is still galloping in others. Amid the second Covid-19 surge, vaccine shortages, and talk of a third wave, what answers do the numbers have for how Covid-19 will behave in India?
Chennai-based researcher Dr Sitabhra Sinha, a professor in the physics group of the Institute of Mathematical Sciences (IMSc), specialises in statistical analysis of epidemiological data - modelling the spread of infectious diseases through contact networks in a population. His team has tracked Covid-19 for over a year now.
Mid-day.com spoke to Sinha to understand the ongoing surge, the missing links in data, the role of vaccinations, and his projections for the pandemic.
Here are the edited excerpts from the interview:
What is the reason for the second surge in India? When is it likely to slow and is there a likelihood of another surge?
The conditions conducive to another outbreak were existing for quite a while. Very few individuals were adhering to any of the norms put in place - like wearing masks, not crowding, practising basic hygiene, not spitting in the open - even though the pathogen had not gone away. Possibly the fact that India fared relatively better than other countries during the first wave (which I partly attribute to the early imposition of a national lockdown -- despite its quite problematic implementation which resulted in high social cost), gave rise to a public perception that Indians are immune to the pathogen.
I would flag this complacency as the key factor that led to people not adhering to basic safety norms which could have prevented the second wave. So, the second wave as such is not surprising.It was a disaster waiting to happen. As to the actual causes that led to the outbreak happening in February, that of course is a stochastic phenomenon - depending on several chance events coming together. What is interesting is that unlike the first wave which broke in different parts of India at different times, the second wave was much more coordinated across India. The earliest was Punjab which showed the rise on February 9, followed by Maharashtra on February 12 and Madhya Pradesh on February 14. But after that, the other states started showing the rise within a few days so that by early March most states had gotten into the second wave.
As of April 24, there is a slight slowing down in the growth of active cases as characterised by R, the effective reproduction number (the average number of people infected by a single infected person, which measures the rate at which the disease is spreading). It has decreased from 1.31 (estimated for the period between March 9 to April 21) to 1.16 for the period April 24 to April 30. This is possibly because Maharashtra has from April 24 actually shown a decrease in the number of active cases (so that R is less than 1). Chhattisgarh has shown such a decreasing number of active cases even earlier, from April 18. But what is really interesting is that I notice that several other states also show a decrease in their R - although as the values are still greater than 1, the number of active cases will continue to increase. However, the states having the highest R at the moment haven't shown this decrease - so we still don't know whether this is the early sign of a flattening. It's too early to say for sure that the wave is dissipating.
Yes, a third wave is a distinct possibility if we cannot convince people to conform to basic norms such as wearing masks.
Your team observed a decline in the transmission rate in Maharashtra. What can we make of this?
For Maharashtra, R has gone down from 1.13 (estimated for the period between April 7 - April 20) to 0.96 in the week starting April 24.
If R is less than 1, it means each infected individual is on average being replaced by less than 1 person, which means that the number of active cases is decreasing. Thus, the condition for an epidemic (the number of active cases keeps increasing) no longer holds in Maharashtra. The same is true for Chhattisgarh which has had its R value of 1.56 estimated between March 10 - April 16 decrease to 0.93 from April 18 onwards. Of course, these values are still close to 1 and liable to be reversed so that the number of active cases might again start increasing - the next weeks will tell us if this is a sustained trend.
The measurement of R is done by statistically fitting exponentially growing or declining curve to the time-series of the number of active cases.
Could the current decline in the number of cases in Maharashtra be linked to lockdown measures?
Typically, any change in R we see for Covid-19 is related to events happening 10 -14 days before that date. So, I would say if a particular event caused the decline in R in Maharashtra it occurred around April 10 - April 14.
How do vaccinations help in Covid containment, and have you noted any changes in the transmission since vaccinations began here?
In general, vaccinations are supposed to help contain an epidemic by reducing the number of susceptible individuals. If a sufficiently high fraction of the population is vaccinated, it also provides protection to unvaccinated individuals, as the pathogen will not be able to find such individuals who are "hidden" in a crowd of vaccinated individuals.
The critical fraction of the population that needs to be vaccinated so as to provide herd immunity is related to R0 (R-nought), the basic reproduction number which measures how quickly the disease spreads in a completely susceptible population. For example, for an epidemic with R0 of 2, at least 50 per cent of the population needs to be vaccinated to ensure that the epidemic doesn't occur. I had estimated R0 for India to be close to 2 (1.83 in fact).
However, Covid-19 is quite complicated as even vaccinated individuals can get infected. This means that the protection is only partial and a much larger fraction of the population - possibly even 100 per cent - needs to be vaccinated to ensure there are no epidemics. As of now, the number of people vaccinated is a few percent of the population of India, and this will not make any practical difference to the rate at which the disease will spread. Unsurprisingly, the data shows no effect as yet of the vaccination.
What do we know about how damaging the variants of the virus are?
We do not have data as to how many infections were caused by which variants, so unfortunately, it's impossible to say what role the variants are playing. It's complete speculation at this point with the data that we have.
What has been India's biggest challenge in dealing with Covid-19 in the last one year? What measures do you think India should be taking to prevent a worse wave in the future?
The challenge is balancing the social and economic needs of the population with the use of drastic measures to contain the epidemic such as lockdown. I personally think that if people could have been convinced to adopt the much easier alternative of wearing masks in public (and wearing it properly, not under the nose or hanging from the chin) and not congregating in high densities for various reasons (religious or political), we would not have needed to go to the extremely drastic option of lockdown which asymmetrically affects the economically weaker sections of the population.
I think it should be clear by now that the most effective way to tackle Covid-19 is simple non-pharmaceutical interventions such as convincing people to always wear masks in public (in Japan, for example, many individuals wear masks in public for hygiene reasons - this was true even in pre-Covid times - so it's just a matter of making it a social norm such that it becomes second nature to people), ensuring that dense gatherings of people cannot happen, and promoting basic hygiene like not spitting in public places, not touching your face with unclean hands, etc. Vaccination is of course also important, but even after vaccination, people need to continue adhering to the above safety norms as the vaccine is not 100 per cent guarantee that someone will not get infected.
Apart from Maharashtra, which are the other states showing an improvement and are there any states showing worrisome trends?
As I mentioned Chhattisgarh is another state which has had its R go less than 1. Among states which still have R>1 but for which R has decreased recently are:
Kerala (R decreased from 1.61 between April 13 - April 21 to 1.35 between April 24 - April 30)
Andhra Pradesh (R decreased from 1.52 between March 11 - April 21 to 1.33 between April 24 - April 30)
Tamil Nadu (R decreased from 1.40 between March 11 - April 21 to 1.09 between April 25 - April 30)
Bihar (R decreased from 1.97 between March 29 - April 16 and 1.62 between April 16 - April 22 to 1.21 between April 24 - April 30)
Delhi (R decreased from 1.60 between March 29 - April 20 to 1.10 between April 25 - April 28)
Telangana (R decreased from 1.45 between April 10 - April 21 to 1.27 between April 24 - April 29)
Jharkhand (R decreased from 1.50 between April 10 - April 21 to 1.21 between April 23 - April 30)
Odisha (R decreased from 1.71 between March 31 - April 21 to 1.31 between April 23 - April 30)
It has been more than a year since the Covid outbreak. How stressful has tracking its spread in India been for your team and what are your learnings?
It's been a humbling experience as we learnt that despite a lot of advances in epidemiological modeling techniques in recent decades the Covid-19 pandemic still behaved in many ways that surprised us. It has taught us that we are yet to know a lot about how epidemics evolve over time and space.
Is enough data being made available to data scientists to study the virus? Would you like more numbers to be released by the government to help with future study?
I use covid19india.org for obtaining data for my active cases. I would really like to thank the team behind this website as they have been a god-send for scientists doing data analysis and modeling of Covid-19. Of course, we would love to have more data - but without covid19india.org and other similar websites we wouldn't even have had that. As I said, there are many questions - like the role of variants - that we simply can't answer without the relevant data. For this, we need to identify the pathogen variant for each active case - which is pretty difficult to expect when hospitals are just getting overwhelmed trying to treat patients. As a government public health official once told me (during the 2009 Influenza H1N1 A pandemic), do you want the doctors to treat patients or collect your data?
