Covid-19 spread in some unique ways in India, new contact tracing data shows5 min read . Updated: 01 Oct 2020, 09:29 AM IST
The world’s largest dataset of cases and contacts shows that superspreading drives the epidemic; shared long travel most risky
Over 70% of those who contract covid-19 do not pass it any further, while a minority act as superspreaders, the first detailed study of SARS-CoV2 transmission patterns in India has shown. The study, which used the largest dataset of contacts of any global study, also shows that children play a bigger role in spreading the virus than earlier believed, and that sustained shared travel is the most high-risk behaviour one could indulge in.
Researchers led by Ramanan Laxminarayan, director of the Washington-based Centre for Disease Dynamics, Economics and Policy, were given access to the full contact tracing dataset of Tamil Nadu and Andhra Pradesh states from the beginning of the pandemic, until 1 August. In all, the two states had seen over 435,000 cases by 1 August, and had reached out to over three million known exposed contacts. However, as in the rest of the country, information was incomplete for the majority of cases and contacts. The researchers had complete epidemiological information and laboratory results for 575,071 tested contacts of 84,965 confirmed cases. Published in Science magazine on 30 September, the study offers five key new insights into how the pandemic is playing out in India.
1. Superspreading defines the disease
The researchers found that positive cases among contacts were concentrated among a minority of cases. Over 70% of index cases - ‘original’ or primary cases found through testing - infected no further people, while fewer than 10% of index cases were responsible for nearly 60% of further infections.
Superspreading as an event is a common misconception, Laxminarayan said. “Superspreading refers to the fact that some proportion of people transmit much more than other people. It doesn’t require an event."
The study confirms other literature so far in superspreading, said Gagandeep Kang, one of India’s foremost viral infectious disease scientists. “Superspreading is not about the virus but about the circumstances that a person finds themselves in," said Kang. “It could be that the person is earlier in their illness and so is shedding more of the virus, it could be that it is a person who indulges in greater social mixing, or someone who has greater opportunity to meet a large number of people, who he or she then infects."
The Tamil Nadu government has been using this insight from their data to focus on isolating positive cases within the first 24-48 hours of their discovery, said B Chandra Mohan, an IAS officer in the state and a member of the state’s epidemic monitoring committee. “From our data, we realized quite early on that if we could identify and isolate all cases within the first 24 hours, we could reduce transmission by 70%," said Mohan, who is a co-author on the paper.
2. Long shared transport carries the greatest risk
The study categorizes contacts into high and low risk. Those who shared a household with the index case, had contact at less than one metre distance without protective equipment, and travelled in the same shared mode of transport within three rows of the index case were classified as high risk contacts. Over 10% of such contacts of index cases tested positive, meaning that they had a one in ten chance of being infected. On the other hand, people who shared the same space with an index case without meeting these high risk contact criteria were categorized as low risk. Under 5% of them tested positive.
By far the most high risk interaction was being in close proximity to an index case in shared transport for six hours or longer - nearly eight out of ten contacts in such cases tested positive. From a policy perspective, it would be important to know whether mask-wearing could reduce the risk on shared transport, Kang said, but this data was not yet available. Living in a shared household carried the next highest risk. But the risk of infection was much lower than in the case of co-travellers. Less than 10% of contacts in shared households tested positive, the study showed.
3. The underestimated role of children in transmission
The age group most likely to result in spreading infections is the 20-44 age group, the study found. Children also play a significant role in the spread of infections, the study shows.
“We found that children transmit to each other, and to the elderly," said Laxminarayan. This finding is particularly of concern since most infections in children are mild, and are likely to be undiagnosed.
“This is consistent with what we know about influenza, where children are among the primary drivers of the infections; it should not be surprising to see this in a respiratory infection," said Kang.
The study also found that the risk of transmission was highest among pairs of contacts of the same age. This was particularly true among children. Of all same-age pairs the likelihood of transmission was among the highest between pairs of children under the age of 14.
While this could have immediate implications for school closures, Kang recommended seeing the finding in the context of how most Indian schools function. “Most schools are much better ventilated than other group spaces, and now in the colder months, taking classes outdoors could be considered," she said.
4. India’s elderly have surprisingly low mortality
The two states’ covid cases were younger than expected, even given the lower median age in India as compared to western countries, the study found. Incidence of cases peaks in the 40-49 years age group in the US, and then declines, only to rise again for the elderly, while in Tamil Nadu and Andhra Pradesh, incidence peaks at an earlier age, and then drops off continuously.
While mortality was higher, as expected, among older age groups in the two states, it plateaued at around 75 years, and there was no further elevated risk of mortality among those older, unlike in the US. This was unexpected, experts said.
This could be on account of a ‘survival effect’, Kang said, where the people who attain very old age are those in relatively better health and without comorbidities. As a result, they might have lower covid mortality than expected.
5. Deaths are occurring worryingly fast
In half of the cases involving a death in which the person tested positive before dying, just six or fewer days elapsed between testing and the person’s death. The median time to death in the two states was far lower than the US, where it was 13 days from the date of hospital admission. In China, the same figure ranged between two to eight weeks from the onset of symptoms, according to estimates by the World Health Organization.
The short time to death is an important finding which highlights the lack of access to care in India, Kang said. “We tend to wait too long to get care, and we have seen this before in other diseases. In the field of diarrhoeal disease where most of my work has been, we see that children getting to government hospitals appear to have waited far longer to access care than in private hospitals, which could mean that poorer people face greater difficulty in access to early care," she said.
Rukmini S. is a Chennai-based journalist.