Models point to India's COVID-19 cases peaking in May, but it depends on policy and people

As a weary India watches the number of daily Covid-19 cases climb up, seemingly inexorably, and cross 2 lakh, the question on everyone’s mind is: how long will the current surge last? One answer could come from mathematical models, used to forecast the progression of epidemics, by epidemiologists and policymakers. In the early days of the pandemic in India, models which relied on global data went far off the mark, predicting that the country might see 200 million to 300 million cases. The actual number so far has turned out to be significantly less, just over 15 million active case, but with more data from India, more accurate forecasting has also been possible though even these come with several caveats.

According to one model developed by a team at the Indian Institute of Science in Bengaluru, and in a scenario where one were to assume a 10x sero prevalence factor (for each detected case, there are nine undetected), India is likely to hit its peak around mid-April, following which cases will decline steadily, touching about 569,000 active cases on June 1. This would be the lower limit of progression, considering the model, using this scenario, predicted 1.1 million active cases on April 11 and the actual number was 1.2 million, says professor Sashikumaar Ganesan, chairman, department of computational and data sciences at IISc and one of the team leads. Here the model assumes that the spread is similar to last March-October and that 3 million are vaccinated every day.

The upper limit, or the worst-case scenario, where the sero prevalence factor is 2x (for every reported case, there is only one unreported case, so more people are susceptible), a first peak is predicted around April 23, when India will see close to 2 million active cases, followed by a slowing down and then another rise to 2 million active cases from the end of May, with a decline only by June end. “I don’t see this happening unless the government is very lax and people are not following any kind of Covid-appropriate behaviour,” says Ganesan. The actual trajectory is likely to be somewhere between the two scenarios, he says.

Dr Bhramar Mukherjee, chair of biostatistics at the University of Michigan’s School of Public Health, says some of their models predict 5 lakh cases, 25,000 hospitalisations and 3,000-4,000 deaths every day in four weeks (mortality was 1,026 on April 13, using a seven-day average), with Maharashtra peaking in early May and India in mid- to late-May. “The rise has been sharp, so the fall is also expected to be sharp, as witnessed in many other countries. However, we have been proven wrong many times and so please take this as things stand right now,” she said in an email. She cautions that it is hard to predict the peak or generate reliable long-term forecasts as human behaviour and policies influence the trajectory.

Some of our models predict about 5 lakh cases, 25,000 hospitalisations and 3,000-4,000 deaths every day in four weeks. Some of these models show Maharashtra peaking in early May, and India mid-late May. The rise has been sharp, so the fall is also expected to be sharp. However, we have been proven wrong many times and so please take this as things stand right now

-Bhramar Mukherjee, Professor of Epidemiology and Global Public Health, University of Michigan

Dr Murad Banaji, a senior lecturer in mathematics at Middlesex University London, who has been tracking India’s Covid-19 epidemic, pegs the best-case scenario at a peak in the next three weeks. “It is too early to be certain, but there are hopeful signs that cases could soon peak in Punjab and Maharashtra which surged early. If this happens, and other states follow suit, then we might see daily cases start to fall by early May,” he says. This too comes with a caveat: it depends on the measures to slow the spread. “When we see huge religious gatherings and election rallies, it’s hard to be optimistic about this,” he adds.

What is of increasing concern to those tracking the spread of the SARS-CoV-2 virus in India this time is, as Mukherjee puts, its velocity and ferocity. “Surges overwhelm the healthcare system, causing more deaths, compared with slow and steady growth in infections.” The reports and visuals from cities like Surat and Lucknow — where patients are scrambling for beds and ambulances are waiting in queues — will attest to this. “District wise data — where it is available — shows that almost no district in India is being spared. Given this pattern, and reports that whole families are testing positive, it is natural to worry that we may be seeing more transmissible variants in circulation,” adds Banaji.

The best-case scenario is that infections peak in the next three weeks. But what makes this very optimistic is that it relies on effective measures to slow the spread across the country. When we see huge religious gatherings and election rallies, it’s hard to be optimistic about this

-Murad Banaji, senior lecturer, Middlesex University London

Epidemiologist Dr Jayaprakash Muliyil says the duration of the surge depends on the susceptibility of individuals — the nature of the virus is such that as long as it can infect subjects, it will do so. “I get the feeling we expect the threshold to be about 60% for urban and 40% for rural. It looks bleak when you look at the graph but each person who is infected, if he escapes death, is immunised,” he says.

If one were to look at how the pandemic is spreading state-wise, Uttar Pradesh, Bihar and Assam are among the large states where the total active cases are growing the fastest, says Rijo M John, health economist and adjunct professor, Rajagiri College of Social Sciences in Kerala. “These states are growing at 20% or higher on a daily basis. That’s alarming because those states also have a relatively weak healthcare system.”

It will be a tough situation at least for some time to come. We will have to really ensure transmission rates come down and we can’t depend on vaccines for that. We will have to make sure large gatherings are totally taboo, and that people are wearing masks properly

-K Srinath Reddy, Public Health Foundation of India

Even as models throw up different scenarios of cases peaking, there is consensus that the emphasis has to be on immediately reducing transmission rates. “We can’t depend on vaccines for that. Right now, we have to make sure large gatherings are totally taboo, and that people are wearing masks properly everywhere. If we do that, we are at least likely to cut down transmission rates,” says Dr K Srinath Reddy, president, Public Health Foundation of India.

The key strategy now should be to reduce hospitalisations and deaths, says Dr Prabhat Jha, chair in global health and epidemiology, University of Toronto. “And the best way to do so will have to be more strict masking/distancing measures, expanding testing and vastly improving data. Without data, there is no way for India to walk out of the pandemic. And secrecy in data can only backfire.”