'Missing’ COVID Deaths Raise Questions About India’s Tracking of the Outbreak

Jayaram Venkatesan, an anti-corruption activist in the southern Indian city of Chennai, Tamil Nadu, recalled hearing rumours of a “big surge” of coronavirus numbers around May 25 in his city. “We started hearing from hospital insiders about the Covid-19 deaths, and how they’re not being counted,” he told VICE. But they didn’t have any evidence.

Then one day, he was sought out by three families who presented death slips of their COVID-19 positive kins, who allegedly died of the virus at government hospitals – the Stanley Medical College and Hospital and Rajah Muthiah Medical College and Hospital. One was a 50-year-old man who died on May 28, the second was a 69-year-old man who passed away on May 31, and the third was an 87-year-old man, who died on June 4. “They all died, but none of them turned up in the hospital’s death records,” he said.

Early this week, more evidence surfaced appearing to support a similar narrative around “missing” COVID-19 deaths in Chennai. On June 10, local news reports revealed that the actual number of deaths in Chennai’s death registry — a city of 4.6 million people — was double the number that the state government’s directorate of public health (DHP) had announced.

At least 460 people died from March until June 8, but the DHP announced only 224. In a single disclosure, Chennai had revealed a hidden number that was just six people short of the total COVID-19 death toll of the whole country of Norway (as on June 12) had not been part of its records.

Even as the report asked the state government to probe the missing numbers and reconcile data, activists and experts warn the issue may be far bigger. “This is just the tip of the iceberg,” said Venkatesan. Giridhar R Babu, the head of Lifecourse Epidemiology at Public Health Foundation of India, a leading Delhi-based non-profit, told VICE that it's difficult to hide such a big number of deaths. "If this is happening in select places, it needs further investigation," he said. "A scientifically conducted verbal autopsy is needed to determine the reasons."

Indian media previously raised questions around relatively low COVID-19 death data. As of June 7, India’s case fatality ratio (CFR) — the number of deaths per 100 cases — was just at 2.8 percent. The global CFR is 5.8 percent.

In one instance, public health officials stated that India had a low fatality rate in comparison to other countries “which is very good,” said Balgram Bhargava, the director-general of Indian Council of Medical Research. “There are several hypotheses on this, but we can’t say anything clearly on any factor. We hope it continues.” There are others who look at deaths per million (DPM) as a better indicator to look at death count. "Currently, the DPM for India is 6.70. Compared to 351 and 608 deaths per million in the US and UK, we are positioned very well," said Babu.

According to the World Health Organisation, while India has the fourth-highest COVID-19 cases in the world, it is 11th on the list of most recorded deaths. Experts and analysts expect that the actual numbers are far higher. The missing deaths, however, are making it difficult to figure out exactly by how much.

Before Chennai, a similar issue was raised in India’s capital Delhi. Some reports from cremation and burial records found that the death toll in Delhi, as of June 10, was twice the official figure. The state government’s health bulletin published 984 deaths, whereas the Municipal Corporation of Delhi said 2,098 people died of COVID-19 between March and June 10. In another case of underreporting, in the east Indian state of West Bengal, a mismatch in the positive case count led to a political controversy and allegations of fudging coronavirus data. Again in the east — this time in the state of Bihar — news reports revealed that the state was underreporting COVID-19 test results by two to three times, which suggested that the positive cases could be far more than the ones officially released. On June 12, the south Indian state of Telangana’s High Court ordered the government to start publishing COVID-19 updates, which it had stopped in the wake of rising cases.

In Chennai, officials dismissed allegations of foul play and blamed the data lag on a “procedural lapse,” home deaths, and deaths in private hospitals. A probe is ongoing, initiated by the Department of Health, and as the state awaits the COVID-19 death audit report, the state government dismissed the incident as deliberate. “We don’t have to hide the death data, we cannot do that,” said Beela Rajesh, then Tamil Nadu’s health secretary. “Following latest reports that there were several deaths that were not reported, we have formed this committee to assess all these alleged COVID-19 deaths in city limits.”

On June 12, Rajesh was transferred out of the Health Department. Some news reports call it a “virtual admission” of failure to control the spread of COVID-19 by the state government, while government sources say there is “nothing controversial” about the transfer. No official reason has been given.

TS Selvavinayagam, Tamil Nadu’s additional director of Public Health and Preventive Medicine, told VICE that the work with the reconciliation committee has already started to investigate the discrepancy. “The data issues,” added Selvavinayagam, “can be easily sorted out.”

Partha Mukhopadhyay, senior fellow at the Delhi-based Centre for Policy Research, told VICE that the cases of missing deaths denote three key issues with how the Indian states are dealing with these cases. The "classification of death", where the states have failed to coordinate on recording this statistic from the beginning; the "date of classification of death", where the increasing deaths mean lags and delays within the government of India’s Death Advisory Committee; and the “actual reporting of the death”.

“[It’s the last system] that seems much less organised,” said Mukhopadhyay. “So, once a person tests positive, tracking that person in a hospital or facility or in home isolation seems not to be in a common database and may be left to district authorities.” The larger implication, the expert added, is that the patients dying are “falling through the cracks”, and that hospitals have limited information on what happens to patients in the hospital.

“The challenges [of data collection] are logistical, administrative and of capacity,” Rukmini Srinivasan, an independent data journalist based in Chennai, told VICE. “The Chennai example showed us that coordination between hospitals and the government, or different government agencies, is still far from smooth. India lags behind in electronic medical records and that's obviously not going to get fixed during a crisis.”

Some experts look at the lack of public health surveillance along with ad hoc mechanisms and temporary workers as the reason behind poor reporting of COVID-19 deaths in India. "Permanent workforce trained in public health surveillance are needed to improve the overall reporting," said Babu.

Tracking deaths during a pandemic is not easy, but researchers maintain that it is key to provide a clearer picture of the crisis, and its actual scale and toll.

India is not alone, though. There are similar stories from across the world that indicate that governments are being selective about collecting and sharing COVID-19 data. News reports from the United States called out the government’s disease-fighting agency, Center for Disease Control and Prevention, for distorting coronavirus data and providing “an inaccurate picture of the state of the pandemic.” In Russia, international news publications called out the government for underreporting the official COVID-19 death count, which was categorically denied by the Russian Foreign Ministry. On June 6, Brazil stopped publishing the running total of COVID-19 deaths and infections, while China continues to dodge allegations of hiding the real figures.

As Chennai rushes into disaster management over the missing deaths, Rukmini added that while it was “unique” of Chennai to discover uncounted deaths from a committee set up by the government itself, it sets an example for other states. “It probably indicates that other states might have data that needs auditing too.”

The crisis of data has significantly impacted the way the scale of the pandemic in India is seen. Will the actual impact of the disease ever truly be determined? “Eventually, yes,” said Rukmini. “But at the moment, we will probably not find out.”

Follow Pallavi Pundir on Twitter.