PHOTO: ATUL LOKE/PANOS PICTURES/PANOS PICTURES/REDUX128 14 JANUARY 2022 • VOL 375 ISSUE 6577 science.org SCIENCEI
ndia, from the earliest days of the pan-
demic, reported far fewer COVID-
deaths than expected given the toll
elsewhere—an apparent death “paradox”
that some believed was real and oth-
ers thought would prove illusory. Now,
a prominent epidemiologist who contended
the country really had been spared the worst
of COVID-19 has led a rigorous new analysis
of available mortality data and concluded he
“got it wrong.”
India has “substantially greater” COVID-
deaths than official reports suggest, says
Prabhat Jha of the University of Toronto—
close to 3 million, which is more than six
times higher than the government has
acknowledged and the largest number
of any country. If true, the finding could
prompt scrutiny of other countries with
anomalously low death rates and would
dramatically push up the current world-
wide pandemic total, estimated by the
World Health Organization (WHO) at some
5.45 million people.
“I think it does call for a recalibration of
the global numbers plus saying, ‘What the
heck is going on in India?’” says Jha, whose
team published the new India analysis onlinelast week in Science. And India’s suffering
could be far from over—the Omicron variant
of the coronavirus is surging there.
At the end of 2021, India reported about
480,000 deaths from SARS-CoV-2 infections.
That’s 340 COVID-19 deaths per million—
about one-seventh the per capita COVID-
mortality tallied in the United States. Jha’s
own early analysis supported the appar-
ently low mortality rate from COVID-19,
but he and his colleagues have now probed
more deeply. They tapped data from an in-
dependent polling agency that surveyed
nearly 140,000 people across the vast coun-
try by telephone, asking whether anyone in
each household had died from COVID-19.
They also analyzed government reports from
hospitals and similar facilities and looked
at officially registered deaths. The result: a
much higher estimate—between 2300 and
2500 deaths per million by September
2021, comparable to the rate in the United
States and pointing to a much higher total
death toll, because India has four times as
many people.
Jha says his early, low estimate was based
on the first wave of infections in the fall
of 2020, which may have been less deadly
than the Delta variant that drove India’s
massive surge in spring of 2021. He also fo-cused on the large cities, where death rates
may have been lower than in the country-
side. And death registration in the country
had been spotty even before the pandemic.
But those factors can’t be the whole story,
he says. “There must be other things that
we still don’t understand.”
One, he says, is politics: He thinks the
administration of Indian Prime Minister
Narendra Modi clouded the true picture
of the pandemic. “The Indian government
very much is trying to suppress the num-
bers in the way that they coded the COVID
deaths,” Jha says. He and others also fault
the government for not releasing data
from what’s known as the Sample Registra-
tion System (SRS), which routinely surveys
1% of India’s population to track births and
deaths. “I think the political pressures were
such that they said, ‘Anything that’s going
to come out is going to be embarrassing.’”
Ramanan Laxminarayan, a Princeton
University epidemiologist and economist,
doesn’t see the undercount as entirely de-
liberate. He notes that SRS data haven’t
been released since 2018, before the pan-
demic, so that suspension may just reflect a
disorganized system. Almost every coun-
try undercounts COVID-19 mortality, he
adds—“I think all governments want toIN DEPTH
By Jon CohenCOVID-India’s pandemic toll far exceeds official count
New analysis bolsters idea that country’s seemingly low death rate was misleading
Images of fields filled with funeral pyres, like these in New Delhi in April 2021, highlighted the massive number of COVID-19 deaths in India.