Skip to main content
Sabrang
Sabrang
Health Politics

Are India’s official Covid-19 mortality numbers grossly under-reported?

Gaslighting the victims of Covid-19 is an all-time political low for any government; official machinery now denying independent data analysis too

Karuna John and Deborah Grey 03 Aug 2021

data analysis

It was only the RJD MP Manoj Jha, who took Parliament by storm when he stood up in the Rajya Sabha on July 20 to remember those who succumbed to Covid-19 that the government seems to have forgotten.

His speech that went viral was in fact an apology, that he reiterated his recent Indian Express column saying, “It is the responsibility of the House to apologise to all those whose bodies were floating in the Ganga. Never in Parliament’s history, between two sessions, we have had to read obituaries of 50 people” adding “This pain is personal, I don’t want to talk about numbers. My number, your number. Look for numbers in your grief.”

However, it is these very numbers that are at the cause of these problems. The Covid-19 reportage, via the official websites, pages and social media handles of the Ministry of Health and Family Welfare (MoHFW), have been challenged by various studies as being grossly under-reported. The most glaring discrepancies have been highlighted by various studies with regard to the total number of deaths that have been reported during Covid-19 peaks in the country, including those who died due to the shortage of ventilators and Oxygen supply in hospitals, and the unavailability of oxygen cylinders for individuals who were critically ill. 

However, according to the government, which has issued denials on the Press Information Bureau (PIB) website, headlining them as “Myth Vs. Facts” India follows ICMR guidelines which in turn “are based on WHO recommended ICD-10 codes for correct recording of all Covid-19 deaths”. The MoHFW states that “Civil Registration System (CRS) ensures institutionalised registration of births and deaths in the country” and as that has been in force “for decades in India, missing out on Covid-19 deaths unlikely”.

This was the government's stand on July 27 in which it completely denied “media reports, based on a yet-to be peer-reviewed study which was uploaded on MedRxiv recently, alleging that at least 2.7 to 3.3 million Covid-19 deaths happened during the two waves of Covid-19 in India, quoting three different databases ‘pointing towards at least 27% excess mortality over a year.’” This alluded to the report authored by Abhishek Anand, Justin Sandefur and Arvind Subramanian, who had reiterated in their IE column that “confronting the scale of the tragedy will help India draw lessons and etch them deep into the nation’s collective consciousness to foster a ‘never again’ resolve.”

The data analysis report by Arvind Subramanian from the Center for Global Development and Brown University, Justin Sandefur from Center for Global Development and Abhishek Anand from Harvard University, concluded that India under-counted deaths. It stated that the actual deaths during the Covid pandemic are likely to “be in the several millions not hundreds of thousands”. This made this “arguably India’s worst human tragedy since partition and independence.”

What does the independent study say?

The study by Subramanian, Sandefur and Anand excess mortality estimates from three different data sources from the pandemic’s start through June 2021. As per the report published on the Center for Global Development website, “First, extrapolation of state-level civil registration from seven states suggests 3.4 million excess deaths. Second, applying international estimates of age-specific infection fatality rates (IFR) to Indian seroprevalence data implies a higher toll of around 4 million.” Finally, the “analysis of the Consumer Pyramid Household Survey, a longitudinal panel of over 8,00,000 individuals across all states, yields an estimate of 4.9 million excess deaths.”

Thus, the government’s official figure of around 4,00,000 deaths is way under, not one, but three figures arrived at by examining three different sources of data.

Here’s a comparative table comparing alternative estimates of All-Cause Excess Mortality (in millions)

Data

Source: Report by Arvind Subramanian, Justin Sandefur and Abhishek Anand

Inherent flaws in each model

The study acknowledges that “uncertainty afflicts all estimates,” and that “the three estimates are quite divergent in their assessment of the timing of Covid-related mortality (keeping in mind that the duration of the two waves is very different, 11 and 3 and a half months, respectively).” It also acknowledges that all three “point to significantly greater deaths than the official estimates.” The authors also conclude, “Relatedly, it seems that the first wave was also more lethal than is widely believed.”

State-wise figures

The authors found data collection challenging because, “Official data on mortality are not collected and reported in a timely manner that would allow excess mortality to be estimated. India’s Sample Registration System (SRS), managed by the central government, conducts annual mortality surveys but has only published numbers through 2019.”

It adds, “In the second wave, as the death toll rose, investigators forced state governments to release data on civil registration of deaths (CRS). As of June 29, 2021, excess deaths in 2020 and 2021 based on death registrations.” The study finds that the disaggregated data for the seven states (Andhra Pradesh, Bihar, Chhattisgar, Karnataka, Kerala, Madhya Pradesh, Tamil Nadu and Uttar Pradesh) accounts for about half of India’s total population.

Here’s a table with data for the seven states:

Data

Source: Report by Arvind Subramanian, Justin Sandefur and Abhishek Anand

Now, even when it comes to this data the authors are cautious because, “First, even for these seven states excess mortality may be mismeasured because CRS typically under-counts final estimated deaths in the SRS. Based on 2019 data, under-counting relative to SRS varies from zero for the Southern states to 37 percent for Uttar Pradesh and 48 percent for Bihar. NFHS-5 data for 2019-20 (but available for only 4 of the seven states) suggests that even the CRS is under-stating the magnitude of the problem: for example, NFHS indicates that the under-counting of deaths in Bihar is 67 not 48 percent and for Andhra Pradesh is 20 percent not zero.”

The authors also find, “Second, in terms of Covid mortality, it is possible that the rest of India is different from the seven states. Sero-prevalence could be different between more rural Bihar and dense, urban Maharashtra. Fatality rates could similarly be different between Kerala’s well-run health system and Rajasthan’s which is more challenged.”

The authors are also wary about UP’s statistics. They say that “the number for Uttar Pradesh in the second wave does not appear to pass the smell test.” Finally given how “CRS data for most states stops in May, so that they almost certainly fail to capture all of the second wave.”

With all these riders, the authors arrive at the following figures:

Data

Source: Report by Arvind Subramanian, Justin Sandefur and Abhishek Anand

When it comes to Infection Fatality Rates (IFR), the study acknowledges that many states had conducted seroprevalence studies. It specifically points out “the third sero-survey done in December 2020- January 2021 and a recent WHO-AIIMS survey covering the period mid-March to early June. The former placed India’s infection rate at about 22 percent. The latter placed India’s infection rate at 57.7 percent, and 63.5 percent in the age group below 18 and above respectively.” But concludes that in the absence of reliable mortality data, they authors were compelled to rely on international estimates of IFRs.

They used the United States’ Center for Disease Control (CDC)’s best estimates of age-specific IFRs which were based on the meta-analyses of Levin et. al. (2021), and combined them with Indian demographic characteristics and the age-pattern of Indian infection rates to derive a plausible measure of IFR for India. The results are tabulated below:

Data

Source: Report by Arvind Subramanian, Justin Sandefur and Abhishek Anand

The authors conclude, “Combining the Indian sero-prevalence data and applying the best estimate of international age-specific IFRs to Indian demography and sero-prevalence patterns, yields excess deaths of 1.5 million and 2.4 million in the two waves, respectively. Together, the toll of Covid until June 2021 is estimated at 4 million. Even though the second wave encompasses three months compared to eleven months for the first, the estimated death toll is significantly greater.”

Government dismisses findings

The MoHFW said that this “mis-informed reports are totally fallacious” to state that India’s Covid death rate may be about 7-8 times higher than the officially reported toll. The ministry stated that the Government “has been transparent in its approach to Covid data management and a robust system of recording all Covid-19 related deaths already exists.” The Centre also put the onus on states and Union Territories who it said had “been entrusted with the responsibility to update the data on a continuous basis.” It stated that the states were “advised to conduct thorough audits in their hospitals and report any cases or deaths that could have been missed with a district and date-wise details so as to guide a data-driven decision making.” Adding that “in addition to this reporting by States/UTs, the robustness of statute based Civil Registration System (CRS) ensures all the births and deaths in the country get registered. The CRS follows a process of data collection, cleaning, collating and publishing the numbers, which although is a long-drawn process, but ensures no deaths are missed out. Because of the expanse and the amplitude of the activity, the numbers are usually published the next year.”

On the delays being alleged in recording the deaths in the devastating second wave of Covid-19, the government stated that “the health system across the country was focused on effective clinical management of cases requiring medical help due to which correct reporting and recording of Covid deaths could have been delayed but later was reconciled by the States/UTs.” It maintains that while some cases could go undetected as per the principles of Infectious Disease and its management, missing out on the deaths is unlikely” and said this was expected as “there shall always be some differences in mortality recorded during a profound and prolonged public health crisis such as COVID pandemic.” 

The complete statement may be read here: https://pib.gov.in/PressReleasePage.aspx?PRID=1739439

The  Union Health and Family Welfare Minister, Mansukh Mandaviya had also informed the Rajya Sabha via a written answer that State Governments and Union Territories did not specifically report any deaths due to lack of oxygen and lack of hospital facilities during the second wave of the coronavirus pandemic. Health Minister Mansukh Mandaviya’s answer read:

“Detailed guidelines for reporting of deaths have been issued by the Union Health Ministry to all states/UTs. Accordingly, all states/UTs report cases and deaths to the Union Health Ministry on a regular basis. However, states have at times reported backlog deaths in case of late reporting from hospitals and districts but, no such reports of patients dying due to shortage of oxygen and lack of hospital facilities in the country have been received by the Ministry from any State/UT.”

This when in May 2021, social media was abuzz with SOS calls for oxygen, and alerts of shortages even from the biggest hospitals in the national capital. Deaths due to shortage were also recorded from Jaipur Golden Hospital in Delhi. Hospitals like Maharaja Agrasen, Batra, Max had moved the Delhi High Court citing lack of oxygen supply. Eight Covid patients including the gastroenterology department’s head died in Batra Hospital due to oxygen shortage just on May 1.

The entire report by Arvind Subramanian, Justin Sandefur and Abhishek Anand may be read here: 

Related:

No Covid-19 deaths due to oxygen shortage reported by States/UTs: Centre in RS
Are obituaries in Gujarat newspapers a better indicator of real Covid-deaths?
Madhya Pradesh falsely reporting Covid deaths? 
EXCLUSIVE: Hundreds die of Covid and data goes missing, UP gov’t remorseless

Are India’s official Covid-19 mortality numbers grossly under-reported?

Gaslighting the victims of Covid-19 is an all-time political low for any government; official machinery now denying independent data analysis too

data analysis

It was only the RJD MP Manoj Jha, who took Parliament by storm when he stood up in the Rajya Sabha on July 20 to remember those who succumbed to Covid-19 that the government seems to have forgotten.

His speech that went viral was in fact an apology, that he reiterated his recent Indian Express column saying, “It is the responsibility of the House to apologise to all those whose bodies were floating in the Ganga. Never in Parliament’s history, between two sessions, we have had to read obituaries of 50 people” adding “This pain is personal, I don’t want to talk about numbers. My number, your number. Look for numbers in your grief.”

However, it is these very numbers that are at the cause of these problems. The Covid-19 reportage, via the official websites, pages and social media handles of the Ministry of Health and Family Welfare (MoHFW), have been challenged by various studies as being grossly under-reported. The most glaring discrepancies have been highlighted by various studies with regard to the total number of deaths that have been reported during Covid-19 peaks in the country, including those who died due to the shortage of ventilators and Oxygen supply in hospitals, and the unavailability of oxygen cylinders for individuals who were critically ill. 

However, according to the government, which has issued denials on the Press Information Bureau (PIB) website, headlining them as “Myth Vs. Facts” India follows ICMR guidelines which in turn “are based on WHO recommended ICD-10 codes for correct recording of all Covid-19 deaths”. The MoHFW states that “Civil Registration System (CRS) ensures institutionalised registration of births and deaths in the country” and as that has been in force “for decades in India, missing out on Covid-19 deaths unlikely”.

This was the government's stand on July 27 in which it completely denied “media reports, based on a yet-to be peer-reviewed study which was uploaded on MedRxiv recently, alleging that at least 2.7 to 3.3 million Covid-19 deaths happened during the two waves of Covid-19 in India, quoting three different databases ‘pointing towards at least 27% excess mortality over a year.’” This alluded to the report authored by Abhishek Anand, Justin Sandefur and Arvind Subramanian, who had reiterated in their IE column that “confronting the scale of the tragedy will help India draw lessons and etch them deep into the nation’s collective consciousness to foster a ‘never again’ resolve.”

The data analysis report by Arvind Subramanian from the Center for Global Development and Brown University, Justin Sandefur from Center for Global Development and Abhishek Anand from Harvard University, concluded that India under-counted deaths. It stated that the actual deaths during the Covid pandemic are likely to “be in the several millions not hundreds of thousands”. This made this “arguably India’s worst human tragedy since partition and independence.”

What does the independent study say?

The study by Subramanian, Sandefur and Anand excess mortality estimates from three different data sources from the pandemic’s start through June 2021. As per the report published on the Center for Global Development website, “First, extrapolation of state-level civil registration from seven states suggests 3.4 million excess deaths. Second, applying international estimates of age-specific infection fatality rates (IFR) to Indian seroprevalence data implies a higher toll of around 4 million.” Finally, the “analysis of the Consumer Pyramid Household Survey, a longitudinal panel of over 8,00,000 individuals across all states, yields an estimate of 4.9 million excess deaths.”

Thus, the government’s official figure of around 4,00,000 deaths is way under, not one, but three figures arrived at by examining three different sources of data.

Here’s a comparative table comparing alternative estimates of All-Cause Excess Mortality (in millions)

Data

Source: Report by Arvind Subramanian, Justin Sandefur and Abhishek Anand

Inherent flaws in each model

The study acknowledges that “uncertainty afflicts all estimates,” and that “the three estimates are quite divergent in their assessment of the timing of Covid-related mortality (keeping in mind that the duration of the two waves is very different, 11 and 3 and a half months, respectively).” It also acknowledges that all three “point to significantly greater deaths than the official estimates.” The authors also conclude, “Relatedly, it seems that the first wave was also more lethal than is widely believed.”

State-wise figures

The authors found data collection challenging because, “Official data on mortality are not collected and reported in a timely manner that would allow excess mortality to be estimated. India’s Sample Registration System (SRS), managed by the central government, conducts annual mortality surveys but has only published numbers through 2019.”

It adds, “In the second wave, as the death toll rose, investigators forced state governments to release data on civil registration of deaths (CRS). As of June 29, 2021, excess deaths in 2020 and 2021 based on death registrations.” The study finds that the disaggregated data for the seven states (Andhra Pradesh, Bihar, Chhattisgar, Karnataka, Kerala, Madhya Pradesh, Tamil Nadu and Uttar Pradesh) accounts for about half of India’s total population.

Here’s a table with data for the seven states:

Data

Source: Report by Arvind Subramanian, Justin Sandefur and Abhishek Anand

Now, even when it comes to this data the authors are cautious because, “First, even for these seven states excess mortality may be mismeasured because CRS typically under-counts final estimated deaths in the SRS. Based on 2019 data, under-counting relative to SRS varies from zero for the Southern states to 37 percent for Uttar Pradesh and 48 percent for Bihar. NFHS-5 data for 2019-20 (but available for only 4 of the seven states) suggests that even the CRS is under-stating the magnitude of the problem: for example, NFHS indicates that the under-counting of deaths in Bihar is 67 not 48 percent and for Andhra Pradesh is 20 percent not zero.”

The authors also find, “Second, in terms of Covid mortality, it is possible that the rest of India is different from the seven states. Sero-prevalence could be different between more rural Bihar and dense, urban Maharashtra. Fatality rates could similarly be different between Kerala’s well-run health system and Rajasthan’s which is more challenged.”

The authors are also wary about UP’s statistics. They say that “the number for Uttar Pradesh in the second wave does not appear to pass the smell test.” Finally given how “CRS data for most states stops in May, so that they almost certainly fail to capture all of the second wave.”

With all these riders, the authors arrive at the following figures:

Data

Source: Report by Arvind Subramanian, Justin Sandefur and Abhishek Anand

When it comes to Infection Fatality Rates (IFR), the study acknowledges that many states had conducted seroprevalence studies. It specifically points out “the third sero-survey done in December 2020- January 2021 and a recent WHO-AIIMS survey covering the period mid-March to early June. The former placed India’s infection rate at about 22 percent. The latter placed India’s infection rate at 57.7 percent, and 63.5 percent in the age group below 18 and above respectively.” But concludes that in the absence of reliable mortality data, they authors were compelled to rely on international estimates of IFRs.

They used the United States’ Center for Disease Control (CDC)’s best estimates of age-specific IFRs which were based on the meta-analyses of Levin et. al. (2021), and combined them with Indian demographic characteristics and the age-pattern of Indian infection rates to derive a plausible measure of IFR for India. The results are tabulated below:

Data

Source: Report by Arvind Subramanian, Justin Sandefur and Abhishek Anand

The authors conclude, “Combining the Indian sero-prevalence data and applying the best estimate of international age-specific IFRs to Indian demography and sero-prevalence patterns, yields excess deaths of 1.5 million and 2.4 million in the two waves, respectively. Together, the toll of Covid until June 2021 is estimated at 4 million. Even though the second wave encompasses three months compared to eleven months for the first, the estimated death toll is significantly greater.”

Government dismisses findings

The MoHFW said that this “mis-informed reports are totally fallacious” to state that India’s Covid death rate may be about 7-8 times higher than the officially reported toll. The ministry stated that the Government “has been transparent in its approach to Covid data management and a robust system of recording all Covid-19 related deaths already exists.” The Centre also put the onus on states and Union Territories who it said had “been entrusted with the responsibility to update the data on a continuous basis.” It stated that the states were “advised to conduct thorough audits in their hospitals and report any cases or deaths that could have been missed with a district and date-wise details so as to guide a data-driven decision making.” Adding that “in addition to this reporting by States/UTs, the robustness of statute based Civil Registration System (CRS) ensures all the births and deaths in the country get registered. The CRS follows a process of data collection, cleaning, collating and publishing the numbers, which although is a long-drawn process, but ensures no deaths are missed out. Because of the expanse and the amplitude of the activity, the numbers are usually published the next year.”

On the delays being alleged in recording the deaths in the devastating second wave of Covid-19, the government stated that “the health system across the country was focused on effective clinical management of cases requiring medical help due to which correct reporting and recording of Covid deaths could have been delayed but later was reconciled by the States/UTs.” It maintains that while some cases could go undetected as per the principles of Infectious Disease and its management, missing out on the deaths is unlikely” and said this was expected as “there shall always be some differences in mortality recorded during a profound and prolonged public health crisis such as COVID pandemic.” 

The complete statement may be read here: https://pib.gov.in/PressReleasePage.aspx?PRID=1739439

The  Union Health and Family Welfare Minister, Mansukh Mandaviya had also informed the Rajya Sabha via a written answer that State Governments and Union Territories did not specifically report any deaths due to lack of oxygen and lack of hospital facilities during the second wave of the coronavirus pandemic. Health Minister Mansukh Mandaviya’s answer read:

“Detailed guidelines for reporting of deaths have been issued by the Union Health Ministry to all states/UTs. Accordingly, all states/UTs report cases and deaths to the Union Health Ministry on a regular basis. However, states have at times reported backlog deaths in case of late reporting from hospitals and districts but, no such reports of patients dying due to shortage of oxygen and lack of hospital facilities in the country have been received by the Ministry from any State/UT.”

This when in May 2021, social media was abuzz with SOS calls for oxygen, and alerts of shortages even from the biggest hospitals in the national capital. Deaths due to shortage were also recorded from Jaipur Golden Hospital in Delhi. Hospitals like Maharaja Agrasen, Batra, Max had moved the Delhi High Court citing lack of oxygen supply. Eight Covid patients including the gastroenterology department’s head died in Batra Hospital due to oxygen shortage just on May 1.

The entire report by Arvind Subramanian, Justin Sandefur and Abhishek Anand may be read here: 

Related:

No Covid-19 deaths due to oxygen shortage reported by States/UTs: Centre in RS
Are obituaries in Gujarat newspapers a better indicator of real Covid-deaths?
Madhya Pradesh falsely reporting Covid deaths? 
EXCLUSIVE: Hundreds die of Covid and data goes missing, UP gov’t remorseless

Related Articles

India

Assam FT deems Assam NRC to be final NRC

Several aspersions, politically as well as officially have been cast on the finality of the NRC which was released on August 31, 2019 as the final NRC. A Foreigners Tribunal in Assam has deemed that this NRC is in fact, the final NRC

India

Assam FT deems Assam NRC to be final NRC

Several aspersions, politically as well as officially have been cast on the finality of the NRC which was released on August 31, 2019 as the final NRC. A Foreigners Tribunal in Assam has deemed that this NRC is in fact, the final NRC


Sunday

03

Jan

Pan-India

Saturday

05

Dec

05 pm onwards

Rise in Rage!

North Gate, JNU campus

Thursday

26

Nov

10 am onwards

Delhi Chalo

Pan India

Theme

Taliban 2021

Taliban in Afghanistan: A look back

Communalism Combat had taken a deep dive into the lives of people of Afghanistan under the Taliban regime. Here we reproduce some of our archives documenting the plight of hapless Afghanis, especially women, who suffered the most under the hardline regime.
2020

Milestones 2020

In the year devastated by the Covid 19 Pandemic, India witnessed apathy against some of its most marginalised people and vilification of dissenters by powerful state and non state actors. As 2020 draws to a close, and hundreds of thousands of Indian farmers continue their protest in the bitter North Indian cold. Read how Indians resisted all attempts to snatch away fundamental and constitutional freedoms.
Migrant Diaries

Migrant Diaries

The 2020 COVID pandemic brought to fore the dismal lives that our migrant workers lead. Read these heartbreaking stories of how they lived before the pandemic, how the lockdown changed their lives and what they’re doing now.
Delhi HC

Hate Speech and Delhi Pogrom 2020

A spate of provocative speeches, that amount to hate speech in law and should be prosecuted allowed blood letting to spill on the streets of north east Delhi in February-March 2020

Campaigns

Sunday

03

Jan

Pan-India

Saturday

05

Dec

05 pm onwards

Rise in Rage!

North Gate, JNU campus

Thursday

26

Nov

10 am onwards

Delhi Chalo

Pan India

IN FACT

Analysis

Taliban 2021

Taliban in Afghanistan: A look back

Communalism Combat had taken a deep dive into the lives of people of Afghanistan under the Taliban regime. Here we reproduce some of our archives documenting the plight of hapless Afghanis, especially women, who suffered the most under the hardline regime.
2020

Milestones 2020

In the year devastated by the Covid 19 Pandemic, India witnessed apathy against some of its most marginalised people and vilification of dissenters by powerful state and non state actors. As 2020 draws to a close, and hundreds of thousands of Indian farmers continue their protest in the bitter North Indian cold. Read how Indians resisted all attempts to snatch away fundamental and constitutional freedoms.
Migrant Diaries

Migrant Diaries

The 2020 COVID pandemic brought to fore the dismal lives that our migrant workers lead. Read these heartbreaking stories of how they lived before the pandemic, how the lockdown changed their lives and what they’re doing now.
Delhi HC

Hate Speech and Delhi Pogrom 2020

A spate of provocative speeches, that amount to hate speech in law and should be prosecuted allowed blood letting to spill on the streets of north east Delhi in February-March 2020

Archives