Data points to gross underestimation of Covid deaths

A July 9 update on the Health Management Information System (HMIS) of the National Health Mission provides fresh evidence that India’s official Covid-19 death toll may be an underestimate. HMIS has recorded 827,597 deaths in April-May 2021, a number that is far greater than what was seen in the last three years. In 2018, 2019 and 2020, these numbers were 355,905, 391,593, and 350,333 deaths over the same period. India’s official death toll from Covid-19 for April-May 2021 is 168,927.

While the HMIS data includes mortality from all causes, the large deviation from previous years and the association of the 2021 deaths with symptoms such as fever and respiratory illnesses strongly suggest that these could be due to Covid-19. The excess mortality reflected in the HMIS numbers in itself could be an underestimate because the database primarily captures data from government facilities in rural India.

What is the HMIS?

The website of National Health Mission describes HMIS as “a Government to Government (G2G) web-based Monitoring Information System that has been put in place by Ministry of Health and Family Welfare (MoHFW), Government of India to monitor the National Health Mission and other Health programmes and provide key inputs for policy formulation and appropriate programme interventions”. Launched in October 2008, the HMIS gets data from around 200,000 health facilities (across all States/UTs) “which upload facility wise service delivery data on monthly basis, training data on quarterly basis and infrastructure related data on annual basis on HMIS web portal”, the official website says.

Because the Civil Registration System (CRS) statistics are published with a significant time lag – the 2019-20 numbers were published in June 2021 – HMIS death numbers are the closest proxy to official death numbers in India. A comparison of total recorded deaths in the CRS and HMIS shows that the latter are significant underestimates. The CRS recorded 7.6 million deaths in 2019-20 compared to just 2.6 million in the HMIS.

What does HMIS data tell us about deaths during the second wave?

The second wave of Covid-19 infections peaked on May 9, if the seven-day rolling average of daily new cases is concerned. In terms of deaths, the peak came much later, on May 23. However, both daily new cases and deaths recorded a big spike in April itself. HMIS data for April and May 2021 shows a sharp rise in the number of deaths – 8.28 lakh compared to less than 4 lakh in the past three years for which data is available.

While deaths normally increase marginally every year, the 2020 numbers actually show a fall, which could be attributed to the lockdowns. This counterintuitive trend could also be a result of lower than usual reporting in the HMIS portal. The number of health facilities which reported this information after March 2020 onwards has still not been published.

What is the likelihood of these deaths being Covid-19 deaths?

It is very high. The HMIS gives age-wise and cause-wise break-up of deaths. We know that Covid-19 affects adults more than children. Adult deaths in April-May 2021 increased by 153% compared to the 2020 figures. The HMIS records adult deaths as separate from maternity related deaths. To be sure, maternity related deaths also increased, perhaps a result of lack of access to health facilities at a time when Covid-19 patients flooded hospitals. For children, deaths decreased by 14%. Deaths due to fever and respiratory problems – the most common aggravated symptom of Covid-19 infections -- saw the biggest jump; more than 500%, in 2021 compared to 2020.