Why India’s COVID-19 death count is low despite skyrocketing cases

According to the article, health experts and epidemiologists who were interviewed have expressed their doubts over the uncertainty in the COVID-19 mortality figures in India.

The Indian Council of Medical Research (ICMR) in its guidelines on the reporting of COVID-19 figures says that people who were suspected to have COVID-19 but were not confirmed should be included in the overall mortality data. This guideline is based on the World Health Organization's (WHO) official code for reporting COVID-19 deaths.

However, as the guidelines are on an advisory level, it is unclear whether the different states in the country have taken these suggestions into account while reporting their daily-weekly-monthly figures on COVID-19 infections and deaths, as there is no official information available out there.

According to a study published in the preprint server medRxiv, the seroprevalence survey which was carried out in Delhi showed an infection rate of about 23.48%, and an implied case fatality rate of about 0.06%. 

Studying the official figures from this serosurveillance and comparing it with the age group-based case fatality rate in France, Spain and the Lombardy region of Italy, the average fatality rate could be far higher than what was estimated in official figures. The study indicated that as many as 1,500-2,000 deaths in the age group of over 60 years due to COVID-19 may be missing from official figures.

The study, which was published in July 2020 also came on the back of some states retrospectively adding more deaths to their official figures. Tamil Nadu added more than 400 deaths to its total; Maharashtra also revised its death toll upwards. West Bengal, which was previously excluding deaths due to comorbidities, began adding those to their overall tally too.

Official government data released by a particular country takes into account the actual number of tests being conducted at the district, state and national levels, followed by the positive and negative results along with the number of deaths that are collected and put on record. But different states in the country, not to mention districts and cities, may be employing different testing models and may have different parameters of reporting their respective figures.

The lack of a uniform set of codes or benchmarks to collect data means that there can be irregularities in the tracking of data, along with misdiagnosis of the cause of infection or deaths, besides the matter of not including those who have not been tested or undergone treatment. Deaths that may have occurred due to a different primary reason or without a direct link of the infection also means that the data being reported from different states can show a completely different picture to the ground reality.

Speaking to the author of The Lancet article, director of the National Centre for Disease Informatics and Research (NCDIR) Prashant Mathur said, "It is important to correctly record the cause of deaths. But it is up to individual states to follow these guidelines. As per the existing law, NCDIR is not required to get data about suspected or probable deaths from states so I can't say whether deaths are being certified."

The differences in the state-based data is also due to the overall quality of healthcare across states and districts, as some economically weaker states may not have the wherewithal to register all the patients getting admitted to the hospital. This is because a majority of India's population lives in rural areas with inadequate access to healthcare, which means many of the deaths (even outside of COVID-19) take place before a patient has been admitted to the hospital. The lack of or delay in registration may result in a huge difference in actual death figures at the national level.

Some experts have also pointed out that the number of deaths that are officially registered or are medically certified stands at a low 22%. According to data released by the office of the Registrar General of India last year, some of the developed states in the country had an official registered deaths of as low as 10%. It must also be noted that a large majority of the total COVID-19 deaths reported in the country have come from just four out of the 28 states and eight union territories—Delhi, Maharashtra, Tamil Nadu and Karnataka—which have accounted for 65% of the national total.

Read more: Serosurveillance to become a monthly affair in Delhi, say authorities

Even national-level surveillance efforts cannot account for the true number of deaths: many of the patients who may have died due to COVID-19 may not have been tested for it.

Another potential reason for underreported COVID-19 figures could be the different methods of doing diagnostic tests by different states, which means there is no uniform method of testing in every state in the country. More and more states have started preferring rapid antibody tests instead of the RT-PCR tests for fear of false negatives.

Thus, there is a call for the health ministry to standardise data reporting from every state and make it transparent to be able to arrive at a more accurate figure of infections and deaths at a national level.