Counting the dead

Covid-19 has exposed India's civil registration lacunae

The Covid-19 pandemic has exposed a massive gap in civil registration data, which has made it impossible to know the real death toll due to the disease. In fact, nobody knows exactly how many Indians have died since 2018, and from what causes. For that matter, there is no way to know exactly how many Indians have been born since 2018. A study from Virginia Commonwealth University, based on the data encompassing parts of 12 states with a population of 825 million, estimates 2.2-2.6 million Indians have died of Covid-19 so far. But the aggregated all-India civil registration data is only available until 2018. The Census of India estimates say it is also incomplete. The Census estimates 92.7 per cent of the births and 92 per cent of the deaths were registered in 2018-19. In several laggard states, only about 50 per cent birth registration took place. Medical certificates exist only for 22 per cent of the registered deaths and in many certificates, the actual cause of death is hard to ascertain because it lists something uninformative like “heart failure”.

The lack of data means health care policy is based on guesswork. Policymakers cannot prioritise, and allocate resources effectively. It is also difficult to review policy decisions, and suggest course corrections. Insurers cannot reliably compute life premiums. Even road and rail safety could improve with good, timely data inputs, given over 200,000 Indians die annually in accidents. Quite apart from all this, hassles arise when it comes to claiming citizenship. The evidence stitched together by analysts suggests massive undercounting of Covid-related deaths. But the data is patchy. The reported figures from states like Tamil Nadu, Andhra Pradesh, Gujarat, Madhya Pradesh, Karnataka, and Uttar Pradesh indicate overall excess deaths in the past two to three months have been anywhere between five times (in Karnataka) and 43 times (in UP) of the official Covid tally. In Bihar, a court-mandated audit indicated serious undercounting.

Some of those excess deaths may not be due to Covid but the causes are unknown. The lack of easily available, timely historical data makes it hard to compute reliable baselines by computing long-term trends of births, deaths, causes, etc. As the population grows, there is a natural increase in deaths. Last year (2020) was unusual due to Covid. A given region may experience excess deaths in any given year due to a natural disaster like an earthquake or cyclone. All these can only be smoothened out by accessing historical data. A comprehensive overhaul of the civil registration system is necessary. These data must also be openly searchable, and released as close to real time as possible. Mandating laws for registration is obviously not enough since those laws already exist; citizens must be encouraged to register births and deaths as a matter of course. Medical certificates must also be more informative with respect to real, and underlying causes.

All this should be entirely possible because the recording processes are already digitised. Given the privacy concerns, such data could also be anonymised before release (though it could be de-anonymised by using the available open-data of the voters’ lists). The pandemic has brought these gaps into sharp focus. The Covid tragedy should provide some impetus for an overhaul of the civil registration system. At least it would allow for a better, long-term health care policy.