By May this year, India had significantly scaled up its testing capacity, but testing centers are mostly in urban areas while SARS-CoV-2 is making inroads into villages and remote regions. (Even cities continue to report delays and difficulty in testing because capacity has not kept up with the rising demand.) The government is not providing a rural-urban breakdown of new cases, deaths or tests conducted, but the local media are reporting vast undercounting, such as a report from Gandhinagar, the capital of Gujarat: While the government claimed 182 people had died of Covid-19 in the entire district, 690 people were reported to have died in just four of the district’s 200-plus villages. Nearly 66 percent of Indians live in villages, and unless innovative ways are found to scale up testing – for instance, door-to-door screening, as has been done with tuberculosis before, or mobile testing vans – the case and death count will always remain underestimated.
Omissions – inadvertent or deliberate
Estimating the death rate during a pandemic is a fraught exercise: At any given time, the total number of Covid-19 cases includes the number of active cases, the number of those who have recovered, and the number of those who have died. But it takes between two and six weeks (or more) for a case to be decided – for the patient to recover, or to die – so including active cases when calculating the death rate is incorrect, and keeps the number artificially low.
With India’s inadequate testing already keeping its Covid-19 case numbers low, its death rate is bound to be an underestimate when combined with the above problem, as IndiaSpend reported in June 2020. At a press conference on May 11, for instance, the central government claimed the death rate from Covid-19 was 1.09 percent, whereas using the more accurate formula (dividing the number of deaths by the number of decided cases) would yield a higher 1.3 percent. This might seem like a difference of a small fraction but translates into a significant number when deaths have exceeded a quarter of a million (again, by the underestimated official count).
Further, our reporting has shown that in official Indian data, only a person who has tested positive prior to death is recorded as having died of Covid-19, which leaves out not only those who were not tested or those who died at home but also those whose test may have returned a false negative (the antigen test, used widely in India, is known to return false negatives up to half the time). It also does not account for those who may have had “long Covid-19” and died of related complications. This is against the WHO’s guidelines for the certification and recording of Covid-19 deaths which specify that “Covid-19 should be recorded on the medical certificate of cause of death for all decedents where the disease caused, or is assumed to have caused, or contributed to death”.