The 14th century bubonic plague, Black Death, known as the most fatal pandemic recorded in human history, resulted in an estimated death toll of 75-200 million in Eurasia and North Africa. The century-old Spanish Flu caused an estimated 17-50 million deaths, and possibly even as high as 100 million. These estimates were put out after wide intervals, and understandably so.
Tracking the infected and recording of the disease and deaths were not very efficient in those days. Well, one could have only expected that the accounting would be much better and more efficient in the ongoing Covid-19 pandemic. But, unfortunately, the reality is different.
It’s widely alleged that deaths from Covid are severely under-reported worldwide. Many experts believe that excess deaths during the coronavirus pandemic in many countries could be 4-5 times, or even a staggering 10 times, the official toll figure. One wonders whether it is really six million or 60 million, as of now.
In early 2020, when Covid was ravaging Italy, an acquaintance, a Professor in Milan, wrote to me : “Many people are dying, who have never been checked for Covid-19, so they do not enter in the official counts.”
It’s not only about Italy, though. It’s widely believed that the official count could have missed deaths that occurred due to overwhelmed hospitals or while healthcare was disrupted, particularly during the devastating virus surge. In his paper, ‘The pandemic’s true death toll: millions more than official counts’, published in Nature at the end-January, science journalist David Adam asserts that the “global excess deaths are estimated at double or even quadruple” the reported figure.
Certainly, estimating the death toll of the pandemic involves dealing with potential outcomes, known unknowns, and uncertainty. A recent report of the US Centers for Disease Control and Prevention (CDC) says that there have been more than one million excess deaths already in America during the pandemic.
The World Health Organization (WHO) also aimed at estimating the excess mortality attributable to Covid as the difference between the total observed deaths for the year and those expected in the absence of Covid. In fact, all-cause excess mortality is one of the most reliable and unbiased ways to look at the effect of a disaster that has been followed in many situations — for example, to find excess deaths attributable to obesity vs normal weight, or to estimate the death toll of a hurricane.
The Economist argues that the estimate of excess deaths around the world should include the people who die while infected but never tested for it, and thus don’t enter the official totals. At the same time, those who had Covid but might have died in a similar timeframe of other ailments should be excluded from the count. However, many people have died of preventable causes during the pandemic, because of the overwhelmed health infrastructure. Although they were not Covid-infected, their death was caused by the pandemic, for sure. Shouldn’t they be counted among Covid casualties?
A July 2020 article in the medical journal JAMA argued that quarantine measures also might have led to a decline in deaths from other causes, such as vehicular crashes because fewer people were driving on the roads. In addition, social distancing measures might have decreased the deaths from causes like flu. However, a simple average of the number of deaths from a few preceding years, that many people have attempted, might not yield a good estimate as it would miss to address the trend component, if any.
One may, instead, consider monthly/yearly mortality data of a city/province/country for a few years before the pandemic, if available, and use a suitable time series model — a pretty standard method though, and use standard software for reasonably reliable predictions of the yearly/monthly mortality figures during 2020, 2021, etc., which would have been the expected death numbers had there been no Covid. The excess numbers can very well be attributed to the deadly pandemic.
But, not that simple in reality though. “Some official data in this regard are flawed, scientists have found. And more than 100 countries do not collect reliable statistics on expected or actual deaths at all, or do not release them in a timely manner,” explained David Adam.
Thus, several methods — ranging from satellite images of cemeteries to door-to-door surveys and machine-learning computer models that try to extrapolate global estimates from available data — are attempted in different corners. The dynamic estimate made by The Economist, for example, shows about 13-23 million excess deaths until now, which is 2-4 times the official toll so far.
Thus, the exact death toll of Covid would remain hazy — even with the technological advancement of the present day. The future generation would invariably have to cite a pretty wide band for the estimated Covid mortality numbers — similar to what we do now for Black Death or Spanish Flu.
The writer is Professor of Statistics, Indian Statistical Institute, Kolkata