Wide disparities in excess mortality estimates show not only how challenging the calculations are, but also how far we still have to go in accurately tracking deaths. We will never know how many people died of COVID.
Too many fatalities worldwide continue to go unreported, therefore we will never be able to determine the precise number of individuals the COVID-19 pandemic has claimed. However, statistical models predict that COVID-19 will surpass coronary heart disease as the main cause of mortality globally in 2021. This conclusion is drawn from estimations of excess mortality, or fatalities that exceed expected levels, rather than from official COVID-19 figures.
Following the release of their initial numbers earlier this year (read below), World Health Organization (WHO)-affiliated researchers this week publish further information on their calculations of excess mortality during the pandemic in Nature1. According to the statistics, excess mortality during 2020 and 2021 was approximately 2.7 times higher than the official death toll, with between 13.2 million and 16.6 million deaths, with 14.8 million deaths being the most likely estimate.
In reality, this seems to be considerably more conservative than other projections. The Institute for Health Metrics and Evaluation (IHME), a global health-research center at the University of Washington in Seattle, reported2 in March a range of 17.1 million to 19.6 million additional deaths between 1 January 2020 and 31 December 2021, with 18.2 million being the most likely figure. The Economist magazine’s team employs a machine-learning technique to provide a regularly updated statistic that was initially around 18 million but is now about 16 million (see go.nature.com/3d5bpc3).
The WHO expects that excess mortality will be between 9 million and 12 million in 2021; fatalities from coronary heart disease are estimated to be around 8.9 million in 2019, the most recent year for which estimations are available. These additional deaths include those associated to the pandemic indirectly, such as ailments that health-care systems could have been able to manage if not for COVID-19-related interruption.
Subscribe to GreatGameIndia
Demographers and data researchers who conduct research on excess mortality are among the first to note that their attempts cannot be more than approximations. Since many nations do not gather or publicly release timely mortality data, statistics must be extrapolated from regional values or survey estimates, or models must be developed based on what is known about the severity of the pandemic in these countries, containment measures used, and numerous proxies for socioeconomic conditions.
Even for the 100 or so nations that report monthly national data on all fatalities, calculating excess deaths requires building models to determine the baseline of ‘normal’ mortality. WHO experts identified errors in their initial estimations for Germany and Sweden earlier this year and corrected their statistics. One study3 including Denmark, Finland, Iceland, Norway and Sweden demonstrated that the IHME results, in particular, are inexplicably out of line with those reported by the WHO and The Economist.
What would be the utility of estimating excess deaths in light of these caveats? First, it underscores the severity of the catastrophe, highlighting that many low- and middle-income countries that, on the surface, experienced few deaths were likely struck just as badly, if not harder, than wealthier nations. Individuals from these countries did not have a covert immunity to COVID-19, even if their deaths were not as meticulously documented as those in higher-income countries.
Second, it emphasizes how much work need to be done to enhance death-recording systems. As part of its Sustainable Development Goals, the United Nations is attempting to track countries’ success in documenting deaths. According to the most recent data, by 2020, 154 of the 188 nations studied had “at least 75% complete” death data. People may have little incentive to disclose deaths in nations with insufficient social safety nets. Many people say they did not realize they needed to when questioned. Census-type surveys can fill in certain gaps later, but tend to concentrate on collecting maternal and child mortality. UNICEF, the UN children’s charity, says that almost 50% of all deaths are not legally recorded; Vital Strategies, a non-profit public-health organization in New York City, thinks that 40% are unregistered.
Improving civil registration and vital statistics (CRVS) systems, which are used to register births and deaths, is critical to improving public health. The WHO is working on a treaty to improve global preparedness for and resilience to future pandemics; improving CRVS systems is not currently included, but it ought to be. More funding should be directed toward initiatives that provide governments with information on how to improve their systems; currently, the WHO, Bloomberg Philanthropies, and the Gates Foundation finance a jumble of advisory committees.
The latest data shows that 58% of COVID-19 deaths in August 2022 were from people who were vaccinated or boosted. Based on past figures and the current trends, we can reasonably estimate that the number of vaccinated/boosted COVID-19 deaths will only rise.
Better, more consistent documentation is a first step towards smoothing out the gaps between estimates, and decreasing the inclination for countries to pick measures that support their own findings. Comparisons between countries will remain tough, making it challenging to assess which strategies were more or less effective at minimizing mortality, or how lethal the virus was in particular populations. Many estimates, including the WHO’s, do not yet completely account for demographic differences between nations, such as not modifying projections for age or gender. To resolve the differences, researchers will need to work together and communicate openly. None of these challenges should detract from the overall endeavor to quantify the devastating impact of this pandemic, not just in regards of those who have died, but also in terms of survivors’ health.
- Msemburi, W. et al. Nature https://doi.org/10.1038/s41586-022-05522-2 (2022). Article Google Scholar
- COVID-19 Excess Mortality Calculators. Lancet 399, 1513–1536 (2022). Article PubMed Google Scholar
- Kepp, K. P. et al. Int. J. Epidemiol. https://doi.org/10.1093/ije/dyac204 (2022). Article Google Scholar
Read the document below: