Excess Mortality: What Can It Reveal About COVID-19?

What does excess mortality measure?

A disease-specific death count depends on identifying and recording that disease. Excess mortality starts with deaths from all causes and compares them with an estimated baseline. During a pandemic, this approach can capture a broader burden, including deaths missed by infection surveillance and deaths associated with disrupted health services. [WHO's explanation of excess mortality](https://www.who.int/news-room/questions-and-answers/item/global-excess-deaths-associated-with-the-COVID-19-pandemic).

In an estimate published on 5 May 2022, WHO calculated approximately 14.9 million excess deaths associated with the pandemic during 2020 and 2021, with a reported range of 13.3 million to 16.6 million. This historical estimate included direct and indirect effects; it was not a count of laboratory-confirmed COVID-19 fatalities. It also reflected offsets from deaths potentially averted during that period, such as fewer fatal traffic injuries. These figures describe the specified two-year period, rather than the cumulative toll through today. [WHO's May 2022 mortality estimate](https://www.who.int/news/item/05-05-2022-14.9-million-excess-deaths-were-associated-with-the-covid-19-pandemic-in-2020-and-2021).

Why do excess-death estimates differ between studies?

The expected number of deaths is an estimate, not an observable alternative history. Researchers must decide how to represent mortality trends and handle missing records. Where registration is incomplete, models may use relationships observed in places with better information. WHO acknowledges that transferring those relationships between populations introduces uncertainty. Comparing two estimates therefore requires checking their methods and time periods before interpreting a difference as disagreement about the underlying deaths. [WHO's methodological explanation](https://www.who.int/news-room/questions-and-answers/item/global-excess-deaths-associated-with-the-COVID-19-pandemic).

Reporting delays create another difficulty. Death certificates take time to complete and process, so the most recent weeks can appear artificially low. The US Centers for Disease Control and Prevention also cautions that excess deaths in its surveillance data cannot automatically be attributed to COVID-19: other causes may contribute. A rise identifies a pattern requiring investigation, but does not establish the cause of each individual death. [CDC's technical notes on excess deaths](https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm).

How can better mortality reporting improve health care?

WHO's SCORE health-data package, introduced on 27 August 2020, emphasized registering deaths, documenting their causes accurately, and examining differences by age and sex. That announcement provides historical context for the mortality-surveillance challenge. Complete records can reveal patterns hidden by an overall total and support decisions about where to direct limited health resources. [WHO's original health-data announcement](https://www.who.int/news-room/feature-stories/detail/new-who-technical-package-to-help-countries-improve-health-data-for-covid-19-response-and-beyond).

A practical implication is to investigate an unusual mortality increase alongside clinical and service information. Reviewing causes of death, hospital demand, and interruptions in care could help distinguish an infection surge from other health problems. That is an application of the surveillance approach, rather than proof that collecting more data alone improves survival. Preliminary findings also need checking as records become more complete: CDC notes that both death totals and cause-of-death information can change after initial reporting. [CDC's guidance on provisional mortality data](https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm).