Quantification of excess mortality related to COVID-19 in the US

Mortality is probably the most important concern in the Wuhan pandemic (aka COVID-19). An important question is whether there was any difference between the risk of mortality related to COVID-19 and the rate of mortality expected in the general population. In this note, I am going to use the US data to address that question.

The key idea is “relative survival“, a well-known concept in the literature of clinical epidemiology. Operationally, relative survival is defined as the observed survival in the study group (where all deaths from any cause are considered an event) divided by the expected survival in the background [general] population of the same age, sex and calendar period. However, in this note, I am going to use mortality instead of survival.

The idea is this: the total observed number of deaths can be attributed to two sources: one due to the virus, and one to other causes. The “other causes” here can be estimated by the population based mortality rate.

Now, we know the total number of deaths related to COVID-19 [from the Johns Hopkins University Dashboard [1], and the age-specific death rate among people infected with SARS-Cov-2 [2]. We also know the age-specific rate of mortality in the US population. Therefore, it is easy to quantify the number of deaths that could be attributable to COVID-19. My student and I have used this idea to estimate the excess mortality related to hip fracture [3].

In the figure below, the mortality rate among COVID-19 patients was — as expected — increased with advancing age. However, it is important to note that the COVID-19 mortality rate was consistently higher than that in the general population, and the difference was highest among those aged between 55 and 74 (with relative mortality ratio being 3.2 to 3.5). Even in the ‘younger’ age group (eg 45 – 54), the relative mortality ratio was as high as 2.7. These data suggest that the loss of life was more pronounced in the productive age groups.

The following table shows the age distribution of the number of confirmed cases, the number of deaths, and population-based mortality rate for 2018. Until April 13, the United States has recorded 550,016 cases infected with the virus; among whom, 21,733 deaths. Now, based on the population mortality rate, the expected number of deaths was 9,887. In other words, the excess mortality associated with the COVID-19 was 2.2 folds. This effect size is comparable to that of hip fracture and cancer.

Another way to understand the magnitude of COVID-19 in the US is that almost 12,000 people (the exact number is 11,846) have died prematurely as a result of the infection. Of course, this number is increasing as the pandemic is still going on.


[1] https://coronavirus.jhu.edu/map.html

[2] https://www.businessinsider.com.au/30-percent-us-coronavirus-cases-people-between-ages-20-44-2020-3?r=US&IR=T

[3] https://www.ncbi.nlm.nih.gov/pubmed/23684802

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s