COVID-19 Age-Related Increase in Mortality Rates

At the time of publishing this article, 82% of COVID-19 fatalities in Canada [1, 2] and an estimated one-half of the fatalities in the U.S. [3] have been in nursing homes due to the frailty of residents and the lack of timely testing and adequate protective gear.
In the U.S. there have been so far 1,435,098 confirmed and probable cases of COVID-19 [4] and I decided to investigate, based on the latest available statistical data in the US, whether being older, in and of itself (without considering the effects of pre-existing conditions), puts someone at a greater risk of dying from COVID-19 as compared to the age-related increased risk of dying [5] without COVID-19. More older people die of COVID-19 than younger people, but how does the difference compare to what we would have expected without COVID-19?

I will start with a brief description of how the analysis was done, followed by the results and conclusions, and finally I will provide detailed step-by-step analysis.

How the analysis was done:

To perform the analysis, I have downloaded age-specific data (COVID-19 deaths, deaths from all causes, population size), from the CDC website (Centers for Disease Control and Prevention).
I have used a definition of "death-rate" as the number of people who die divided by the population size (also known as Population Mortality Rate).
To find how the death-rates (or probability of dying) changes with age, I used (for both COVID-19 deaths and non-COVID-19 deaths) the age-group of 15 to 24-year-olds as a benchmark, and compared the death-rates of older age-groups to that of their respective 15-24 years old group. For example, the death-rate (or probability of dying) of the 55-64 population from non-COVID-19 causes (e.g. heart disease, cancer, and any other cause except COVID-19) is 14 times that of the 15-24 population (see table 4 below).
Finally, I compared, for each age-group, the COVID-19 increase in death-rate to the "expected" (i.e. non-COVID-19) increase in death-rate.

Results:

The following table (Table 4, copied from the bottom of this page) displays, for each age-group, and using the 15-24 group as a benchmark, the increase in death-rates from COVID-19 and non-COVID-19 causes, and their ratio.

Table 4

Conclusion:

Using the death-rate of 15-24-year-olds (either from COVID-19 or non-COVID-19 causes) to measure the age-related increased death-rate of older groups, it was found that the ratio does generally become worse with age, and specifically, that the COVID-19 death-rates of people in the 45-54 group, or older, is 9.7 to 11.5 times what we would have "expected" from non-COVID-19 causes. Most startlingly, the probability of COVID-19 death for the 85+ years old group, is 2,169 times that of the 15-24 group.

Analysis Step-by-Step:

1. I downloaded from the CDC website [6] a data file (CSV format) that contained the number of COVID-19 deaths, number of deaths from all causes, and population size, for the various age-groups. The number of deaths is for the period of January 26th to April 18th, 2020:

Table 1: Original data from the CDC website

Note: The CDC has stopped releasing data of COVID-19 deaths by age-groups, and the information is no longer available on its website.

2. I calculated, for each age-group, the number of non-COVID-19 deaths by subtracting the COVID-19 deaths from the deaths from all causes:

Table 2: Non-COVID-19 deaths

3. I calculated, for each age-group, the death-rate by dividing the "number of people who die" by the "population size." I did it for both COVID-19 and non-COVID-19 deaths:

Table 3: COVID-19 and non-COVID-19 death-rates

Note: The letter "e" represents power of 10 (scientific notation). For example, 1.23e-2 equals 0.0123 .

4. Using the 15-24-year-olds group as a benchmark, I calculated the increase in death-rates of older age-groups as compared to the 15-24 group, by dividing their respective death-rate by the death-rate of the 15-24 group. I did it for both COVID-19 and non-COVID-19 deaths. To find the ratio between the COVID-19 increase in death-rate and the "expected" (i.e. non-COVID-19) increase in death-rate, I divided the two numbers:

Table 4: Using the 15-24 group as a benchmark, the increase in death-rates of older age-groups from COVID-19 and non-COVID-19 causes, and their ratio