Case fatality rates for COVID-19 are higher than case fatality rates for motor vehicle accidents for individuals over 40 years of age

The death toll of the COVID-19 pandemic has been unprecedented, due to both the high number of SARS-CoV-2 infections and the seriousness of the disease resulting from these infections. Here, we present mortality rates and case fatality rates for COVID-19 over the past year compared with other historic leading causes of death in the United States. Among the risk categories considered, COVID-19 is the third leading cause of death for individuals 40 years old and over, with an overall annual mortality rate of 325 deaths per 100K individuals, behind only cancer (385 deaths per 100K individuals) and heart disease (412 deaths per 100K individuals). In addition, for individuals 40 years old and over, the case fatality rate for COVID-19 is greater than the case fatality rate for motor vehicle accidents. In particular, for the age group 40-49, the relative case fatality rate of COVID-19 is 1.5 fold (95% CI: [1.3, 1.7]) that of a motor vehicle accident, demonstrating that SARS-CoV-2 infection may be significantly more dangerous than a car crash for this age group. For older adults, COVID-19 is even more dangerous, and the relative case fatality rate of COVID-19 is 29.4 fold (95% CI: [23.2, 35.7]) that of a motor vehicle accident for individuals over 80 years old. On the other hand, motor vehicle accidents have a 4.5 fold (95% CI: [3.9, 5.1]) greater relative case fatality rate compared to COVID-19 for the age group of 20-29 years. These results highlight the severity of the COVID-19 pandemic especially for adults above 40 years of age and underscore the need for large-scale preventative measures to mitigate risks for these populations. Given that FDA-authorized COVID-19 vaccines have now been validated by multiple studies for their outstanding real-world effectiveness and safety, vaccination of all individuals who are over 40 years of age is one of the most pressing public health priorities of our time.


Introduction
Here, we assess the health risks and disease burden for COVID-19 and other leading causes of death in the United States over the past year, including: flu, cancer, heart disease, motor vehicle accidents, suicide, and homicide. We report the overall mortality rates and case fatality rates for 10-year age groups of the population, and we compute the relative risks for each of the non-COVID-19 causes of death versus COVID-19. For each age group, we present the absolute and relative risks of COVID-19. By placing the mortality risk of COVID-19 in the context of other prominent causes of death, within each age group, this analysis provides a useful point of reference for the danger of COVID-19.

Data Sources
Publicly available data was used for this analysis. Population data was obtained from the US Census Bureau 7 . National-level COVID-19 data was obtained from the Centers for Disease Control and Prevention (CDC) public use case surveillance data 8 . We restricted our analysis to the CDC-reported cases and deaths for the year spanning from March 15, 2020 to March 14, 2021. The CDC WONDER public access database was used to get overall mortality risks for a number of familiar causes of death, including heart disease and drug overdoses 9 . The mortality databases as well as cancer data were queried. In total, data from 4 of the datasets were used:  11 . Overall mortality by motor vehicle accident was computed using the publicly available Fatality Analysis Reporting System (FARS) from the National Highway Traffic Safety Administration (NHTSA), which contains detailed information about all reported fatal accidents 12 . Case fatality rates (i.e. probability of dying conditioned on being in an accident) was derived from the publicly available NHTSA Crash Report Sampling System (CRSS) 13 . The Substance Abuse and Mental Health Services Administration (SAMHSA) National Survey on Drug Use and Health (NSDUH) was used to get survey-based age-specific estimates of annual drug misuse and misuse rates (including opioids and heroin) by age. Data was accessed through the survey's Public-use Data Analysis System 14 . This was joined with the CDC WONDER drug fatality data, and the fatality numbers were divided by the estimated users to estimate the "conditional risk" of heroin use/opioid misuse. Note that the broadest category of the survey is used for the denominator, so frequent drug users likely have higher risk. Estimated deaths, illnesses, and hospitalizations by age group for influenza in the United States for the 2017-18 and 2018-2019 flu seasons were obtained from a recent CDC report 15,16 .

Statistical analysis
For each age group and risk category, we computed the following measures: • Prevalence = (Number of cases) / (Total population), • Overall mortality = (Number of deaths) / (Total population), • Case fatality rate = (Number of deaths) / (Number of cases).
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 13, 2021 For the estimates of the overall mortality and case fatality rates, individuals with missing or unknown death status were considered to be alive. The age group specific prevalence and overall mortality rate for COVID-19 were calculated by scaling up the number of cases and deaths captured by the CDC public use case surveillance dataset to the total number of cases and deaths reported in the CDC COVID data tracker 1 during the study period (March 15, 2020 to March 14, 2021). Case fatality rates, along with standard errors, for motor vehicle accidents were computed using the survey package 17 in R in order to account for the complex survey design. Case fatality rates for individuals in 4-door sedans using no restraint during an accident were computed using the same method. For cancer, 5-year mortality was used in place of the case fatality rate. In particular, the 5-year mortality rates for cancer along with standard errors were estimated from individuals who were diagnosed with cancer between 2002-2016 using the Ederer I method 18,19 . To compare the overall mortality and case fatality rates between other risks and COVID-19, we computed the following relative risks: • Overall mortality relative risk = (Overall mortality for risk category) / (Overall mortality for COVID-19) • Case fatality rate relative risk = (Case fatality rate for risk category) / (Case fatality rate for  In addition, 95% confidence intervals for the relative risks were computed using the delta method 20 . Rates were considered to be significantly different if the 95% confidence interval for the relative risk values did not contain one. This corresponds to a statistical significance test with pvalue < 0.05.
We note that for influenza, estimates of case fatality and mortality data were only available for age groups of 0-4, 5-17, 18-49, 50-64, and 65+ years old. 16 In addition, no confidence intervals were reported for the overall mortality and case fatality rate estimates for influenza because only uncertainty intervals were provided from the original data sources.

COVID-19 is among the top-3 causes of death for individuals over 40 years old
Recently released statistics on the leading causes of death in the United States revealed that COVID-19 was the third most prevalent cause of death in 2020 (377,883 deaths, 11.3%), behind only Heart Disease and Cancer. 3 We have stratified the mortality rates (deaths by cause per 100k individuals) of a diverse set of leading causes of death (defined in Table S1): Heart Disease, Cancer, COVID-19, Cerebrovascular Disease, Suicide, Motor Vehicle Accidents, Firearm Homicide, and Drug Overdoses, within each 10-year age interval. This analysis reveals that the high overall mortality of COVID-19 is mainly driven by high mortality rates for individuals over the age of 40 years (Figure 1). COVID-19 (red dots in Figure 1) is consistently among the top three causes of death in individuals aged 40 years or older. While COVID-19 ranks third among the causes of mortality in both the 40-49 age range as well as the 80+ age range, the overall mortality for the 40-49 age group is 21 deaths per 100K individuals, while it is 1,028 deaths . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 13, 2021. ; https://doi.org/10.1101/2021.04.09.21255193 doi: medRxiv preprint per 100K individuals for the age group over 80 years old. These results show that most deaths from COVID-19 are among older age groups, which is consistent with previous studies 21,22 .

Drug overdoses, suicide, and motor vehicle accidents are among the leading causes of death for individuals younger than 40 years old
For individuals under the age of 40, the leading causes of death are primarily environmental and mental health-related (Figure 1, inset). Drug overdoses are the leading cause of death for the age groups 30-39 and 20-29 (37 deaths and 25 deaths per 100K individuals, respectively). Suicide is the second leading cause of death for the age groups 10-19 and 30-39 (6.6 deaths and 17 deaths per 100K individuals, respectively), and third for the age group 20-29 (17 deaths per 100K individuals). Motor vehicle accidents are the leading cause of death for the age group 10-19 (7 deaths per 100K individuals) and the second-leading cause of death for the age group 20-29 (17 deaths per 100K individuals). For all age groups below 40 years old, the overall mortality rate of COVID-19 is low and it declines with each decade of age. There are 13 deaths per 100K individuals due to COVID-19 for the age group 30-39 years old, and for the age group 0-9 years old the overall mortality rate is less than one death per 100K individuals.

COVID-19 has a higher case fatality rate than motor vehicle accidents for individuals over 40 years old
To further contextualize the risk of COVID-19 for individuals in varying age groups we analyzed the case fatality rate (CFR). This quantity is defined as the percentage of occurrences of a given event, such as a motor vehicle accident or contracting a disease, that result in death. Specifically, we compare the CFR of COVID-19 to that of motor vehicle accidents, specific cancers with high mortality, 23 and the 2018-2019 flu (Figure 2). Strikingly, while for individuals in the younger age groups the CFR of motor vehicle accidents is significantly higher than that of COVID-19, the opposite is true for the older age groups. The largest difference, as quantified by the ratio of the motor vehicle accident CFR and the COVID-19 CFR, is observed between the 10-19 age group and the 80+ age group. Individuals in the 10-19 age group involved in a motor vehicle accident have a 13-fold (95% CI: [10,17]) increased risk of death compared to individuals in the same age group who contract COVID-19 (Table S2). Conversely, individuals in the 80+ age group who contract COVID-19 have a 29-fold (95% CI: [24, 36]) increased risk of death compared to individuals in the same age group who are involved in a motor vehicle accident (Table S2). For individuals over 60 years old, the CFR of COVID-19 also surpasses that of prostate cancer (Figure 2), further emphasizing the dramatically increased risk of COVID-19 for older individuals. While the CFR for flu is also age dependent, it is significantly lower than the CFR for COVID-19 in comparable age-groups. Together, these data show that the risk of death upon contracting COVID-19 is greatly increased in older patients.

Discussion
Here we have compared the mortality rate and case fatality rate of COVID-19 to other leading causes of death 3 , in the United States, stratified by 10-year age-groups (Figure 1). We find that COVID-19 mortality rates show a strong age dependence, with COVID-19 mortality increasing from less than 1 death per 100K individuals, for individuals under 20 years old, to 1,900 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted April 13, 2021. ; https://doi.org/10.1101/2021.04.09.21255193 doi: medRxiv preprint deaths per 100K individuals, for individuals over 80 years old. This increase in mortality can largely be attributed to high CFR in age groups over 40 years (Figure 2), where the risk of death for individuals with COVID-19 is greater than that of same aged individuals involved in a motor vehicle accident.
This study also corroborates previous studies on the leading causes of death in the United States. A recent study 24 in December 2020 found that for the age group 85 years and older, the mortality rates for heart disease, COVID-19, and malignant neoplasms were 2,453 deaths, 1,070 deaths, and 1,044 deaths per 100K individuals respectively, which is similar to reported rates in this study for the age group 80 years and older (2,700 deaths, 1,900 deaths, and 1,500 deaths per 100K individuals, respectively). In addition, the case fatality rates for motor vehicle accidents reported here are consistent with data from the AAA Foundation's American Driving Survey 25 , given that the motor vehicle accident CFR can be approximated as: (Number of Fatal Crashes) / (Total Number of Crashes). Although less pronounced than in the case of COVID-19, pancreatic cancer, lung and bronchus cancer, motor vehicle accidents, and the flu also show a near monotonic increase in CFR with age. On the other hand, breast cancer, prostate cancer, and leukemia have local maxima in CFR for younger age groups, consistent with previous studies that identified younger patients have poor prognosis for these types of cancer 26,27 .
There were a number of data challenges which limited the analysis in this study. During the first few months of 2020, counts of COVID-19 reported cases and deaths may be lower than the actual values due to underreporting. Case fatality rate data was sparse for the non-COVID-19 risk categories considered, so we focused the analysis on those categories with data available (e.g. motor vehicle accidents, cancer subtypes, influenza). Outcome measures for different risk categories were available for earlier time periods (e.g. cancer data was taken from 2002-2016), while COVID-19 outcome measures were taken from the past year. As a result, the outcome measures for the non-COVID-19 risk categories may be slightly different during the study time period from the past year. For example, the case fatality rates for cancer may be overestimated due to advances in cancer treatment 28 .
There are multiple areas for future work. Here, we consider COVID-19 overall mortality and case fatality rates as fixed, but it would be interesting to see how these outcome measures have changed over time. For example, we may expect that the mortality rates for COVID-19 have declined as therapeutics such as antiviral treatments 29 and monoclonal antibodies 30,31 have become available. As more follow-up data becomes available, we can report mortality rates for COVID-19 for longer time horizons. For example, in several years, we could conduct a follow-up study comparing the 5 year mortality rates from long-COVID with the 5 year mortality rates from cancer. In addition, one limitation of this study is that we restricted the analysis to mortality rates in the United States only. It would be interesting to replicate this analysis on data from developing countries which may have significantly different leading causes of death and mortality rates for COVID-19 and other risk categories.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted April 13, 2021. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.  Comparison of causes of death per 100k people in a given age group, across all age groups. To better visualize mortality rates in the low-mortality age groups (ages 0 -39), we have added a zoomed-in insert.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.  . Vertical bars show 95% confidence intervals, and the shaded region is a visual guide to indicate CFR values lower than COVID-19 CFR, per age group. Datapoints are not shown if a cause of death has a mortality of less than 0.1 deaths/100k individuals within an age group. Flu data is shown as horizontal lines due to flu data being stratified in different age groups 16 than the other data shown. This data shows that for individuals in the age groups 40-49 years old and older, contracting COVID-19 results in a higher CFR than getting involved in a motor vehicle accident.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted April 13, 2021. For these causes, we did not obtain additional data in order to estimate prevalence or risk-of-death, but we did include their more easily available overall mortality measure.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted April 13, 2021  . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted April 13, 2021. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted April 13, 2021 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 13, 2021. ; https://doi.org/10.1101/2021.04.09.21255193 doi: medRxiv preprint Supplementary Table 3: Estimated case fatality risk and overall mortality for flu stratified by age groups, along with estimated relative risks compared to COVID-19. This is a continuation of Supplementary Table 2 for the case of flu for the age groups 0-4, 5-17, 18-49, 50-64, and 65+ years old as reported by the CDC. In the first five rows, we provide data for the 2017-18 flu season, and in the last five rows we provide data for the 2018-19 flu season. 15 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 13, 2021. ; https://doi.org/10.1101/2021.04.09.21255193 doi: medRxiv preprint