SARS-CoV-2 Coronavirus Case fatality Ratio

Driving the anxiety and uncertainty about the current outbreak of COVID-19 is the case fatality ratio (CFR) being thrown about carelessly by not just the press, but also WHO and other organizations. However the CFR is not a one-size-fits all, and is influenced by many factors.

The case fatality ratio for a disease is the number of fatalities divided by the number of confirmed infections. The latter number is typically obtained by laboratory diagnosis, in the current outbreak via RT-PCR, which detects viral nucleic acids. As of this writing this ‘crude’ global CFR for COVID-19 is 3214 deaths divided by 94,250 cases which is 3.4%. This crude CFR is high: for comparison, the CFR for seasonal influenza is 0.1%. However, as I will show below, this number is not a one-size-fits all, and is influenced by many factors. Please do not look at 3.4% as an indicator of your risk of dying from COVID-19!

As I’ve written previously, data from the COVID-19 outbreak in China we know that age has a substantial impact on CFR. Based on a summary of 44,672 cases earlier in the outbreak (published by China CDC Weekly), no child between 0-9 years of age died of infection; and in individuals less than 50 years of age, the CFR is less than 1%. Only in older individuals does the CFR increase.

We also know that the CFR varied within China based on geographical location and even when in the outbreak the ratio is calculated. The graph included below (from the joint WHO-China mission on COVID-19 report) shows that at the beginning of January, the CFR was much higher in Wuhan than in other areas of China. This skewed CFR is due to lack of hospital care for the many infected individuals. In areas of China other than Wuhan, patients with serious disease could obtain excellent hospital care. Furthermore, the CFR declined during the course of the outbreak in China. This trend has been attributed to the evolving standard of care during the outbreak.

The CFR also varies by country, likely due to differences in the standard of care. In South Korea there have been 28 deaths among 5,621 infected individuals, for a CFR of 0.4%! In Singapore there have been 110 cases with no deaths. In comparison, in Italy and in Iran the CFR so far is 3.1%. These differences are not due to circulation of different viruses in these countries, but a combination of the quality of health care available and the age of the patients.

So far there have been 9 deaths out of 128 confirmed cases in the United States of America, a fatality ratio of 7%. Many of these deaths have been in elderly individuals in Washington, which skews the CFR. Furthermore, it is obviously very early in this outbreak in the U.S. Given the overall good quality of health care in the U.S. I suspect that the actual number will be much lower. However, if the hospitals in the United States of America (or any other country) are overwhelmed, the CFR could be higher.

Finally, I do think that we are missing many infections, not just in China but in other countries, and therefore the CFR may be even lower. There are likely many individuals with mild respiratory symptoms (typical in the winter) who are infected but never diagnosed. Only well after this outbreak is over, when serological studies are done to determine the extent of infection, will we have an accurate CFR.

I am not attempting to minimize the seriousness of SARS-CoV-2 and COVID-19: there will certainly be many infections in the U.S. and there will be deaths due to the infection. However, please understand that a CFR of 3.4% does not apply to everyone; what happens to you if infected depends on your age, health status, and where you live.

#COVID19