Probability/Chance Question

[cloves2020]

Hi everyone, I apologize as my rusty brain has forgotten how to figure something out and was hoping to get some help

I am trying to figure out the chance of passing away if you get sick from a virus. So we have X amount of recovered in our state and Y who have died. So if one gets sick, what is the probability of it being terminal? Probability came to mind, I was going to use the total population but using actual people who are sick made more sense. Thoughts?

 

[Dr.Peterson]

There are various probabilities you could calculate, depending on what you use as your denominator. If what you want is the probability of dying if you get the virus, then you want to divide the total who have died over the total who have had the virus. But since those who are currently sick may or may not die, you shouldn't include them. I'd divide the number who have died over the sum of those who have died and those who have recovered (that is, the number whose cases have been resolved). This, depending on where the numbers come from, is the "case fatality rate" or the "infection fatality rate".

But these numbers keep changing; early in the pandemic, relatively few had resolved, so this calculation was not very accurate. And even now, the rate varies considerably depending on age and other conditions. Perhaps even more important, calculations vary depending on whether you count (in both numerator and denominator) everyone who actually had the virus (unknowable), or every who was tested and positive, or everyone who had symptoms, ... .

Here is one article (from a month ago) about some of these issues: https://www.nytimes.com/2020/04/17/us/coronavirus-death-rate.html

For many details (updated recently), see: https://www.worldometers.info/coronavirus/coronavirus-death-rate/

For comparison of several different rates, see https://en.wikipedia.org/wiki/Case_fatality_rate#Terminology

 

[cloves2020]

There are various probabilities you could calculate, depending on what you use as your denominator. If what you want is the probability of dying if you get the virus, then you want to divide the total who have died over the total who have had the virus. But since those who are currently sick may or may not die, you shouldn't include them. I'd divide the number who have died over the sum of those who have died and those who have recovered (that is, the number whose cases have been resolved). This, depending on where the numbers come from, is the "case fatality rate" or the "infection fatality rate".

But these numbers keep changing; early in the pandemic, relatively few had resolved, so this calculation was not very accurate. And even now, the rate varies considerably depending on age and other conditions. Perhaps even more important, calculations vary depending on whether you count (in both numerator and denominator) everyone who actually had the virus (unknowable), or every who was tested and positive, or everyone who had symptoms, ... .

Here is one article (from a month ago) about some of these issues: https://www.nytimes.com/2020/04/17/us/coronavirus-death-rate.html

For many details (updated recently), see: https://www.worldometers.info/coronavirus/coronavirus-death-rate/

For comparison of several different rates, see https://en.wikipedia.org/wiki/Case_fatality_rate#Terminology

Thanks Dr Peterson! It's a moving target for sure and a lot of variables as you mentioned. We will never fully know how many people actually have had the virus either since the country is so large (329 million aprox). I don't think we will ever see 329 million tests. I am going to try to narrow it down to a few states that have been hardest hit. Thanks for the direction!