May 1, 2020
Before we talk about your mortality, let’s review the four topics so far in the “for dummies” series. We now know about herd immunity (the time when COVID will no longer be able to freely spread in the population), about R0 (a way to communicate the contagiousness of COVID), about prevalence (the number of people that have or have had the illness). Finally, we discussed the fact that we simply don’t know the mortality (the chance you will die from COVID in a period of time) or the true case fatality rate (the chance you will die if you get COVID) because we don’t know the number of asymptomatic cases or the exact number of deaths associated with COVID. Now let’s consider YOUR mortality.
Below, I will not consider health care capacity, the morbidity of COVID, or the idea that self-isolation is a spectrum. It’s a hard enough topic already without introducing those variables.
I ended yesterday with the one fact I feel very confident will hold up in time: the older you are and the more medical problems you have the higher your mortality will be. I then said if you are in a group that has a higher case fatality rate (ie you are older and have medical problems), it is vital that you not get the virus until after we develop better treatments or a vaccine.
Well, what is your mortality and is it high? Let’s see if we can give an approximate case fatality rate (mortality for this conversation) with the best data we have. Here are three sources, one, two, and three. They all have similar numbers.
|Wordometers||CDC Feb 12 - Mar 16 US|
|80+ years old 14.8%||≥85 (144) 10.4–27.3|
|70-79 years old 8.0%||75–84 (210) 4.3–10.5|
|60-69 years old 3.6%||65–74 (409) 2.7–4.9|
|50-59 years old 1.3%||55–64 (429) 1.4–2.6|
|40-49 years old 0.4%||45–54 (429) 0.5–0.8|
|30-39 years old 0.2%|
|20-29 years old 0.2%||20–44 (705) 0.1–0.2|
|10-19 years old 0.2%||0–19 (123) 0|
|0-9 years old 0%|
Lancet based on 1,023 Chinese cases
Now keep in mind, these numbers are likely going to do down once we understand the number of asymptomatic cases. I know yesterday I referenced an article where these numbers may be under representing death as well. So we really don’t know. But, if I had to guess right now, the case fatality rate will drop. You can adjust these numbers up or down based on your own health or other data (please share with me any other data set you have found) or if you simply want to already adjust for the presumed huge amounts of asymptomatic cases. I’m personally going with the current data and will adjust my odds ratios below accordingly.
Now, let’s figure out your chance of dying in the next year not considering COVID. This data is well known because of the actuarial tables from insurance companies. So go here and put your age in.
I’m 44. So, my chance of dying in the next year is 0.28%. From the tables above, currently, the best guess of my case fatality rate with COVID is 0.4%.
So, if I get COVID, the chance of me dying goes up about 240% (0.68% / 0.28% = 2.4).
Another way to look at this is I go from a 1 in 357 (100/.28 = 357) chance of dying, if I don’t get COVID, to a 1 in 147 (100/(.28+.4) = 147) chance of dying with COVID.
Note, I added the two percentages there. I did that because COVID does not eliminate the other chances of dying. Sure, we could make an argument that I am driving less and trauma is a major cause of death for a 44 yr old, but let’s keep this simple.
Let’s do the same scenario for a 75 yr old who has a 3.63% chance of dying in one year. Let’s go with 7.5% as the case fatality rate based on the above data.
So, if a 75 yr old gets COVID, the chance of them dying goes up about 300% (11.13 / 3.63% = 3.06 ).
Another way to look at it is they go from a 1 in 28 (100 / 3.63 = 27.5%) chance of dying if they don’t get COVID, to a 1 in 9 (100/(7.5 + 3.63) = 9) chance of dying with COVID.
Now, I haven’t told you what you should do. That is ultimately your decision and will be related to YOUR risk tolerance. Since starting to practice medicine, I have gained an appreciation that people’s risk tolerances vary widely. When I’m talking to patients now about being admitted to the hospital for low-risk chest pain, I may simply let them know that there is say a 2% chance their chest pain is from heart disease. Some people are relieved to hear that percentage and some people are terrified to hear that percentage.
Now you start to understand why I asserted in one of my very first letters that you have two questions to ask yourself.
One, how important is it to do that which you’re leaving your house to do?
Two, what is your risk of having a bad outcome if you get COVID?
Once you can answer those two questions, based on your risk tolerance, you will likely find your answer about how strict you should be with your own personal social distancing practices.
For instance, two examples regarding heading out into the world and interacting with other humans…
Are you going to work? And if you don’t go to work you can’t eat? Or you can’t buy critical medications?
Or, are you going out to eat with friends? And life would go on if you don’t go?
For instance, two examples regarding the impact of getting COVID…
Are you young and the chance of a bad outcome with a COIVD infection is very low?
Or, are you 80 yrs old and have heart disease and the chance of a bad outcome is very high?
Keep in mind, if you are isolating with others you need to make sure you consider the worst-case fatality rate of all the people you are isolating with. This is why it’s awful for teenagers. They have an almost zero chance of dying of COVID and are old enough to leave the house by themselves but are living with people with a much higher chance of a bad outcome if they get COVID. Sorry, that is the way the cornbread crumbs on this one. I didn’t make up the science.
OK but for how long? Well, if you can make a living at home and don’t mind the quality of life, then you may answer forever. Or you may say, I’ll do this until the case fatality rate is below a certain number.
Let’s walk through a few scenarios. Let’s say we ultimately are able to reduce the case fatality rate by 50% by the end of the summer. Let’s take out your normal risk of death and just look at the risk of death from COVID. How important will that be for you?
If you had a 0.2% chance of dying from COVID, you now have a 0.1% chance. So you went from 1 in 500 to a 1 in 1,000 chance of dying from the disease. You might be saying that’s not that big of a deal. I went from a really small number to an even smaller one.
If you had a 15% chance of dying from an illness, you now have a 7.5% chance. So you went from a 1 in 7 to a 1 in 13 chance. That change in risk may be more meaningful to you.
Or, what if a vaccine took you to a 1:100,000 or 1:1,000,000 chance of dying? Would you self-isolate to hold out for those numbers?
I know this is a long one. I’m sorry. I wanted to provide enough context to explain how I am thinking through the decision with my family. I hope I provided you a rational way to think and talk about how much you should personally self-isolate. One not simply based on fear and emotion. This allows a real conversation about your individual values, risk tolerance, and personal circumstances to help you determine how rigorously or leniently to self-isolate.
Now of course, if people don’t follow the guidelines and we start heading back into an exponential growth situation that will outstrip healthcare capacity, we are back in a situation where shelter at home may again be necessary to avoid a lot of unnecessary death. Let’s hope people take a sensible approach to the easing of restrictions. We don't want to slide backward.
If you are not working this weekend, have a relaxing one and until next time,
Stay emotionally connected and physically distant,
PS: Now with all your super-duper epidemiologic knowledge on pandemics and COVID, you can follow this great conversation with the epidemiologist Jeff Shaman PhD. It’s from the TWiV podcast that I previously mentioned. I listened last night and he was able to provide a lot more detail and nuance to many of these "for dummies" topics in a one-hour conversation than I can in these letters.