On coronavirus and life days lost

My family is facing a conundrum.

My grandfather is 94 years old and due to strict lockdown policies in his assisted living facility, his quality of life has deteriorated. He’s not allowed to leave the building and can only interact with visitors through a pane of glass. Community meal service has been suspended and he must stay 6 feet away from other residents at all times. He can’t visit his ‘ladyfriend’ and can no longer volunteer as a mediator. He’s essentially imprisoned in his own home.

Should we stage a jailbreak or keep him there, safely out of harm from COVID-19? After all, aren’t these social isolation policies there for a reason?

Like all important questions, I want to answer it as objectively and intelligently as possible. This may quite literally be a life or death decision.

Fortunately, I have a background in statistics from running marketing research firms for over a decade, and have already explored similar questions in my previous posts. I even have the necessary data handy from my posts on mortality displacement and the age-adjusted fatality risk of COVID-19.

As a starting point, we have to consider how many precious years of life my 94 year old grandfather might have left. According to the Social Security Administration’s actuarial tables, he has just 3.61 years remaining (obviously this is just an average) which we wouldn’t want to squander.

Next we have to consider his risk of dying from COVID-19 should he be infected.

We don’t yet know for sure because the Infection Fatality Rate (IFR) of this virus is still an open question due to ascertainment bias in testing. However, a slew of recent population antibody studies seem to suggest its IFR is somewhere in the 0.1% to 0.5% range, and I believe towards the middle of that range in most geographies.

Sadly, we also need to factor in that elderly people like my grandfather face far worse COVID-19 fatality odds than young people. In a previous post I calculated the IFR of someone in the 85+ age bracket to be somewhere in the 1.5% to 7% range, which means up to 1 in 14 people my grandfather’s age might die if infected. Not good.

However if we combine these two competing factors, his relatively low life expectancy with his relatively high COVID-19 fatality risk, it leads us to an interesting conclusion… my grandfather should only expect to lose 20 to 98 days of life if infected with COVID-19

Here’s the simple math: 3.4 years life expectancy x 1.5% to 7% IFR x 365 days

Better yet, venturing out of his assisted living facility doesn’t guarantee him a COVID-19 infection, it just increases the risk (or perhaps not, but we’ll assume that it will for the purpose of this math exercise)

I can’t calculate what increase risk he may face as confidently as the numbers above, but can confidently say that every 1% below 100% infection risk reduces his overall COVID-19 fatality risk proportionately. So if we think he has a 50% chance of getting a COVID-19 infection “in the wild” then he would only expect to lose 10 to 49 days of life if we jailbreak him.

Whether he’s willing to risk losing these days of life in exchange for his freedom is a question only my grandfather can answer, but since he already served in and survived WWII, I think I know how he’ll answer it.

We should also consider that social isolation itself can be deadly, especially in old age, not to mention detrimental to life quality. I’m not factoring in this risk for the sake of simplicity, but it’s entirely possible that he could lose more life days from social isolation than the virus itself.

Since most of my readers aren’t in their 90’s, I’ll share a math table (see columns S to W in the Data Tables tab) where you can look up your own expected life days lost if you’re infected with COVID-19 at whatever IFR assumption you’d like. Here is the graphical version:

As you’ll see, people in their 80’s stand to lose the most. That’s because the virus is already quite fatal at that age yet they still have a decent number of expected life years ahead of them. (Ignore the notches in the graph, they’re due to fatality risk category artifacts that I couldn’t smooth out all the way)

I suspect that a lot of younger people will be surprised at how few life days they’d expect to lose if infected with COVID-19. Even though they have a lot of expected life years ahead of them to risk, their fatality risk from this virus is very low so the net effect is quite benign.

For example, I’m 42 years old and if I get infected with COVID-19, I’d expect to lose 3 to 17 days of life (yes, that’s days and not years). If my 6 year old niece is infected, she would expect to lose just 0.1 to 0.5 days of life… better expressed as 2.4 to 12 hours.

Personally, I’m comfortable with an expected loss of 3 to 17 days of life (perhaps 1.5 to 8.5 days with 50% infection odds) in exchange for living my other days and years to the fullest.

Of course, everyone should make their own risk/reward decisions, but no one should make them without a clear understanding of the odds.

Techie and investor. Founder at Rebel Fund and previously Pioneer Fund. Chairman of Infosurv/Intengo and CrowdMed (YC W13). Former Bain consultant. Data nerd.