How likely are you to die from flying in a plane, or taking the bus? What about getting your head stuck in a hole? We’ve evolved to be very attuned to physical threats - animal attacks, heights, and the deep ocean are some of the most common phobias. However, six million years of evolution did not prepare us for the modern world; We know logically that cars, planes, automatic machinery, and invisible illnesses are all dangerous, but we aren’t wired in the same way to be scared of them. Yet, as a society we’re still careful of these risks. We wear seatbelts, stay grounded in bad weather, use safety equipment, and wash our hands to mitigate those risks. We’ve been brought up to deal with these risks, despite them not being ingrained in our evolutionary history.

COVID-19 has a unique position, since it is a risk that is neither evolutionary nor is it ingrained in our culture. That makes it hard to position it among other life risks.

With that in mind, we wanted to determine the individual risk of COVID-19 to situate it in our daily lives. On a population scale, the risk is quite formidable - with a reported 37M cases and 623,000 deaths in 2020 in the US alone.

Despite these numbers, it’s really hard to contextualize the risk of Covid in a meaningful way. 37 million people infected is a little more than 11% of the population. How dangerous is that really? As it turns out, humans aren’t actually that good at being scared of things that kill them. More on this later.

COVID RISK VS CAR CRASH

We wanted to find an objective way of determining current COVID-19 risk. Our first thought was to situate COVID-19 by standardizing deaths per day, and comparing the mortality rate of the disease to another common risk: car crashes. Here’s a quick summary of the calculations we did and what we found for Acton, MA:

First, we calculated the frequency of both traffic crashes and crash fatalities. With a little math, we found that there were 1.17 days per crash event and 608 days per crash fatality.

Second, we calculate COVID-19 case and fatality rate. We found that there were 2.35 days per case in Acton, and 84.3 days per COVID-19 fatality in Acton.

Finally, we calculate the risk when vaccinated. According to Covid ActonNow, at the time of writing, 68.5% of people are vaccinated (thus 31.5% unvaccinated) in Middlesex County. We assume a vaccine efficacy of 90%, since variants are becoming present, and find that there are 472 days per COVID-19 fatality among vaccinated people.

According to these calculations, being a vaccinated person in Acton means that I am 11.3 times more likely to get in a crash than develop a case of COVID-19, but my risk of death from COVID-19 is about 1.3 times higher than it is from a crash.

QUANTIFYING RISK - MICROMORTS

Digging a little deeper into risk quantification, we found an easier way to standardize risk: the micromort. Ronald A. Howard, the pioneer of decision analysis, came up with the concept of the micromort (micro-mortality). A micromort is a unit of risk defined as a one-in-a-million chance of death. This is calculated from the population level. Meaning that if one person in every million who participates in an activity dies every year, the risk for that activity would be counted as one micromort.

We went ahead and calculated some micromort values:

Living in the USA (2018): 2.8M deaths out of 327M people 8,000 micromorts per year

Skydiving: 11 deaths per 2.8M jumps 4 micromorts per jump

Murder + Manslaughter in USA (2019): 14,014 deaths per 328M people 42 micromorts per year

Climbing everest: 223 deaths per 5879 attempts 38,000 micromorts per attempt

Back to our COVID calculations:

COVID-19 in USA (2020): 377,883 deaths per 331M people 1,141 micromorts per year

Fatal Car Accidents in USA(2019): 33,244 deaths per 328M people 101 micromorts per year

COVID-19 in Middlesex County, MA (2021): 448 deaths per 1.6M people 280 micromorts per year

Fatal Car Accidents in MA (2018): 355 deaths per 6.9M people 51 micromorts per year

Micromorts are a simple way to quantify risk for the population level, and allow for easy comparison between different activities . It also leads to some insights: skydiving once a week (208 micromorts a year) carries less risk than COVID-19 does in 2021. You’re more likely to die in a car crash than you are to get murdered (depending on where you live), and you’d have to live through 33 years of the pre-vaccine 2020 pandemic to reach the same risk level of climbing Mount Everest once.

Phobias

All of this led me to wonder: what is the relationship between fear and actual risk? How useful is our evolution in the modern world? Let’s start by looking at predators and common phobias.

An Ipsos poll found that 38% of people were scared to swim in the ocean due to fear of sharks. If you look at the numbers, you’re over a thousand times more likely to drown in a swimming pool (1.2 micromorts/year) than you are to die from a shark attack (0.001 micromorts/year). In fact, you’re more likely to be struck and killed by lightning (0.08 micromorts/year) than die from a shark attack.

Of course, these numbers aren’t adjusted for the population of people who are actually reasonably exposed to these risks (I expect more people swim in pools than swim in the ocean), but the comparison is still striking. People are afraid of water as well as sharks. Doing a little more digging, I found that somewhere between 10% and 45% of Americans are afraid of the water.

The most common phobias are, as you might expect by now, threats that are very tangible: animals, heights, flying, closed spaces, water, storms, and blood. In contrast, the leading cause of death in the USA under the age of 55, according to the CDC, is unintentional injury (278 micromorts/year). The most common of these being poisoning (including drug OD), motor accidents, falls, choking, and drowning. Strikingly, not many of our common phobias make the list. If you look at the population as a whole, heart disease (1979 micromorts/year) and cancer (1821 micromorts/year) greatly surpass all kinds of accidents.

The closest analogy for fear of various illnesses is probably mysophobia, the fear of germs. Despite disease being the biggest killer of humans, I wasn’t able to find concrete specifics on how many people are mysophobic, presumably because it’s a relatively rare condition.

This isn’t me telling you that you shouldn’t be afraid of accidents; I’m very squeamish around spiders, despite the fact that their risk is less than that of sharks (0.0007 micromorts/year), and that the spiders likely to kill you aren’t as common in the US as they are in other parts of the world. It’s natural (literally) to be fearful of those things. Years of evolution told us to be. The biggest take away from all this, for me, is that we aren’t very good at assessing risk objectively, and what we’re scared of doesn’t necessarily align with what is dangerous.

Wrapping up

Let’s get back to COVID-19. How do we situate this illness in our everyday lives? Our comfort level clearly is not a good analogy; we are disproportionately afraid of things that evolution tells us to be scared of, while our physiological responses are not caught up to the 21st century. Micromorts give us the tools to make these analyses. COVID is more dangerous than driving and skydiving, but less dangerous than heart disease and cancer. In 2021, it isn’t substantially riskier than many other things we do, but it cannot be discounted either. Moving forward, it seems the risk is no longer so large that caution about the disease is disproportionate to everything else in our lives. I think it is prudent to be careful and take the proper precautions to mitigate COVID-19 risk - just like we wear seatbelts or lifejackets, we should be aware of the dangers and follow the procedures by those whose jobs it is to determine the proper safety precautions.