"Micromorts": A New Way to Talk About Risks

Thinking in terms of micromorts--one chance in a million of dying each day--lets us see, for instance, that 230 miles in a car equals six on a motorcycle. 

Playing chess

A recent article in Wired surprised me with its pessimistic outlook about nuclear war in Europe. It said that one group of "superforecasters" put the likelihood of a nuclear weapon being detonated somewhere in Europe by April 30 of next year at 9.1%. 

While a detonation is a truly scary prospect, the article also conducted an analysis based on "micromorts." It says that a young person has a risk level of roughly one micromort -- one chance in a million of dying each day -- just by being alive. And the prospect of a nuclear detonation adds only one micromort per day to the risk for someone living in London, which is the same additional risk that comes from traveling 230 miles by car or six miles on a motorcycle. Going on a scuba diving trip adds five micromorts. 

Now, insurers are in the business of quantifying risk all the time for all manner of activities, and I'm not sure that the concept of micromorts need change anything about underwriting, but it strikes me that micromorts could be a useful way to communicate to the public both about what is, and isn't, risky behavior.

The Wired article used a lot of its real estate to explore the implications of nuclear attacks and to explain the methodology of superforecasters, but I can imagine much more prosaic calculations.

Lots of people seem to worry about being hit by lightning. Might it help to tell them that, with only roughly 20 people killed by lightning each year in the U.S., the danger amounts to just .00000000002 micromort for each of us? Ride six miles on a motorcycle, and you've incurred 50 billion times as many micromorts. And the chance of being killed by a shark is roughly 1/1,000th the micromorts of the risk from lightning.

I certainly found the estimate about scuba diving interesting, given that I came reasonably close to being a statistic while just minding my own business. My wife and I had done a somewhat aggressive dive at the Blue Hole about 50 miles off the coast of Belize, getting to a depth of 140 feet, which is a bit beyond what's considered recreational diving because we had to do a long decompression stop on the ascent. All was fine -- until the captain of the dive boat capsized it three miles offshore. The boat sank so fast that I, napping down below, had to climb out through a porthole. Our equipment went down with the boat, and we only rescued a few life vests for the nine divers and four crew, one of whom couldn't swim. The boat sank at dusk, and nobody was going to miss us soon enough to find us before morning, so we faced the prospect of swimming toward shore or hoping to tread water all night. Fortunately, some Rastafarians were out for a sail and spotted us. They had a dinghy attached to their homemade boat and ferried us in batches to land. 

Having done about 50 dives, I have my own sense of risk and wouldn't let a few micromorts influence me, but I still could imagine using the stat in conversation with my daughters, who are certified but have done just a handful of dives. (Of course, my main advice is to avoid incompetent Belizean captains who claim to be experts on the reef system but clearly aren't.)

Increasingly, insurers are trying to use data from wearables (steps, heart rate, etc.) and from scales (not just weight but body mass index, bone density and more) to steer people away from behaviors that put their health at risk. Why not use measures of micromorts to make sure people understand the risks associated with their activities, too? 

The measurements of micromorts aren't perfect, any more than the data from wearables is, but the concept of micromorts sounds to me like a good place to start.