Dear Straight Dope:
If you were to make a histogram of human life expectancy, I imagine it would start out somewhat high (infant/early childhood mortality), decline, and then gradually rise beyond age 40 before peaking and tapering down to zero after 110. My question is, excluding extremely long life spans, at what age is a person least likely to die?
This is a tougher question to answer than you might think. Not for nothing do they make actuaries, the people who figure out life insurance tables and such, take a series of qualifying examinations that would stump a particle physicist. But give me a minute and I’ll see what I can do.
“Life expectancy” in actuarial terms is a relative, not absolute, concept. It’s keyed to current age. For example, at age 15, having survived the diseases of childhood, an American male can expect to live another 60.1 years to age 75.1, while an American female can figure to live another 66.6 years to 81.6. At age 70, a man can expect to live another 11.9 years to age 81.9 while a woman can expect another 15.3 years to 85.3. In other words, a person who has already lived to age 70 has a much better chance of living to age 80 (hence a longer total life expectancy) than does a teenager.
A useful number to look at is mortality probability, which is usually expressed as “the chance that someone who has reached age n will die before reaching age n+1," usually expressed as qn. Here are some figures from the 1994 Group Annuity Mortality Table:
|Age n||Chance of Dying Before Age n+1|
These numbers have been subject to various adjustments that life insurance companies typically make, and the 1994 table, the most recent I have on hand, is now somewhat out of date. Still, we can make a few observations:
From ages 1 through 8, mortality decreases each year–the rate for a male dying in the year after attaining age 1 is .000430, after age 2 is .000357, etc. Childhood deaths tend to hit heaviest in year 0 (before reaching first birthday), and in fact in the first few days after birth. As the child survives, the risk of childhood death lessens slightly; up till age 9 or 10 (varies by male/female, etc.). At that point, the expected mortality curve kicks in, and at each birthday thereafter, the chance of dying the next year increases.
There is a significant change around age 40, as well. If you graph age vs. mortality rate, the curve stays fairly constant (although slightly increasing) until age 40-ish for males (slightly later for females), at which point the line begins to curve sharply upward and ever steeper.
Now to your question: at what age is a person least likely to die?
I can give two answers. For you personally, the least likelihood is death in the next year. That is, if you are already age 33, then the age at which you are least likely to die is 33 (assuming you are in good health, don’t work for a skydiving company with a sloppy safety record or engage in other perilous activities, etc.). If you are age 47, then that is the age at which you are least likely to die. Your best odds are always that you will survive the year–that’s true even if you are age 103.
OTOH, if you are asking at what age the rate of death is lowest for the population as a whole, the answer would be around ten, for both males and females.
Follow all that? Excellent. You just passed your first actuarial exam.
Send questions to Cecil via firstname.lastname@example.org.
STAFF REPORTS ARE WRITTEN BY THE STRAIGHT DOPE SCIENCE ADVISORY BOARD, CECIL'S ONLINE AUXILIARY. THOUGH THE SDSAB DOES ITS BEST, THESE COLUMNS ARE EDITED BY ED ZOTTI, NOT CECIL, SO ACCURACYWISE YOU'D BETTER KEEP YOUR FINGERS CROSSED.