Intimations of mortality
The past few weeks have been notable for the number of people I know who have died or are about to. A colleague yesterday recollected having heard that there’s a peak time of death in the early 50s and that if you live past that, then you’re likely to live for a relatively long time. I decided to check it out as best I could. This web site at the CDC has quite a bit of data, although not exactly what I was looking for. The graph (click on the thumbnail to see a bigger version) shows the number of deaths reported in the U.S. in 2002 for each year-cohort. It’s not corrected by the number of people of that age (and therefore at risk of dying that year), which is obviously the source for the drop in deaths after about age 80.
It’s interesting that the number of deaths of infants less than 1 year old (28034) isn’t exceeded until the age 63 year cohort (28551).
I thought there might be an anniversary effect, in which people hold out until some round number age, but 80 is the only year where there seems to be evidence of that. There is an apparent jump in deaths at age 55 and 59. The most striking change occurs at a point where there are relatively few deaths. At age 16, when American children start driving, the number of deaths goes up by 65% in one year, which is very striking (although then it keeps going up to a higher plateau around age 20.
There is a bit of a dip after age 55, but I wonder how stable that is—it’s a drop of about 13% after an increase of 9% the previous year.
Of course, without knowing the number of people at risk of dying in a given year-cohort, these data aren’t that meaningful.