We’ve long been obsessed with the super-elderly. How do some people make it to 100 or even 110 years old? Why do some regions — say, Sardinia, Italy, or Okinawa, Japan —produce dozens of these “supercentenarians” while other regions produce none? Is it genetics? Diet? Environmental factors? Long walks at dawn?
A new working paper released on bioRxiv, the open access site for prepublication biology papers, appears to have cleared up the mystery once and for all: It’s none of the above.
Instead, it looks like the majority of the supercentenarians (people who’ve reached the age of 110) in the United States are engaged in — intentional or unintentional — exaggeration.
The paper, by Saul Justin Newman of the Biological Data Science Institute at Australian National University, looked at something we often don’t give a second thought to: the state of official record-keeping.
Across the United States, the state recording of vital information — that is, reliable, accurate state record-keeping surrounding new births — was introduced in different states at different times. A century ago, many states didn’t have very good record-keeping in place. But that changed gradually over time in different places.
Newman looks at the introduction of birth certificates in various states and finds that “the state-specific introduction of birth certificates is associated with a 69-82% fall in the number of supercentenarian records.”
In other words, as soon as a state starts keeping good records of when people are born, there’s a 69 to 82 percent fall in the number of people who live to the age of 110. That suggests that of every 10 supposed supercentenarians, seven or eight of them are actually younger than that, but we just don’t know it because of poor record-keeping.
This doesn’t mean that any of these false supercentenarians are lying. It could be that they lost track of their age a long time ago, accidentally double-counted some years, or were told the wrong birth year. But it does mean that the majority of people claiming to be supercentenarians, born in areas that didn’t keep reliable, accurate birth records, are probably not quite as old as they say they are.
As a result, most of the studies we’ve conducted on them — trying to divine the secrets of old age from genetic tests and diet surveys — may be no good. But this isn’t just a funny little accident of old-age science: It actually illustrates a serious challenge in science.
Why we may have to question what we know about supercentenarians
The paper also looks at the phenomenon in Italy and Japan, where something different seems to be happening.
Italy keeps better vital statistics than the United States does, and has had reliable vital statistics across the country for hundreds of years — yet in Italy, too, there are clusters of the country where lots of supercentenarians pop up. Maybe the Italian supercentenarians are for real?
Newman’s analysis suggests not. He starts out by noticing something fishy: The parts of Italy that claim the most supercentenarians overall have high crime rates and low life expectancy. Isn’t that weird? Why would an area generally have low life expectancy but also produce an extremely disproportionate share of the world’s oldest people?
The same pattern repeats itself in Japan: Okinawa has the greatest density of super-old people, despite having one of the lowest life expectancies in the country and generally poor health outcomes.
The paper puts forward a controversial proposal. It seems unlikely that living in high-crime, low-life-expectancy areas is the thing that makes it likeliest to reach age 110. It seems likelier, the paper concludes, that many — perhaps even most — of the people claiming to reach age 110 are engaged in fraud or at least exaggeration. The paper gives a couple of examples of how this might come about; some of it might be reporting error, and some of the supercentenarians might be produced by pension fraud (someone might be claiming a dead person is still alive for pension benefits, or claiming the identity of a parent or older sibling).
Newman’s overall conclusion: “Remarkable age attainment is predicted by indicators of error and fraud,” and isn’t correlated with things like a healthy population of 80-year-olds or high-quality access to medical care. “As a result, these findings raise serious questions about the validity of an extensive body of research based on the remarkable reported ages of populations and individuals.”
In other words, all of our research into the biomarkers, habits, and diets that predict extreme old age? Probably worthless, because a significant share of the sample was not actually as old as we thought.
The paper still needs to undergo peer review, but if its findings hold, it does illustrate an interesting statistical phenomenon: When you’re looking for something exceptionally rare, your data set will be dominated by errors and false positives. For example, if you’re looking for a disease that affects only one in a million people, and your test for the disease is 99.99 percent accurate, then it’ll turn up 100 false positives for every true positive. Even though you used a highly accurate test, most of your “positives” don’t have the disease!
Similarly, supercentenarians are extremely rare. Only about one in 1,000 people who live to the age of 100 make it to 110. The vast majority of people would never impersonate their parent or older sibling for benefits, or forge a birth certificate, or participate in identity theft, or get confused about how old they even are. But if one in 1,000 people would do that, then fraudulent supercentenarians will be more common than bona fide supercentenarians. When you’re looking at an exceptionally rare phenomenon, you have to be exceptionally careful — or you’ll mostly find yourself studying something else entirely.
Sign up for the Future Perfect newsletter. Twice a week, you’ll get a roundup of ideas and solutions for tackling our biggest challenges: improving public health, decreasing human and animal suffering, easing catastrophic risks, and — to put it simply — getting better at doing good.