Is there a mistake?
An Australian scientist suspected that the "plateau of aging" discovered this summer is a statistical error
According to his calculations, for the sample with which the researchers worked, literally one error is enough to draw a "plateau"
Polina Loseva, "The Attic"
There is no precise definition of aging as a process. Scientists often use Gompertz's law, which states: with age, the risk of dying increases exponentially. In one form or another, this pattern is true for a variety of species of living organisms. But many organisms also have a peculiar limit of aging: after reaching a certain age, the curve reaches a plateau, that is, the risk of dying further becomes constant.
Such a plateau was also found in humans. So, this summer, a group of scientists published in Science presented its results of observations from 2009 to 2015 for almost 4,000 super-long-livers (people over 105 years old), based on which it stated that the plateau is real: the increase in the risk of death slows down after 80, and starting from 105 years stops altogether. The authors of the article did not provide a biological explanation for this phenomenon – in other words, it is unclear what exactly changes in the human body at 105 years old. However, these demographic data suggest to us that aging – at least if we understand it purely statistically, as the dependence of the risk of dying on age – at some point "stops". This means that nothing prevents a person from living longer, keeping the risk of death constant.
But for every optimist there is always a pessimist. Australian scientist Saul Newman has found a vulnerability in such statistical calculations, as he writes in the journal PLoS Biology. In his article, he shows how errors of various kinds – incorrect dating of birth and death or an assessment of the subject's own age – affect the possibility of distorting the final aging curve. According to his calculations, one error per 10,000 is enough to significantly change the trajectory of the graph. Given that the observed super-long-lived Italians were born at the turn of the XIX-XX centuries, one can expect, Newman believes, that the percentage of errors in their documents is higher.
How errors in data "bend" the survival curve
As expected, the authors of the summer statistical article disagree with Newman: in their comments sent to the editorial office of PLoS Biology on the publication of the Australian, they vouch for the "purity" of the Italian registers and claim that Newman's estimates of the level of errors are "improbably overstated."
However, as far as the reality of the plateau phenomenon in general is concerned, Newman is not completely without optimism – at least in relation to other animals. Here in drosophila flies, he writes, the graph really reaches a plateau. Therefore, some kind of natural cessation of aging processes may take place, but we still have to check and explain this.
Portal "Eternal youth" http://vechnayamolodost.ru