The longest a human has ever lived — that we know of — is 122 years, a record set by a Frenchwoman called Jeanne Calment. She died in 1997, and no one has outlived her despite considerable advances in medicine and living standards since then. Some have seen this as a sign that we may have reached the absolute limit of human longevity.
However, a new study suggests that the human lifespan has not reached its ceiling, but could rather be in a temporary plateau phase with an uptick in lifespan expected once today’s 60- and 70-year-olds advance in age past 100. As such, Calment’s record might not last very long, according to the two authors, David McCarthy of the University of Georgia and Po-Lin Wang of the University of Southern Florida.
Table of contents
Probabilities and lifespan: not just a game of dice
McCarthy had previously developed mathematical models that estimate mortality parameters using a Bayesian approach with important applications in the insurance industry and pension funds. But he soon realized that the same model could be applied to explaining past extreme longevity as well.
Bayesian analysis is a statistical method that allows scientists to combine prior information about a population parameter with evidence from just a sample of the population to make inferences about the larger group. In more relatable terms, the Bayesian approach is a way of thinking about how we can update our beliefs or knowledge based on new evidence.
Imagine you’re trying to figure out the probability of something, like whether it will rain tomorrow. You might start with a prior belief, which is what you currently think is the probability of rain based on past experience or information, such as the current season or whether or not it rained for the past week.
Now, let’s say you look at the weather forecast and see that there’s a 50% chance of rain tomorrow. The Bayesian approach says that you should update your prior belief based on this new evidence. Specifically, you should adjust your belief towards the new information you received, by taking into account both your prior belief and the new evidence.
The age of mortality postponement
The researchers analyzed historical and current population mortality data in 19 currently-industrialized countries to understand whether the human lifespan is reaching a maximum limit or not. Bayesian statistics were applied simultaneously on a large number of parameters, resulting in a tremendous amount of data that required weeks to crunch with the supercomputer at the University of Georgia.
The researchers also applied a variant of the Gompertz mortality law to this research, which describes the exponential increase in mortality rates with age.
“The Gompertz law dates back to the early 19th century. It states that after a certain age, annual mortality probabilities increase at a constant percentage with each year of age. For instance, in the US, we estimate that male mortality probabilities increase at roughly 8% for each year of age after age 50. This means that with each decade of life, mortality probabilities approximately double,” McCarthy told ZME Science in an e-mail.
The researchers found that while the dominant historical pattern has been one of mortality compression, where the maximum age does not change but more people reach older ages, there have been occasional episodes of mortality postponement, where the maximum attainable age seems to have risen.
“We seem to be in one of these episodes of mortality postponement,” McCarthy says.
“We show that the Gompertz law, despite its simplicity and age, provides a very good fit to historical mortality data. For instance, in cohorts born in 1900, the Gompertz law explains something like 99.5% of the variation in mortality rates between ages 50 and 100 in our median cohort. Other researchers often believe that it doesn’t provide a good fit after age 80. We show that when you fit the Gompertz law to birth cohorts (ie. you look at how the mortality of the same group of individuals changes as they age), it fits very well,” the scientist added.
This means that while some samples of the population born between around 1900 and 1950 are experiencing historically unprecedented mortality postponement, they are still too young to break longevity records. In other words, more people are living longer than before, but we may not have reached the absolute limit of human lifespan yet.
“We show that what we call the Gompertzian maximum age – that is, the age at which we assume that mortality rates stop increasing at a constant percentage by year of age – did not change for long periods. For instance, we show that Swedish males reached an annual mortality probability of around 50% at around age 100 whether they were born in 1780 or 1900. But for cohorts born after 1900, this historical pattern appears to have changed dramatically, and there appear to be significant increases at the age at which individuals reach these very high mortality probabilities on the horizon,” McCarthy said.
This finding suggests that there is still potential for longevity records to rise by the year 2060 as younger cohorts reach advanced old age.
There’s a 50/50 chance someone in Japan will turn 130 in the future
So where does this leave us? That’s still difficult to say. When asked to estimate the maximum age that will be reached this century, McCarthy couldn’t come up with a straightforward answer due to the amount of uncertainty involved, but the analysis lends confidence that Calment’s record, which has persisted for more than 25 years, could be shattered soon enough. However, it is important to note that this latest study does not provide a definitive answer on whether there is a maximum limit to the length of life or not.
“Our method only provides estimates we regard as reasonably reliable up to cohorts born in around 1950. But The model suggests that the oldest Japanese woman born in 1940 has a 50% chance of living past 130. Of course, this depends on whether our model is an accurate description of how old-age morality will change, and whether there is a stable economic, political and environmental environment that continues to support extreme longevity. But our model fits past data extremely well, and so we believe that it provides a good basis for projection. But projections remain just that: projections,” the researcher told me.
The notion that we’re still far from hitting a ceiling in human longevity is supported by other studies. For instance, when Stanford biologists analyzed birth and mortality data for individuals aged 65 or older from 1960 to 2010, they found that the average age of death for people older than 65 increased by three years every 25-year period or every generation. This means that people can generally expect to live six years longer than their grandparents, on average.
Consider the fact that Calment along with the other runner-ups (the second oldest person that ever lived was 119 years old) were virtually all born in the 19th century. Then consider the key roles that nutrition, epigenetics, access to healthcare, and other environmental factors — all of which have improved tremendously in the last century — play in longevity. It now seems quite likely that humans could live past age 130.
The new challenges of pushing lifespan to its limits
And while the limit of the human span is currently murky, what is rather certain is that an increasingly sizable proportion of the population will be comprised of the elderly. By one estimate, the population of adults 85 and older is projected to increase 351 percent by 2050, while those older than 100 will increase 10-fold between 2010 and 2050.
This means that people will have the opportunity to experience life with their families and loved ones for longer than ever before. But this also comes with many challenges, such as how to support a rapidly aging population that strains the healthcare system and, perhaps more importantly, the pension and social security systems. It’s no wonder then that the Stanford Center on Longevity forecasts that those who might live to 100 could expect to work for 60 years or more — two extra decades than people work during their lifetimes on average today.
“Longer lifespans also have implications for how people will reallocate work across their lives, possibly starting work later (allowing them to get more education), and ending work later. It will also change the structure of families, where the prospect of multiple generations alive all at once raises both opportunities and challenges. In the long run, if large numbers of people begin to reach old age, it would be difficult to understate the societal and individual implications of this change,” McCarthy says.
The new study appeared today in the journal PLoS One.