Scientists Push Back Against Controversial Paper Claiming a Limit To Human Lifespans – Gizmodo
Humans don’t like dying, they don’t like the idea of dying, and most have made not dying an important part of their life. Lots of folks are interested in making us not die for longer, so it was a real bummer last year when a team of researchers said that the maximum human lifespan has plateaued at around 115 years of age. Some folks might live to be older, but those oldies are outliers.
When the scientists behind this idea published their research in Nature last October, it sparked a whole lot of press coverage. It also brought debate, accusations that the study was flawed, and questions as to whether it was based on enough data. Today, the journal Nature is publishing five rebuttals from researchers who have problems with the original study—who think that a harder look at the data is warranted, and that the authors’ original conclusions might be incorrect.
One researcher who reviewed the first Nature paper thinks the controversy misses the point. “The authors of the rebuttals quibble about how to deal with the mathematics of small numbers at extreme old age,” S. Jay Olshansky from the School of Public Health University of Illinois at Chicago wrote in an email to Gizmodo. One of the points they fail to realize, he thought, was “if only they would look up long enough from their mathematical formulas attempting to model trends in small numbers, they would realize that… death always occurs, and it does so in a consistent fashion in humans because there is a limit to the duration of life.”
Essentially, last year, researchers from Albert Einstein College of Medicine in New York collected the ages of the single oldest people to die in a given year in the United States, France, Japan, and the United Kingdom, based on information in the International Database on Longevity. When they crunched all the data, it appeared to them that the maximum reported age of death increased until the 1990s, and has plateaued since then, averaging out at 115. They did other analyses looking at the second through fifth oldest ages at death, and added data from other sources. The paper’s authors concluded we may be hovering around the limit to human longevity.
What followed, naturally, was loud press coverage. The study’s principal investigator Jan Vijg told me that he didn’t mean for it to sound like there was an absolute limit to human longevity, but rather that we’d reached a limit that advancements in his own field, the genetics of aging, or medicine might eventually surpass. “I can never rule out that we’ll see this ceiling broken,” he told Gizmodo. “Maybe we can be successful in generating new drugs that work against diseases. Work against the aging process overall deserves way more publicity.”
Others didn’t see it that way. All of the popular press “allows people to say that Uncle Jack can’t live past 115. That was the impression everybody got. There’s a limit to how much you can say you didn’t mean to give the impression that” there’s a hard age limit, one of the rebuttals authors, Nick Brown from the University Medical Center Groningen in the Netherlands told Gizmodo. “It seemed to me that the people didn’t try hard to correct that impression.”
“I can never rule out that we’ll see this ceiling broken”
So Brown’s team, as well as four other teams, re-analyzed the Nature paper and found lots of problems. Brown’s own analysis found that the existence of the plateau at 115 years depends on the age and death date of the oldest person ever, France’s Jeanne Louise Calment, who lived to be 122. Others mentioned the increase in the number of people living past 100 will make it more likely to see more folks live past 115, or even 122. Still others found problems with the statistics and methods used to analyze such small sample sizes or argued that we don’t have enough data to be sure. One paper noted that breaking the data up into individual years that people died in is arbitrary, since years are an arbitrary division of time. Maybe the presence of a 25-year plateau is itself a statistical fluctuation.