Why biological clocks are confusing our ‘real age’ – and how artificial intelligence could help us

When I first started writing about aging years ago, there was talk of something called the biological clock, also known as the aging clock or measuring “true age.” Basically, they’re pretty simple: we all have a chronological age, the number of years since birth, but that doesn’t necessarily reflect how far down the slope we are from birth to decrepitude.

On average, this is a fairly predictable trajectory, with a gradual decline in almost all physical and mental traits over the course of adulthood. When we judge how old someone is, we intuitively pick up on many of these signs we see—wrinkles and gray hair, or changes in posture, gait, voice, mental acuity, and so on.

The goal of measuring biological age is to capture this decline in a single metric, scientifically evaluated and expressed in years. The results tell us something we know intuitively: some people age better than others.

Most people are biologically within a few years on either side of their chronological age. But the two can differ wildly. A fifty-six-year-old (which is me) may have the biological age of a typical thirty-something (which I almost certainly do not), while another may have the biology of a seventy-year-old (similarly). Crucially, biological age may rise more slowly than chronological age and may even decline.

Biological age is a useful measure. It can provide individuals with concise, easy-to-understand information about their overall health, encourage them to make lifestyle changes, and tell them whether any resulting interventions, such as diet and exercise, are working. Judging by the number of commercial companies offering biological age testing, there is considerable hype for such information, even if it is expensive.

It’s a useful tool for scientists testing anti-aging interventions to see what works and what doesn’t, without having to wait years to see if their guinea pigs (human or otherwise) decline and die at different rates. And for those concerned with the basic biology of aging, measuring biological age can help us understand what happens in our bodies as we age.

So what’s not to like? Quite a lot, as usual. Biological age is sound in principle, but leaves a lot to be desired in practice.

The first biological clocks were based on epigenetic markers. These are molecular markers added or removed from nuclear DNA that influence patterns of gene expression. About ten years ago, scientists led Fr Steve Horvath — the father of the biological clock based at the University of California, Los Angeles — found that while there is a lot of individual variation, epigenetic markers change predictably over the average lifespan. Measure the right ones, input the data through a complex algorithm, and soon an estimate of someone’s biological age will appear.

But epigenetics is not the only way to make an estimate. In recent years, many other clocks have been developed based on various other biological markers, including blood proteins, the length of DNA caps at the ends of chromosomes called telomeres, urine metabolites, facial imaging and chest X-rays. This wouldn’t be a problem if everyone came up with roughly the same answer, but they don’t.

As one example we can see this in a recent analysis of a human clinical trial called CALERIEwhich examined the impact of long-term caloric restriction—a proven antiaging intervention in many organisms, although whether it applies to us remains up in the air. The CALERIE study used five different clocks of aging on 220 adults. Two of the watches showed a significant reduction in biological age among the calorie-restricted participants. Three of them don’t. What are we to believe? This is a problem that plagues anyone—whether individual or scientist—who uses an aging clock.

Another problem with aging watches is the illusion of accuracy. Most spit out a single number with no error bars, despite the inherent levels of uncertainty in the data and analyses. According to a recent paper in the diary npj Agingthat’s just the tip of the iceberg. Overall, the researchers concluded that existing clocks don’t do what they say on the tin and risk instilling in people undue confidence or unnecessary anxiety about the state of their health.

Does this mean that aging clocks are useless? Not quite. The authors of the article headed by Dmitry Kriukov at the Skolkovo Institute of Science and Technology in Russia say that “all limitations of the aging clock are hypothetically solvable”. But whether it is worth solving is another question.

This is partly because of a new and very promising approach that is coming. The current aging clock needs biological samples. The new approach doesn’t, instead relying on—you guessed it—artificial intelligence, specifically something called large health models (LHMs). These are essentially large language models—like those that power AI chatbots like ChatGPT—trained on vast amounts of health data to predict two of the aging clock’s main targets: an individual’s risk of dying and their risk of developing age-related diseases. ON a recent paper in Natural medicine stated that this approach outperforms existing clocks.

LHMs are still under development, and while we’re still making them faster, issues with existing clocks may be fixed. But the main message is: if you are tempted to have your biological age measured, think twice. Or, if you’ve already done so, take the results with a grain of salt. In return, I promise that the next time I write a story about aging, I will be much more skeptical of the research that uses them. Older, wiser.

topics: