What can a biological age test actually tell you?

How old are you? How old do you feel? Are those two numbers the same?

Biological age tests have validated what many of us sense intuitively: Our cells don’t always match our birth certificates.

As these tests have evolved over the past decade, longevity researchers have started using them to identify what behaviours and conditions cause people to age faster and whether there are interventions that can slow the process down.

And although the tests were intended to be used in laboratory research, more and more consumers are shelling out a few hundred dollars to wellness and longevity companies in order to reveal their “true” age.

Here’s what experts say the tests can tell you – and when, if ever, it’s worth taking one.

A PROMISING CONCEPT

Biological age tests claim to inform you about how your body is faring by telling you what age your blood test results most resemble.

“In a way, when the biological age is older than the chronological age, you are at high risk of adverse health outcomes,” said Dr Luigi Ferrucci, the scientific director of the National Institute on Aging in the US. “When your biological age is younger than the chronological age, you’re kind of protected against those health outcomes.”

The most prominent tests look for modifications to the genome – a process called DNA methylation – that influence whether a given gene is active or not. The discovery that some of these modifications occur reliably with age led to the development of the Horvath “clock” in 2013 by geneticist Steve Horvath, which is widely regarded as the first biological age test.

Today, there are three “generations” of this type of test. Generation one – the Horvath clock and others like it – are “literally predicting what your age should be based on your DNA methylation,” said Raghav Sehgal, a principal investigator at the Marcus Institute for Aging Research, which is affiliated with Harvard Medical School. “So it will tell you your DNA methylation looks like that of a 20-year-old, or it looks like that of a 40-year-old.”

The second-generation clocks, with names like PhenoAge and GrimAge, were introduced roughly five years later and look for methylation sites that are associated with a person’s relative mortality risk. The results are again given as an age, based on whether a person’s risk of death looks more like, say, the average 50-year-old’s or the average 70-year-old’s.

Third-generation clocks, like DunedinPACE, which came out in 2022, measure how fast someone is aging. Researchers have used a car analogy to describe the differences between the tests: The first- and second-generation models act like an odometer, measuring your age or proximity to death. The third-generation tests act like a speedometer.

The first-generation clocks “put this field on the map, made it famous, and remain probably the best known,” said Daniel Belsky, an associate professor of epidemiology at the Columbia University Mailman School of Public Health who developed DunedinPACE. But the second- and third-generation clocks have turned out to be more informative when it comes to predicting long-term health.

These newer clocks are also more sensitive to health-related behaviours and treatments, Sehgal said. In a recent study, he pitted 16 biological age tests against one another. The second- and third-generation clocks were more likely to change in response to various longevity-related interventions, like diets and medications.

SOME KINKS TO WORK OUT

Even with the best tests, there are concerns about accuracy and reliability, especially when the tests are used on a personal level.

Dr Albert Higgins-Chen, an assistant professor of psychiatry and pathology at Yale School of Medicine, said factors like time of day, when you had your last meal or how well you slept the night before appear to influence the results. This isn’t a unique problem for biological age tests, but it is something that scientists need to figure out how to manage, and that people may need to take into account when taking a test.

Morgan Levine, who developed PhenoAge while she was a postdoctoral researcher at UCLA, pointed out that if you take multiple biological age tests, each one will give you a different answer. That’s in part because the tests are all measuring slightly different things. But it’s also because they were developed using different data sets to establish what the “average” for any given age looks like.

“I don’t think people should put too much emphasis on that number that they’re getting,” said Levine, who is launching an artificial intelligence biotechnology startup. (Levine developed and licensed another biological age test with Sehgal and Higgins-Chen.)

To Dr Mohammad Arfan Ikram, a professor of epidemiology at Erasmus Medical Center in Rotterdam, Netherlands, the biggest shortcoming of the tests is that they don’t tell you what might be going wrong in the body to cause a higher age, or, consequently, how to fix the problem.

Returning to the car metaphor, Ikram pointed out that just knowing a car has 100,000 miles on it doesn’t tell you whether the brakes or the engine or the clutch will fail first.

That issue partly stems from what the tests are measuring: When scientists isolate DNA from a blood sample, they’re only testing the DNA of white blood cells.

“You can measure the changes in these blood cells,” said Tony Wyss-Coray, a professor of neurology and neurological sciences at Stanford University, but “you don’t see changes that might happen in neurons in your brain, or in liver cells or in the pancreas.”

To try to solve for this, instead of looking at DNA, Wyss-Coray and other researchers have started testing blood samples for thousands of proteins that are released from different tissues. That enables them to evaluate how multiple types of cells or organs are doing, rather than inferring someone’s head-to-toe health from just their white blood cells.

Some experts think this approach will provide more valuable and actionable information, because while the idea of distilling the health of your entire body into one number is seductive, it’s probably more useful from a medical standpoint to know which organ is going to cause problems first. However, this technique is more expensive and not widely available to the public. (Wyss-Coray is a co-founder of a direct-to-consumer company that aims to sell the test in the next year or two.)

WHAT THEY'RE GOOD AT

Despite the tests’ shortcomings, the experts said they can be valuable research tools when used to evaluate large data sets. For example, scientists are utilising the tests as a new way to quantify how life experiences, like pregnancy or losing a loved one, affect people’s health.

Perhaps most promising, the tests can be used to assess, in a short period of time, whether a drug or other intervention being studied in a clinical trial is able to slow ageing. This is a major benefit for longevity research, since following someone for decades to find out whether a treatment altered their life span isn’t feasible.

The tests have already been used for this purpose in a few trials, including one that examined the potential health benefits of common supplements and another that evaluated whether calorie restriction might improve longevity in humans.

But in some cases, the tests have raised more questions than they have answered. In the supplement trial, for instance, the intervention significantly slowed some of the biological age clocks but didn’t affect more concrete health metrics, like blood pressure. Belsky, who was involved in the study, hypothesised that this discrepancy may have arisen because the tests could be detecting subtle changes that emerge before more established signs of disease.

There is also a possibility that the tests measure something related to ageing or health that doesn’t make a meaningful difference if it is changed. For example, people’s hair turns grey and their skin wrinkles as they age, but fixing either of those two things won’t make people younger or healthier.

FOR CONSUMERS, TIME WILL TELL

Several experts painted a rosy picture of a future in which people could purchase and use the tests to effectively track themselves long-term and to alert them to early warning signs that their health is declining.

“I think the biological age tests are the promised land of the preventive care version of medicine,” Sehgal said. But, he added, “that’s not where we are yet.”

Higgins-Chen said he saw a potential benefit now: Consumers could choose one test and take it periodically to track their score as they age, or after they start a new medication or health routine. “Ageing is a longitudinal process that unfolds over very long periods of time,” he said, so the real benefit of the tests is detecting when and how your health changes over the years.

In general, though, the experts agreed that the direct-to-consumer tests available today, especially when used as a one-off, aren’t really that informative.

To Ikram, people are better off taking standard doctor’s office tests, like those for cholesterol and glucose. “That already gives you a much better picture” and identifies things “you can directly intervene on,” he said.

At best, biological age tests can give you a loose understanding of your health, Levine said. “People who generally score lower on it, it signals to some degree that they’re probably healthier than people who are generally scoring high on it.”

But, she added, most people already have a rough sense of how healthy they are. If you exercise, eat a balanced diet and don’t smoke, your biological age score will likely be lower – and you probably already know that.

By Dana G Smith © The New York Times Company

The article originally appeared in The New York Times.