Biological age

1

What is biological age?

An estimate of how worn your body is, not how many birthdays you have had. Same-age people can differ a lot.

Biological age is the wear on your organs, such as the lungs, kidneys, heart and liver, set against calendar age. You can estimate it, but you cannot see it.¹

In Dunedin, New Zealand, about a thousand people born in the early 1970s have been followed since birth. At 38 they were all the same age by the calendar. A formula built on a US health survey turned their body measurements into a biological age for each. Most sat near 38, but the range ran from 28 to 61.¹

A second measure tracked 18 body measurements over 12 years to see how fast each person was changing. One body had hardly changed. Another had aged nearly 3 years for every calendar year. Those who were aging faster were less physically able, showed cognitive decline, felt less healthy and looked older. The links were modest: a leaning, not a rule.¹

2

How is a clock built?

Researchers measure many things in many people, then fit a formula that predicts age or risk of dying. That formula is the clock.

Nobody can read your body's age from one dial. One review sorted the candidates into six kinds, such as chemical tags on DNA and proteins in the blood. The one it found most promising is the epigenetic clock, which reads those tags. It added that the clock's true value still needs long-term confirmation.²

An early DNA clock, from 2013, was trained to predict calendar age from the tags, using thousands of samples. It was built on 51 kinds of healthy tissue and cell. In test data it typically missed real age by about 3.6 years: a median miss, so half of the people missed by more.³

Later clocks were trained on health, not just on age. PhenoAge used a mix of clinical measures that capture differences in lifespan and healthspan to pick its DNA tags, and its authors report that it predicted death, cancers and physical functioning better than earlier clocks. GrimAge was trained to predict time to death, from DNA stand-ins for seven blood proteins and for smoking. DunedinPACE was trained on how fast 19 organ measurements declined over two decades in one group, and needs one blood sample. Another family is built from what a body can do, such as its physical abilities.^4,5,6,7

3

Why do clocks disagree?

Each clock looks at a different part of the body, was built on different people, and carries some noise.

Swedish researchers followed 845 adults for 20 years, measuring nine different biological ages again and again. They agreed with one another to varying degrees, mostly because everyone gets older.⁷

One body is not one age, either. Researchers used blood proteins that come from particular organs to estimate the age of 11 organs in 5,676 adults. Nearly a fifth of people showed one organ ageing strongly ahead of its calendar age.⁸

Then there is noise. DNA clocks come from a lab test, and the same sample measured twice can give different answers. In one study, technical noise moved six well-known DNA clocks by up to 9 years between repeat runs. So a small gap between a clock and your birthday may mean little.⁹

If two tests give you two different ages, that is what to expect. They measured different parts of you, in different ways.^7,9

4

What can change it?

Sometimes, for a while. In mice, stress raised it and recovery brought it back; one trial moved one of three clocks a little.

In young mice, being surgically joined to an older mouse raised their biological age. It was restored after the two were separated. Temporary rises also showed up during major surgery, pregnancy and severe COVID-19, in people or mice. So a reading may show how your body was on the day, not only its long-run wear.¹⁰

A randomised trial called CALERIE assigned 220 adults without obesity to cut calories by a quarter, or to eat as usual. The cutters managed about half of that. After two years, one clock, DunedinPACE, showed a pace of ageing about 2 to 3% slower. Two others, PhenoAge and GrimAge, showed no clear change. The effects were small, and this was a post hoc analysis, not the trial's main test. The authors say a conclusive test needs trials that follow people long enough to see disease and death.¹¹

Life leaves its mark as well. DunedinPACE indicated a faster pace of ageing in young adults who had faced hardship in childhood, a link seen in people, not proof of cause.⁶

5

What does it say about your health?

Across large groups, people whose clocks run ahead have more illness and die sooner. For one person it only shifts the odds.

In the same Swedish group, eight of the nine ages were linked with dying sooner, whatever the person's calendar age. The exception was telomere length, the protective caps on the ends of chromosomes. The strongest were GrimAge and a frailty index, a tally of a person's health problems. Two DNA clocks and the frailty index each added something, so they seem to complement each other.⁷

Organ ages carry weight too. Accelerated ageing of an organ went with 20 to 50% higher risk of death. That is a share of the risk, not 20 to 50 extra deaths in every 100 people.⁸

Some newer clocks measure speed, not level. In five other datasets DunedinPACE went with illness, disability and death, and it added to GrimAge's predictions.⁶

All of this is about crowds. A clock running ahead tells you that people with similar readings did worse on average. These studies watched people, so they cannot say that closing the gap would change the outcome. In 2023 a large group of researchers wrote that the lack of agreed standards for what makes a reliable ageing biomarker hinders their development and validation.^7,12

6

Can a watch tell your biological age?

A research clock estimates age from the pulse at the wrist, and its gap went with disease. It is a study tool.

A watch cannot read your blood or your DNA. It sees your pulse and your movement. Researchers built PpgAge to predict calendar age from the pulse signal a consumer wearable records at the wrist, in the Apple Heart & Movement Study: more than 200,000 people and over 149 million days of data. People whose PpgAge ran above their real age had higher rates of heart disease, heart failure and diabetes. The gap also predicted new heart disease events. The study was observational, and its authors are employed by Apple.¹³

The gap was also linked with smoking, exercise and sleep, and it rose sharply during pregnancy. That fits the idea that a reading follows how your body is doing now. Whatever a watch-based age shows, it is built from what a wrist can see, which is one slice of you.^10,13

7

What can you do with this?

Any biological age is a rough, changeable estimate: a direction, not a verdict.

1. One reading is one day. In mice and a few human cases, a reading rose with stress and came back with recovery, and lab noise alone can move it by years.^9,10
2. Clocks differ. Different clocks measure different things, so follow yourself with the same clock from the same kind of test, and expect some noise even then.^7,9
3. It is not a verdict. A clock shows where people with similar readings ended up on average. Nobody yet knows whether changing your own number changes your future.¹¹
4. Fitness and strength can be tracked directly. VO2 max and grip strength each have their own article.