You probably have a fairly good idea of how old you are.
It is printed on forms, attached to your date of birth, quietly advancing every year whether you pay attention to it or not. But there is another kind of age that is much harder to see.
It appears in the way you move through a room, how regularly you sleep, how quickly you recover from exertion, how active you are at different times of day and how your ordinary routines change over the years.
- You probably have a fairly good idea of how old you are.It is printed on forms, attached to your date of birth, quietly advancing every year whether you pay attention to it or not. But there is another kind of age that is much harder to see.It appears in the way you move through a room, how regularly you sleep, how quickly you recover from exertion, how active you are at different times of day and how your ordinary routines change over the years.
- Across five years of everyday life, they may look very different.
- Living longer is one achievement.
You rarely notice these changes while they are happening. A walk becomes a little slower. You stop taking the stairs quite as often. You sleep at slightly different hours. A weekend that once involved several outings becomes one in which staying home feels more appealing. None of these things necessarily means that anything is wrong. They are simply small pieces of a life changing.
Now imagine that a machine could observe those patterns for years.
Not diagnose you from a single reading, but learn what your ordinary life looks like and recognise when the pattern begins to change.
This is one of the more interesting developments emerging from digital health. Researchers are increasingly exploring the possibility that ordinary behaviour recorded by wearable devices can act as a kind of digital biomarker, providing clues about biological ageing, physical resilience and changes in health long before those changes become obvious during an occasional medical appointment. Recent longitudinal research has found that patterns in everyday rest and activity recorded by consumer wearables are associated with trajectories of biological ageing.
The important idea is not that your watch can tell you your “real age.”
It is that ageing may leave traces in ordinary behaviour long before we consciously recognise them.
For generations, medicine has relied heavily on snapshots. You visit a doctor, have your blood pressure measured, give a blood sample, perhaps undergo imaging, and the resulting information becomes a picture of your health at that moment.
There is good reason for this. Medical measurements need to be accurate, interpretable and clinically useful. A single blood test can reveal something enormously important.
But a snapshot has a limitation.
It does not tell you very much about the direction in which you are moving.
Consider two people who are both 55. Their blood pressure, cholesterol and routine laboratory results may look broadly similar. Yet one may have maintained remarkably stable sleep, activity and mobility for years, while the other has experienced a gradual decline in movement, increasingly irregular routines and reduced physical capacity.
At the clinic, they may look surprisingly similar.
Across five years of everyday life, they may look very different.
That difference is where digital biomarkers become intriguing.
The idea is still developing, and there is a great deal to learn before these measures become routine clinical tools. But researchers are beginning to investigate whether patterns captured passively through sensors can reveal changes in physical and cognitive health that conventional, occasional assessments can miss. Digital measures of mobility, activity, sleep and other behaviours are being studied as potential indicators of ageing and disease risk.
This changes the meaning of a health measurement.
Instead of asking only, “What is my heart rate today?” we might eventually ask, “How has my heart rate behaved over the last three years?”
Instead of asking whether you can walk a certain distance in a clinic, we might be able to understand how your everyday walking has changed across thousands of ordinary journeys.
Instead of asking you to remember whether you have been sleeping differently, a long-term record could show that your sleep timing has gradually become less regular.
There is something almost strange about this.
The most valuable health information about you may increasingly come from moments when you are not thinking about your health at all.
Walking to the shop.
Going upstairs.
Sleeping.
Getting out of a chair.
Moving around the house.
Going for an evening walk.
Or simply not moving as much as you used to.
Medicine has traditionally been very interested in what happens when you enter the medical system. Digital health is beginning to make it possible to study what happens before you enter it.
That could become particularly important as populations age.
We often speak about ageing as though it were a number. Someone is 60, 70 or 80, and the number becomes a rough shorthand for what we assume their body might be capable of doing.
But two people of the same chronological age can have dramatically different levels of physical resilience.
One 75-year-old may travel, climb stairs, garden, carry groceries and recover quickly from a minor illness. Another may struggle with ordinary movement and have much less physiological reserve.
Age is not irrelevant.
It is simply incomplete.
This is why researchers are increasingly interested in measuring function and resilience rather than relying on age alone. A recent body of work on digital biomarkers of ageing considers whether data from everyday life can provide information across multiple physiological systems rather than treating ageing as a single number.
That distinction could become increasingly important as medicine moves from treating disease to preserving healthspan.
Living longer is one achievement.
Remaining capable while living longer is another.
And capability is surprisingly measurable.
How quickly do you walk?
How steadily do you move?
How much do you move during the day?
How does your activity vary between weekdays and weekends?
How regular are your sleep and waking patterns?
How quickly do you return to your normal activity after an illness?
Some of these measures may eventually become as interesting to preventive medicine as traditional laboratory values.
But there is a subtle danger in this future.
Once something becomes measurable, we have a tendency to turn it into a score.
We already see this with sleep scores, readiness scores, fitness scores and various forms of biological age estimation. A number appears on a screen and suddenly something as complicated as human health begins to look like a school examination.
That is not necessarily helpful.
Ageing is not a competition, and the body is not a machine that receives a failing grade because one week was less active than another.
The real value of continuous measurement may be found in patterns rather than scores.
A single bad night’s sleep is ordinary life.
A persistent change in sleep over several months is more interesting.
A week of reduced activity after an illness is hardly surprising.
A gradual decline that continues long after recovery might deserve attention.
The difference is context.
This is also where artificial intelligence could become useful, although the technology is sometimes described too grandly.
The interesting role for AI is not necessarily to announce that you are “ageing badly.” It is to process years of messy, continuous information and identify patterns that would be almost impossible for a human being to notice.
Imagine having five years of movement data.
Nobody wants to examine five years of graphs.
An algorithm can.
It can look for changes in regularity, pace, recovery, sleep timing or interactions between different measures. It can potentially compare your current pattern with your own history rather than simply comparing you with a population average.
That last distinction may be one of the most important.
Personalisation in medicine is often discussed as though it means giving different people different treatments. But there is another form of personalisation that may arrive first.
Understanding what is normal for you.
Your normal resting pattern.
Your normal activity pattern.
Your normal sleep.
Your normal recovery.
Your normal fluctuations.
Once that baseline exists, a change becomes easier to recognise.
Recent work in digital health is already moving toward combining continuous sensor measurements with clinical data and AI models. Research has shown, for example, that wearable time-series information combined with routine biomarkers can help predict insulin resistance, suggesting how behavioural data might eventually complement conventional medical measurements.
Other research is exploring wearable data for brain health, including whether passive measurements of behaviour and activity might eventually help identify changes in cognition or neurological function earlier than conventional assessments. The evidence is promising but still developing, and researchers continue to face questions about accuracy, validation and whether a detected pattern actually improves clinical outcomes.
That last point matters enormously.
Detection is not the same thing as diagnosis.
Prediction is not the same thing as prevention.
And an algorithmic signal is not the same thing as a medical conclusion.
The temptation will be to treat every new digital biomarker as another window into the body. Some will prove genuinely useful. Others will turn out to be noisy, misleading or too dependent on the particular device or population from which the data was collected.
The field is therefore likely to advance through a mixture of impressive discoveries and disappointing ones.
That is normal science.
The more interesting question is what happens if some of these signals prove reliable enough to matter.
Healthcare could gradually become less episodic.
Instead of discovering that something has changed when you happen to have a medical examination, your health record might contain a long-running picture of your physical and behavioural trajectory.
A doctor could see not merely that your mobility is below a reference value, but that it has changed significantly compared with your own previous baseline.
An older person recovering from an operation might be monitored at home rather than relying entirely on occasional appointments.
A person at risk of declining mobility might receive support before everyday independence has been lost.
A treatment could potentially be evaluated not only by whether a laboratory value improved, but by whether ordinary function improved in real life.
Even clinical trials are increasingly exploring wearable measurements because they can capture physiological and behavioural changes continuously in people’s normal environments rather than only during scheduled visits.
That could eventually change what counts as evidence of health.
For much of medical history, the body became visible to medicine when the patient entered the clinic.
The emerging digital body is visible through patterns of everyday life.
That does not mean we should surrender our privacy to every sensor available. Quite the opposite. The more valuable health data becomes, the more important questions about consent, ownership, security, access and interpretation become. There is also a genuine risk that continuous monitoring could turn ordinary variations in human behaviour into unnecessary medical concerns.
A healthy future will therefore require restraint as much as innovation.
Not everything worth measuring needs to be measured.
Not every detectable difference requires treatment.
And not every prediction deserves to become a diagnosis.
For the individual, the most useful shift may be surprisingly modest.
Think less about your health as a collection of isolated numbers and more as a trajectory.
You do not need to obsess over every night’s sleep or every step. What matters is noticing meaningful changes that persist. If something about your energy, movement, sleep or physical capacity has clearly changed and stayed changed, that is information worth discussing with a qualified professional rather than dismissing simply because you still feel “mostly fine.”
The deeper change is philosophical.
We have spent much of modern medicine asking whether you have a disease.
Increasingly, we may be able to ask a different question.
How is your capacity changing?
That is a more subtle question, but perhaps a more useful one for a world in which people are living longer.
Because ageing does not happen in a single dramatic moment. It happens quietly, through thousands of ordinary mornings and evenings. A slightly slower walk. A little less recovery. A different sleep pattern. A different rhythm to the day.
For years, those details have simply disappeared into life.
Technology may soon make them visible.
The challenge will be to use that visibility wisely.
Not to turn life into a dashboard, or ageing into a score, but to notice the direction in which a human being is moving while there is still time to influence the journey.
Perhaps the most interesting health record of the future will not tell you how old you are.
It will show you how you are becoming.