Can Walking Speed Help Predict How Long You'll Live?
How fast you walk is a useful clue to health in later life. A new Whitehall II study in JAMA Network Open compared a timed clinic walk with days of wrist tracking. Faster clinic walkers and people with a brisker everyday step rate both died less often, but only the clinic walk significantly improved prediction beyond standard risk factors. Neither result hands you a pace to hit. It does support treating a slowing walk as a signal worth checking, and building activity, strength, and balance around what you can do now.
Both ways of measuring walking pointed toward longer survival
The link itself isn't new. Earlier pooled analyses of cohort studies found that faster walkers had a substantially lower risk of death than slower ones. Earlier Whitehall II work showed walking speed starting to fall faster years before death, with slower walkers carrying more silent artery disease. This analysis compared, in the same people, a supervised clinic walk with free-living cadence measured during ordinary life. Speed is distance over time and cadence is steps per minute, so the two can diverge. This study measured both in 4,067 civil servants with an average age of 69 in 2012 and 2013. It then followed them through national death records, and 739 of them died.
| Walking measure | What it captured | Associated difference in death rate |
|---|---|---|
| Clinic walking speed | Usual pace over a 2.44-meter course, timed three times and averaged | About 10% lower for every 0.1 m/s faster |
| Typical everyday cadence | Median steps per minute while walking, from a wrist sensor worn for nine days | About 17% lower for every 10 steps per minute faster |
These figures come from the fully adjusted models, so they already account for the measured health, behavioral, and social differences between people. The cadence results also held after accounting for total daily steps. Some earlier US cohorts, by contrast, saw step-rate associations fade once step totals were included. The pattern generally held across age groups, body weight, and chronic disease status. It also held after excluding deaths in the first two years and under several alternative ways of defining a walking bout.
What the clinic walk added
A measure can track with risk without improving a prediction you'd otherwise make, so the researchers also assessed how well each model ranked people by who died first. Adding clinic speed to standard risk factors produced a small, statistically significant gain. Adding typical or brisk cadence did not significantly improve prediction. An exploratory analysis found no significant extra gain from adding cadence to a model that already had clinic speed and all measured risk factors.
The authors themselves describe the two approaches as having comparable associations and similar predictive ability. They suggest the choice for research depends on what a study needs. The clinic walk takes a few minutes of hallway instead of nine days of wear. The two measures were also only weakly correlated, so they aren't interchangeable readings of one underlying number, and this study doesn't establish why they diverge.
Some limits apply to all of this together. About three quarters of participants were men and more than nine in ten were White, all from one British civil-service cohort. The people with sensor data were healthier than the wider cohort, and walking was measured only once. The prediction models weren't tested in an independent group, and there's no standard method for deriving cadence from a wrist device. The research-grade sensor used here also isn't the watch on your wrist, so these results don't tell you how a consumer device's cadence reading would perform.
What a walking result can guide
This Whitehall II analysis was observational and captured walking at a single point. It can't say whether someone who deliberately speeds up changes their risk. Randomized trials also haven't shown that telling people to walk faster than their usual pace improves survival. That leaves no pace or cadence target from this study that you can treat as life-extending.
Walking programs have randomized evidence for health benefits. A meta-analysis of 37 randomized trials in inactive healthy adults found walking programs improved blood pressure, fasting glucose, and aerobic fitness. Two UK primary-care trials used pedometers to increase daily walking and followed participants for four years. Their pooled follow-up analysis recorded fewer cardiovascular events and fractures in the walking groups, though those programs targeted step volume and didn't isolate pace. The UK Chief Medical Officers' guidance, updated in July 2026, asks adults 65 and over to work toward at least 150 minutes a week of moderate activity. It also asks them to include strength and balance work on at least two days and to build up gradually from wherever they start.
The more personal use of this study is as a warning light. You may have noticed that you walk more slowly than you used to, or that others now pull ahead of you. That change is worth an in-person assessment, since slowing has been tied to underlying disease and functional decline in this cohort and others. An assessment can look for treatable causes. There's no reason from these data to buy a wearable for risk prediction. If you already own one, cadence is a reasonable thing to watch as long as you don't read it as a survival score.
Wrist-measured cadence may still prove useful. That would take standardized methods tested in more diverse cohorts and on consumer devices. It would also take evidence that cadence sharpens individual risk estimates beyond a clinic walk and established factors. Whether walking faster, as distinct from walking more, extends life would need a large, long randomized trial that separates the two and counts deaths and harms.