Clinical walking speed, free-living walking cadence, and all-cause mortality
Walking speed is a well-known health indicator, particularly in older adults, and is associated with the risk of disability, chronic disease and death. It is most often measured with a short walking test in a clinic or examination centre and is a snapshot that does not necessarily reflect how a person actually walks in daily life.
With the development and popularization of accelerometers, it is now possible to record participants’ activity over several days in their everyday environment and to derive their walking cadence, that is, the number of steps per minute of walking. Cadence is a component of walking speed and could be an alternative that is easier to collect on a large scale. But are these two measures equally informative for identifying people at higher risk of death?
Our study aimed to answer this question by comparing the association of clinically measured walking speed and free-living walking cadence with all-cause mortality, as well as their ability to distinguish people who die from those who survive.
We used data from the British Whitehall II cohort, in which civil servants have been followed since the late 1980s. The study included 4,067 participants whose walking speed was measured during a short walking test at their usual pace in 2012-2013. These participants also wore an accelerometer on the wrist for about nine days, which allowed us to derive two cadence indicators: the median cadence and the cadence of their fastest walking (95th percentile). Their mortality was then followed for about twelve years using the national mortality register, and we accounted for many factors that can influence the risk of death including sociodemographic, lifestyle, and health-related factors.
Our results show that a higher clinical walking speed, as well as a higher free-living walking cadence, is associated with a lower risk of death, in women and men. Importantly, the three indicators had a very similar ability to discriminate people at risk. Correlations between clinical walking speed and free-living walking cadence metrics were weak, suggesting that they describe different aspects of walking and are not interchangeable. Only clinical walking speed improved the discrimination of a model that already contained all known risk factors, and combining both types of measures brought no additional benefit once health status was taken into account.
These findings show that both approaches can be used in population-based studies, the choice depending on research objectives and practical constraints. The walking test is quick, inexpensive and easy to implement, whereas the accelerometer allows monitoring over several days and can be used in studies that were not designed to measure walking. The growing availability of consumer wearables could also, in the future, allow walking cadence to be integrated into medical follow-up.
Our study nevertheless has limitations: there is no standardized method for calculating walking cadence, and our results have not been verified in an independent cohort. Further work is therefore needed to confirm them and to better understand the mechanisms linking clinical and free-living walking to health.
By Rayane Haddadj, PhD candidate (Norwegian University of Science and Technology)
Photo credit: Davis Brown – Pexels – https://www.pexels.com/fr-fr/photo/homme-femme-marcher-buisson-15913972/
