As people age, their bodies inevitably change, affecting strength, mobility, and sensory function. Tasks that once seemed effortless become more challenging. Muscle mass tends to diminish, visual acuity declines and joints may become less flexible, all of which contribute to a gradual decrease in stability. This decline results in a startling statistic: nearly one in three individuals over 65 suffer a fall each year. The consequences of these falls are far from trivial. They often lead to broken bones, head injuries, or prolonged immobility, and in some cases, they can be fatal. In the United States alone, the financial toll of treating fall-related injuries amounts to billions of dollars annually. While ageing is unavoidable, falling does not have to be an inevitable consequence of getting older.
Recognising the need for early intervention, Jiaen Wu of Stanford University and his colleagues sought to explore whether balance impairments could be detected before they led to dangerous incidents. “One big challenge is that small balance impairments can go unnoticed until someone actually falls,” Wu explains. “So, we wanted to ask: Can we detect these impairments before someone gets hurt?” Together with collaborators Michael Raitor, Guan Tan, Kristan Staudenmayer, Scott Delp, Karen Liu, and Steven Collins, Wu investigated whether tracking the way people walk—long before any signs of frailty appear—could help forecast an individual’s risk of falling later in life. Their research, recently published in the Journal of Experimental Biology, provides compelling evidence that subtle variations in walking patterns may hold the key to fall prevention.
The researchers recruited ten healthy young adults between 24 and 31 to test their hypothesis. Each participant was fitted with a waist harness, attached to ropes and reflective markers positioned at the body’s front, back, and sides. This configuration enabled a sophisticated array of eleven cameras to record precise movements as each person walked on a treadmill at a steady pace of 1.25 metres per second. The team examined numerous aspects of gait, such as the consistency of foot placement and the degree of lateral movement in the person’s centre of mass. These initial measurements served as a baseline for each individual’s natural walking behaviour under normal, unimpaired conditions.
Next, the researchers introduced simulated impairments that mimicked the challenges associated with ageing. Participants walked again while wearing ankle braces to restrict joint mobility, a mask that limited vision, or pneumatic jets that disrupted their balance with bursts of air. These added hindrances caused noticeable disruptions in walking patterns. Step widths became more erratic, the timing between steps grew less regular, and the predictability of movement decreased. These findings mirrored the kinds of instability often seen in elderly individuals, indicating that visual obstructions or reduced joint function could profoundly affect a person’s ability to walk safely and steadily.
The team analysed the data and found that not all walking metrics were equally valuable for predicting fall risk. Of the six variables measured during normal walking, only three reliably signalled a person’s vulnerability to balance disturbances: variability in step width, irregularity in step timing, and the placement of each foot. Each of these indicators proved to be over 86 per cent effective in predicting whether someone would struggle to maintain balance when impaired. Surprisingly, when the researchers added a recovery test—tugging on the ropes attached to each walker’s harness to simulate a sudden loss of balance—the new data did not significantly enhance the accuracy of fall predictions. “We thought that seeing how people recover from a pull would reveal more about their balance ability,” Wu reflected, “but in this study, the normal walking data was just as informative in most cases.”
Perhaps the most striking insight from the study was that comparing an individual’s gait to their baseline proved far more predictive than comparing it to a group average. This finding suggests that the key to assessing fall risk is personalised, longitudinal monitoring rather than one-off clinical evaluations. Current practice typically involves assessing gait only after mobility issues have already emerged. Wu and his colleagues propose a paradigm shift: by measuring walking patterns earlier in adulthood—when movement is still largely unimpaired—clinicians could identify subtle deviations that indicate future risk. Such an approach could provide critical early warnings, enabling timely interventions that prevent falls before they occur. In doing so, it could reduce healthcare costs and preserve countless older adults’ quality of life and independence.
More information: Jiaen Wu et al, Detecting artificially impaired balance in human locomotion: metrics, perturbation effects and detection thresholds, Journal of Experimental Biology. DOI: 10.1242/jeb.249339
Journal information: Journal of Experimental Biology Provided by The Company of Biologists
