Home-based assessments may detect early mobility risks

Predicting whether a healthy 45-year-old will struggle to climb stairs or walk a decade later has long been a challenge for geriatric medicine. Now, a study published in JMIR Aging, a leading open access journal from JMIR Publications, reveals that early mobility decline can be predicted using a simple set of home-based measurements and artificial intelligence.

The research was conducted by a multi-institutional team including Alberto Conde Freniche, Mo Chen, Dantong Wang, and Denis Breuillé from the Nestlé Institute of Health Sciences (Switzerland) and Wei Hu and Yu-ming Chen from Sun Yat-sen University (China).

The silent decline

Mobility limitations, such as difficulty walking or climbing stairs, are often the first signs of functional decline, leading to a loss of independence and higher healthcare costs. However, in the early stages, many adults unconsciously adapt their movements to compensate for physical weakness, making the decline "silent" until it is too late to easily reverse.

The research team followed 1,344 healthy, mobility-intact adults aged 45 and older in Guangzhou, China, for nearly seven years. By the end of the study, 15.3% of the participants had developed early mobility limitations (EMLs).

Six keys to predicting the future

Using machine learning, the researchers identified six essential factors that accurately predict the onset of these limitations:

1. Age

2. Sex

3. Body Mass Index (BMI)

4. Sit-to-Stand (STS) Power: A measure of leg muscle power calculated from a standard chair-rise test.

5. Adherence to a Mediterranean Diet: A nutritional score based on healthy food intake.

6. Dietary Calcium Intake

Among these, the study found that advanced age, lower muscle power (STS), a higher BMI, and poor adherence to a Mediterranean-style diet were the most significant predictors of future mobility struggles.

A tool for home self-assessment

Unlike traditional diagnostic tools that require clinical visits or expensive equipment, the model developed in this study relies on simple-to-obtain data. The Sit-to-Stand test, for example, requires only a standard chair and a stopwatch.

"This study shows that early mobility limitations can be predicted using easy-to-obtain physical performance measures and specific nutritional factors," the researchers conclude. This opens the door for digital health tools or mobile apps that allow individuals to self-assess their risk at home.

The findings suggest that even for middle-aged adults who feel perfectly healthy, subtle deficits in muscle power or poor nutrition are early warning signs. Identifying these risks early allows for a window of opportunity where lifestyle interventions, such as targeted exercise and improved diet, can effectively change an individual's aging trajectory.