JMIR Publications

ABSTRACT

Background:

With the aging of the global population, preventing the onset of mobility limitations is considered a worldwide public health priority.

Objective:

This study aimed to develop a predictive model for incident early mobility limitations (EML) in late middle-aged and older adults, based on a simple functional test and modifiable lifestyle factors to facilitate a home-based self-assessment of early mobility decline and promote lifestyle intervention strategies.

Methods:

Our study population was community-dwellers aged 45 years and above, who participated in the second and fourth waves of the Guangzhou Nutrition and Health Study. Included participants were healthy, non-frail adults reporting no limitations in activities of daily living (ADLs) at baseline. At six-year follow-up, participants with poor physical performance (walking speed < 1 m/s or handgrip strength < 28 kg for male and < 18 kg for female) or reporting some difficulty walking and/or climbing stairs were classified as experiencing EML. Least absolute shrinkage and selection operator (LASSO) was used to identify predictors from various factors, and six machine learning (ML) models were trained and evaluated for EML prediction, employing bootstrap-based techniques to address class imbalance. Predictive ability was quantified using the area under the receiver operating characteristic curve (AUC). Variable importance analysis was used to identify key predictors.

Results:

A total of 1344 participants were included in the analysis, of which 206 (15.33%) developed EML after a median follow-up of 6.67 years. Those who developed EML were older, had a higher BMI, a lower Mediterranean diet score (aMDS), and poorer performance in the sit-to-stand test (STS) test as well as lower estimated muscle power from the STS at baseline. The final models included 6 out of 9 predictors: age, sex, BMI, aMDS, STS power, and dietary calcium intake. Three ML models (logistic regression, LASSO, and neural network) achieved an acceptable AUC value of ≥ 0.70 in the testing dataset, with the neural network performing best (AUC 0.71, 95% CI 0.64-0.78). Bootstrapping did not improve classification performance. Advanced age, lower adherence to Mediterranean diet, and lower muscle power estimated from STS test at baseline were identified as the most important predictors of EML.

Conclusions:

This study suggests that the onset of early mobility limitations in Chinese adults aged 45 years and above can be predicted by an easy-to-obtain physical performance measure, age, sex, body mass index, and specific nutritional factors. The combination of prediction models and variable importance analysis provides valuable insights for the early identification and intervention of EML. More efforts are needed to validate our findings in external cohorts.