Nonlinear and threshold effects of built environment on older adults' walking duration: do age and retirement status matter?

Introduction: Walking plays a crucial role in promoting physical activity among older adults. Understanding how the built environment influences older adults' walking behavior is vital for promoting physical activity and healthy aging. Among voluminous literature investigating the environmental correlates of walking behaviors of older adults, few have focused on walking duration across different age groups and life stages, let alone examined the potential nonlinearities and thresholds of the built environment. Methods: This study employs travel diary from Zhongshan, China and the gradient boosting decision trees (GBDT) approach to disentangle the age and retirement status differences in the nonlinear and threshold effects of the built environment on older adults' walking duration. Results: The results showed built environment attributes collectively contribute 57.37% for predicting older adults' walking duration, with a higher predicting power for the old-old (70+ years) or the retired. The most influencing built environment attribute for the young-old (60-70 years) is bus stop density, whereas the relative importance of population density, bus stop density, and accessibility to green space or commercial facilities is close for the old-old. The retired tend to walk longer in denser-populated neighborhoods with better bus service, but the non-retired are more active in walking in mixed-developed environments with accessible commercial facilities. The thresholds of bus stop density to encourage walking among the young-old is 7.8 counts/km Discussion: Overall, the findings provide nuanced and diverse interventions for creating walking-friendly neighborhoods to promote walking across different sub-groups of older adults.