The Relationship Between Changes in Walking Practice Rate Disparities and Deprivation Index: Using a Generalized additive mixed model

The walking practice rate is a health indicator that can be influenced by the physical infrastructure and collective social functions of the places where individuals live and work. It may be affected by various factors, such as individual changes, health policy shifts, improvements in community infrastructure, and analyzing how these factors evolve over time and establishing appropriate strategies is essential. Thus, this study employed a generalized additive mixed model to analyze changes in the walking practice within Busan, including both individual and area-level factors while accounting for the non-linearity of yearly trends. As a result, the findings revealed varying effects of annual changes on regional disparities and identified a cross-level interaction between education level and deprivation index. Especially, the group with education level ranging from no education to middle school, predominantly comprising elderly individuals, demonstrated the physical conditions of areas with high levels of deprivation could potentially have negative effects on the walking practice. The generalized additive mixed model would be a valuable tool for identifying area factors influencing various health-related indicators exhibiting non-linear change patterns.