This study revealed the relationship between steps counts and medical cost and estimated the effect of neighborhood environment and change of travel behavior on medical cost.
The study used data of 1,028 older citizens from surveys which conducted from 2016 to 2018. Medical cost was gathered from claims history of national health insurance. GPS log data and steps counts data were collected in panel survey in 2016 and 2018.
The relationship between medical cost and steps counts was examined by cross-lagged panel modeling. The model was estimated by full information maximum likelihood (FIML) method. The result shows that medical cost of older adults who walk more tend to decrease. On the other hand, there is no relationship that medical cost affects steps counts.
In addition, the effects of neighborhood environment and change of travel behavior on medical cost mediated by step counts are quantitatively estimated by structural equation modeling. For example, the effect of accessibility to commerce area and the change of frequency of stay in city center by cycling or public transportation on medical cost were evaluated. The model is also estimated by a full information maximum likelihood (FIML) method. The findings reinforce that urban design policy which promotes to live around the commerce area and the transport policy which increase of frequency of visit to commerce area and park by walk and to city center by cycle and public transportation will reduce quite a few medical costs.
View full abstract