遊歩道整備が周辺居住高齢者の歩数に与える影響

横浜市における縦断的研究

抄録

 In recent years, both life expectancy and “healthy life expectancy” (i.e., the average life period in good health without limitation of activity in daily life or incapacities) have been increasing in Japan. However, the difference between life expectancy and healthy life expectancy has remained roughly unchanged over the last 15 years. This remains a social issue from the viewpoint of the quality of life of older adults as well as social security expenditure. In order to mitigate this issue, it is important to create a pedestrian-friendly walkable environment in which older adults can maintain physical activity. The construction of walking routes as a population-based approach to improve public health can contribute to creating such an environment.

 This study aimed to examine the effect of walking routes on the physical activity of older residents as a natural experiment utilizing four-wave panel data of step counts obtained from a large sample in Yokohama, Japan. The sample included older adults aged between 65 and 84 participating in the Yokohama Walking Point Program (YWPP) who recorded their step counts for more than 20 days every March from 2016 to 2019. The construction of three pedestrian-friendly walking routes in Yokohama began in April 2016 and was completed in March 2018. The effective distance of walking routes was changed from 100 to 500 meters (m) in increments of 100 m and residents within each distance were assigned to the intervention group. A longitudinal analysis was conducted employing a generalized linear mixed model (GLMM) to examine the effect of the walking routes on step counts of older residents by gender.

 Step count data from 9,133 men and 8,974 women living in 759 neighborhoods were analyzed. First, the built environment (BE) variables associated with step counts were selected by comparing the AIC of GLMM. As a result, three BE variables (i.e., distance to a station, adequacy of parks, and proportion of public buildings) were included in the AIC-minimum model for men, and two additional variables (i.e., population density and distance to a bus stop) were included for women. Second, changes in step counts between the intervention group and the control group were compared using GLMM with a Group x Year interaction term controlling for sample characteristics and BE variables included in the AIC-minimum model. Consequently, there was no significant difference in step counts between the two groups for the men, whereas the intervention group for the women had significantly more step counts than those in the control group immediately after the completion of the walking routes (i.e., in the third wave). This was significant when the effective distance was set at 200 to 500 m from the walking routes. The intervention/control ratio of step counts in women was approximately 1.04, which was equivalent to 200-300 steps. The shorter the effective distance, the greater the ratio. However, the ratio decreased to almost 1.0 one year after the completion of the walking routes (i.e., in the fourth wave).

 In conclusion, the newly constructed walking routes appeared to only have an effect for women with the effect no longer in place one year after the completion of construction. Future studies are needed to clarify what factors cause gender differences in relation to environmental improvement and what measures are effective in maintaining this effect.