<p>Objective　Maintaining or increasing walking provides several health benefits to older adults. However, the mid-term evaluation of Health Japan 21 [second term] showed that older adults' daily steps did not meet the goal. Therefore, this study emphasized primordial prevention, which is different from previous preventive approaches and focuses on the relationship between the built environment and physical activities, including daily steps. This study investigated the relationship between changes in the number of food stores and walking time.</p><p>Methods　This longitudinal study utilized the self-administered mail survey data between 2016 and 2019 from the Japan Gerontological Evaluation Study (JAGES). Older adults aged ≥65 years and residing in 27 independent municipalities were recruited. The dependent variable was a change in the walking time at two-time points (increase or not). Our explanatory variable was the change in the number of food stores at two-time points, reported on a 5-point scale, including no store (reference), increased stores, store available, decreased stores, and I don't know. Equivalently, it was defined as the self-reported change in the number of food stores (stores that sell meats, fish, fruits, and vegetables) within the walking distance of participants' homes (within ～1 km) from 2016 to 2019. The covariates included demographic factors, health behavior factors, environmental factors, and health factors in 2016. We used Poisson regression analysis (5% significance level) to calculate the cumulative incidence rate ratio (CIRR) and 95% confidence interval (CI) for an increase in walking time compared to no increase in walking time. The multivariate normal imputation method supplemented missing data of the dependent variable, explanatory variable, and covariates. Additionally, respondents' answer of “other” for the covariates was supplemented.</p><p>Results　Three years later, 13,400 (20.4%) respondents had increased their walking time. Older adults who reported increased number of stores (5,311, 8.1%) had more walking time than those who reported no stores (6,577, 10.0%) (CIRR=1.12; 95% CI: 1.03-1.21).</p><p>Conclusion　Participants who reported an increase in the number of fresh food stores within the walking distance had 12% more walking time three years later. A built environment might be used to measure primordial prevention that increases the amount of walking in daily life. Our results may provide evidence for policymakers and stakeholders to consider healthy urban planning.</p>