Snow-Related Injuries in Aomori Prefecture, Japan

Snow-related injuries are a recurrent public-safety concern in regions with heavy snowfall, yet the temporal effects of snowfall and their heterogeneity across accident mechanisms remain insufficiently understood. This study examines snow-related accidents in Aomori and Hirosaki, two cities in one of Japan’s heaviest-snowfall regions, using detailed accident records for fiscal years 2013–2022 linked to daily weather data. We estimate Poisson distributed-lag models of daily accident counts, allowing snowfall to affect accident incidence contemporaneously and for up to six subsequent days. We further estimate mechanism-specific models and an ordered logit model of injury severity. The results reveal a distinct temporal pattern. Same-day snowfall is associated with a lower incidence of snow-related accidents, whereas snowfall on preceding days is generally associated with higher accident incidence, particularly over the first several days following snowfall. Greater snow depth is also associated with increased accident incidence, while relatively warmer conditions are associated with fewer accidents. These relationships vary across accident mechanisms: high-place falls, same-level slips and falls, and injuries caused by falling snow or ice respond differently to current and lagged snowfall and snow depth. In the severity analysis, greater same-day snowfall is associated with less severe outcomes, while snow depth shows little association with severity. These findings demonstrate that snow-related injury risk depends not only on the amount of snow but also on its timing, accumulation, and the mechanism through which injuries occur. Accounting for these dimensions may improve the timing and targeting of warnings and preventive measures in heavy-snowfall communities.