Day type and heatstroke-related ambulance transports in Tokyo, Japan: a time-series study

Heatstroke is a major public health problem during summer, and meteorological conditions, particularly wet bulb globe temperature (WBGT), strongly influence its occurrence. However, even under similar heat conditions, heatstroke risk may vary according to behavioural patterns on weekdays, weekends/holidays, and post-holiday days. This study aimed to examine whether heatstroke-related ambulance transports in Tokyo vary according to day type after adjusting for WBGT.

We conducted a time-series analysis using daily heatstroke-related ambulance transports in Tokyo, Japan, during the warm season (May-September) from 2015 to 2025. The days were classified as weekday, weekend/holiday, or post-holiday. The WBGT data for seven locations in Tokyo were used, and the highest daily maximum WBGT served as the heat exposure indicator. Modified Poisson regression models were used to estimate risk ratios (RRs) and 95% confidence intervals (CIs) according to day type, adjusting for the WBGT category at lags 0 and 1 and the calendar year. The same models were applied by age group, severity, and place of occurrence. Four sensitivity analyses, including the use of alternative heat exposure indicators and exclusion of the coronavirus disease 2019 period, were conducted.

After adjusting for WBGT and calendar year, the occurrence of heatstroke-related ambulance transports differed according to day type. Overall, ambulance transports tended to be more frequent on post-holiday days than on weekdays (RR, 1.08; 95% CI, 0.97-1.22). Among adults aged ≥65 years, transports were more frequent on post-holiday days (RR, 1.13; 95% CI, 1.00-1.28) and less frequent on weekends/holidays (RR, 0.87; 95% CI, 0.79-0.96). Among individuals aged <18 years, transports were markedly more frequent on weekends/holidays (RR, 1.80; 95% CI, 1.61-2.03). Weekend/holiday increases were also observed in mild cases and in public indoor and outdoor spaces and educational institutions, whereas cases in residences and workplaces were less frequent. Sensitivity analyses yielded similar results.

Heatstroke-related ambulance transports in Tokyo varied according to day type even after adjusting for WBGT, with distinct patterns across age groups, severity categories, and places of occurrence. Weekly behavioural patterns may contribute to heatstroke risk and warrant consideration in preventive strategies.