Study findings
We conducted a case-crossover study using case data on heat illness incidents in junior high and high school sports club activities in Japan, along with WBGT data from nearby locations. The analysis of both pre- and post-stratification data confirmed a significant relationship between heat illness incidence and WBGT (WBGT-Incident and WBGT-Prev). This finding is consistent with previous research showing that many heat illness cases in Japan occur at a WBGT of 28°C or higher [13] and that the odds of heat illness in junior high and high school of major Japanese cities increase with higher WBGT [14]. By applying a case-crossover approach to case data from the JSC-System, which covers most of the child and student population [21], we were able to confirm the significant relationship between the WBGT and heat illness incidents during school sports club activities, mitigating the risks of selection bias and ecological fallacy.
In the stratified analysis, significant differences in odds ratios were observed within groups stratified by club, location, year, month, and DOW. This indicates that applying uniform HSTs may not be appropriate. The higher odds ratios in the early summer months (April to June) and in 2019, when the number of heat illness cases was high not only in the mid-summer [22], suggest that heat illness risk increases when heat acclimatization is insufficient. This aligns with findings that heat illness is more prevalent in June than in July due to lower temperatures and humidity [13], and higher latitude areas in Japan have more heat illness cases than lower latitude areas at a maximum daily WBGT of 33°C [23]. The analysis also revealed that the primarily outdoor sports clubs, as well as locations such as playgrounds and stadiums, were associated with higher odds ratios, suggesting increased heat illness risk with greater outdoor heat exposure. Additionally, higher odds ratios on Sundays suggest increased heat illness risk, likely due to prolonged activities on non-school days.
These findings highlight the need for the flexible application of HSTs. Specifically, the threshold of 28°C for "ceasing strenuous exercise" and that of 31°C for "ceasing all exercise" should be adjusted considering factors such as club, location, year, month, and DOW.
Clinical implications
Every year, thousands of heat illness cases in Japanese junior high and high school sports clubs highlight the need for enhanced countermeasures. This study reinforces the need for adherence to current HSTs, particularly the need to "cease strenuous exercise" when the WBGT is 28°C or higher but below 31°C and to "cease all exercise" when the WBGT is 31°C or higher. Key measures include consistent WBGT monitoring and responsive actions at club activity. In a prefecture in Japan's Kanto region, approximately 90% of junior high schools measure WBGT, with 64.1% taking action based on the measurement [24]. However, a broader nationwide understanding is still needed.
The stratified analysis in this study suggested that the risk of heat illness varies according to factors such as heat acclimatization, location, and duration. This underscores the need for flexible application of HSTs in school sports clubs, adapting to specific contexts. It is recommended to lower the HSTs by up to 3-4°C lower (roughly one rank) during cooler summer months such as April to June, on days with an unusually high WBGT, in outdoor settings, and on long activity days. Although not directly addressed in this study, lowering the HSTs may also be beneficial in relatively cool regions, after high WBGT days, or with increasing fatigue. In case lowering standards are difficult, enhancing cooling measures to suppress body temperature increases before, during, and after activities [25], promoting indoor practices in air-conditioned environments, and implementing more frequent and shorter breaks are also important strategies.
Limitations
This study analyzed heat illness (mainly EHI) incidents from 2011 to 2019 during junior high and high school sports club activities in Japan. Therefore, it has not been confirmed whether the observed relationship between the WBGT and heat illness is applicable to other years, age groups, countries, or classic heat illness. Future research should expand to a broader demographic and context, especially considering potential changes after the COVID-19 pandemic. The analyzed data lacked details on conditions during heat illness incidents, such as specific exercise, metabolic rates, cooling measures, rest periods, duration, or individual’s physical/mental conditions. Hence, the results may be influenced particularly by factors such as rigorous training, unusual activities (such as punishment runs), exercising in poor health, or effective cooling. Additionally, in stratified analysis, some strata had small sample sizes, which may have affected confidence intervals and statistical significance. Future studies could enhance reliability by incorporating new data on heat illness incidents and detailed circumstances as explanatory variables. In addition, the matching of heat illness cases to WBGT data at the municipality level may not reflect the exact location of the incident, thus affecting the results. More accurate analyses might result from using detailed data on each incident and local WBGT measurements. Finally, since WBGT may underestimate heat stress in environments where sweat evaporation is limited, such as high humidity and low wind speed [26], future research could assess heat illness risk more accurately by considering humidity and wind speed directly in the analysis.