Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological study

Daisuke Nakamura; Hiroki Kinoshita; Kazuo Asada; Takuma Arimitsu; Mikinobu Yasumatsu; Takayuki Ishiwata

doi:10.1371/journal.pone.0275641

. 2022 Nov 7;17(11):e0275641. doi: 10.1371/journal.pone.0275641

Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological study

Daisuke Nakamura ^1,^2,^3,^*, Hiroki Kinoshita ¹, Kazuo Asada ¹, Takuma Arimitsu ⁴, Mikinobu Yasumatsu ^2,^3,⁵, Takayuki Ishiwata ^2,⁵

Editor: Ho Ting Wong⁶

PMCID: PMC9639828 PMID: 36342929

Abstract

Heatstroke is a serious heat-related illness that can even cause death. Heat alert systems play an important role in reducing the number of patients experiencing heat illness, as they encourage preventive actions such as the use of air conditioning, hydration, or other strategies. However, to date, the Japanese hazard classification has not considered seasonal and regional variations, despite clear differences in meteorological conditions across different regions in Japan. Moreover, several studies have reported a difference in thermoregulation between older and younger adults, implying that the hazard classification should also consider age differences. This study examined the relationship between the number of ambulance dispatches related to heat illness (ADRHI) and the Japanese heat hazard classification from 2010 to 2019, focusing on monthly and regional differences. Data from 47 prefectures during the 10-year period were collected and analyzed. ADRHI and wet bulb globe temperature (WBGT) data were collected from Japan’s Ministry of Internal Affairs and Communications and the Ministry of the Environment Heat Illness Prevention Information website, respectively. The findings showed a significant relationship between ADRHI and WBGT_max (p < 0.05, r = 0.74). ADRHI per 100,000 people showed significant differences across months. The post hoc test detected the first steep increase in ADRHI at a WBGT_max of 23°C than at 22°C in June, and at a WBGT_max of 26°C, 27°C, and 25°C in July, August, and September, respectively. Moreover, the first significant increase in ADRHI per 100,000 people at WBGT_max differed across each region, at a WBGT_max of 24°C in Hokkaido-Tohoku, 25°C in Kanto, Kansai, and Chugoku, 26°C in Chubu, 27°C in Shikoku, and 28°C in Kyushu-Okinawa. Further, Poisson regression analysis revealed that the relative risks differed across each region and month. These results imply that the hazard classification should be adjusted according to region and month in Japan.

Introduction

Heatstroke is known to be a serious heat-related illness that may even lead to death [1, 2]. Heat illnesses are complex conditions, and both intrinsic (e.g., health, hydration, acclimation) and extrinsic (e.g., environment, equipment, work-to-rest ratios) factors can influence risk [3]. Among these risk factors, there is a well-established association between high ambient temperature (i.e., heat) and higher rates of mortality and morbidity worldwide [4–7].

Although many studies have reported the relationship between the meteorological condition and the mortality related to heat illness [7–9], few studies have focused on the relationship between morbidity and heat illness [6, 10]. Ambulance dispatches related to heat illness (ADRHI) can approximately represent the morbidity of this condition. In fact, 692 Japanese people died of heat illnesses from June to September 2012 [11], whereas the total number of ADRHI was 45,701 [12], with a mortality rate of 1.5%. In addition, approximately 30% of all ADRHI in Japan’s Kanto area was found in people older than 65 years (Old) [10].

Few studies have examined the relationship between ADRHI and the wet bulb globe temperature (WBGT). WBGT was designed to integrate the various meteorological variables that can influence human heat balance via gains or losses of energy through evaporative cooling (wet bulb temperature), radiant heating (globe temperature), and convection of sensible heat (dry bulb temperature). Currently, in Japan, WBGT is used as the index for heat alert systems.

Previous studies have reported the effectiveness of heat alert systems to decrease the number of heat illnesses. These systems are effective because they can both raise alertness about the possibility of heat illnesses and trigger preventive actions, such as the use of air conditioning, hydration, or other strategies [9, 13–15]. Thus, the criteria or cut-off values of these alert systems play an important role in heat illness prevention. For example, if the criteria do not match the actual mortality or morbidity related to heat illnesses, or when the actual meteorological conditions are different from the forecast, the effectiveness of these alert systems is likely to be weakened [9, 10]. However, despite the clear differences in meteorological condition throughout the year and among different regions in Japan, the hazard classification has not considered seasonal and regional differences. To support this issue, Ng et al. [10] further suggested that the hazard classification in Japan may not capture the vulnerabilities through which heat illnesses can occur at lower temperatures, which occur before the summer. The rainy season in Japan, which typically lasts from June to the end of July, is characterized by hot and humid environments, and people who are not acclimated to heat are exposed to a high risk of heat illnesses, even at a lower air temperature or WBGT. Aging contributes to a weakened heat-dissipation system, suggesting a higher risk for ADRHI in older rather than younger adults (YA) [16].

Given that the causal association between meteorological conditions and incidence of mortality has been demonstrated [7, 8, 17, 18], it is rational for governments and academic societies [19, 20] to have developed and implemented heat warning systems, such as heat alert systems and heat hazard levels, to prevent heat illnesses. For example, the US National Weather Service (NWS) issues heat advisories and warnings (collectively known as “heat alerts”) in advance to forecast extreme heat events. Similarly, since 2006, the Ministry of the Environment of Japan has issued a hazard classification based on the WBGT with the aim to prevent heat illnesses, make people aware of heat events, and encourage preventive actions [21]. The hazard classification comprises five scales, ranging from “danger” (on days with WBGTs over 31°C) to “almost safe” (with WBGTs under 21°C). Moreover, since 2021, for days when the weather forecast indicates that the WBGT will be over <33°C, the ministry issues a heat alert either the day before or early in the morning on the relevant day. However, this classification is not likely to match the actual number of ADRHI in Japan since it does not consider differences in meteorological conditions according to region or time of year.

Accordingly, exploring the characteristics of heat illness occurrences using the number of ADRHI could provide useful information to improve guidelines for hazard classification or heat alert systems. Moreover, although several studies have focused on the relationship between ADRHI and meteorological conditions in Japan, they did not use WBGT, even though this measurement is used in the hazard classification. Therefore, it is worth investigating the relationship between the occurrence of heat illness and the hazard classification—used for the prevention of heat illnesses in Japan—from the perspective of heat illness morbidity using big data analysis. This study aimed to examine the relationship between the number of ADRHI and the Japanese heat hazard classification from 2010 to 2019 in Japan, focusing on monthly and regional differences.

Methods

Study sites

To examine the relationship among the number of ADRHI and the WBGT data, we analyzed data from all prefectures and from the seven regions in Japan.

Data sources

Ambulance dispatches related to heat illness

The data on ADRHI were acquired from the Fire and Disaster Management Agency under the Ministry of Internal Affairs and Communications, Japan [12]. They provided daily statistics of the number of people transported by ambulance in each of the 47 prefectures in Japan from June to September in every year from 2010 to 2019. As this period is summer in Japan, data on both ADRHI and WBGT are published. In this data set, the age of patients was recorded as follows: newborns are those aged less than 28 days, infants are aged from 28 days to 6 years, adolescents are aged from 7 to 17 years, adults are aged between 18 and 64 years, and older adults are aged 65 years and older. Days when no ADRHI transports were recorded were included in the analysis and counted as zero. Consequently, the total data set comprised 57,340 ADRHI and dispatches from 47 prefectures over 10 years.

Wet bulb globe temperature

The data on WBGT for the study period were acquired from the Ministry of Environment Heat Illness Prevention Information website [21]. Information on the exact time and location of each ADRHI in each prefecture was not available through the data set; thus, we used maximal WBGT data from the relevant meteorological station in the prefectural capital. As Tokyo was the only prefectural capital that did not have a meteorological station, we used data from the nearest station, which was about 10 km from the capital. In this study, the maximum WBGT of a given day was defined as WBGT_max. On the Ministry of the Environment Heat Illness Prevention Information website, four WBGT data points were not recorded; hence, we used 57,336 recordings of WBGT_max in the analysis.

Japanese heat hazard classification

The Japanese heat hazard classification is defined by the following five categories: Danger, denoting that exercise is prohibited due to a WBGT ≥ 31°C; Severe Warning, denoting that heavy exercise is prohibited due to a WBGT 28°C–31°C; Warning, denoting that rest should be provided often due to a WBGT 25°C–28°C; Caution, denoting that water should be replenished often due to a WBGT 21°C–25°C; and Almost Safe, denoting that appropriate water replenishment should be provided due to a WBGT < 21°C. This heat hazard classification is published by the Japan Sports Association [22] and used by the Ministry of the Environment to prevent excessive heat illnesses.

Population data and regional settings

We acquired population data from each prefecture from the Ministry of Health, Labour and Welfare website [23], which uploads data annually. Table 1 shows the average population (in thousands) in all prefectures, the seven regional settings during the study period, and locational data.

Table 1. Average population (in thousands) data for all prefectures and seven regions.

Regions	Prefectures	Population (in thousands)	Longitude	Latitude
Hokkaido-Tohoku	Hokkaido	5387	43.06	141.35
Hokkaido-Tohoku	Aomori	1313	40.82	140.74
Hokkaido-Tohoku	Iwate	1279	39.7	141.15
Hokkaido-Tohoku	Miyagi	2327	38.27	140.87
Hokkaido-Tohoku	Akita	1029	39.72	140.1
Hokkaido-Tohoku	Yamagata	1126	38.24	140.36
Hokkaido-Tohoku	Fukushima	1927	37.75	140.47
Kanto	Ibaraki	2918	36.34	140.45
Kanto	Tochigi	1974	36.57	139.88
Kanto	Gunma	1975	36.39	139.06
Kanto	Saitama	7262	35.86	139.65
Kanto	Chiba	6223	35.6	140.12
Kanto	Tokyo	13488	35.69	139.69
Kanto	Kanagawa	9115	35.45	139.64
Chubu	Niigata	2305	37.9	139.02
Chubu	Toyama	1069	36.7	137.21
Chubu	Ishikawa	1155	36.59	136.63
Chubu	Fukui	788	36.07	136.22
Chubu	Yamanashi	838	35.66	138.57
Chubu	Nagano	2103	36.65	138.18
Chubu	Gifu	2035	35.39	136.72
Chubu	Shizuoka	3705	34.98	138.38
Chubu	Aichi	7476	35.18	136.91
Chubu	Mie	1820	34.73	136.51
Kansai	Shiga	1413	35	135.87
Kansai	Kyoto	2611	35.02	135.76
Kansai	Osaka	8838	34.69	135.52
Kansai	Hyogo	5535	34.69	135.18
Kansai	Nara	1368	34.69	135.83
Kansai	Wakayama	965	34.23	135.17
Chugoku	Tottori	574	35.5	134.24
Chugoku	Shimane	696	35.47	133.05
Chugoku	Okayama	1921	34.66	133.93
Chugoku	Hiroshima	2837	34.4	132.46
Chugoku	Yamaguchi	1406	34.19	131.47
Shikoku	Tokushima	759	34.07	134.56
Shikoku	Kagawa	978	34.34	134.04
Shikoku	Ehime	1389	33.84	132.77
Shikoku	Kochi	732	33.56	133.53
Kyushu-Okinawa	Fukuoka	5094	33.61	130.42
Kyushu-Okinawa	Saga	834	33.25	130.3
Kyushu-Okinawa	Nagasaki	1380	32.74	129.87
Kyushu-Okinawa	Kumamoto	1786	32.79	130.74
Kyushu-Okinawa	Oita	1168	33.24	131.61
Kyushu-Okinawa	Miyazaki	1107	31.91	131.42
Kyushu-Okinawa	Kagoshima	1657	31.56	130.56
Kyushu-Okinawa	Okinawa	1426	26.21	127.68

Open in a new tab

Statistical analysis

The WBGT was divided per 1°C, from 20°C to 35°C. The ADRHI was evaluated by absolute value and per 100,000 people (ADRHI_ALL), per 100,000 people aged 65 and older (ADRHI_OLD), and other than ADRHI_OLD (ADRHI_YA). To analyze the relationship between ADRHI and WBGT, we examined Spearman’s correlation coefficient by TIBCO Spotfire® [24]. To compare the differences for the number of ADRHI by month and region, we conducted the Kruskal–Wallis test. When a significant difference was detected, the Dann-Bonferroni test was used for post hoc analysis. In addition, to clarify the relationship between ADRHI and WBGT_max, the WBGT_max of the first significant increase in ADRHI compared to the previous WBGT_max was defined as the first significant increase point. Moreover, to investigate the effects on ADRHI of a one unit increase in WBGT in each region, a Poisson regression analysis was performed using SPSS (version 27.0; SPSS Inc, Chicago, IL). In the Poisson model, the ADRHI was set as the response value, and regions, month, WBGTmax, and proportion of people aged over 65 were set as a predictor. Population size was set as an offset value. The statistical significance was set at p < 0.05.

Results

Table 2 shows the total annual number of ADRHI and the average total annual number of ADRHI_ALL throughout the study period in each prefecture. The highest total annual number of ADRHI per year appeared in Tokyo, with 4,286.7 cases, while the highest total annual number of ADRHI_ALL appeared in Okayama, having 66.0 cases throughout the study period (Table 2). Fig 1 shows the changes in ADRHI by year. The highest number of ADRHI appeared in 2018 (92,710 cases) and the lowest appeared in 2014 (40,048 cases). Fig 2 shows the ADRHI ratios stratified by age and year.

Table 2. The average total number of ADRHI per year (ADRHI / year) and average total number of ADRHI per 100,000 people / year (ADRHI_ALL / year) throughout the 10-year (2010–2019) study period, stratified by prefecture.

Regions	Prefectures	ADRHI / year	ADRHI_ALL / year
Hokkaido-Tohoku	Hokkaido	988.1	18.4
Hokkaido-Tohoku	Aomori	375.3	28.6
Hokkaido-Tohoku	Iwate	485.4	38.0
Hokkaido-Tohoku	Miyagi	916.9	39.4
Hokkaido-Tohoku	Akita	439.2	42.8
Hokkaido-Tohoku	Yamagata	455.1	40.6
Hokkaido-Tohoku	Fukushima	1023.2	53.3
Kanto	Ibaraki	1418.3	48.7
Kanto	Tochigi	899.6	45.6
Kanto	Gunma	1133.9	57.5
Kanto	Saitama	3539.0	48.7
Kanto	Chiba	2546.6	40.9
Kanto	Tokyo	4286.7	31.7
Kanto	Kanagawa	2720.3	29.8
Chubu	Niigata	1165.2	50.6
Chubu	Toyama	387.0	36.3
Chubu	Ishikawa	525.7	45.6
Chubu	Fukui	350.5	44.6
Chubu	Yamanashi	375.7	45.0
Chubu	Nagano	782.8	37.3
Chubu	Gifu	1059.0	52.2
Chubu	Shizuoka	1467.7	39.7
Chubu	Aichi	3689.1	49.3
Chubu	Mie	987.6	54.4
Kansai	Shiga	654.9	46.3
Kansai	Kyoto	1518.4	58.2
Kansai	Osaka	3850.1	43.6
Kansai	Hyogo	2601.5	47.1
Kansai	Nara	764.8	56.1
Kansai	Wakayama	602.7	62.6
Chugoku	Tottori	351.0	61.4
Chugoku	Shimane	378.1	54.4
Chugoku	Okayama	1266.6	66.0
Chugoku	Hiroshima	1405.9	49.6
Chugoku	Yamaguchi	615.1	43.9
Shikoku	Tokushima	382.0	50.6
Shikoku	Kagawa	565.3	57.9
Shikoku	Ehime	761.1	55.0
Shikoku	Kochi	457.0	62.7
Kyushu-Okinawa	Fukuoka	2175.5	42.7
Kyushu-Okinawa	Saga	484.5	58.3
Kyushu-Okinawa	Nagasaki	695.0	50.6
Kyushu-Okinawa	Kumamoto	1104.4	62.0
Kyushu-Okinawa	Oita	627.2	53.9
Kyushu-Okinawa	Miyazaki	607.9	55.2
Kyushu-Okinawa	Kagoshima	1039.7	63.1
Kyushu-Okinawa	Okinawa	753.3	52.7

Open in a new tab

Fig 3 shows the relationship between the number of ADRHI and WBGT_max. A significant relationship between ADRHI and WBGT_max (p < 0.05, r = 0.74, CI: 0.736–0.744, n = 57,336) was observed. The distribution of WBGT_max stratified by month and region is shown in Figs 4 and 5, respectively. The WBGT_max in each month differed significantly, with the differences in all months having differed significantly from other months (p < 0.05, Fig 4). The WBGT_max also differed significantly between regions (p < 0.05, Fig 5).

Fig 4 — * p < 0.05 vs July, August, and September. # p < 0.05 vs June, August, and September. $ p < 0.05 vs June, July, and September. † p < 0.05 vs June, July, and August.

Fig 5 — * p < 0.05 vs all other regions. # p < 0.05 vs all other regions except for Chubu. $ p < 0.05 vs all other regions except for Kanto and Chugoku. † p < 0.05 vs all other regions except for Chubu.

Fig 6 shows ADRHI_ALL stratified by month. The values were 0.12 ± 0.20 (CI: 0.117–0.123), 0.67 ± 0.75 (CI: 0.659–0.683), 0.66 ± 0.64 (CI: 0.645–0.666), and 0.12 ± 0.23 (CI: 0.117–0.125) in June, July, August, and September, respectively. A significant difference was observed across the months, with the post hoc test detecting that all months differed significantly from other months (p < 0.05, Fig 6). The first steep increase in ADRHI_ALL was seen at a WBGT_max of 23°C rather than at 22°C in June, and at a WBGT_max of 26°C, 27°C, and 25°C in July, August, and September, respectively, when compared to the previous WBGT_max (Fig 7).

No significant difference was observed between regions in the number of ADRHI_ALL throughout the study period (p > 0.05). Conversely, when the data were analyzed across regions, the first significant increase in ADRHI_ALL at a WBGT_max differed across each region, at a WBGT_max of 24°C in Hokkaido-Tohoku, 25°C in Kanto, Kansai, and Chugoku, 26°C in Chubu, 27°C in Shikoku, and 28°C in Kyushu-Okinawa, compared to the previous WBGT_max (Table 3). The first steep increases in the number of ADRHI_ALL, ADRHI_OLD, and ADRHI_YA at WBGT_max of all regions are shown in Table 3. Moreover, the relative risks differed for each region and month are shown in Table 4 (LRχ2 = 697742.8, p < 0.001). The relative risks also differed among regions and months. Fig 8 shows the ratio of ADRHI in each hazard level in each region and month. In June, approximately 64.1% of the ADRHI that occurred in the Hokkaido-Tohoku region were related to a WBGT_max of > 25°C, which is categorized as “Caution”—the 2nd lowest level in the hazard classification, whereas only 15.0% of ARDHI occurred in the “Caution” level at Kyushu-Okinawa. However, from July to September, most ADRHI occurred at levels over “Warning”.

Table 3. The first steep increase in ADRHI_ALL, ADRHI_OLD, and ADRHI_YA at WBGT_max in each region.

Regions	ADRHI_ALL (°C)	ADRHI_OLD (°C)	ADRHI_YA (°C)
Hokkaido-Tohoku	24	24	25
Kanto	25	26	26
Chubu	26	26	27
Kansai	25	26	26
Chugoku	25	26	26
Shikoku	27	28	28
Kyushu-Okinawa	28	28	28

Open in a new tab

Table 4. The relative risks for each region, month, WBGTmax, and proportion of people aged over 65.

Predictor	Relative risks
Intercept	< 0.01
Regions
Ref. value Hokkaido-Tohoku	1
Kanto	.859 (.848 - .868)
Chubu	.748 (.740 - .757)
Kansai	.829 (.820 - .839)
Chugoku	.638 (.629 - .647)
Shikoku	.361 (.355 - .367)
Kyushu-Okinawa	.472 (.467 - .478)
Month
Ref. value September	1
June	1.898 (1.87–1.92)
July	2.373 (2.35–2.40)
August	1.748 (1.73–1.77)
WBGTmax	1.492 (1.490–1.495)
Proportion of people aged over 65	.992 (.991 - .993)

Open in a new tab

The values in parentheses are in the 95% probability range.

Discussion

This study aimed to examine the relationship between the number of ADRHI and WBGT data from 2010 to 2019 in Japan, focusing on monthly and regional differences.

To the best of our knowledge, this was the first study to examine the relationship between ADRHI and WBGT in Japan using big data analysis. As expected, the number of ADRHI showed a significant relationship with WBGT_max (p < 0.05, r² = 0.55; Fig 3). This result is similar to the previous study, which reported that r² was 0.48 in three big prefectures in Japan [6]. During our experimental period, the average WBGT_max was highest in July 2018, while the number of ADRHI cases were highest in 2018.

Moreover, our findings showed that among all ADRHI, YAs accounted for 50% of the heat illness (Fig 2); this value was much bigger than the average of 30% in big cities in Kanto shown in a prior study [10]. This difference between studies may be because of different data sources and study settings; the current study collected the ADRHI data from the Fire and Disaster Management Agency in Japan, whereas the cited previous study collected these data from fire departments within each city. Another possible reason is that the examined area in the previous study had a lower percentage of older patients per 100,000 people compared to other areas with a higher percentage in Japan. Thus, since our study covered all regions of Japan, there was a great likelihood that the two studies would yield different percentages regarding older adults.

One of the new findings of this study is the significant difference by month in ADRHI_ALL. Specifically, the highest frequencies of ADRHI_ALL (Fig 6) and ADRHI_OLD were observed in August while the lowest frequencies were observed in June. Those of ADRHI_YA in July and August significantly differed from the other months, whereas those in June and September were not significantly different. The primary reason for these significant differences among the months may be related to meteorological condition variations throughout the year. Indeed, the average WBGT_max in this study was 24.1 ± 2.9°C in June, 28.4 ± 2.7°C in July, 29.3 ± 2.5°C in August, and 25.5 ± 3.3°C in September (Fig 4). This result is partly supported by previous findings, which described a causal relationship between the meteorological condition and the number of heat illnesses [7, 8, 17, 18].

The current result also indicated the potential for the relationship between heat acclimation and ADRHI. Notably, the first steep increase in ADRHI_ALL in June was seen at a WBGT_max of 23°C, while that in July and August were at a WBGT_max of 26°C and 27°C, respectively. It is well known that factors other than weather are involved in the occurrence of heat illness, such as economic conditions, health conditions, and heat acclimation [3, 8]. Among these factors, heat acclimation is the best-known protection against heat illnesses [19, 25]. In general, heat acclimation requires several days of heat exposure [25] or exposure and exercise. As heat acclimation helps to decrease heat stress for the body, it also supports the reduction of the physiological burden that is caused by such weather conditions. Conversely, previous studies suggested that older people are likely to have modest improvement following the heat acclimation program, such as sweat rate response [16] or body fluid regulation [26]. For these reasons, the difference in the first steep increase of WBGT may be assumed to be based on age categories; however, that of the WBGT in June was the same between ADRHI_OLD and ADRHI_YA at 23°C WBGT_max. These results suggested that, at least in June, the hazard levels in Japan should be considered based on not being sufficiently prepared to deal with heat stress because of insufficient heat exposure and heat acclimation, regardless of age. Although there were no data regarding heat acclimation in our study, it can be inferred that heat acclimation may be important for preventing heat illnesses, and shifting the hazard level might serve as an early precautionary measure when warm weather is forecasted in early summer.

While a significant difference was observed in the number of ADRHI among months (Fig 6), no significant differences were found across the regions. However, when analyzed across each region, the first steep increase of WBGT_max was a maximum of 4°C difference between regions. For example, the first steep increase in ADRHI_ALL in Hokkaido-Tohoku (see Table 3) was at 24°C, while that of Kyushu-Okinawa was at 28°C. This trend was also observed in the other categories, that is, ADRHI_OLD and ADRHI_YA (Table 3). Additionally, the relative risks by Poisson regression analysis indicated that the values differed for each region (Table 4). This result was similar to a previous study that revealed different relative risks in the Kanto area in Japan [10]. In contrast, there was no significant difference in ARDHI across regions in this study, supporting a previous finding that the temperature at which heat-related mortality started to increase was lower at higher latitudes (i.e., north Finland), but the trend of the increase in daily mortality rate above that temperature was not significant between regions, such as in north Finland, London, and Athens [5]. The results of the previous study and the current study could suggest the mechanism of heat illness; that is, the fact that when the temperature rises heatstroke occurs. To support this result, when the ADRHI is summarized based on the Japanese heat hazard classification in each month for each region (Fig 8), about 64.1% of the ADRHI that occurred in June in the Hokkaido-Tohoku region was related to a WBGT_max of > 25°C, which categorized “Caution” as the 2nd lowest level of the hazard rank, whereas only 15.0% of ARDHI occurred at the “Caution” level in the Kyushu-Okinawa region. Thus, these results indicated the need to adjust the hazard classification based on the actual occurrence of ARDHI in each region and each month.

Previous studies have suggested that such heat warning systems are important to trigger preventive actions to deal with heat illnesses [13, 27, 28]. Fouillet et al. [13] found that during a 2006 heat wave in France, and given the characteristics of the heat wave, there were approximately 4,400 fewer observed excess deaths than expected; the authors suggested that this finding may have been related to the implementation of adaptive measures, including an early heat warning system that was developed following the severe European heat wave of 2003. Meanwhile, Weinberger et al. [9] reported the effectiveness of NWS heat alerts in preventing mortality across 20 American cities, concluding that the NWS heat alerts were not associated with lower mortality in most cities because of the missed opportunity to implement a heat response plan to avert heat-related deaths. For example, the heat alert only covered 6% of the actual eligible days of the alert in Fayetteville, NC. In addition, other than preventing action for heat illness, previous studies also showed the significance of campaigns for stroke by the National Stroke Foundation of Australia [29]. These preventive actions or implementation of campaigns might contribute to having a large impact not only on a decrease in the number of patients or dispatches but also on the awareness for heat illness or stroke. In Japan, to prevent an excessive increase in heat illnesses, the Ministry of the Environment provides an online forecast of a heat stress index based on the WBGT and information on the hazard levels of heat-related illnesses. However, the hazard classification of heat illnesses is based on a fixed set of criteria derived from the WBGT. In this guideline, a Warning occurs when the WBGT goes over < 25°C and Danger when the WBGT goes over < 31°C. This guideline, notwithstanding, may not capture the actual number of ADRHI which can occur at a lower temperature during the early period of warming weather, such as what happens in June in Japan. In fact, the ADRHI which occurred at a Danger classification in June accounted for 5.1% of all ADRHI in all regions; meanwhile, the ADRHI which occurred under a Caution classification (Almost safe and Caution) in the Hokkaido-Tohoku area in June accounted for 64.1% of all ADRHI in the region. Additionally, further analysis showed differences in the relative risks (Table 4), which implies the difference in risk when there is a one unit increase in WBGT_max. From these notions, to reduce the occurrence of heat illness, it is important: to set an optimal hazard or warning system that matches the actual occurrence of heat illness; to ensure that heat-related information is always up-to-date and accessible via, for example, the internet or a community information transportation tool; and to use this information for developing and implementing heat illness action plans, such as promoting the use of air conditioning, hydration, and measures to avoid an excessive increase in core temperatures [30] (e.g., to reduce heat illness). Nonetheless, in the month of June in Japan, it would likely be better for the hazard levels to be set to lower temperatures because the factors that influence the occurrence of heat illnesses are complex; such lowering of the threshold could make the heat warning system more effective. However, this study could not evaluate the actual effect of the hazard level on ARDHI; thus, future studies should investigate the awareness of the current hazard level and how it affects the implementation of action plans.

Our study has several limitations. First, our collected data did not allow for pinpointing the exact time and location of ADRHI occurrence; however, this does not mean that the occurrence of heat illnesses was unaffected by the local meteorological condition and did not differ by hours. Additionally, the WBGT data used were forecasted live using an estimated value calculated by the following formula; WBGT = 0.735×Ta+0.0374×RH+0.00292×Ta×RH+7.619×SR−4.557×SR2−0.0572×WS−4.064 [31]. Moreover, we acknowledge that our analyzed data are not widely translatable because the occurrence of ADRHI depends on whether the individual is inside a house or not; particularly, an average of 39% of all ADRHI cases reportedly occurred inside the individual’s house during 2017–2019 [12]. Second, we did not have access to factors other than WBGT data. As mentioned, several factors are assumed to be related to ADRHI, such as air pollution, economic status, and health status. However, we could not obtain data corresponding to heat stroke data for this experiment. Addressing this limitation is a concern for future research. Third, this study indicates that meteorological conditions other than WBGT, including daily maximum or apparent temperature, may increase ARDHI risk, particularly the time lag effect of maximum temperature or average apparent temperature [10]. However, this was because the current study aimed to investigate the relationship between the number of ARDHI and Japanese heat hazard levels based on WBGT. Despite several study limitations, our findings provided more realistic WBGT values, and it is believed that the data can help decision-makers in making more well-informed decisions about the development and implementation of action plans; this may ensure more effective measures against heat illnesses in Japan.

In summary, our results showed a significant relationship between ADRHI and WBGT_max (p < 0.05, r = 0.74). Additionally, the number of ADRHI_ALL people showed significant differences by month, with the post hoc test detecting the first steep increase in ADRHI at a WBGT_max of 23°C when compared with that of 22°C in June. For July, August, and September, the WBGT_max were 26°C, 27°C, and 25°C when compared with the previous WBGT_max (Fig 7). In addition, the first significant increase in the number of ADRHI_ALL occurred at 24°C in Hokkaido-Tohoku, 25°C in Kanto, Kansai, and Chugoku, 26°C in Chubu, 27°C in Shikoku, and 28°C in Kyushu-Okinawa, compared with the previous WBGT_max. Further analysis revealed that the relative risks differed for each region and month. These results imply the need to update the hazard classification criteria, adjusting it by month and region.

Acknowledgments

We thank Koya Suzuki (Faculty of Health and Sports Science, JUNTENDO University, Japan), Mr. Kazuhiro Tajima (NTT Com Online Marketing Solutions Corporation, Tokyo, Japan), and Satoshi Inose (Weathernews Inc, Chiba, Japan) for helpful statistical advice.

Data Availability

All ADRHI data are available from The Fire and Disaster Management Agency database (https://www.fdma.go.jp/disaster/heatstroke/post3.html). All WBGT data are available from The Ministry of the Environment database (https://www.wbgt.env.go.jp/record_data.php?region=03&prefecture=44&point=44132). All population data are available from Ministry of Internal Affairs and Communications, Statistics Bureau (https://www.stat.go.jp/data/jinsui/2.html).

Funding Statement

The author(s) received no specific funding for this work.

References

1.Coris EE, Ramirez AM, Van Durme DJ. Heat illness in athletes: the dangerous combination of heat, humidity and exercise. Sports Med. 2004;34:9–16. doi: 10.2165/00007256-200434010-00002 . [DOI] [PubMed] [Google Scholar]
2.Fuhrmann CM, Sugg MM, Konrad CE, Waller A 2nd. Impact of extreme heat events on emergency department visits in North Carolina (2007–2011). J Community Health. 2016;41:146–56. doi: 10.1007/s10900-015-0080-7 . [DOI] [PubMed] [Google Scholar]
3.Leon LR, Bouchama A. Heat stroke. Compr Physiol. 2015;5:611–47. doi: 10.1002/cphy.c140017 [DOI] [PubMed] [Google Scholar]
4.Gasparrini A, Guo Y, Hashizume M, Lavigne E, Zanobetti A, Schwartz J, et al. Mortality risk attributable to high and low ambient temperature: a multicountry observational study. Lancet. 2015;386:369–75. doi: 10.1016/S0140-6736(14)62114-0 . [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Keatinge WR, Donaldson GC, Cordioli E, Martinelli M, Kunst AE, Mackenbach JP, et al. Heat related mortality in warm and cold regions of Europe: observational study. BMJ. 2000;321:670–3. doi: 10.1136/bmj.321.7262.670 . [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Kodera S, Nishimura T, Rashed EA, Hasegawa K, Takeuchi I, Egawa R, et al. Estimation of heat-related morbidity from weather data: a computational study in three prefectures of Japan over 2013–2018. Environ Int. 2019;130:104907. doi: 10.1016/j.envint.2019.104907 . [DOI] [PubMed] [Google Scholar]
7.Li T, Ding F, Sun Q, Zhang Y, Kinney PL. Heat stroke internet searches can be a new heatwave health warning surveillance indicator. Sci Rep. 2016;6:37294. doi: 10.1038/srep37294 . [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Basu R. High ambient temperature and mortality: a review of epidemiologic studies from 2001 to 2008. Environ Health. 2009;8:40. doi: 10.1186/1476-069X-8-40 . [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Weinberger KR, Zanobetti A, Schwartz J, Wellenius GA. Effectiveness of National Weather Service heat alerts in preventing mortality in 20 US cities. Environ Int. 2018;116:30–8. doi: 10.1016/j.envint.2018.03.028 . [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Ng CF, Ueda K, Ono M, Nitta H, Takami A. Characterizing the effect of summer temperature on heatstroke-related emergency ambulance dispatches in the Kanto area of Japan. Int J Biometeorol. 2014;58:941–8. doi: 10.1007/s00484-013-0677-4 . [DOI] [PubMed] [Google Scholar]
11.Ministry of Health, Labour and Welfare. Vital Statistics in Japan [Internet]. [cited 2020 Sep 2]. Available from: https://www.mhlw.go.jp/toukei/saikin/hw/jinkou/geppo/m2018/09.html.
12.The Fire and Disaster Management Agency. Heat stroke information [Internet]. [cited 2019 Dec 12]. Available from: https://www.fdma.go.jp/disaster/heatstroke/post3.html.
13.Fouillet A, Rey G, Wagner V, Laaidi K, Empereur-Bissonnet P, Le Tertre A, et al. Has the impact of heat waves on mortality changed in France since the European heat wave of summer 2003? A study of the 2006 heat wave. Int J Epidemiol. 2008;37:309–17. doi: 10.1093/ije/dym253 . [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Heudorf U, Schade M. Heat waves and mortality in Frankfurt am Main, Germany, 2003–2013: what effect do heat-health action plans and the heat warning system have? Z Gerontol Geriatr. 2014;47:475–82. doi: 10.1007/s00391-014-0673-2 . [DOI] [PubMed] [Google Scholar]
15.Nitschke M, Tucker G, Hansen A, Williams S, Zhang Y, Bi P. Evaluation of a heat warning system in Adelaide, South Australia, using case-series analysis. BMJ Open. 2016;6:e012125. doi: 10.1136/bmjopen-2016-012125 . [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Gerrett N, Amano T, Inoue Y, Kondo N. The sweat glands’ maximum ion reabsorption rates following heat acclimation in healthy older adults. Exp Physiol. 2021;106:302–15. doi: 10.1113/EP088486 . [DOI] [PubMed] [Google Scholar]
17.Armstrong BG, Chalabi Z, Fenn B, Hajat S, Kovats S, Milojevic A, et al. Association of mortality with high temperatures in a temperate climate: England and Wales. J Epidemiol Community Health. 2011;65:340–5. doi: 10.1136/jech.2009.093161 . [DOI] [PubMed] [Google Scholar]
18.Goldberg MS, Gasparrini A, Armstrong B, Valois MF. The short-term influence of temperature on daily mortality in the temperate climate of Montreal, Canada. Environ Res. 2011;111:853–60. doi: 10.1016/j.envres.2011.05.022 . [DOI] [PubMed] [Google Scholar]
19.American College of Sports Medicine, Armstrong LE, Casa DJ, Millard-Stafford M, Moran DS, Pyne SW, et al. American College of Sports Medicine position stand. Exertional heat illness during training and competition. Med Sci Sports Exerc. 2007;39:556–72. doi: 10.1249/MSS.0b013e31802fa199 . [DOI] [PubMed] [Google Scholar]
20.Japan Society of Biometeorology. Guide to preventing heatstroke in everyday life. Version 3. [Internet]. 2013. [cited 2020 Aug 10]. Available from: https://seikishou.jp/cms/wp-content/files/news/shishin.pdf. [Google Scholar]
21.Ministry of the Environment. Heat illness prevention information [Internet]. [cited 2020 Sep 5]. Available from: https://www.wbgt.env.go.jp/en/wbgt.php.
22.Japan Sports Association. Guidebook for the prevention of heatstroke during sports activities. Japan Sports Association; 1994. [Google Scholar]
23.Ministry of Internal Affairs and Communications Statistics Bureau. [Internet]. [cited 2020 Aug 8]. Available from: https://www.stat.go.jp/data/jinsui/2.html.
24.Allden SJ, Ogger PP, Ghai P, McErlean P, Hewitt R, Toshner R, et al. The transferrin receptor CD71 delineates functionally distinct airway macrophage subsets during idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 2019;200:209–19. doi: 10.1164/rccm.201809-1775OC . [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Daanen HAM, Racinais S, Périard JD. Heat acclimation decay and re-induction: a systematic review and meta-analysis. Sports Med. 2018;48:409–30. doi: 10.1007/s40279-017-0808-x . [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Takamata A, Ito T, Yaegashi K, Takamiya H, Maegawa Y, Itoh T, et al. Effect of an exercise-heat acclimation program on body fluid regulatory responses to dehydration in older men. Am J Physiol. 1999;277:R1041–50. doi: 10.1152/ajpregu.1999.277.4.R1041 . [DOI] [PubMed] [Google Scholar]
27.Martinez GS, Imai C, Masumo K. Local heat stroke prevention plans in Japan: characteristics and elements for public health adaptation to climate change. Int J Environ Res Public Health. 2011;8:4563–81. doi: 10.3390/ijerph8124563 . [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Wendt D, van Loon LJ, van Marken Lichtenbelt WD. Thermoregulation during exercise in the heat: strategies for maintaining health and performance. Sports Med. 2007;37:669–82. doi: 10.2165/00007256-200737080-00002 . [DOI] [PubMed] [Google Scholar]
29.Bray JE, Mosley I, Bailey M, Barger B, Bladin C. Stroke public awareness campaigns have increased ambulance dispatches for stroke in Melbourne, Australia. Stroke. 2011;42:2154–7. doi: 10.1161/STROKEAHA.110.612036 . [DOI] [PubMed] [Google Scholar]
30.Nakamura D, Muraishi K, Hasegawa H, Yasumatsu M, Takahashi H. Effect of a cooling strategy combining forearm water immersion and a low dose of ice slurry ingestion on physiological response and subsequent exercise performance in the heat. J Therm Biol. 2020;89:102530. doi: 10.1016/j.jtherbio.2020.102530 . [DOI] [PubMed] [Google Scholar]
31.Ono M, Tonouchi M. Estimation of wet-bulb globe temperature using generally measured meteorological indices. Jpn J Biometeor. 2014;50:147–57. doi: 10.11227/seikisho.50.147 [DOI] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0275641.r001

Decision Letter 0

Ho Ting Wong

31 Aug 2021

PONE-D-21-20370

Trends in ambulance dispatches related to heat illness from 2010 to 2019; A secondary data analysis

PLOS ONE

Dear Dr. Nakamura,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Oct 15 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ho Ting Wong, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

3. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data

4. We note that Figure 1 in your submission contain map images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

1. You may seek permission from the original copyright holder of Figure 1 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

2. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The paper aimed to identify the association between the number of ambulance dispatches related to heat illness (ADRHI) and WBGT. And the authors want to propose the hazard classification needed to be adjusted in different regions. The idea is very good but the current results cannot support your conclusion so far. I would like to suggest some directions for authors to improve your results.

Major revision:

1. In your data, you have daily ADRHI data and also age group information. If you can apply zero-inflated Poisson model on your data, you can know how the effects on ADRHI by one unit increase of WBGT and also identify the effects from different age groups. This contribution will be very large because the heat illness might affect those vulnerable population. The alert signals can be issued to those specific population.

2. In addition, the regions have different weather conditions, you can stratify the models by seven regions and compare their effect’s differences. Then, you can propose different alerting model in different regions.

3. I will encourage the authors to use the daily data rather monthly data to show the effect from WBGT.

4. If you want to summarized the effects from seven regions, you can apply meta-regression to summarized their total effect.

Minor revision:

1. In the abstract, WBGT is needed to spell out the full name at the first occurrence.

2. The study design in the title can replace “a secondary data analysis” with “an ecological study”.

Reviewer #2: This study examines the relationship between ADHI and WBGT. I believe this is an important topic, with both academic and policy implications. I have the following suggestions to the authors:

1. Abstract: State the full name (wet bulb globe temperature) for WBGT in the abstract.

2. Methodology: Currently this analysis examines only two variables, ADHI and WBGT. But as the authors' have stated, there are many other factors that influence heat illness. The authors can consider using other statistical methods that can have multiple independent variables (including WBGT and others) and one dependent variable (ADHI), so that we can understand even when other factors are considered or controlled, WBGT still have positive relationship with ADHI.

3. other factors to be considered: this is related to point #2. I am wondering if the following factors also influence ADHI: (1) percentages of elderly, percentages of living alone, and/or percentages of elderly living alone. (2) the ratio between numbers of hospitals to population sizes (e.g., how many hosptials per, say, 1 million people). (3) poverty rate. (4) percentages of households/families that have air conditioners. (5) population density. (6) gender.

4. Heat awareness influence ADHI: The following paper suggest that heat awareness influence ADHI. Although the authors have listed this as limitations, I think that the authors can still think about how to incorporate this factor in their analysis.

https://www.ahajournals.org/doi/full/10.1161/STROKEAHA.110.612036

5. Lit review: Currently the manuscript does not have good discussion on the factors that influence heat illness or ADHI. I strongly suggest the authors to include this part, so that readers can evaluate the results of this study. For example, I am wondering whether the max daily temperature in one day or the duration of high temperature (high temp that last several days) matter more for heat illness. A proper lit review can clarify the questions.

6. fig 5: fig 5 suggests that the max temp do not result in most ADHI. That is very interesting, and the authors can elaborate more on this.

7.Fig 9: the analysis suggested that there are no regional differences between regions, but I am wondering how factors listed in point #3 influence the results. For example, how the percentage of elderly influence the results?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2022 Nov 7;17(11):e0275641. doi: 10.1371/journal.pone.0275641.r002

Author response to Decision Letter 0

24 Oct 2021

Reviewer #1

Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological study

Thank you for your constructive review of our manuscript. Please note that our major revision of the statistical analysis is based on comments from both reviewers. The major revision is as follows:

1. We analyzed the ARDHI data in three groups—All population categories, older people categories (ADRHI_Old), and other than older people categories (ADRHI_YA).

2. To indicate the regional difference in ARDHI, we added the Poisson regression analysis.

3. We added the Fig.6.

The issues you pointed out were all very important. We have made revisions accordingly. Our point-by-point responses are written below each comment. We hope that these revisions will address your concerns.

We highlighted under the line and red font where we modified in the revised manuscript.

Reviewer #1 comments

The paper aimed to identify the association between the number of ambulance dispatches related to heat illness (ADRHI) and WBGT. And the authors want to propose the hazard classification needed to be adjusted in different regions. The idea is very good but the current results cannot support your conclusion so far. I would like to suggest some directions for authors to improve your results.

Major revision:

Ans.

Thank you for your very constructive comments. As you pointed out, we conducted Poisson regression analysis to investigate the effect on ARDHI by one degree of WBGT across each age category. The result of exponential β value is shown in Table 4 (P.16, L.255), and other parts for the explanation for the analysis. (P.10, L.183-185)

【P.10, L.183-185】

To analyze the relationship between ADRHI and WBGT, we performed Spearman’s correlation coefficient. To compare the differences for the number of ADRHI by month and region, we conducted Kruskal–Wallis test. When a significant difference was detected, to find where the first steep and significant increase in ADRHI occurred, a comparison circle was performed. Moreover, to investigate the effects on ADRHI by one unit increase of WBGT, Poison regression analysis was performed using SPSS (version 27.0; SPSS Inc, Chicago, IL). The statistical significance was set at p < 0.05.

Ans.

Thank you for your pertinent comment. Similar to the answer above, we analyzed the data across each region and modified the paragraph in the discussion section related to the regional differences, from P.18 L.309 to P.19 L.338, to describe the importance of applying different alerting models by each region from this analysis. And we also added the Table.3 (P.15, L.248)

【P.18 L.309 to P.19 L.338】

These results suggested that, at least in June, the hazard levels in Japan should be considered based on not being sufficiently prepared to deal with heat stress because of insufficient heat exposure and heat acclimation. Although there were no data regarding heat acclimation in our study, it can be inferred that heat acclimation may be important for preventing heat illnesses, and shifting the hazard level might serve as an early precautionary measure when warm weather is forecasted in early summer.

While a significant difference was observed in the number of ADRHI for July and August compared to June and September (Fig. 7), no significant differences were found across the regions. However, when analyzed across each region, the first steep increase of WBGTmax was a maximum of 2℃ apart between regions. For example, the first steep increase of ADRHI_ALL in Hokkaido-Tohoku and Chugoku (see Table 3), while that of Kanto, Kansai, Shikoku, and Kyushu-Okinawa was 28℃. This trend was also observed in the other categories, that is ADRHI_Old and ADRHI_YA (Table. 3). Additionally, to investigate the effects on ADRHI by one unit increase of WBGT, the exponential β value was calculated, and the values differed across each age category in each region (Table. 4). This result corroborates with a previous study that revealed different relative risks in Kanto area in Japan [10]. In contrast, there is no significant difference in ARDHI across regions in the current study, supporting a previous finding that the temperature at which heat-related mortality started to increase was lower at higher latitudes (i.e., north Finland), but the trend of the increase in daily mortality rate above that temperature was not significant between the regions such as in north Finland, London, and Athens [5]. The results of the previous study and the current study could suggest the mechanism of heat illness, that is, the fact that when the temperature rises heatstroke occurs. To support this result, when the ADRHI is summarized based on the Japanese heat hazard classification in each month for each region (Fig.9), about 64.1% of the ADRHI that occurred in June in the Hokkaido-Tohoku region was related to a WBGTmax of > 25°C, which categorized “Caution” as 2nd lowest level of the hazard rank, whereas only 15.0% of ARDHI occurred in the “Caution” at the Kyushu-Okinawa. Thus, these results indicated the need to adjust the hazard classification based on the actual occurrence of ARDHI in each region and each month.

3. I will encourage the authors to use the daily data rather monthly data to show the effect from WBGT.

Ans.

Thank you for this pertinent comment. As you pointed out, the changes in daily meteorological conditions should have an impact on the occurrence of ARDHI. On the other hand, this paper is focusing on the analysis of the trend of occurrence of each hazard level from the point of view of the big data. We agree to the importance of daily-based analysis on ARDHI, but the purpose of the current study was not to investigate the occurrence of ARDHI from the daily aspect.

However, we agree that future research requires an analysis based on your suggestion.

4. If you want to summarized the effects from seven regions, you can apply meta-regression to summarized their total effect.

Ans.

Thank you for your thoughtful suggestion. We agree that the use of meta-analysis is useful if we are to summarize the effects from seven regions. However, our study focused on the trend in the occurrence of each hazard level using big data, and showing the real number of ARDHI by each region as the first priority. We have added the result of the Poisson regression model to clearly indicate the regional difference in the occurrence of ARDHI.

Minor revision:

1. In the abstract, WBGT is needed to spell out the full name at the first occurrence.

Ans.

We have revised it accordingly (P2, L38)

【P2, L38】

ADRHI and wet bulb globe temperature (WBGT) data

2. The study design in the title can replace “a secondary data analysis” with “an ecological study”.

Ans.

Thank you for your suggestion. We modified the title from “Trends in ambulance dispatches related to heat illness from 2010 to 2019; A secondary data analysis” to “Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological study” (See the title)

【See the title】

Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological study

Reviewer #2

Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological study

Thank you for your very constructive review of our manuscript. Please note that our major revision of the statistical analysis is based on comments from both reviewers, the major revision was as follows:

1. We analyzed the ARDHI data in three groups—All population categories, older people categories (ADRHI_Old), and other than older people categories (ADRHI_YA).

2. To indicate the regional difference in ARDHI, we added the Poisson regression analysis.

3. We added the Fig.6.

All issues you pointed out were all very important. We have made revisions accordingly. Our point-by-point responses are written below each comment. We hope that these revisions will address your concerns.

We highlighted under the line and red font where we modified in the revised manuscript.

Comments by Reviewer #2

This study examines the relationship between ADHI and WBGT. I believe this is an important topic, with both academic and policy implications. I have the following suggestions to the authors:

1. Abstract: State the full name (wet bulb globe temperature) for WBGT in the abstract.

Ans.

We have expanded the full name accordingly (P2, L38)

【P2, L38】

ADRHI and wet bulb globe temperature (WBGT) data

Ans. (For the answer Q2 and Q3)

Thank you for your important suggestion. As you pointed out, several factors are related to the incidence of ARDHI, including poverty rate, the ratio between numbers of hospitals to population sizes, or percentages of households/families that have air conditioners. Unfortunately, we did not find these data in each prefecture and every year. We could only find the population data and WBGT data, and these data can be linked to the ARDHI data, which were used in the current study. In addition, we were also interested in the gender difference of ARDHI, but the data source of ARDHI did not include the information on gender. On the other hand, we actually performed multiple regression analysis to investigate factors affecting the occurrence of ARDHI. We used WBGT, Month, and the percentage of elderly people as an independent variable. However, the result was not interesting; WBGT and Month highly affected the ARDHI. If we have data on living situation, living alone or not or use of air condition or not, we can expect an interesting result if we apply multiple regression analysis; but, we could not find these data. Similar to your concern, we acknowledge that providing significant alerting information for heat illness as well as effects of other factors on the occurrence of ARDHI other than meteorological conditions is very important.

Ans.

Thank you for your useful comment and suggestion. We have deleted the description on the awareness of “the potential effect of implementing heat illness prevention plans” and added information based on your suggestion (P20, L350-355).

【P20, L350-355】

In addition, other than preventing action for heat illness, previous studies also showed the significance of campaigns for stroke by the National Stroke Foundation of Australia [29]. These preventive actions or campaigns implementation might contribute to having a large impact not only on a decrease in the number of patients or dispatches but also on the awareness for heat illness or stroke.

Ans.

Thank you for your meaningful suggestion. As you mentioned above, the occurrence of heat illness related to the max daily temperature in one day or the duration of high temperature (high temp that lasts several days) matters more for heat illness, and this hypothesis is supported by several studies. Actually, in the current study, there is a significant relationship between ARDHI and -1day WBGTmax and similar results in the main text. However, because the purpose of the current study was to investigate the relationship between the number of ARDHI and Japanese heat hazard level using WBGT, we could not incorporate your suggestion; rather, we recommend it for future studies.

(P.22 L395-404)

【P.22 L395-404】

Third, the current study indicates that meteorological conditions other than WBGT, including daily maximum or apparent temperature, may increase ARDHI risk; particularly the time lag effect of maximum temperature or average apparent temperature [10]. However, this was because the current study aimed to investigate the relationship between the number of ARDHI and Japanese heat hazard levels based on WBGT. Despite several limitations of the study, our findings provided more realistic WBGT values, and it is believed that the data can help decision makers in making more well-informed decisions about the development and implementation of action plans; this may ensure more effective measures against heat illnesses in Japan.

6. fig 5: fig 5 suggests that the max temp do not result in most ADHI. That is very interesting, and the authors can elaborate more on this.

Ans.

Thank you for your comment, we elaborated on this in the main manuscript. (P.17 L.289 to P.18 L.294)

【P.17 L.289 to P.18 L.294】

Interestingly, despite the strong relationships between ARDHI and WBGT (Fig. 4), most ADRHI did not occur in the highest WBGTmax (Fig.6). This corresponds with each of the given WBGT. For example, a total of 2 days were < 35℃ WBGTmax during the study period while 28℃ WBGTmax was observed for 39739 days. Additionally, the ratio of ARDHI per day in each case was 182 (364 cases/2 days) and 8.0 (39739 cases/4958 days).

Ans.

Thank you for your constructive comments. First, to clarify the age category effect, we analyzed the ARDHI data in three categories, ADRHI_ALL, ADRHI_Old, and ADRHI_YA.

In addition, as the other reviewer also suggested, we performed Poisson regression analysis to clarify the effect on ARDH by one unit increase of WBGT across each age category. As expected, the result of the Poisson regression analysis shows the difference of exponential β value across each region and age category. This result is shown in Table. 4. (P.16 L.255)

However, the first steep increase in WBGTmax in ARDHI showed no significant difference between ADRHI_Old, and ADRHI_YA in June (P18, L.304-314). We added these very meaningful results in line with your pertinent comment. We appreciate your comments.

【P18, L.304-314】

Conversely, previous studies suggested that Older people are likely to have modest improvement following the heat acclimation program, such as sweat rate response [16] or body fluid regulation [26]. For these reasons, it is likely to the difference in first steep increase WBGT between the age categories, however, that of the WBGT in June was the same between ADRHI_Old and ADRHI_YA at 24℃ WBGTmax. These results suggested that, at least in June, the hazard levels in Japan should be considered based on not being sufficiently prepared to deal with heat stress because of insufficient heat exposure and heat acclimation. Although there were no data regarding heat acclimation in our study, it can be inferred that heat acclimation may be important for preventing heat illnesses, and shifting the hazard level might serve as an early precautionary measure when warm weather is forecasted in early summer.

Attachment

Submitted filename: Reply_to_the_Reviewer_2.docx

Click here for additional data file.^{(24.9KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0275641.r003

Decision Letter 1

Ho Ting Wong

15 Nov 2021

PONE-D-21-20370R1Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological studyPLOS ONE

Dear Dr. Nakamura,

Please submit your revised manuscript by Dec 30 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ho Ting Wong, PhD

Academic Editor

PLOS ONE

Journal Requirements:

Additional Editor Comments (if provided):

After the first round of revision, a lot of problems can still be observed in the manuscript. Some are listed below.

1. Table 2: it is not necessary to show both Total ADRHI and ADRHI/year, as they are just related by a factor of 1/10.

2. Table 2: the name of the first row should be clearer, it is difficult to remember the difference between Total ADRHI and ADRH_ALL/ year. The variable names used in the statistical software should be translated into human-readable names for putting on the manuscript.

3. Figure 3, it is difficult to distinguish between the colors, perhaps bar charts will be better.

4. Figures 5 and 6, again, the figures are atypical graphs that are not easy to be understood by general readers.

5. Line 203-219: The paragraphs are not clear which have to be rewritten.

6. Figure 7, and Line 210: they are talking about two different things.

7. Table 4, 95%CI and significant levels should be given.

In general, the manuscript should be amended to reduce the graphs and tables which are not necessary, the current style is not smooth enough.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #1: The authors have fully addressed my concerns.

One minor suggestion:

1. The variables' name (i.e. ADRHI_ALL, ADRHI_Old, and ADRHI_YA) can be removed from the abstract.

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

PLoS One. 2022 Nov 7;17(11):e0275641. doi: 10.1371/journal.pone.0275641.r004

Author response to Decision Letter 1

29 Nov 2021

Reviewer #1

Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological study

Thank you again for your careful review of our manuscript. We modified the manuscript based on your comments as indicated below.

1. The variables' name (i.e. ADRHI_ALL, ADRHI_Old, and ADRHI_YA) can be removed from the abstract.

Response

Per our suggestion, we deleted the variables' names, i.e., ADRHI_ALL, ADRHI_Old, and ADRHI_YA from the abstract.

Attachment

Submitted filename: Reply_to_the_Reviwer_1_final.docx

Click here for additional data file.^{(22.3KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0275641.r005

Decision Letter 2

Ho Ting Wong

7 Feb 2022

PONE-D-21-20370R2Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological studyPLOS ONE

Dear Dr. Nakamura,

Please submit your revised manuscript by Mar 24 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ho Ting Wong, PhD

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Additional Editor Comments (if provided):

I have also made some minor comments in addition to reviewer 1's comments

Line 181: What is a comparison circle?

Figure 5: it is better to show all months. This is because you mentioned the difference between months in the main text, but the figure does not show the difference. The situation is the same in Figure 7.

Figure 6: You mentioned that some places have the difference, but the figure shows that all the places are quite similar. How about you highlight the difference in the figure and let readers know the statistics test result?

For figure 8, it is not good for readers to guess the meaning of “*” because you are showing people a figure instead of a typical statistical table.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #1: Minor suggestion:

Abstract

1.(Old), (YA) can be removed from the abstract.

2. β value can be replaced by "the coefficients".

In Table 4: Old means "elderly"? Please change the term and define the age of elderly in the table's footnote.

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

PLoS One. 2022 Nov 7;17(11):e0275641. doi: 10.1371/journal.pone.0275641.r006

Author response to Decision Letter 2

7 Mar 2022

Reviewer #1

Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological study

Thank you again for your careful review of our manuscript. We modified the manuscript based on your comments as indicated below.

1.　(Old), (YA) can be removed from the abstract.

Response

We have removed (Old), (YA) from the abstract.

【Abstract】

Moreover, several studies have reported an age difference in thermoregulation between older and younger adults, implying that the hazard classification should also consider age differences. 　　

2. β value can be replaced by "the coefficients".

Response

We have replaced β value with the coefficients in the manuscript. (L.49, 245, 253, 322, 367, 414.)

・Further, Poisson regression analysis revealed that the exponential of the coefficients differed（L.49）　

・Moreover, the exponential of the coefficients differed in each region and in each age category (Table. 4). （L.243）　

・Table. 4 The exponential of the coefficients by each region and age（L.251）　

・Additionally, to investigate the effects on ADRHI by one unit increase of WBGT, the exponential of the coefficients was calculated, （L.315）　

・　Additionally, further analysis showed differences in the exponential of the coefficients (Table. 4) （L.360）　

・Further analysis revealed that the exponential of the coefficients differed in each region and age category. （L.407）　

3.　In Table 4: Old means "elderly"? Please change the term and define the age of elderly in the table's footnote.

Response

In the text, we have defined “Old” as those aged 65 years old and above, along with the definition for all the age categories. Thus, we have used “Old” in all figures and tables. I hope this is sufficient.

Attachment

Submitted filename: Reply_to_the_Reviwer_1.docx

Click here for additional data file.^{(24.8KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0275641.r007

Decision Letter 3

Ho Ting Wong

24 Mar 2022

PONE-D-21-20370R3Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological studyPLOS ONE

Dear Dr. Nakamura,

Referring to Figure 6, you mentioned that * P<0.05 vs July, August, and September. However, visually the bar represents June looks the same as the bar represents September. I would like to confirm if there is any mistake. Moreover, please also have a check on the same issue on other figures.

Regarding Figure 7, please add the unit for the x-axis. (i.e. WBGTmax). Please also have a check on the same issue on other figures.

Please submit your revised manuscript by May 08 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ho Ting Wong, PhD

Academic Editor

PLOS ONE

Journal Requirements:

Additional Editor Comments:

Regarding Figure 7, please add the unit for the x-axis. (i.e. WBGTmax). Please also have a check on the same issue on other figures.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

PLoS One. 2022 Nov 7;17(11):e0275641. doi: 10.1371/journal.pone.0275641.r008

Author response to Decision Letter 3

29 Mar 2022

nothing

Attachment

Submitted filename: Reply_to_the_Editor_5th.submission_final.docx

Click here for additional data file.^{(23.1KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0275641.r009

Decision Letter 4

Ho Ting Wong

11 Apr 2022

PONE-D-21-20370R4Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological studyPLOS ONE

Dear Dr. Nakamura,

==============================

Regarding the problem in Figure 6, how about you explicitly write down the 95% CI in an appropriate location such as the main text or figure? I think this can make things transparent. Please do the same on other figures with the same problem.

==============================

Please submit your revised manuscript by May 26 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ho Ting Wong, PhD

Academic Editor

PLOS ONE

Journal Requirements:

Additional Editor Comments:

Regarding the problem in Figure 6, how about you explicitly write down the 95% CI in an appropriate location such as the main text or figure?

I think this can make things transparent.

Please do the same on other figures with the same problem.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

PLoS One. 2022 Nov 7;17(11):e0275641. doi: 10.1371/journal.pone.0275641.r010

Author response to Decision Letter 4

17 Apr 2022

Attachment

Submitted filename: Reply_to_the_Editor_6th.docx

Click here for additional data file.^{(24.1KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0275641.r011

Decision Letter 5

Ho Ting Wong

9 May 2022

PONE-D-21-20370R5Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological studyPLOS ONE

Dear Dr. Nakamura,

Please submit your revised manuscript by Jun 23 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ho Ting Wong, PhD

Academic Editor

PLOS ONE

Additional Editor Comments:

I am sorry that because the statistics part of your manuscript is still unclear to me, I have asked a statistical expert to future comment on your manuscript.

His comment is attached for your further action, please.

Thank you for your patience.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: I Don't Know

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #3: Yes

**********

6. Review Comments to the Author

Reviewer #3: How was the “first significant increase” defined? Was it through statistical testing?

Line 136- June 2010 to September 2019 is 9 years and 122 days.

Table 1 is not directly relevant to the main result and can be a supplement table.

ADRHI_old and ADRHI_YA- how were they defined? Number of ADRHI in old or young per 100,000 old or young? That is, both the numerator and denominator are limited the subgroup (old or young)?

How was the Poison regression set up? Was ADRHI count used as the response variable? Has the model accounted for the difference in population size for different prefecture? Were separate models fitted for different regions and different age groups? What were the coefficients in Table 4? Coefficient for WBGT as a continuous variable? There be may be curvature in the relationship and WBGT-squared term may also be needed. Were any covariates included? Potential covariates of interest to include are region, the proportion of 65 and above in the population, some measurements of socioeconomic levels of the prefecture, year, month of the year. By fitting all the data in the same model you can perform statistical test to see if the effects of WBGT differ for different regions or differ for different month.

What happened in 2018? Are there explanations for the sudden increase of ADRHI? Is this due to a particular region?

Figure 1. maybe overlay the histogram with a line to indicate the average WBGT across the years to see if the changes correlate?

Figure 2. what is shown here? the proportion of ADRHI in each age category for each of the 10 years? This figure does not seem to contain much information and there was not much text in the manuscript describing the Figure either. Maybe it should be a supplement?

Figure 3 is very informative and it shows the relationship between ADRH and WBGT. Similar plots showing ADRH in older and younger populations will be helpful.

Figure 4 and 5 are not very informative to me. They only show the distribution of WBGT but the primary interest is the relationship between WBGT and ADRH. It will be more of interest to show ADRH vs WBGT stratified by month and ADRHI_ALL vs WBGT stratified by region instead.

Figure 7. How was the comparison done? Each ADHRI level vs the previous level? Is this how “first significant increase” was defined? Please add the details in the method section.

Figure 8 was mentioned in the discussion. The relevant text should be in the result section instead. In the Poisson regression model what if the heat hazard classification is used as a predictor, will there be an interaction between region and heat hazard classification?

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No

PLoS One. 2022 Nov 7;17(11):e0275641. doi: 10.1371/journal.pone.0275641.r012

Author response to Decision Letter 5

21 May 2022

Reviewer #3:

Thank you for your review of our manuscript.

We have further modified our manuscript according to your suggestions.

We hope you are satisfied with our response.

To clearly indicate our revisions, the revised parts of the manuscript are in red colored font.

Q. 1．How was the “first significant increase” defined? Was it through statistical testing?

Response

Thank you for your comment. We have added a description of the definition of the first significant increase to the manuscript as follows:

Lines 175–177

To compare the differences for the number of ADRHI by month and region, we conducted Kruskal–Wallis test. When a significant difference was detected, the Dann-Bonferroni test was used for post hoc analysis. In addition, to clarify the relationship between ADRHI and WBGTmax, the WBGTmax of the first significant increase in ADRHI compared to the previous WBGTmax was defined as the first significant increase point. Moreover, to investigate the effects on ADRHI by one unit increase of WBGT, Poisson regression analysis was performed using SPSS (version 27.0; SPSS Inc, Chicago, IL). The statistical significance was set at p < 0.05.

Q. 2.

Line 136- June 2010 to September 2019 is 9 years and 122 days.

Response

We used data from the years 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, and 2019. Hence, we considered the period to be 10 years.

Q. 3.

Table 1 is not directly relevant to the main result and can be a supplement table.

Response

As you pointed out, Table 1 is not directly related to the main result; however, this table is very important to understand the results of the study. Because the investigation of this research focused on the regional differences in ADRHI, we think Table 1 is needed to understand the geographical locations of the regions being analyzed. To clarify, each region’s location and the specific location data helps to understand the meteorological conditions, which are very important factors in the occurrence of heatstroke. Thus, we do not think this table should be a supplement. We hope you understand.

Q. 4. ADRHI_old and ADRHI_YA- how were they defined? Number of ADRHI in old or young per 100,000 old or young? That is, both the numerator and denominator are limited the subgroup (old or young)?

Response

Thank you for your comment. We have included definitions of ADRHI_Old and ADRHI_YA in the Methods section as follows:

Lines 168–170

The WBGT was divided per 1℃, from 20℃ to 35℃. The ADRHI was evaluated by absolute value and per 100,000 people (ADRHI_All), per 100,000 people aged 65 years and older (ADRHI_Old), and other than ADRHI_Old (ADRHI_YA).

Q.5

Response

Thank you for your comment. Before answering this question, we would like to provide an excerpt of the communication between another reviewer, the editor, and us.

Regarding the statistical analysis, we were provided some suggestions by Reviewer #2 during the first peer review process almost 6 months ago. We have attached our responses from that round of revisions below:

Q from Reviewer #2:

Response from us:

Thank you for your very constructive comments. As you pointed out, we conducted Poisson regression analysis to investigate the effect on ARDHI by one degree of WBGT across each age category. The result of exponential β value is shown in Table 4, and other parts for the explanation for the analysis.

After the first revision, Reviewer #2 was satisfied with our response. The editor required clarifications in the 2nd review process 5 months ago as follows:

Q from the Editor:

7. Table 4, 95%CI and significant levels should be given.

Response from us:

We added the 95%CI in Table 4 in parenthesis.

Regions All age categories Old Adults Adolescents Infants Newborns

Hokkaido-Tohoku 1.43

(1.43–1.43) 1.42

(1.42–1.43) 1.46

(1.45–1.47) 1.39

(1.37–1.40) 1.31

(1.27–1.35) NS

Kanto 1.56

(1.56–1.57) 1.56

(1.55–1.56) 1.61

(1.60–1.61) 1.5

(1.49–1.51) 1.39

(1.36–1.41) NS

Chubu 1.54

(1.53–1.54) 1.53

(1.53–1.54) 1.58

(1.57–1.59) 1.47

(1.46–1.49) 1.39

(1.35–1.43) NS

Kansai 1.56

(1.55–1.56) 1.56

(1.55–1.57) 1.59

(1.58–1.60) 1.47

(1.46–1.49) 1.44

(1.40–1.48) NS

Chugoku 1.37

(1.37–1.38) 1.36

(1.35–1.37) 1.40

(1.39–1.41) 1.35

(1.34–1.37) 1.34

(1.28–1.40) No-ADRHI

Shikoku 1.55

(1.53–1.56) 1.53

(1.51–1.54) 1.60

(1.57–1.62) 1.50

(1.47–1.53) 1.37

(1.26–1.48) No-ADRHI

Kyushu-Okinawa 1.58

(1.57–1.58) 1.56

(1.55–1.57) 1.63

(1.62–1.64) 1.54

(1.52–1.56) 1.44

(1.38–1.50) NS

ALL 1.52

(1.52–1.52) 1.50

(1.57–1.57) 1.57

(1.57–1.57) 1.47

(1.46–1.47) 1.38

(1.37–1.40) 1.13

(1.02–1.26)

After the 2nd revision, Reviewer #2 had a few minor comments which were addressed 3 months ago as below:

Q from Reviewer #2:

2. β value can be replaced by "the coefficients".

Response from us:

We have replaced β value with the coefficients in the manuscript.

・Further, Poisson regression analysis revealed that the exponential of the coefficients differed（Line 46）　

・Moreover, the exponential of the coefficients differed in each region and in each age category (Table 4). （Line 242）　

・Table 4 The exponential of the coefficients by each region and age（Line 253）　

・Additionally, to investigate the effects on ADRHI by one unit increase of WBGT, the exponential of the coefficients was calculated （Line 317）　

・　Additionally, further analysis showed differences in the exponential of the coefficients (Table 4) （Line 362）　

・Further analysis revealed that the exponential of the coefficients differed in each region and age category. （Line 410）　

Response

We hope that these communications have clarified your questions. Please find point-by-point responses to your comments below:

・Was ADRHI count used as the response variable?

Yes.

・Has the model accounted for the difference in population size for different prefecture?

Yes, we accounted for the population size in the prefectures; however, it was not realistic to collect data from all 47 prefectures. As mentioned above, this study focused on the regional differences in ADRHI; hence, we conducted analysis on each region. Regarding your comment, you proposed that Table 1 be moved to the supplement file; however, to include prefecture difference in the analysis, the information in Table 1 is important. Hence, we have retained Table 1 in the manuscript. We hope you understand our point of view.

・Were separate models fitted for different regions and different age groups?

Table 4 presents our analysis of the different regions and age categories.

・What were the coefficients in Table 4?

Please refer to the communication with the other reviewer and editor presented above.

・Potential covariates of interest to include are region, the proportion of 65 and above in the population, some measurements of socioeconomic levels of the prefecture, year, month of the year.

Thank you for your comment. As mentioned above, this study aimed to investigate data using big data analysis across 10 years to reconsider the Japanese hazard classification for heatstroke in Japan. We acknowledge the importance of the potential covariates recommended by you to be included in the analysis. As you pointed out, the socioeconomic levels of the participants have the potential to affect ADRHI. However, we did not have these data. Similarly, we did not think an analysis by year was suitable for the purpose of this study, as we wanted to address the need to reconsider the Japanese hazard classification for heatstroke based on big data analysis, and not every year’s data.

Again, we thank you for your comments as an expert in statistical analysis. We hope you understand the standing purpose of this study and our point of view. We agree that further analysis, including the potential covariates you pointed out, are required. Accordingly, we have already added this to the limitation section of the Discussion as follows:

(Lines 389–392)

As mentioned, several factors are assumed to be related to ADRHI, such as air pollution, economic status, and health status. However, we could not obtain heat stroke data for this experiment. Addressing this limitation is a concern for future research.

Q.6

What happened in 2018? Are there explanations for the sudden increase of ADRHI? Is this due to a particular region?

Answer

Thank you for your comment. We are grateful for your careful review of our data. As you stated, ADRHI in 2018 was higher than in other years. One of the reasons for this was that the average WBGT was higher compared to other years. However, as you know, data were collected from June to September and ADRHI occurred for many reasons. Hence, the exact reason for the sudden increase of ADRHI in 2018 is not known. However, the weather conditions were related to ADRHI significantly; thus, we have added information to the main manuscript to clarify this point.

(Lines 268–269)

To the best of our knowledge, this was the first study to examine the relationship between ADRHI and WBGT in Japan using big data analysis. As expected, the number of ADRHI showed a significant relationship with WBGTmax (p < 0.05, r2 = 0.55; Fig 3). This result is similar to the previous study, which reported that r2 was 0.48 in three big prefectures in Japan [6]. During our experimental period, the average WBGTmax was highest in July 2018, while the number of ADRHI cases were highest in 2018.

Q.7

Figure 1. maybe overlay the histogram with a line to indicate the average WBGT across the years to see if the changes correlate?

Response

Thank you for your suggestion. We have added a line denoting the average WBGT in Fig 1.

Q.8

Response

Thank you for your comment. As mentioned above, this study aimed to investigate the hazard classification in Japan, and we also wished to analyze the age differences in this context. Therefore, we believe that the proportion of the ADRHI in each age category is one of the main results in the study. It supports the importance of reconsidering the hazard classification based on the age categories. In addition, this study first demonstrated that the age proportion of ADRHI is similar across the years from the point of view of Japan as a whole. In previous studies that focused on heatstroke, the age categories are the most important component, because the ability of thermoregulation of the body is closely related to age. Hence, we think this figure is necessary in the main manuscript, rather than in the supplement.

Q.9

Figure 3 is very informative and it shows the relationship between ADRH and WBGT. Similar plots showing ADRH in older and younger populations will be helpful.

Response

Thank you for your suggestion. We were also interested in analyzing the relationships as per your suggestions. Hence, we conducted the analysis; however, the results were not significantly different between age groups. While we agree with your suggestion, we decided to exclude this information from the manuscript as it did not add much to our Discussion, while increasing the word count of our paper.

Q.10

Response

Thank you for your comment.

In the field of research regarding heatstroke or mortality related to meteorological conditions, the data about the meteorological conditions is the most important to understand the occurrence of heatstroke. Hence, it is generally included in the Results because it is correlated with the incidence of heatstroke. [1-3]

We hope you understand our point of view and hope our response addresses your concerns. Regarding your latter suggestion, we analyzed the data and found that the ADRHI coefficient value was higher in July and August; however, this was related to the number of ADRHI and the range of WBGT compared to June or September. For further clarification, please refer to our response to Q9.

1. Armstrong BG, Chalabi Z, Fenn B, Hajat S, Kovats S, Milojevic A, et al. Association of mortality with high temperatures in a temperate climate: England and Wales. J Epidemiol Community Health. 2011;65: 340-345. doi: 10.1136/jech.2009.093161. PMID: 20439353.

2. Ng CF, Ueda K, Ono M, Nitta H, Takami A. Characterizing the effect of summer temperature on heatstroke-related emergency ambulance dispatches in the Kanto area of Japan. Int J Biometeorol. 2014;58: 941-948. doi: 10.1007/s00484-013-0677-4. PMID: 23700200.

3. Basu R. High ambient temperature and mortality: a review of epidemiologic studies from 2001 to 2008. Environ Health. 2009; 8: 40. doi: 10.1186/1476-069X-8-40. PMID: 19758453.

Q.11

Figure 7. How was the comparison done? Each ADHRI level vs the previous level? Is this how “first significant increase” was defined? Please add the details in the method section.

Response

Thank you for your comment. Please refer to our response to Q1.

Q.12

Figure 8 was mentioned in the discussion. The relevant text should be in the result section instead.

Response

Thank you for your comment. We have moved the text and Fig 8 to the Results section.

(Lines 245–249)

In June, approximately 64.1% of the ADRHI that occurred in the Hokkaido-Tohoku region were related to a WBGTmax of > 25°C, which is categorized as “Caution”—the 2nd lowest level in the hazard classification, whereas only 15.0% of ARDHI occurred in the “Caution” level at Kyushu-Okinawa. However, from July to September, most ADRHI occurred at levels over “Warning”.

Q.13

In the Poisson regression model what if the heat hazard classification is used as a predictor, will there be an interaction between region and heat hazard classification?

Response

Thank you for your comment. As mentioned in the main text, hazard classification was set regardless of region. On the other hand, the Poisson regression model was used as to clarify the differences between the regions. Thus, we did not check for interactions. As mentioned above, we examined the regional differences and monthly differences using big data in order to help with the reconsideration of Japanese hazard classification. Hence, we believe that investigating the interaction would not have served our purpose. I hope you understand our point of view.

Attachment

Submitted filename: response_to_reviwer_3.docx

Click here for additional data file.^{(43.8KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0275641.r013

Decision Letter 6

Ho Ting Wong

26 May 2022

PONE-D-21-20370R6Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological studyPLOS ONE

Dear Dr. Nakamura,

Please submit your revised manuscript by Jul 10 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ho Ting Wong, PhD

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #3: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #3: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #3: (No Response)

**********

6. Review Comments to the Author

Reviewer #3: Some of my comments were address in this revision but not all. In particular I would like the following comments be addressed.

As the authors listed June 2010 to September 2019 is 10 years (or 10 122-day periods) not 10 years and 122 days.

Regarding the Poisson regression. All relevant information should show in the manuscript. I still have trouble figuring out how were the models set up. Was separate Poisson model used for each age category and each region combination? The cell for “ALL” and “All age categories” was the relative risk estimate for all data? Were any covariates adjusted for in the model?

The model with all the data included while adjusting for covariates such as age group, region, month, and potential interaction will be more informative and will help to clarify the relationship between WBGT and ARDHI. This is also consistent with the goal of this study. If there is an interaction between WBGT and month this supports the idea that the same WBGT will influence ARDHI differently for different month, hence the heat hazard classification needs to month specific. Similarly, the comment I made last time “In the Poisson regression model what if the heat hazard classification is used as a predictor, will there be an interaction between region and heat hazard classification?” will help us understand if the heat hazard classification needs to be region specific.

The authors mentioned that they “accounted for the population size in the prefect”. Please clarify how was this achieved in the model? It is conceivable that the number of ARDHI will be directly link to the population size. One way to account for the population size is to set the population size of each prefecture as an offset in the Poisson regression model.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No

PLoS One. 2022 Nov 7;17(11):e0275641. doi: 10.1371/journal.pone.0275641.r014

Author response to Decision Letter 6

15 Jun 2022

Reviewer #3:

Thank you for your review of our manuscript. We have gained valuable insights from your suggestions. As per your comments, we have performed Poisson analysis by setting the population size as the offset value; therefore, we used the absolute data for ADRHI in the analysis. Thus, the difference among regions is more clearly indicated, which is supported by another study result mentioned in the manuscript. We are deeply grateful for your expert knowledge on the Poisson regression analysis.

However, regarding question 3, we are unable to adequately understand your concern. Therefore, kindly provide more explanation regarding the insufficiency of our analysis in meeting our study objective and the suggested model.

We apologize for the inconvenience and deeply appreciate your assistance in this regard.

The responses to your comments are presented below and the changes are marked in red font in the manuscript.

Q. 1．As the authors listed June 2010 to September 2019 is 10 years (or 10 122-day periods) not 10 years and 122 days.

Response

Thank you for your comment. We have modified the text as follows:

Consequently, the total data set comprised 57,340 ADRHI and contained dispatches from 47 prefectures for 10 years and 122-day period. (L.136)

Q. 2．

Response

Thank you for your comment. We apologize for the unclear description. We have modified the Methods section to clearly explain the statistical analysis process. (L177-181)

We have used the absolute value in the Poisson model, and the average population was set as the offset value. We hope that these changes address your concern.

Moreover, to investigate the effects on ADRHI by one unit increase of WBGT in each region, Poisson regression analysis was performed using SPSS (version 27.0; SPSS Inc, Chicago, IL). The ADRHI, WBGTmax, and the average population was set as an objective value, covariates, and the offset value, respectively. The statistical significance was set at p < 0.05.

Q. 3．

Response

Thank you for your comment. Unfortunately, we are unable to clearly understand your suggestion. In Question 4, as per your suggestion, we have modified the analysis and the results seem to provide more information regarding the difference among regions for the occurrence of ARDHI. Moreover, we hope that these changes address some of your concerns raised here. Similar to your guidance in Question 4, we would appreciate it if you could suggest specific methods to help us address the issue mentioned here.

In particular, we do not understand the following sentence because the heat hazard classification was not set depending on each region, “will there be an interaction between region and heat hazard classification?” Could you explain this further and suggest methods of analysis to achieve this as you have done in Question 4: “One way to account for the population size is to set the population size of each prefecture as an offset in the Poisson regression model.”

Q. 4．

Response

Thank you for your comment, we agree with your opinion that the number of ARDHI will be directly linked to the population size. Thus, in the Poisson model, we performed the analysis by setting the average population size as the offset value. (Table 4)

In addition, according to the modified analysis, we have revised the text describing the Poisson analysis.

L47

across each region and age →　across each region

L245-L246

Moreover, the exponential of the coefficients differed in each region and in each age category (Table 4).　

→　Moreover, the exponential of the coefficients differed for each region and in each age category (Table 4).

L328

the exponential of the coefficients was calculated, and the values differed across each age category in each region (Table 4).

　→　the exponential of the coefficients was calculated, and the values differed for each age category in each region (Table 4).　

L420

Further analysis revealed that the exponential of the coefficients differed in each region and age category.　

→　Further analysis revealed that the exponential of the coefficients differed for each region and age category.

Attachment

Submitted filename: response_to_reviwer_3_Ver.2_final.docx

Click here for additional data file.^{(27.8KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0275641.r015

Decision Letter 7

Ho Ting Wong

27 Jun 2022

PONE-D-21-20370R7Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological studyPLOS ONE

Dear Dr. Nakamura,

Please submit your revised manuscript by Aug 11 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ho Ting Wong, PhD

Academic Editor

PLOS ONE

Journal Requirements:

Additional Editor Comments (if provided):

I am pleased to let you know that the statistical reviewer recommended a minor revision. As the decision of minor revision does not guarantee acceptance, it is also recommended that you should seek professional statistical support to help you revise the manuscript.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #3: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #3: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #3: (No Response)

**********

6. Review Comments to the Author

Reviewer #3: I just realized that there was a typo in my original comment 1 for the last revision. I meant to say "9 years and 122 days" or "10 years" since from 2010 to 2019 (including both 2010 and 2019) are 10 years. Sorry about that.

Response to question 2:

"ADRHI, WBGTmax, and the average population was set as an objective value, covariates,

and the offset value, respectively."

"objective value" should be "response variable". "predictor" is probably a better choice than "covariates" here since WBGTmax is the only predictor in the model, right? Does "average population" mean "population size"?

Here is a little bit more details about Question 3- A multivariate Poisson regression can be used while fitting all data in the model. In addition to WBGTmax, the predictors can include other covariates such as age group, region, month and interaction terms between WBGTmax and age group, region, and month. A multiplicative interaction is basically a product term between two variables, e.g.,WBGTmax*age_group, WBGTmax*region, WBGTmax*month (note that age group, region and month should be categorical variables here). The interaction terms can tell you if the effect of WBGTmax differs across different categories of the interacting variables, e.g., effect of WBGTmax is different for different age group or not.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No

**********

PLoS One. 2022 Nov 7;17(11):e0275641. doi: 10.1371/journal.pone.0275641.r016

Author response to Decision Letter 7

2 Jul 2022

Reviewer #3:

Thank you for your review of our manuscript and your kind advice and significant insight which has helped us improve the manuscript further. To respond to your request, we have consulted with a specialist in statistical analysis who majors in artificial intelligence and have now included him as one of the authors.

The responses to your comments are presented below and the changes are marked in red font in the manuscript.

I just realized that there was a typo in my original comment 1 for the last revision. I meant to say “9 years and 122 days” or “10 years” since from 2010 to 2019 (including both 2010 and 2019) are 10 years. Sorry about that.

Response

Thank you for your comment. We have modified the text as follows:

→

Consequently, the total data set comprised of 57,340 ADRHI and dispatches from 47 prefectures for 10 years. (L134)

“ADRHI, WBGTmax, and the average population was set as an objective value, covariates,

and the offset value, respectively.”

“objective value” should be “response variable”. “predictor” is probably a better choice than “covariates” here since WBGTmax is the only predictor in the model, right? Does “average population” mean “population size”?

Response

Thank you for your comment. We have modified the text as follows:

→

The ADRHI, WBGTmax, and population size were set as the response value, predictor, and offset value, respectively. (L178)

A multivariate Poisson regression can be used while fitting all data in the model. In addition to WBGTmax, the predictors can include other covariates such as age group, region, month and interaction terms between WBGTmax and age group, region, and month. A multiplicative interaction is basically a product term between two variables, e.g.,WBGTmax*age_group, WBGTmax*region, WBGTmax*month (note that age group, region and month should be categorical variables here). The interaction terms can tell you if the effect of WBGTmax differs across different categories of the interacting variables, e.g., effect of WBGTmax is different for different age group or not.

Response

Thank you for your detailed explanation of the analysis. According to your advice, we have obtained informative data to depict the regional differences. The value of Akaike's Information Criterion (AIC) and deviance was greatly decreased when a multivariate Poisson regression was conducted using WBGTmax and region, compared to when using WBGTmax and month. WBGTmax*region did not show good AIC and deviance. Hence, we added the comments on this in the discussion section. We hope this revision meets your request, and the right process and understanding of the data in multivariate Poisson regression has now been presented.

(L324-333)

→

Additionally, to investigate the effects on ADRHI by one unit increase of WBGT, the exponential of the coefficients was calculated, and the values differed for each region (Table 4). This result corroborates with a previous study that revealed different relative risks in Kanto area in Japan [10]. Moreover, the value of Akaike's Information Criterion (AIC) and deviance were greatly decreased when a multivariate Poisson regression was conducted using WBGTmax and region compared to that when using WBGTmax and month as predictors. Furthermore, it was indicated that the ratio of older people (> 65 years) was an important predictor of the effect of the value of AIC and deviance in multivariate Poisson regression. This result might lead to the consideration of hazardous levels, based on the age difference.

PLoS One. doi: 10.1371/journal.pone.0275641.r017

Decision Letter 8

Ho Ting Wong

19 Jul 2022

PONE-D-21-20370R8Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological studyPLOS ONE

Dear Dr. Nakamura,

Please submit your revised manuscript by Sep 02 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ho Ting Wong, PhD

Academic Editor

PLOS ONE

Journal Requirements:

Additional Editor Comments (if provided):

As the decision of minor revision does not guarantee acceptance, it is also recommended that you should seek professional statistical support to help you revise the manuscript.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #3: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #3: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #3: (No Response)

**********

6. Review Comments to the Author

Reviewer #3: It is not clear how the Poisson regression model was set up. Have region, month, proportion of age over 65 been considered as covariates? They were mentioned in the discussion but not in the method or result. Have likelihood ratio tests been performed to see if each variable is statistically significant? What are the parameter estimates? The newly added paragraph in the discussion belongs to the result and more details about how the statistical modeling was performed need to be provided.

What is shown in Table 4? Poisson regression result? Was region the only predictor in the model? Risk estimates from the final Poisson model should be shown instead of a model without adjusting for other covariates.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No

**********

PLoS One. 2022 Nov 7;17(11):e0275641. doi: 10.1371/journal.pone.0275641.r018

Author response to Decision Letter 8

1 Aug 2022

Reviewer #3:

Thank you for your thoughtful review of our manuscript. The responses to your comments are presented below and the changes are marked in red font in the manuscript.

It is not clear how the Poisson regression model was set up. Have region, month, proportion of age over 65 been considered as covariates? They were mentioned in the discussion but not in the method or result. Have likelihood ratio tests been performed to see if each variable is statistically significant? What are the parameter estimates? The newly added paragraph in the discussion belongs to the result and more details about how the statistical modeling was performed need to be provided.

Response

Thank you for your helpful comments and suggestions. We apologize that we did not understand your question correctly. We modified the text regarding the Poisson regression analysis in the method and result sections, and revised Table 4. We hope that this modification adequately addresses your concerns.

Method section (p. 176-178)

Moreover, to investigate the effects on ADRHI of a one unit increase in WBGT in each region, a Poisson regression analysis was performed using SPSS (version 27.0; SPSS Inc, Chicago, IL) to investigate the effects on ADRHI of a one unit increase in WBGT in each region. In the Poisson model, the ADRHI was set as the response value, and regions, month, WBGTmax, and proportion of people aged over 65 were set as a predictor. Population size was set as an offset value. The statistical significance was set at p < 0.05.

Results section (p. 239-241)

Moreover, the exponential values of the coefficients differed for each region and month are shown in Table 4 (LRχ2 = 697742.8, p < 0.001). The value of the exponential of the coefficients also differed among regions and months.

Table 4.

Predictor b SE Waldχ2 The exponential of the coefficients

Intercept -9.63 .03 145630.9*** < 0.01

Regions 33902.3***

Ref. value

Hokkaido-Tohoku 0 1

Kanto -0.15 .01 685.7*** .859

Chubu -0.29 0.01 2536.2*** .748

Kansai -0.19 0.01 1066.5*** .829

Chugoku -0.45 0.01 4273.4*** .638

Shikoku -1.02 0.01 14950.8*** .361

Kyushu-Okinawa -0.75 0.01 14777.1*** .472

Month 29617.7***

Ref. value

September 0 1

June 0.64 0.01 7958.0*** 1.898

July 0.86 0.01 24603.3*** 2.373

August 0.56 0.01 9878.0*** 1.748

WBGTmax 0.40 0.00 280255.3*** 1.492

Proportion of people aged over 65 -0.01 0.00 330.2*** .992

***; P < .001.

Attachment

Submitted filename: response_to_reviwer_3_Ver.5_final.docx

Click here for additional data file.^{(28.5KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0275641.r019

Decision Letter 9

Ho Ting Wong

25 Aug 2022

PONE-D-21-20370R9Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological studyPLOS ONE

Dear Dr. Nakamura,

Please submit your revised manuscript by Oct 09 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ho Ting Wong, PhD

Academic Editor

PLOS ONE

Journal Requirements:

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #3: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #3: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #3: (No Response)

**********

6. Review Comments to the Author

Reviewer #3: In a Poisson regression the exponential of the coefficient is the relative risk. Table 4- the current columns are not very helpful for a paper. Maybe only show the point estimate of relative risk and its 95% confidence interval is enough. Given the large sample size, p values do not carry as much information as the 95% confidence intervals.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No

**********

PLoS One. 2022 Nov 7;17(11):e0275641. doi: 10.1371/journal.pone.0275641.r020

Author response to Decision Letter 9

25 Aug 2022

Reviewer #3:

Thank you for your thoughtful review of our manuscript. The responses to your comments are presented below and the changes are marked in red font in the manuscript.

In a Poisson regression the exponential of the coefficient is the relative risk. Table 4- the current columns are not very helpful for a paper. Maybe only show the point estimate of relative risk and its 95% confidence interval is enough. Given the large sample size, p values do not carry as much information as the 95% confidence intervals.

Response

Thank you for your kind advice. We modified Table 4. (L252-266)

Attachment

Submitted filename: response_to_reviwer_3_Ver.6_final.docx

Click here for additional data file.^{(26.5KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0275641.r021

Decision Letter 10

Ho Ting Wong

12 Sep 2022

PONE-D-21-20370R10Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological studyPLOS ONE

Dear Dr. Nakamura,

Please submit your revised manuscript by Oct 27 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ho Ting Wong, PhD

Academic Editor

PLOS ONE

Journal Requirements:

Additional Editor Comments (if provided):

The reviewer request you make the following minor change. Once you have amended your manuscript, I will make an acceptance decision. Thank you.

Please use "relative risk" instead of "exponential of coefficient" for the interpretation of the Poisson model throughout the paper.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #3: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #3: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #3: (No Response)

**********

6. Review Comments to the Author

Reviewer #3: Please use "relative risk" instead of "exponential of coefficient" for the interpretation of the Poisson model throughout the paper.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No

**********

PLoS One. 2022 Nov 7;17(11):e0275641. doi: 10.1371/journal.pone.0275641.r022

Author response to Decision Letter 10

14 Sep 2022

Reviewer #3:

Thank you for your thoughtful review of our manuscript. The responses to your comments are presented below and the changes are marked in red font in the manuscript.

Please use "relative risk" instead of "exponential of coefficient" for the interpretation of the Poisson model throughout the paper.

Response

Thank you for your kind advice. We used "relative risk" instead of "exponential of coefficient" for the interpretation of the Poisson model throughout the paper.

Attachment

Submitted filename: response_to_reviwer_3_Ver.7_final.docx

Click here for additional data file.^{(24.6KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0275641.r023

Decision Letter 11

Ho Ting Wong

21 Sep 2022

Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological study

PONE-D-21-20370R11

Dear Dr. Nakamura,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Ho Ting Wong, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

PLoS One. doi: 10.1371/journal.pone.0275641.r024

Acceptance letter

Ho Ting Wong

28 Oct 2022

PONE-D-21-20370R11

Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological study

Dear Dr. Nakamura:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Ho Ting Wong

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Attachment

Submitted filename: Reply_to_the_Reviewer_2.docx

Click here for additional data file.^{(24.9KB, docx)}

Attachment

Submitted filename: Reply_to_the_Reviwer_1_final.docx

Click here for additional data file.^{(22.3KB, docx)}

Attachment

Submitted filename: Reply_to_the_Reviwer_1.docx

Click here for additional data file.^{(24.8KB, docx)}

Attachment

Submitted filename: Reply_to_the_Editor_5th.submission_final.docx

Click here for additional data file.^{(23.1KB, docx)}

Attachment

Submitted filename: Reply_to_the_Editor_6th.docx

Click here for additional data file.^{(24.1KB, docx)}

Attachment

Submitted filename: response_to_reviwer_3.docx

Click here for additional data file.^{(43.8KB, docx)}

Attachment

Submitted filename: response_to_reviwer_3_Ver.2_final.docx

Click here for additional data file.^{(27.8KB, docx)}

Attachment

Submitted filename: response_to_reviwer_3_Ver.5_final.docx

Click here for additional data file.^{(28.5KB, docx)}

Attachment

Submitted filename: response_to_reviwer_3_Ver.6_final.docx

Click here for additional data file.^{(26.5KB, docx)}

Attachment

Submitted filename: response_to_reviwer_3_Ver.7_final.docx

Click here for additional data file.^{(24.6KB, docx)}

Data Availability Statement

[pone.0275641.ref001] 1.Coris EE, Ramirez AM, Van Durme DJ. Heat illness in athletes: the dangerous combination of heat, humidity and exercise. Sports Med. 2004;34:9–16. doi: 10.2165/00007256-200434010-00002 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref002] 2.Fuhrmann CM, Sugg MM, Konrad CE, Waller A 2nd. Impact of extreme heat events on emergency department visits in North Carolina (2007–2011). J Community Health. 2016;41:146–56. doi: 10.1007/s10900-015-0080-7 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref003] 3.Leon LR, Bouchama A. Heat stroke. Compr Physiol. 2015;5:611–47. doi: 10.1002/cphy.c140017 [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref004] 4.Gasparrini A, Guo Y, Hashizume M, Lavigne E, Zanobetti A, Schwartz J, et al. Mortality risk attributable to high and low ambient temperature: a multicountry observational study. Lancet. 2015;386:369–75. doi: 10.1016/S0140-6736(14)62114-0 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0275641.ref005] 5.Keatinge WR, Donaldson GC, Cordioli E, Martinelli M, Kunst AE, Mackenbach JP, et al. Heat related mortality in warm and cold regions of Europe: observational study. BMJ. 2000;321:670–3. doi: 10.1136/bmj.321.7262.670 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0275641.ref006] 6.Kodera S, Nishimura T, Rashed EA, Hasegawa K, Takeuchi I, Egawa R, et al. Estimation of heat-related morbidity from weather data: a computational study in three prefectures of Japan over 2013–2018. Environ Int. 2019;130:104907. doi: 10.1016/j.envint.2019.104907 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref007] 7.Li T, Ding F, Sun Q, Zhang Y, Kinney PL. Heat stroke internet searches can be a new heatwave health warning surveillance indicator. Sci Rep. 2016;6:37294. doi: 10.1038/srep37294 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0275641.ref008] 8.Basu R. High ambient temperature and mortality: a review of epidemiologic studies from 2001 to 2008. Environ Health. 2009;8:40. doi: 10.1186/1476-069X-8-40 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0275641.ref009] 9.Weinberger KR, Zanobetti A, Schwartz J, Wellenius GA. Effectiveness of National Weather Service heat alerts in preventing mortality in 20 US cities. Environ Int. 2018;116:30–8. doi: 10.1016/j.envint.2018.03.028 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0275641.ref010] 10.Ng CF, Ueda K, Ono M, Nitta H, Takami A. Characterizing the effect of summer temperature on heatstroke-related emergency ambulance dispatches in the Kanto area of Japan. Int J Biometeorol. 2014;58:941–8. doi: 10.1007/s00484-013-0677-4 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref011] 11.Ministry of Health, Labour and Welfare. Vital Statistics in Japan [Internet]. [cited 2020 Sep 2]. Available from: https://www.mhlw.go.jp/toukei/saikin/hw/jinkou/geppo/m2018/09.html.

[pone.0275641.ref012] 12.The Fire and Disaster Management Agency. Heat stroke information [Internet]. [cited 2019 Dec 12]. Available from: https://www.fdma.go.jp/disaster/heatstroke/post3.html.

[pone.0275641.ref013] 13.Fouillet A, Rey G, Wagner V, Laaidi K, Empereur-Bissonnet P, Le Tertre A, et al. Has the impact of heat waves on mortality changed in France since the European heat wave of summer 2003? A study of the 2006 heat wave. Int J Epidemiol. 2008;37:309–17. doi: 10.1093/ije/dym253 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0275641.ref014] 14.Heudorf U, Schade M. Heat waves and mortality in Frankfurt am Main, Germany, 2003–2013: what effect do heat-health action plans and the heat warning system have? Z Gerontol Geriatr. 2014;47:475–82. doi: 10.1007/s00391-014-0673-2 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref015] 15.Nitschke M, Tucker G, Hansen A, Williams S, Zhang Y, Bi P. Evaluation of a heat warning system in Adelaide, South Australia, using case-series analysis. BMJ Open. 2016;6:e012125. doi: 10.1136/bmjopen-2016-012125 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0275641.ref016] 16.Gerrett N, Amano T, Inoue Y, Kondo N. The sweat glands’ maximum ion reabsorption rates following heat acclimation in healthy older adults. Exp Physiol. 2021;106:302–15. doi: 10.1113/EP088486 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref017] 17.Armstrong BG, Chalabi Z, Fenn B, Hajat S, Kovats S, Milojevic A, et al. Association of mortality with high temperatures in a temperate climate: England and Wales. J Epidemiol Community Health. 2011;65:340–5. doi: 10.1136/jech.2009.093161 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref018] 18.Goldberg MS, Gasparrini A, Armstrong B, Valois MF. The short-term influence of temperature on daily mortality in the temperate climate of Montreal, Canada. Environ Res. 2011;111:853–60. doi: 10.1016/j.envres.2011.05.022 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref019] 19.American College of Sports Medicine, Armstrong LE, Casa DJ, Millard-Stafford M, Moran DS, Pyne SW, et al. American College of Sports Medicine position stand. Exertional heat illness during training and competition. Med Sci Sports Exerc. 2007;39:556–72. doi: 10.1249/MSS.0b013e31802fa199 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref020] 20.Japan Society of Biometeorology. Guide to preventing heatstroke in everyday life. Version 3. [Internet]. 2013. [cited 2020 Aug 10]. Available from: https://seikishou.jp/cms/wp-content/files/news/shishin.pdf. [Google Scholar]

[pone.0275641.ref021] 21.Ministry of the Environment. Heat illness prevention information [Internet]. [cited 2020 Sep 5]. Available from: https://www.wbgt.env.go.jp/en/wbgt.php.

[pone.0275641.ref022] 22.Japan Sports Association. Guidebook for the prevention of heatstroke during sports activities. Japan Sports Association; 1994. [Google Scholar]

[pone.0275641.ref023] 23.Ministry of Internal Affairs and Communications Statistics Bureau. [Internet]. [cited 2020 Aug 8]. Available from: https://www.stat.go.jp/data/jinsui/2.html.

[pone.0275641.ref024] 24.Allden SJ, Ogger PP, Ghai P, McErlean P, Hewitt R, Toshner R, et al. The transferrin receptor CD71 delineates functionally distinct airway macrophage subsets during idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 2019;200:209–19. doi: 10.1164/rccm.201809-1775OC . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0275641.ref025] 25.Daanen HAM, Racinais S, Périard JD. Heat acclimation decay and re-induction: a systematic review and meta-analysis. Sports Med. 2018;48:409–30. doi: 10.1007/s40279-017-0808-x . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0275641.ref026] 26.Takamata A, Ito T, Yaegashi K, Takamiya H, Maegawa Y, Itoh T, et al. Effect of an exercise-heat acclimation program on body fluid regulatory responses to dehydration in older men. Am J Physiol. 1999;277:R1041–50. doi: 10.1152/ajpregu.1999.277.4.R1041 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref027] 27.Martinez GS, Imai C, Masumo K. Local heat stroke prevention plans in Japan: characteristics and elements for public health adaptation to climate change. Int J Environ Res Public Health. 2011;8:4563–81. doi: 10.3390/ijerph8124563 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0275641.ref028] 28.Wendt D, van Loon LJ, van Marken Lichtenbelt WD. Thermoregulation during exercise in the heat: strategies for maintaining health and performance. Sports Med. 2007;37:669–82. doi: 10.2165/00007256-200737080-00002 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref029] 29.Bray JE, Mosley I, Bailey M, Barger B, Bladin C. Stroke public awareness campaigns have increased ambulance dispatches for stroke in Melbourne, Australia. Stroke. 2011;42:2154–7. doi: 10.1161/STROKEAHA.110.612036 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref030] 30.Nakamura D, Muraishi K, Hasegawa H, Yasumatsu M, Takahashi H. Effect of a cooling strategy combining forearm water immersion and a low dose of ice slurry ingestion on physiological response and subsequent exercise performance in the heat. J Therm Biol. 2020;89:102530. doi: 10.1016/j.jtherbio.2020.102530 . [DOI] [PubMed] [Google Scholar]

[pone.0275641.ref031] 31.Ono M, Tonouchi M. Estimation of wet-bulb globe temperature using generally measured meteorological indices. Jpn J Biometeor. 2014;50:147–57. doi: 10.11227/seikisho.50.147 [DOI] [Google Scholar]

PERMALINK

Trends in ambulance dispatches related to heat illness from 2010 to 2019: An ecological study

Daisuke Nakamura

Hiroki Kinoshita

Kazuo Asada

Takuma Arimitsu

Mikinobu Yasumatsu

Takayuki Ishiwata

Roles

Abstract

Introduction

Methods

Study sites

Data sources

Ambulance dispatches related to heat illness

Wet bulb globe temperature

Japanese heat hazard classification

Population data and regional settings

Table 1. Average population (in thousands) data for all prefectures and seven regions.

Statistical analysis

Results

Table 2. The average total number of ADRHI per year (ADRHI / year) and average total number of ADRHI per 100,000 people / year (ADRHI_ALL / year) throughout the 10-year (2010–2019) study period, stratified by prefecture.

Fig 1. Changes in ADRHI and average WBGTmax in each year throughout the study period.

Fig 2. ADRHI ratios stratified by age and year.

Fig 3. The relationship between the number of ADRHI and WBGTmax.

Fig 4. The average WBGTmax by month.

Fig 5. The average WBGTmax by region.

Fig 6. Changes in ADRHI_ALL by month.

Fig 7. Changes in ADRHI_ALL stratified by month and WBGTmax from <21°C to 35°C.

Table 3. The first steep increase in ADRHI_ALL, ADRHI_OLD, and ADRHI_YA at WBGTmax in each region.

Table 4. The relative risks for each region, month, WBGTmax, and proportion of people aged over 65.

Fig 8. ADRHI ratio in Japanese heat hazard classification by region and month.

Discussion

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Ho Ting Wong

Roles

Author response to Decision Letter 0

Decision Letter 1

Ho Ting Wong

Roles

Author response to Decision Letter 1

Decision Letter 2

Ho Ting Wong

Roles

Author response to Decision Letter 2

Decision Letter 3

Ho Ting Wong

Roles

Author response to Decision Letter 3

Decision Letter 4

Ho Ting Wong

Roles

Author response to Decision Letter 4

Decision Letter 5

Ho Ting Wong

Roles

Author response to Decision Letter 5

Decision Letter 6

Ho Ting Wong

Roles

Author response to Decision Letter 6

Decision Letter 7

Ho Ting Wong

Roles

Author response to Decision Letter 7

Decision Letter 8

Ho Ting Wong

Roles

Author response to Decision Letter 8

Decision Letter 9

Ho Ting Wong

Roles

Author response to Decision Letter 9

Decision Letter 10

Ho Ting Wong

Roles

Fig 1. Changes in ADRHI and average WBGT_max in each year throughout the study period.

Fig 3. The relationship between the number of ADRHI and WBGT_max.

Fig 4. The average WBGT_max by month.

Fig 5. The average WBGT_max by region.

Fig 7. Changes in ADRHI_ALL stratified by month and WBGT_max from <21°C to 35°C.

Table 3. The first steep increase in ADRHI_ALL, ADRHI_OLD, and ADRHI_YA at WBGT_max in each region.