Next Article in Journal
Voice Disorder Multi-Class Classification for the Distinction of Parkinson’s Disease and Adductor Spasmodic Dysphonia
Next Article in Special Issue
Exposure to Noise and Vibration of Vocational Education Training Teachers
Previous Article in Journal
A Synergic Application of High-Oxygenated E-Fuels and New Bowl Designs for Low Soot Emissions: An Optical Analysis
Previous Article in Special Issue
Biological Monitoring via Urine Samples to Assess Healthcare Workers’ Exposure to Hazardous Drugs: A Scoping Review
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 13
Issue 15
10.3390/app13158561
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Novel Technological Advances to Protect People Who Exercise or Work in Thermally Stressful Conditions: A Transition to More Personalized Guidelines

by
Leonidas G. Ioannou
1,*,
Urša Ciuha
1,
Jason T. Fisher
1,
Lydia Tsoutsoubi
1,
Kunihito Tobita
2,
Ana Bonell
3,
James D. Cotter
4,
Glen P. Kenny
5,
Andreas D. Flouris
6 and
Igor B. Mekjavic
1
1
Department of Automation, Biocybernetics and Robotics, Jozef Stefan Institute, 1000 Ljubljana, Slovenia
2
Department of Psychology, Osaka Metropolitan University, 1-1, Gakuen-cho, Naka-ku, Sakai, Osaka 599-8531, Japan
3
Medical Research Council Unit the Gambia at London School of Hygiene and Tropical Medicine, Fajara, Banjul P.O. Box 273, The Gambia
4
School of Physical Education, Sport and Exercise Sciences, University of Otago, 55 Union Street West, North Dunedin, Dunedin 9016, New Zealand
5
Human and Environmental Physiology Research Unit, School of Human Kinetics, Faculty of Health Sciences, University of Ottawa, 125 University Ave., Room 367, Montpetit Hall, Ottawa, ON K1N 6N5, Canada
6
FAME Laboratory, Department of Physical Education and Sport Science, University of Thessaly, 42100 Trikala, Greece
*
Author to whom correspondence should be addressed.
Appl. Sci. 2023, 13(15), 8561; https://doi.org/10.3390/app13158561
Submission received: 22 May 2023 / Revised: 21 July 2023 / Accepted: 22 July 2023 / Published: 25 July 2023
(This article belongs to the Special Issue Recent Research in Occupational Exposure Assessments and Hazard Control Measures)
Browse Figures
Versions Notes

Abstract

:
Background: Prevention plays a key role in ensuring health and safety and is particularly important in scenarios when life is threatened. Adverse thermal conditions are experienced by billions of people daily, affecting the human capacity for thermoregulation and increasing the risks of life-threatening accidents, diseases, and fatalities. The aim of this study was to develop and validate a new, freely accessible method that will ultimately allow health, as well as exercise and labour organizations, to predict and potentially mitigate the physiological strain experienced by people who exercise or work in thermally stressful environmental conditions. Methods: First, we used concurrent technological advances and thermophysiological modelling to (i) develop a mobile phone application that predicts the physiological heat strain experienced by individuals conducting physical activity in adverse environmental conditions, and (ii) provide them with individualized heat mitigation strategies. Second, to examine the construct validity of the newly developed mobile phone application, core body temperature was recorded using gastrointestinal thermometry in 37 healthy soldiers during different activities. These data were used to examine the predictive capacity of our application in pre-classifying individuals with an increased risk of experiencing elevated physiological heat strain during work based on the guidelines (core body temperature ≥ 38 °C) of the World Health Organization. Results: The core body temperature predictions made by the mobile phone application were positively related (r = 0.57, p < 0.05) with the actual physiological measurements taken by our participants (mean absolute error: 0.28 °C). More importantly, our application correctly predicted 93% of occurrences of elevated physiological heat strain and 90% of those that were not (overall accuracy: 92%). Conclusions: Mobile phone applications integrating thermophysiological models can predict the physiological heat strain experienced by an individual, but it remains to be studied whether the suggested heat mitigation strategies can reduce or prevent adverse impacts.

1. Introduction

“An ounce of prevention is worth a pound of cure.” This famous phrase attributed to Benjamin Franklin underscores the importance of having a proactive approach to mitigating potential threats. The value of prevention is undeniable and plays a key role in many aspects of our daily routine, but it becomes even more important when life and well-being are directly threatened. For instance, adverse environmental conditions, such as the ones experienced by many workers on a daily basis [1], often exceed the human capacity for thermoregulation and this, in turn, increases the overall risk of life-threatening accidents, diseases, and fatalities [2,3,4,5,6,7]. Similarly, heat-related illnesses are among the most frequent causes of death in sport, with a considerable increase over the last three decades [8,9]. This is mainly because the majority of the metabolic energy required to perform manual work is released in the human body in the form of heat (metabolic heat production), whereas only a small fraction of this energy is converted to motion and thus external work [10,11]. Aside from the significant impact of ambient conditions and metabolic heat production, one may also experience thermally stressful conditions due to wearing highly insulative clothing, as well as due to other intra- and inter-individual factors [3]. In light of this, thermally stressful conditions are not solely defined by ambient conditions that may reduce heat loss to or increase heat gain from the environment, but also by other factors, underscoring human capability to maintain thermal homeostasis. To maintain thermal equilibrium, and thus deep body temperature within safe limits, heat must be transferred from the human body to the surrounding environment through four available pathways: conduction, convection, evaporation, and radiation [3].
All four heat transfer pathways can be altered by changes in human physiological [12,13,14] and behavioural [15,16] responses. From a physiological point of view, involuntary responses, such as sweating and changes in vasomotor tone, make a major contribution to dissipating the excess heat from the body [3]. Likewise, behavioural responses, which are often considered the first line of defence in the maintenance of heat balance [15,16,17,18,19], encompass a wide range of actions that act to reduce metabolic heat production or increase heat loss, including via changes in clothing, posture, orientation to the sun, and physical activity (metabolic rate). In simple terms, human thermal balance is the result of metabolism and the transfer of heat between the human body and the surrounding thermal environment. From a modelling perspective, a good estimate of the physiological heat strain experienced by an individual can be made based on the physical work performed, as well as the thermal (e.g., environment) and nonthermal (e.g., anthropometric) factors that modulate the heat transfer from the human body to the surrounding environment.
During the past decade, significant advancements have been made in mobile phone technology that have revolutionized the way we interact with the world around us. Advances in sensor technology [20] and mobile computing [21] have enabled new capabilities for mobile phones, which are now equipped with a range of sensors and connectivity to meteorological data that can be used for prevention and early heat warning alerts. Intelligent mobile phone applications can be valuable tools in preventing heat-related illnesses by providing weather updates, reminders to stay hydrated, and location-based alerts [22,23,24]. When equipped with technology that is not only widespread but also intelligent and can consider changes in the needs of the user, the quality of preventive care can be greatly improved, supporting the notion that investing in prevention is much cheaper than the cost of treatment per se [25]. This is especially evident in light of the increasing reliance on mobile phones, with the aid of intelligent applications, as a vital platform for delivering health interventions [26]. In fact, half of smartphone owners employ their devices to obtain health information, while one-fifth of smartphone owners have health applications installed on their phones [27]. This increasing prevalence of smartphone usage to access health information presents a unique opportunity to introduce more personalized approaches to mitigate the deleterious impacts of physiological heat strain on human health and performance. Indeed, the importance of shifting the focus from out-of-date one-size-fits-all guidelines to more individualized heat-health guiding principles has been emphasised before [3,28,29]. Therefore, the aim of the present study was to develop an intelligent mobile phone application and present new methods that will ultimately allow health, exercise, and labour organizations to provide personalized heat-health guidance to people who perform in thermally stressful conditions.

2. Materials and Methods

2.1. Procedures

This study involved two connected stages. Firstly, the development of a mobile phone application that predicts the physiological heat strain experienced by an individual (Section 2.2) and provides them with feedforward individualized heat-mitigation guidance (Section 2.2.1, Section 2.2.2., Section 2.2.3 and Section 2.2.4). Secondly, a field evaluation to examine the effectiveness of the newly developed mobile phone application to correctly predict individuals who may experience high physiological heat strain or not (known as “construct validity”) and provide them with personalized heat-mitigation guidance. The two stages are detailed below.

2.2. Mobile Phone Application

A new mobile phone application named “Heat-Health” was developed in the B4X programming language. The Heat-Health application can be freely downloaded from the following link: https://heat-health.org/application.html (accessed on 22 July 2023). This mobile phone application uses geolocation and Application Programming Interface (API) services to locate the nearest weather station to the end-user in order to obtain current and forecast weather data from an online meteorological service (OpenWeatherMap: https://api.openweathermap.org, accessed on 22 July 2023). These weather data include air temperature, relative humidity, wind speed, wind direction, cloud coverage, and weather description on an hourly basis. Solar radiation is estimated [30] using time and geolocation data through the global positioning system (GPS) sensor integrated into the mobile phone of the end-user while correcting for cloud coverage using published literature [31]. The obtained hourly values are converted to minute-by-minute values using the gradient by computing the rate of change between the hours and using it to calculate the values for each minute. The Wet-Bulb Globe Temperature (WBGT) was adopted in the newly developed mobile phone application because it is considered the best meteorology-based indicator for quantifying the physiological heat strain experienced by workers [5,32,33,34], and it is calculated using previously published methodology [35]. For the calculation of the WBGT thermal stress indicator, the Wet-Bulb Temperature and Globe Temperature are estimated using the approach described by Liljegren et al. [36] because it is considered the most accurate method for calculating WBGT from meteorological data [37]. Thereafter, the obtained (i.e., air temperature, relative humidity, and wind speed) and calculated (i.e., Globe Temperature) environmental data, along with individualized input variables that describe the anthropometric (body mass and height) and general (i.e., duration, intensity, and worn clothing) characteristics of the work/exercise performed are used to estimate the anticipated physiological heat strain of an individual (Figure 1). In addition to the features described above, the Heat-Health application provides the end-user with the ability to simulate work/exercise scenarios that allow for behavioural changes in clothing and work/exercise intensity using the principles described in the relevant subsections below. Specifically, the obtained environmental and input data are used to estimate the response of an individual’s body core temperature (Tcore) and body water loss using a modified version of the Predicted Heat Strain (PHS) model [10,38] developed by the FAME Laboratory, Greece [39]. This modified model accounts for sequential time periods, as well as the mechanical efficiency [40] that characterizes the work/exercise performed. It was recently used to estimate the physiological heat strain reported in athletic [41], occupational [39,42,43], and modelling [44] studies. Extensive beta-testing of the Heat-Health application was performed for quality assurance by experts in thermal and occupational physiology based in countries situated in Africa, Asia, America, Europe, and Oceania. While the present study was neither designed nor conducted with the purpose of examining the validity of the PHS model, which has been extensively assessed in previous studies reported in the literature, we found it necessary to investigate the sensitivity of the model incorporated in the Heat-Health application in line with international practices. Therefore, the sensitivity of the modified version of the PHS model, incorporated into the Heat-Health application, was tested against the normative data presented in the relevant international standard (Annex F: ISO 7933:2004) [10]. This testing demonstrated that the predicted values for the tested conditions (37.5 to 41.2 °C Tcore) fall within a normal physiological range for various metabolic, clothing, and environmental scenarios. However, it is important to note that this modified version of the PHS model accounts for mechanical efficiency rather than being considered negligible, as suggested in the original PHS model. Consequently, there is an average reduction of ~0.6 °C in the Tcore predictions made by the modified version of the model compared to the predictions made by the original PHS model. Also, the Heat-Health application supports many languages, and more are planned to be available in the future. All predictions can be exported and shared with anyone in numeric format (*.csv) and as animated illustrations (*.gif).

2.2.1. Estimation of Optimal Clothing Ensemble

An iterative model-solving approach was employed to estimate the optimal clothing ensemble for an individual conducting an activity with a predetermined metabolic demand under certain environmental conditions. Specifically, a condition-controlled loop is executed either while a termination condition (clo = 0.5) is satisfied or while the physiological heat strain condition is not satisfied (Tcore > 37.5 °C). In other words, this loop keeps solving the modified version of the Predicted Heat Strain model [39] with the same environmental and input data while using different clothing insulation values (starting from 4.0 clo to 0 clo, with an iterative step of −0.1 clo) either until it attains a Tcore ≤ 37.5 °C (i.e., normothermic state) or until the clothing insulation is equal to 0.5 clo (i.e., very light clothing). Thereafter, the mobile phone application identifies the optimal apparel to wear from a predetermined list of available clothing ensembles. The recommended clothing ensemble will have an insulation value nearest to the one required. The list of available clothing ensembles can be modified as needed by the end-user. In very windy ambient conditions (wind speed > 12 m/s), the mobile phone application suggests the end-user wear a clothing ensemble that includes a windproof jacket, only if Tcore is projected to remain at safe normothermic levels (≤37.5 °C). This is because a 3-layer clothing ensemble, such as the ones often used in occupational settings, with low air permeability loses about two-thirds of its thermal insulation in winds > 12 m/s [45]. Additionally, if the weather forecast is for rain, a clothing ensemble that includes a waterproof jacket is recommended. It is important to note that in the present study, individual clothing items that are designed for ballistic protection of the head (e.g., helmet) and torso (e.g., bulletproof jacket) are not included in the analysis, as they are worn during missions in hostile enemy territory and only on the orders of the commanding officer. A similar approach should be followed with any ensemble involving protective equipment worn by soldiers, firefighters, rescuers, police officers, and any other professions that may require the use of such apparel.

2.2.2. Estimation of Safe Work/Exercise Intensity

To estimate the optimal work–rest cycles required to keep an individual’s Tcore within safe limits, an iterative model-solving approach was employed. Specifically, a condition-controlled loop is executed either while a termination condition (metabolic rate < 60 W/m2) is satisfied or while the physiological heat strain condition is not satisfied (Tcore > 37.5 °C). In other words, this loop keeps solving the modified version of the Predicted Heat Strain model [39] with the same environmental and input data while using different metabolic rate values (starting from 1000 W/m2 to 50 W/m2, with an iterative step of 10 W/m2) until either a Tcore ≤ 37.5 °C (i.e., normothermic state) is attained or until the required metabolic rate for safe work intensity is lower than 60 W/m2 (i.e., resting state).

2.2.3. Estimation of Water Intake

The suggested water intake is estimated using the prediction of total body water loss with a modified version of the Predicted Heat Strain model that accounts for sequential time periods and work efficiency [39]. Precise information regarding all the calculations involved in the estimation of total body water loss can be found in the relevant international standard [10].

2.2.4. Estimation of Work–Rest Cycles

To estimate the optimal work–rest cycles required to keep an individual’s Tcore within safe limits, an iterative model-solving approach was employed. A condition-controlled loop is executed either while a termination condition (break = 60 min/h) is satisfied or while the physiological heat strain limit (Tcore > 38.0 °C) of the WHO [46] is not satisfied. In other words, this loop keeps solving a modified version of the Predicted Heat Strain model [39] with the same environmental and input data while using different work–rest cycles (starting with a resting time of 0 min/h to 60 min/h, with an iterative step of 1 min/h) over and over until Tcore < 38.0 °C (i.e., safe Tcore limit) is attained or until the rest time is equal to 60 min/h (i.e., no work at all). The resting metabolic rate is assumed to be equal to one metabolic equivalent (MET), while the metabolic rate that characterizes the work performed is an input value entered by the end-user.

2.3. Human Trials

The experimental protocol was approved by the National Committee for Medical Ethics of the Republic of Slovenia in accordance with the Declaration of Helsinki (92/12/04). The minimum required sample size for investigating “Correlation: Bivariate normal model” was calculated using the results of a previous study, which identified a moderate relationship (r = 0.573) between the measured Tcore and predicted Tcore using thermal modelling [39]. Assuming an α of 0.05 and β of 0.90, a minimum of 27 participants were needed to provide sufficient power to detect a correlation of a similar magnitude (G*Power Version 3.1.9.2) [47]. Therefore, the present study involved monitoring 37 healthy participants (Table 1). Of these, 21 participants were measured during a hiking session and 7 during guard duty, while 10 of them were measured both during a hiking session and guard duty. Overall, physiological and anthropometric data were collected during 48 trials (hiking: 31; guard duty: 17). Written informed consent was obtained from all volunteers after a detailed explanation of all the procedures involved.
Baseline anthropometric (height and body mass) and self-reported demographic (age and gender) data were collected from our participants. Body surface area was calculated from anthropometric data [48]. During the experimental trials, continuous Tcore data were collected at a 30 s sampling rate using ingestible telemetric capsules (BodyCap, Caen, France) taken 90 min prior to testing. Recent findings indicate that the time following ingestion does not influence the validity of telemetry pill measurements of Tcore in the absence of ingested food or fluids [49]. Moreover, the validity and reliability of the ingestible telemetric capsules utilized in the present study were previously confirmed for an older version of the capsule [50]. This was reconfirmed for the latest version, revealing an average systematic bias lower than −0.04 °C [51]. This bias is considered negligible for observing potential physiological impacts. The actual metabolic rate (i.e., oxygen consumption) during each trial was measured using a portable gas analyzer (k4b2, Cosmed, Rome, Italy).
The experimental trials involved monitoring soldiers during 3 h guard duty sessions and 1.5 to 3 h mountain hiking activities. The trials took place at three rural locations in Slovenia: (i) Ankaran, latitude: 45.57953; longitude: 13.73436; elevation: ~50 m; (ii) Pokljuka, latitude: 46.35617; longitude: 14.02467; elevation: ~1250 m; and (iii) Planica, latitude: 46.47477; longitude: 13.72285; elevation: ~750 m. All guard duty sessions were characterized by 2.0 METs based on the #21040 code (i.e., “walking, less than 2.0 mph, very slow”) of the compendium of physical activities [52,53], and thus were considered easy work based on the guidelines of the American Conference of Governmental Industrial Hygienists [54]. Hiking activities ranged from moderate (4.0 METs) to hard (≥6.0 METs) work [54] based on codes #17133 (i.e., “walking: stair climbing, slow pace”; 4.0 METs), #17080 (i.e., “walking: hiking, cross country”; 6.0 METs), and #17010 (i.e., “walking: backpacking”; 7.0 METs) of the compendium of physical activities [52,53].
Although we appreciate the importance of clothing [12,15,55,56] and hydration [3,15,57,58,59] on the physiological heat strain experienced by an individual, we avoided interfering with the usual clothing and water consumption routine of our participants in an attempt to examine the capacity of the newly developed mobile phone application to correctly predict the physiological heat strain experienced by people in real-life scenarios. Nevertheless, the clothing worn by our participants was recorded and used as an input value in the mobile phone application. Specifically, eighteen different clothing ensembles were provided by the Ministry of Defence of the Republic of Slovenia, and their clothing insulation values were measured on the Jozef Stefan Institute’s thermal manikin following the methods described in ISO 15831:2004 [60]. The tested clothing ensembles were similar to those worn by soldiers in other North Atlantic Treaty Organization (NATO) countries [61]. Basic clothing insulation or “intrinsic clothing insulation” was determined as the insulation from the skin surface to the clothing surface and was calculated by subtracting the insulation of the naked manikin (0.40 clo) from the average clothing insulation measured for each ensemble [62]. All the relevant information regarding the individual clothing items that together characterize each clothing ensemble can be found in Table A1. The intrinsic insulation eliminates the contribution of the air layer surrounding the manikin to the overall insulation value. The calculated intrinsic clothing insulation values for each clothing ensemble can be found in Table A1.

2.4. Data Analysis

The raw Tcore data collected were used to calculate the numeric average and find the maximum Tcore values that were attained in each trial. The average values for each trial were calculated for the estimated WBGT values. To clearly present the physiological and metabolic demands that characterized the tested work activities, a descriptive analysis was performed on all variables (metabolic rate, clothing insulation, trial duration, weather data, Wet-Bulb Globe Temperature, as well as average and maximum Tcore values). Additionally, Pearson’s correlation coefficient was used to examine the potential associations among the variables measured during each trial. Paired t-test and Cohen’s d-effect size were used to examine potential statistically significant differences and the associated magnitude of the difference between the metabolic rates selected in the mobile phone application and the actual ones measured during each trial.
To examine the validity of the newly developed mobile phone application, a series of correlation measures and error indices were utilized. Specifically, Pearson’s correlation coefficient and linear regression models were used to examine the relationship between the measured Tcore values and those predicted by the mobile phone application. The root-mean-square error, mean bias error, and mean absolute error were used to examine the magnitude of the differences between the measured Tcore values and those predicted by the mobile phone application. Additionally, Willmott’s refined index of agreement [63] was computed to determine the degree of agreement between the measured and predicted Tcore values (averages and maximum of each trial). In this index of agreement, a value of “0” indicates no agreement at all and “1” corresponds to perfect agreement [64]. The Bland–Altman 95% limits of agreement and associated per cent coefficient of variation were used to further assess and visualize the differences between the measured Tcore values and those predicted by the mobile phone application. Finally, sensitivity, specificity, and accuracy analyses were conducted to examine the predictive capacity of the newly developed mobile phone application in pre-classifying individuals with an increased risk of experiencing elevated physiological heat strain, which was determined as a Tcore higher than the World Health Organization (WHO) [46] guidelines (≥38 °C).
Statistical analyses were conducted using both SPSS v28.0 (IBM, Armonk, NY, USA) and Excel software (Microsoft Office, Microsoft, Washington, DC, USA). The level of significance for these analyses was set at p < 0.05.

3. Results

3.1. Weather Conditions and Physiological Strain

The environmental conditions throughout the experiments fluctuated considerably, including the air temperature (3.3 ± 7.3 °C; −6.8 to 23.2 °C), relative humidity (58.8 ± 21.1%; 24.0 to 100.0%), absolute humidity (3.9 ± 3.0 g/m3; 1.1 to 15.1 g/m3), wind speed (3.6 ± 7.3; 0.0 to 23.5 m/s), and solar radiation (92.6 ± 48.8 W/m2; 8.8 to 215.3 W/m2). Similarly, the ambient thermal stress experienced by the participants in the present study varied significantly (1.7 ± 8.3 °C WBGT; −11.5 to 26.9 °C WBGT), with 90% of trials taking place in cold-to-very-cold conditions (<10 °C WBGT) and 10% in neutral-to-warm conditions (>20 °C WBGT). No experiments were conducted in hot or very hot climatic conditions. The average ambient thermal stress experienced by the participants in the present study was similar during guard duty (1.1 ± 4.7 °C WBGT; −7.6 to 6.7 °C WBGT) and mountain hiking (2.1 ± 9.8 °C WBGT; −11.5 to 26.9 °C WBGT), yet only hiking sessions were measured in warm environmental conditions. Ambient conditions had a significant impact on the behavioural responses. Specifically, the air temperature (r = −0.30) and WBGT (r = −0.30) were inversely related to the clothing insulation worn by the participants in the present study (all p < 0.05), indicating that they wore less insulative clothing when the ambient conditions were warm. No other statistically significant relationships were identified between the ambient conditions recorded and the physiological responses of the monitored participants (p > 0.05).
The actual metabolic rate of our participants during guard duty (106 ± 23 W/m2; 69 to 172 W/m2) was approximately three times lower compared to mountain hiking (341 ± 93 W/m2; 171 to 602 W/m2). There were no statistically significant differences (p > 0.05) and only negligible effect sizes (d = 0.06) between the predetermined options for the metabolic rates selected in the mobile phone application (265 ± 121 W/m2) and the actual ones measured using the portable gas analyser (258 ± 136 W/m2).
The clothing insulation of the participants in the present study (3.00 ± 0.86 clo; 1.50 to 3.73 clo) varied considerably, as they had the option to freely choose what to wear among the 18 different clothing ensembles presented in Appendix A (Table A1). Clothing insulation was negatively correlated (all p < 0.05) with the physiological heat strain experienced during the monitored trials (Tcore: r = −0.61; max Tcore: r = −0.75), as expected. Also, a positive relationship was identified between clothing insulation and the duration of the trial/session, indicating that participants chose to wear less insulative clothing when they knew that they had to perform for a longer period. Moreover, a very strong inverse association was identified between clothing insulation and the metabolic rate that characterized the work tasks performed by the participants in the present study (r = −0.86; p < 0.05), indicating that when they performed tasks of higher intensity, they chose to wear less insulative apparel.

3.2. Predictive Capacity of the Heat-Health Application

The predictive capacity of the Heat-Health application in predicting physiological heat strain and pre-classifying individuals with a high risk of experiencing elevated physiological heat strain was confirmed. Specifically, strong associations were evident between the Tcore predictions made by the Heat-Health application (average: 37.8 ± 0.3 °C; 37.3 to 38.1 °C; maximum: 37.9 ± 0.3 °C; 37.4 to 38.3 °C) and the actual Tcore measurements obtained from the participants (average: 37.7 ± 0.4 °C; 36.7 to 38.6 °C; maximum: 38.2 ± 0.6 °C; 37.0 to 39.4 °C) in the present study (Figure 2). It is important to note that while the present study did not involve experiments in very hot ambient conditions, nearly two-thirds (65%) of our participants experienced elevated physiological heat strain (>38 °C Tcore). This suggests that, as expected, the combined influences of the ambient, metabolic, and clothing conditions tested were indeed thermally stressful, effectively allowing us to address the primary aim of this study. Moreover, the correlation measures and error indices further confirmed that the Heat-Health application was able to predict the physiological heat strain experienced by the monitored participants (Figure 3). The calculated mean bias errors between the Tcore predictions made by the Heat-Health application and the actual Tcore measurements obtained from the participants showed that the application tended to slightly overestimate the average physiological heat strain experienced by an individual (+0.1 °C Tcore), while it slightly underestimated the maximum physiological heat strain that the same individual might experience (−0.2 °C Tcore). However, it is important to note that the observed mean errors between the predicted and measured Tcore values were influenced by the input clothing insulation in the Heat-Health application. This indicated that although the monitored participants had started their work/exercise trials wearing highly insulative clothing, many had changed the insulative capacity of their ensembles by unbuttoning their shirts, rolling up their sleeves, and/or taking off their jackets, but these adjustments were not updated in the application. Also, an increase in the moisture content of the clothing layers as a consequence of absorbing sweat may have further reduced the insulative value of primarily the next-to-skin clothing layer. Nevertheless, the observed heteroscedasticity did not cause significant bias in the application’s predictive, capacity as indicated by the computed error indices (Figure 3). Importantly, the Heat-Health application correctly predicted more than nine out of every ten participants who experienced elevated physiological heat strain (Figure 4). Similarly, the Heat-Health application was able to correctly predict more than nine out of every ten participants who did not experience elevated physiological heat strain (Figure 4). Overall, the application demonstrated very good accuracy (92%) in pre-classifying individuals with a high risk of experiencing elevated physiological heat strain (Figure 4).

4. Discussion

In this study, we used concurrent technological advances and thermophysiological modelling to develop a mobile phone application that predicts the physiological heat strain experienced by an individual using simple information that can be easily understood by non-experts in thermal physiology and modelling. The validity of the Heat-Health application was confirmed by experts who are based in countries situated in Africa, Asia, America, Europe, and Oceania. Similarly, the predictive capacity of the Heat-Health application in pre-classifying individuals with an increased risk of experiencing elevated physiological heat strain was extensively tested and confirmed in the present study.
When faced with a decision that has only two possible outcomes, the chance of randomly predicting the correct option is 50%. For instance, flipping a coin has an equal chance of landing heads or tails, but flipping it 100 times does not necessarily mean that the toss will result in 50 heads and 50 tails. This is because the probability of randomly predicting multiple consecutive outcomes correctly decreases with each subsequent prediction, as the odds of accurately predicting a series of independent events decreases exponentially. In the present study, we found that the Heat-Health application was able to correctly predict 44 out of the 48 trials examined, but this does not indicate, by any means, that the present approach will be equally accurate with a different population sample. There is always a probability of 1 in every 6000 coin flips for an edge landing [65], and there is an even higher probability of someone’s thermophysical responses not being well explained with the current thermophysiological modelling knowledge. This is especially true, considering that 2 out of the 28 participants who experienced high physiological heat strain in the present study were not predicted correctly. Applying our findings to the coin flip analogy, the sole use of the Heat-Health application without additional precautions could lead to a misclassification of 429 out of 6000 potential workers. Nevertheless, when comparing the present approach to those currently used in the few parts of the world where there is relevant heat-health legislation [3], this is undoubtedly a more sophisticated and accurate seemingly approach to adopt for helping those who exercise or work in adverse environmental conditions. If we were to follow the WBGT Threshold Limit Values (TLVs), as suggested by the American Conference of Governmental Industrial Hygienists [54], there would be no heat exposure recommendations for the acclimated participants of the present study, even after accounting for clothing adjustment factors, since all our experiments took place in environments much cooler than the 27.5 °C WBGT, where the first TLV limit is set.
Even though we did not conduct experiments in hot or very hot environmental conditions, it is reasonable to assume that the effectiveness of the Heat-Health application in pre-classifying individuals with an increased risk of experiencing elevated physiological heat strain will be similarly accurate or even enhanced in warmer conditions. This is because the PHS model, which is the core of the newly developed mobile phone application, was extensively tested and validated across a wide range of environmental conditions (air temperature: 15 to 50 °C; humidity: 0 to 4.5 kPa; radiation: 0 to 60 (radiant temperature − air temperature) °C; and air velocity: 0 to 3 m/s) in laboratory and field settings [66], as well as repeatedly assessed in agriculture [39], mining [42], and petrochemical [43] industries. Nevertheless, it is important to note that the accuracy of the prediction is not only a matter of the capacity of the thermophysiological model to correctly estimate physiological heat strain but also relies on the accuracy (including proximity) of the weather data provided by the online meteorological service. In light of this, a recent assessment [67] was carried out to examine the accuracy of OpenWeatherMap weather data for two cities in Brazil, revealing that the average differences between the on-site and online weather data were 1 °C for air temperature and roughly 4% for relative humidity. While these differences in ambient conditions are relatively small, the implementation of artificial intelligence technology holds great potential to further improve the quality of weather forecasts in the near future [68].
The accuracy of predictions made by the Heat-Health application is also dependent on the precision of user input data. Accurate and precise user input data, such as metabolic rate and clothing insulation, can aid in making well-informed decisions and reducing the likelihood of errors. Conversely, inaccurate or imprecise user input data can result in modelling errors and a reduction in prediction confidence, impairing the capacity of a mobile phone application in pre-classifying individuals with an increased risk of experiencing elevated physiological heat strain. The predictions made by the Heat-Health application in the present study were founded upon the precise measurements of clothing insulation and metabolic rate that were obtained using cutting-edge equipment. However, it is important to acknowledge that the end-user is unlikely to have access to such precise information, which may hinder the ability of the Heat-Health application to make accurate predictions. Another important limitation of the Heat-Health application that should be acknowledged is the inability of the core algorithm to consider several intra- and inter-individual factors that are known to significantly impact the physiological heat strain experienced by an individual [3,69,70]. For instance, several factors that are currently not considered, such as pregnancy [71], menstrual cycle [72], age [73,74,75,76,77,78,79,80], sex [80,81], medication [70], medical conditions [82,83,84], and consecutive shifts [14,77,85,86], can potentially impact the predictive capacity of the Heat-Health application. Unfortunately, it was not possible to test the separate and/or combined contributions of these factors on the accuracy of the Tcore predictions made by the Heat-Health application, as our measurements were performed on young, healthy individuals across a limited number of trials. In addition to the above limitation, the Heat-Health application can predict the physiological heat strain experienced by people who exercise or work only outdoors. This is because it obtains weather information from an online service that employs weather stations designed to measure and record only outdoor atmospheric conditions. Indoor temperature readings are complicated to estimate accurately due to various factors such as building size, insulation, and heating/cooling systems. Moreover, the present version of the Heat-Health application was designed for research purposes and thus does not currently provide real-time risk updates. We recognize the importance of this feature for heat-health assessment, particularly in dynamic situations where a swift risk evaluation is crucial. Therefore, we fully intend to incorporate real-time updates into future revisions of the application after extensive validation processes with diverse populations and across different environmental conditions.
Despite the many limitations of thermophysiological modelling, mobile phone technology with the aid of intelligent applications has the capacity to transform heat-health guidance by enhancing its accessibility, affordability, and convenience for people. Individuals who often experience increased physiological heat strain can take charge of their well-being, and healthcare providers can remotely prevent severe health problems, as more intelligent code becomes readily available. Although there are still hurdles to overcome, such as sensor accuracy and modelling limitations, the prospect of mobile phone applications in heat-health guidance seems encouraging. In fact, the Ministry of Defense of the Republic of Slovenia has adopted a modified version of the Heat-Health application that specifically caters to the armed forces, aiming to provide tailored heat-health guidance and protect the health and welfare of those who perform in unsuitable conditions. This exemplifies the significant positive impact that mobile phone applications can have in the field of heat-health guidance.
While the potential of mobile applications that integrate thermophysiological models to predict physiological heat strain is evident, there is still a crucial research gap that needs to be addressed. Specifically, it currently remains unknown whether the heat mitigation strategies suggested by the Heat-Health application can effectively reduce or altogether prevent the adverse impacts associated with heat strain. Moreover, in light of these promising developments, future studies should focus on expanding the capabilities of mobile phone applications in heat-health guidance. This could include incorporating real-time data from a wider range of environmental sensors, such as temperature and humidity measurements, to provide more comprehensive and accurate guidance. Additionally, the Heat-Health application could significantly benefit from incorporating feedback from actual end-users. Such feedback can provide invaluable insights into the application’s usability and effectiveness in real-world conditions, improving the overall user-friendliness and efficiency of the application. Moreover, researchers could investigate the integration of machine learning algorithms to evaluate the suitability of the 38 °C Tcore criterion for determining elevated strain, as well as to improve the personalization of heat-health recommendations. By considering individual factors, such as age, fitness level, and pre-existing health conditions, these algorithms could offer more tailored and effective guidance for maintaining optimal health during heat exposure. Further studies could also investigate the effectiveness of the Heat-Health application in diverse populations and settings, such as among the elderly, children, or those in low-resource environments, to ensure that the benefits of these technologies are accessible and equitable for all. Finally, collaboration between public health organizations and researchers is crucial for establishing standardized protocols and best practices for the development, evaluation, and implementation of these mobile applications to ensure their safety, reliability, and effectiveness in promoting heat-health guidance.

5. Conclusions

The Heat-Health application is a sophisticated and accurate approach for helping those who often exercise or work in thermally unsuitable conditions, especially when compared to what is currently employed in the few parts of the world that have enacted relevant heat-health legislation. That is, mobile phone applications integrating thermophysiological models may be used in the future as a means to provide individualized heat-health guidance, safeguarding the health and well-being of those who often experience increased physiological heat strain, including athletes, workers, and the general public.

Author Contributions

Conceptualization, L.G.I.; methodology, L.G.I.; software, L.G.I.; validation, L.G.I., U.C., L.T., K.T., A.B., J.D.C., G.P.K. and A.D.F.; formal analysis, L.G.I.; investigation, L.G.I., U.C., J.T.F. and I.B.M.; resources, I.B.M.; data curation, L.G.I.; writing—original draft preparation, L.G.I.; writing—review and editing, all authors; visualization, L.G.I.; supervision, I.B.M.; project administration, I.B.M.; funding acquisition, I.B.M. All authors have read and agreed to the published version of the manuscript.

Funding

The work was supported, in part, by the Ministry of Defence of the Republic of Slovenia (project ReMOS: 4330-433/2019-3).

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the National Committee for Medical Ethics of the Republic of Slovenia (92/12/04).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Data are available upon request from the corresponding author.

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A

Table
Table A1. Clothing items and clothing insulation (clo) values of the various ensembles tested, sorted based on their thermal insulative capacity.
Table A1. Clothing items and clothing insulation (clo) values of the various ensembles tested, sorted based on their thermal insulative capacity.
cloPictureClothing DescriptioncloPictureClothing Description
0.40Applsci 13 08561 i001
  • Naked manikin
1.11Applsci 13 08561 i002
  • Cotton underwear
  • Polo shirt
  • Summer pants
  • Baseball hat
1.41Applsci 13 08561 i003
  • Functional underwear
  • Summer socks
  • Shirt
  • Summer pants
  • Baseball hat
1.50Applsci 13 08561 i004
  • Cotton underwear
  • Summer socks
  • Shirt
  • Summer pants
  • Baseball hat
1.65Applsci 13 08561 i005
  • Thermal underwear
  • Tactical shirt
  • Tactical pants
  • Flame-retardant under cap
  • Helmet
  • Body armour
1.74Applsci 13 08561 i006
  • Cotton underwear
  • Summer socks
  • Tactical shirt
  • Tactical pants
  • Helmet
  • Body armour
1.88Applsci 13 08561 i007
  • Cotton underwear
  • Turtleneck
  • Winter socks
  • Thermovelour (fleece) jacket
  • Winter pants
  • Baseball hat
1.95Applsci 13 08561 i008
  • Thermal underwear
  • Winter socks
  • Shirt
  • Thermovelour (fleece) jacket
  • Winter pants
  • Baseball hat
2.11Applsci 13 08561 i009
  • Cotton underwear
  • Summer socks
  • Summer pants
  • Raincoat jacket
  • Water-resistant pants
2.16Applsci 13 08561 i010
  • Cotton underwear
  • Summer socks
  • Shirt
  • Wind jacket
  • Summer pants
  • Baseball hat
2.19Applsci 13 08561 i011
  • Cotton underwear
  • Turtleneck
  • Winter socks
  • Shirt
  • Thermovelour (fleece) jacket
  • Winter pants
  • Baseball hat
2.19Applsci 13 08561 i012
  • Cotton underwear
  • Summer socks
  • Shirt
  • Thermovelour (fleece) jacket
  • Summer pants
  • Baseball hat
  • Summer boots
2.23Applsci 13 08561 i013
  • Thermal underwear
  • Turtleneck
  • Thermovelour (fleece) jacket
  • Tactical pants
  • Flame-retardant under cap
  • Helmet
  • Body armour
2.28Applsci 13 08561 i014
  • Thermal underwear
  • Turtleneck
  • Winter socks
  • Shirt
  • Thermovelour (fleece) jacket
  • Winter pants
  • Baseball hat
2.34Applsci 13 08561 i015
  • Cotton underwear
  • Summer socks
  • Tactical shirt
  • Tactical pants
  • Raincoat jacket
  • Water-resistant pants
  • Helmet
  • Body armour
2.76Applsci 13 08561 i016
  • Thermal underwear
  • Turtleneck
  • Wind jacket
  • Tactical pants
  • Flame-retardant under cap
  • Helmet
  • Body armour
3.04Applsci 13 08561 i017
  • Thermal underwear
  • Turtleneck
  • Wind jacket
  • Tactical pants
  • Raincoat jacket
  • Water-resistant pants
  • Flame-retardant under cap
  • Helmet
  • Body armour
3.42Applsci 13 08561 i018
  • Thermal underwear
  • Turtleneck
  • Undergarment
  • Winter socks
  • Thermovelour (fleece) jacket
  • Winter pants
  • Knitted cap
3.73Applsci 13 08561 i019
  • Thermal underwear
  • Turtleneck
  • Undergarment
  • Winter socks
  • Thermovelour (fleece) jacket
  • Wind jacket
  • Winter pants
  • Knitted cap

References

  1. Kjellstrom, T.; Maître, N.; Saget, C.; Otto, M.; Karimova, T. Working on a Warmer Planet: The Impact of Heat Stress on Labour Productivity and Decent Work; Publications Production Unit, International Labour Organization: Geneva, Switzerland, 2019; ISBN 978-92-2-132968-8. [Google Scholar]
  2. Anderson, B.G.; Bell, M.L. Weather-related mortality: How heat, cold, and heat waves affect mortality in the United States. Epidemiology 2009, 20, 205–213. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  3. Ioannou, L.G.; Foster, J.; Morris, N.B.; Piil, J.F.; Havenith, G.; Mekjavic, I.B.; Kenny, G.P.; Nybo, L.; Flouris, A.D. Occupational heat strain in outdoor workers: A comprehensive review and meta-analysis. Temperature 2022, 9, 67–102. [Google Scholar] [CrossRef] [PubMed]
  4. García-Trabanino, R.; Jarquín, E.; Wesseling, C.; Johnson, R.J.; González-Quiroz, M.; Weiss, I.; Glaser, J.; José Vindell, J.; Stockfelt, L.; Roncal, C.; et al. Heat stress, dehydration, and kidney function in sugarcane cutters in El Salvador—A cross-shift study of workers at risk of Mesoamerican nephropathy. Environ. Res. 2015, 142, 746–755. [Google Scholar] [CrossRef] [Green Version]
  5. Flouris, A.D.; Ioannou, L.G.; Nybo, L. Working in a warming world: Translating thermal physiology to policy-relevant information. Temperature 2022, 9, 223–226. [Google Scholar] [CrossRef]
  6. Gubernot, D.M.; Anderson, G.B.; Hunting, K.L. Characterizing occupational heat-related mortality in the United States, 2000–2010: An analysis using the census of fatal occupational injuries database. Am. J. Ind. Med. 2015, 58, 203–211. [Google Scholar] [CrossRef] [Green Version]
  7. Bonauto, D.; Anderson, R.; Rauser, E.; Burke, B. Occupational heat illness in Washington State, 1995–2005. Am. J. Ind. Med. 2007, 50, 940–950. [Google Scholar] [CrossRef] [PubMed]
  8. Boden, B.P.; Breit, I.; Beachler, J.A.; Williams, A.; Mueller, F.O. Fatalities in high school and college football players. Am. J. Sports Med. 2013, 41, 1108–1116. [Google Scholar] [CrossRef]
  9. Gamage, P.J.; Fortington, L.V.; Finch, C.F. Epidemiology of exertional heat illnesses in organised sports: A systematic review. J. Sci. Med. Sport 2020, 23, 701–709. [Google Scholar] [CrossRef]
  10. ISO 7933; Ergonomics of the Thermal Environment—Analytical Determination and Interpretation of Heat Stress Using Calculation of the Predicted Heat Strain. International Organization for Standardization: Geneva, Switzerland, 2004.
  11. Whipp, B.J.; Wasserman, K. Oxygen uptake kinetics for various intensities of constant-load work. J. Appl. Physiol. 1972, 33, 351–356. [Google Scholar] [CrossRef] [PubMed]
  12. Ioannou, L.G.; Tsoutsoubi, L.; Mantzios, K.; Gkikas, G.; Piil, J.F.; Dinas, P.C.; Notley, S.R.; Kenny, G.P.; Nybo, L.; Flouris, A.D. The impacts of sun exposure on worker physiology and cognition: Multi-country evidence and interventions. Int. J. Environ. Res. Public Health 2021, 18, 7698. [Google Scholar] [CrossRef]
  13. Meade, R.D.; Lauzon, M.; Poirier, M.P.; Flouris, A.D.; Kenny, G.P. An evaluation of the physiological strain experienced by electrical utility workers in north America. J. Occup. Environ. Hyg. 2015, 12, 708–720. [Google Scholar] [CrossRef] [PubMed]
  14. Meade, R.D.; D’Souza, A.W.; Krishen, L.; Kenny, G.P. The physiological strain incurred during electrical utilities work over consecutive work shifts in hot environments: A case report. J. Occup. Environ. Hyg. 2017, 14, 986–994. [Google Scholar] [CrossRef]
  15. Ioannou, L.G.; Mantzios, K.; Tsoutsoubi, L.; Nintou, E.; Vliora, M.; Gkiata, P.; Dallas, C.N.; Gkikas, G.; Agaliotis, G.; Sfakianakis, K.; et al. Occupational heat stress: Multi-country observations and interventions. Int. J. Environ. Res. Public Health 2021, 18, 6303. [Google Scholar] [CrossRef]
  16. Ioannou, L.G.; Tsoutsoubi, L.; Samoutis, G.; Bogataj, L.K.; Kenny, G.P.; Nybo, L.; Kjellstrom, T.; Flouris, A.D. Time-motion analysis as a novel approach for evaluating the impact of environmental heat exposure on labor loss in agriculture workers. Temperature 2017, 4, 330–340. [Google Scholar] [CrossRef]
  17. Flouris, A.D.; Schlader, Z.J. Human behavioral thermoregulation during exercise in the heat. Scand. J. Med. Sci. Sports 2015, 25, 52–64. [Google Scholar] [CrossRef]
  18. Schlader, Z.J.; Coleman, G.L.; Sackett, J.R.; Sarker, S.; Chapman, C.L.; Johnson, B.D. Activation of autonomic thermoeffectors preceding the decision to behaviourally thermoregulate in resting humans. Exp. Physiol. 2016, 101, 1218–1229. [Google Scholar] [CrossRef] [Green Version]
  19. Schlader, Z.J.; Sackett, J.R.; Sarker, S.; Johnson, B.D. Orderly recruitment of thermoeffectors in resting humans. Am. J. Physiol.-Regul. Integr. Comp. Physiol. 2018, 314, R171–R180. [Google Scholar] [CrossRef]
  20. Paek, J.; Kim, J.; Govindan, R. Energy-efficient rate-adaptive GPS-based positioning for smartphones. In Proceedings of the 8th International Conference on Mobile Systems, Applications, and Services, San Francisco, CA, USA, 14–18 June 2010; pp. 299–314. [Google Scholar]
  21. Ahmed, A.; Ahmed, E. A survey on mobile edge computing. In Proceedings of the 2016 10th International Conference on Intelligent Systems and Control (ISCO), Coimbatore, India, 7–8 January 2016; pp. 1–8. [Google Scholar]
  22. Kingma, B.R.M.; Steenhoff, H.; Toftum, J.; Daanen, H.A.M.; Folkerts, M.A.; Gerrett, N.; Gao, C.; Kuklane, K.; Petersson, J.; Halder, A.; et al. ClimApp—Integrating Personal Factors with Weather Forecasts for Individualised Warning and Guidance on Thermal Stress. Int. J. Environ. Res. Public Health 2021, 18, 11317. [Google Scholar] [CrossRef] [PubMed]
  23. Folkerts, M.A.; Boshuizen, A.W.; Gosselink, G.; Gerrett, N.; Daanen, H.A.M.; Gao, C.; Toftum, J.; Nybo, L.; Kingma, B.R.M. Predicted and user perceived heat strain using the ClimApp mobile tool for individualized alert and advice. Clim. Risk Manag. 2021, 34, 100381. [Google Scholar] [CrossRef]
  24. Eggeling, J.; Rydenfält, C.; Kingma, B.; Toftum, J.; Gao, C. The usability of ClimApp: A personalized thermal stress warning tool. Clim. Serv. 2022, 27, 100310. [Google Scholar] [CrossRef]
  25. Sungmee, P.; Jayaraman, S. Enhancing the quality of life through wearable technology. IEEE Eng. Med. Biol. Mag. 2003, 22, 41–48. [Google Scholar] [CrossRef] [PubMed]
  26. Klasnja, P.; Pratt, W. Healthcare in the pocket: Mapping the space of mobile-phone health interventions. J. Biomed. Inf. 2012, 45, 184–198. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  27. Fox, S.; Duggan, M. Mobile Health 2012; Pew Research Center’s Internet & American Life Project: Washington, DC, USA, 2012. [Google Scholar]
  28. Notley, S.R.; Flouris, A.D.; Kenny, G.P. Occupational heat stress management: Does one size fit all? Am. J. Ind. Med. 2019, 62, 1017–1023. [Google Scholar] [CrossRef] [PubMed]
  29. Notley, S.R.; Flouris, A.D.; Kenny, G.P. On the use of wearable physiological monitors to assess heat strain during occupational heat stress. Appl. Physiol. Nutr. Metab. 2018, 43, 869–881. [Google Scholar] [CrossRef]
  30. Khatib, T.; Elmenreich, W. Modeling of Photovoltaic Systems Using Matlab: Simplified Green Codes; John Wiley & Sons: Hoboken, NJ, USA, 2016. [Google Scholar]
  31. Kasten, F.; Czeplak, G. Solar and terrestrial radiation dependent on the amount and type of cloud. Sol. Energy 1980, 24, 177–189. [Google Scholar] [CrossRef]
  32. Ioannou, L.G.; Dinas, P.C.; Notley, S.R.; Gofa, F.; Gourzoulidis, G.A.; Brearley, M.; Epstein, Y.; Havenith, G.; Sawka, M.N.; Brode, P.; et al. Indicators to assess physiological heat strain—Part 2: Delphi exercise. Temperature 2022, 9, 263–273. [Google Scholar] [CrossRef]
  33. Ioannou, L.G.; Mantzios, K.; Tsoutsoubi, L.; Notley, S.R.; Dinas, P.C.; Brearley, M.; Epstein, Y.; Havenith, G.; Sawka, M.N.; Brode, P.; et al. Indicators to assess physiological heat strain—Part 1: Systematic review. Temperature 2022, 9, 227–262. [Google Scholar] [CrossRef] [PubMed]
  34. Ioannou, L.G.; Tsoutsoubi, L.; Mantzios, K.; Vliora, M.; Nintou, E.; Piil, J.F.; Notley, S.R.; Dinas, P.C.; Gourzoulidis, G.A.; Havenith, G.; et al. Indicators to assess physiological heat strain—Part 3: Multi-country field evaluation and consensus recommendations. Temperature 2022, 9, 274–291. [Google Scholar] [CrossRef] [PubMed]
  35. Yaglou, C.; Minaed, D. Control of heat casualties at military training centers. Arch. Indust. Health 1957, 16, 302–316. [Google Scholar]
  36. Liljegren, J.C.; Carhart, R.A.; Lawday, P.; Tschopp, S.; Sharp, R. Modeling the Wet Bulb Globe Temperature Using Standard Meteorological Measurements. J. Occup. Environ. Hyg. 2008, 5, 645–655. [Google Scholar] [CrossRef] [PubMed]
  37. Lemke, B.; Kjellstrom, T. Calculating Workplace WBGT from Meteorological Data: A Tool for Climate Change Assessment. Ind. Health 2012, 50, 267–278. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  38. Malchaire, J.B.M. Occupational Heat Stress Assessment by the Predicted Heat Strain Model. Ind. Health 2006, 44, 380–387. [Google Scholar] [CrossRef] [Green Version]
  39. Ioannou, L.G.; Tsoutsoubi, L.; Mantzios, K.; Flouris, A.D. A free software to predict heat strain according to the ISO 7933:2018. Ind. Health 2019, 57, 711–720. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  40. Fiala, D. Dynamic Simulation of Human Heat Transfer and Thermal Comfort; De Montfort University: Leicester, UK, 1998. [Google Scholar]
  41. Piil, J.F.; Kingma, B.; Morris, N.B.; Christiansen, L.; Ioannou, L.G.; Flouris, A.D.; Nybo, L. Proposed framework for forecasting heat-effects on motor-cognitive performance in the Summer Olympics. Temperature 2021, 8, 262–283. [Google Scholar] [CrossRef] [PubMed]
  42. Lazaro, P.; Momayez, M. Validation of the Predicted Heat Strain Model in Hot Underground Mines. Min. Met. Explor. 2019, 36, 1213–1219. [Google Scholar] [CrossRef]
  43. Monazzam, M.R.; Golbabaei, F.; Hematjo, R.; Hosseini, M.; Nassiri, P.; Dehghan, S.F. Evaluation of DI, WBGT, and SWreq/PHS Heat Stress Indices for Estimating the Heat Load on the Employees of a Petrochemical Industry. Int. J. Occup. Hyg. 2015, 6, 6–10. [Google Scholar]
  44. Campbell-Staton, S.C.; Walker, R.H.; Rogers, S.A.; De León, J.; Landecker, H.; Porter, W.; Mathewson, P.D.; Long, R.A. Physiological costs of undocumented human migration across the southern United States border. Science 2021, 374, 1496–1500. [Google Scholar] [CrossRef]
  45. Holmer, I.; Nilsson, H.O.; Anttonen, H. Prediction of wind effects on cold protective clothing. In Proceedings of the RTO HFM Symposium on “Blowing Hot and Cold: Protecting Against Climatic Extremes”, Dresden, Germany, 8–10 October 2001; pp. 6-1–6-6. [Google Scholar]
  46. World Health Organization. Technical Report Series No. 412: Health Factors Involved in Working under Conditions of Heat Stress; World Health Organization: Geneva, Switzerland, 1969; pp. 1–32. [Google Scholar]
  47. Faul, F.; Erdfelder, E.; Lang, A.G.; Buchner, A. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav. Res. Methods 2007, 39, 175–191. [Google Scholar] [CrossRef]
  48. Du Bois, D.; Du Bois, E.F. Clinical calorimetry: Tenth paper—A formula to estimate the approximate surface area if height and weight be known. Arch. Intern. Med. 1916, XVII, 863–871. [Google Scholar] [CrossRef] [Green Version]
  49. Notley, S.R.; Meade, R.D.; Kenny, G.P. Time following ingestion does not influence the validity of telemetry pill measurements of core temperature during exercise-heat stress: The journal Temperature toolbox. Temperature 2021, 8, 12–20. [Google Scholar] [CrossRef] [PubMed]
  50. Bogerd, C.P.; Velt, K.B.; Annaheim, S.; Bongers, C.; Eijsvogels, T.M.H.; Daanen, H.A.M. Comparison of two telemetric intestinal temperature devices with rectal temperature during exercise. Physiol. Meas. 2018, 39, 03nt01. [Google Scholar] [CrossRef] [PubMed]
  51. Service, T.W.; Junker, K.; Service, B.; Coehoorn, C.J.; Harrington, M.; Martin, S.; Stuart-Hill, L.A. An assessment of the validity and reliability of the P022–P version of e-Celsius core temperature capsules. J. Therm. Biol. 2023, 112, 103486. [Google Scholar] [CrossRef] [PubMed]
  52. Ainsworth, B.E.; Haskell, W.L.; Herrmann, S.D.; Meckes, N.; Bassett, D.R., Jr.; Tudor-Locke, C.; Greer, J.L.; Vezina, J.; Whitt-Glover, M.C.; Leon, A.S. 2011 Compendium of Physical Activities: A second update of codes and MET values. Med. Sci. Sports Exerc. 2011, 43, 1575–1581. [Google Scholar] [CrossRef] [Green Version]
  53. Ainsworth, B.E.; Haskell, W.L.; Leon, A.S.; Jacobs, D.R.; Montoye, H.J.; Sallis, J.F.; Paffenbarger, R.S. Compendium of physical activities: Classification of energy costs of human physical activities. Med. Sci. Sports Exerc. 1993, 25, 71–80. [Google Scholar] [CrossRef] [PubMed]
  54. ACGIH. TLVs and BEIs: Based on the Documendation of the Threshold Limit Values for Chemical Substances and Physical Agents & Biological Exposure Indices; ACGIH: Cincinnati, OH, USA, 2020. [Google Scholar]
  55. Havenith, G. Clothing and thermoregulation. Curr. Probl. Dermatol. 2003, 31, 35–49. [Google Scholar] [PubMed]
  56. Havenith, G. Interaction of clothing and thermoregulation. Exog. Dermatol. 2002, 1, 221–230. [Google Scholar] [CrossRef] [Green Version]
  57. Piil, J.F.; Lundbye-Jensen, J.; Christiansen, L.; Ioannou, L.; Tsoutsoubi, L.; Dallas, C.N.; Mantzios, K.; Flouris, A.D.; Nybo, L. High prevalence of hypohydration in occupations with heat stress—Perspectives for performance in combined cognitive and motor tasks. PLoS ONE 2018, 13, e0205321. [Google Scholar] [CrossRef] [Green Version]
  58. Tawatsupa, B.; Lim, L.L.Y.; Kjellstrom, T.; Seubsman, S.-a.; Sleigh, A.; Thai Cohort Study Team. Association Between Occupational Heat Stress and Kidney Disease Among 37 816 Workers in the Thai Cohort Study (TCS). J. Epidemiol. 2012, 22, 251–260. [Google Scholar] [CrossRef] [Green Version]
  59. Sawka, M.N.; Montain, S.J.; Latzka, W.A. Hydration effects on thermoregulation and performance in the heat. Comp. Biochem. Physiol. Part A Mol. Integr. Physiol. 2001, 128, 679–690. [Google Scholar] [CrossRef] [PubMed]
  60. ISO 15831: 2004; Clothing—Physiological Effects—Measurement of Thermal Insulation by Means of a Thermal Manikin. International Organization for Standardization: Geneva, Switzerland, 2004.
  61. Potter, A.W.; Gonzalez, J.A.; Carter, A.J.; Looney, D.P.; Rioux, T.P.; Srinivasan, S.; Sullivan-Kwantes, W.; Xu, X. Comparison of Cold Weather Clothing Biophysical Properties: US Army, Canadian Department of National Defence, and Norwegian Military; U.S. Army Research Institute of Environmental Medicine: Natick, MA, USA, 2018. [Google Scholar]
  62. Ioannou, L.G.; Mantzios, K.; Tsoutsoubi, L.; Panagiotaki, Z.; Kapnia, A.K.; Ciuha, U.; Nybo, L.; Flouris, A.D.; Mekjavic, I.B. Effect of a Simulated Heat Wave on Physiological Strain and Labour Productivity. Int. J. Environ. Res. Public Health 2021, 18, 3011. [Google Scholar] [CrossRef]
  63. Willmott, C.J.; Robeson, S.M.; Matsuura, K. A refined index of model performance. Int. J. Climatol. 2012, 32, 2088–2094. [Google Scholar] [CrossRef]
  64. Willmott, C.J. On the validation of models. Phys. Geogr. 1981, 2, 184–194. [Google Scholar] [CrossRef]
  65. Murray, D.B.; Teare, S.W. Probability of a tossed coin landing on edge. Phys. Rev. E 1993, 48, 2547–2552. [Google Scholar] [CrossRef]
  66. Malchaire, J.; Piette, A.; Kampmann, B.; Mehnert, P.; Gebhardt, H.; Havenith, G.; den Hartog, E.; Holmer, I.; Parsons, K.; Alfano, G.; et al. Development and validation of the predicted heat strain model. Ann. Occup. Hyg. 2001, 45, 123–135. [Google Scholar] [CrossRef] [Green Version]
  67. Musah, A.; Dutra, L.M.M.; Aldosery, A.; Browning, E.; Ambrizzi, T.; Borges, I.V.G.; Tunali, M.; Başibüyük, S.; Yenigün, O.; Moreno, G.M.M.; et al. An evaluation of the OpenWeatherMap API versus INMET using weather data from two Brazilian cities: Recife and Campina Grande. Data 2022, 7, 106. [Google Scholar] [CrossRef]
  68. Jian, S.; Zhuo, C.; Heng, L.; Simeng, Q.; Xin, W.; Limin, Y.; Wei, X. Application of artificial intelligence technology to numerical weather prediction. J. Appl. Meteorol. Sci. 2021, 32, 1. [Google Scholar] [CrossRef]
  69. Flouris, A.D.; McGinn, R.; Poirier, M.P.; Louie, J.C.; Ioannou, L.G.; Tsoutsoubi, L.; Sigal, R.J.; Boulay, P.; Hardcastle, S.G.; Kenny, G.P. Screening criteria for increased susceptibility to heat stress during work or leisure in hot environments in healthy individuals aged 31–70 years. Temperature 2018, 5, 86–99. [Google Scholar] [CrossRef] [Green Version]
  70. Ioannou, L.G.; Gkikas, G.; Mantzios, K.; Tsoutsoubi, L.; Flouris, A.D. Chapter 32—Risk assessment for heat stress during work and leisure. In Toxicological Risk Assessment and Multi-System Health Impacts from Exposure; Tsatsakis, A.M., Ed.; Academic Press: Cambridge, MA, USA, 2021; pp. 373–385. [Google Scholar]
  71. Bonell, A.; Sonko, B.; Badjie, J.; Samateh, T.; Saidy, T.; Sosseh, F.; Sallah, Y.; Bajo, K.; Murray, K.A.; Hirst, J.; et al. Environmental heat stress on maternal physiology and fetal blood flow in pregnant subsistence farmers in The Gambia, west Africa: An observational cohort study. Lancet Planet. Health 2022, 6, e968–e976. [Google Scholar] [CrossRef] [PubMed]
  72. Baker, F.C.; Siboza, F.; Fuller, A. Temperature regulation in women: Effects of the menstrual cycle. Temperature 2020, 7, 226–262. [Google Scholar] [CrossRef]
  73. Meade, R.D.; Notley, S.R.; Kenny, G.P. Aging and human heat dissipation during exercise-heat stress: An update and future directions. Curr. Opin. Physiol. 2019, 10, 219–225. [Google Scholar] [CrossRef]
  74. Stapleton, J.M.; Poirier, M.P.; Flouris, A.D.; Boulay, P.; Sigal, R.J.; Malcolm, J.; Kenny, G.P. Aging impairs heat loss, but when does it matter? J. Appl. Physiol. 2015, 118, 299–309. [Google Scholar] [CrossRef]
  75. Notley, S.R.; Poirier, M.P.; Hardcastle, S.G.; Flouris, A.D.; Boulay, P.; Sigal, R.J.; Kenny, G.P. Aging Impairs Whole-Body Heat Loss in Women under Both Dry and Humid Heat Stress. Med. Sci. Sports Exerc. 2017, 49, 2324–2332. [Google Scholar] [CrossRef] [PubMed]
  76. Wright, H.E.; Larose, J.; McLellan, T.M.; Miller, S.; Boulay, P.; Kenny, G.P. Do older firefighters show long-term adaptations to work in the heat? J. Occup. Environ. Hyg. 2013, 10, 705–715. [Google Scholar] [CrossRef]
  77. Notley, S.R.; Meade, R.D.; D’Souza, A.W.; Friesen, B.J.; Kenny, G.P. Heat loss is impaired in older men on the day after prolonged work in the heat. Med. Sci. Sports Exerc. 2018, 50, 1859–1867. [Google Scholar] [CrossRef] [PubMed]
  78. Kenny, G.P.; Yardley, J.; Brown, C.; Sigal, R.J.; Jay, O. Heat stress in older individuals and patients with common chronic diseases. Can. Med. Assoc. J. 2010, 182, 1053–1060. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  79. Kenny, G.P.; Larose, J.; Wright-Beatty, H.E.; Boulay, P.; Sigal, R.J.; Flouris, A.D. Older firefighters are susceptible to age-related impairments in heat dissipation. Med. Sci. Sports Exerc. 2015, 47, 1281–1290. [Google Scholar] [CrossRef] [PubMed]
  80. D’Souza, A.W.; Notley, S.R.; Kenny, G.P. The Relation between Age and Sex on Whole-Body Heat Loss during Exercise-Heat Stress. Med. Sci. Sports Exerc. 2020, 52, 2242–2249. [Google Scholar] [CrossRef]
  81. Gagnon, D.; Kenny, G.P. Does sex have an independent effect on thermoeffector responses during exercise in the heat? J. Physiol. 2012, 590, 5963–5973. [Google Scholar] [CrossRef]
  82. Stansberry, K.B.; Shapiro, S.A.; Hill, M.A.; McNitt, P.M.; Meyer, M.D.; Vinik, A.I. Impaired Peripheral Vasomotion in Diabetes. Diabetes Care 1996, 19, 715–721. [Google Scholar] [CrossRef]
  83. Carter, M.R.; McGinn, R.; Barrera-Ramirez, J.; Sigal, R.J.; Kenny, G.P. Impairments in local heat loss in type 1 diabetes during exercise in the heat. Med. Sci. Sports Exerc. 2014, 46, 2224–2233. [Google Scholar] [CrossRef]
  84. Notley, S.R.; Akerman, A.P.; Friesen, B.J.; Sigal, R.J.; Flouris, A.D.; Boulay, P.; Kenny, G.P. Exercise-heat tolerance in middle-aged-to-older men with type 2 diabetes. Acta Diabetol. 2021, 58, 809–812. [Google Scholar] [CrossRef] [PubMed]
  85. Notley, S.R.; Meade, R.D.; D’Souza, A.W.; McGarr, G.W.; Kenny, G.P. Cumulative effects of successive workdays in the heat on thermoregulatory function in the aging worker. Temperature 2018, 5, 293–295. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  86. Notley, S.R.; Meade, R.D.; Friesen, B.J.; D’Souza, A.W.; Kenny, G.P. Does a prolonged work day in the heat impair heat loss on the next day in young men? Med. Sci. Sports Exerc. 2018, 50, 318–326. [Google Scholar] [CrossRef] [PubMed]
Applsci 13 08561 g001 550
Figure 1. Graphical illustration depicting the strategy of the Heat-Health mobile phone application, specifically the input variables obtained by the app and the output derived by the thermoregulatory model presented as simple guidelines and recommendations to prevent heat strain.
Figure 1. Graphical illustration depicting the strategy of the Heat-Health mobile phone application, specifically the input variables obtained by the app and the output derived by the thermoregulatory model presented as simple guidelines and recommendations to prevent heat strain.
Applsci 13 08561 g001
Applsci 13 08561 g002 550
Figure 2. Relationships between the predicted and measured core body temperatures. Dashed lines and shaded areas indicate the linear regression model and 95% confidence interval for the average (top graph) and maximum (bottom graph) core body temperatures measured during each trial.
Figure 2. Relationships between the predicted and measured core body temperatures. Dashed lines and shaded areas indicate the linear regression model and 95% confidence interval for the average (top graph) and maximum (bottom graph) core body temperatures measured during each trial.
Applsci 13 08561 g002
Applsci 13 08561 g003 550
Figure 3. Error indices and average differences between measured and predicted core body temperatures (Tcore). Pink and green circles represent the average and maximum Tcore measured during each trial, respectively. Dark thick and light-grey dashed lines represent mean bias and 95% confidence interval, respectively. Pink and green dashed lines represent the associations between participants’ clothing insulation and the difference between their measured and predicted Tcore during each trial.
Figure 3. Error indices and average differences between measured and predicted core body temperatures (Tcore). Pink and green circles represent the average and maximum Tcore measured during each trial, respectively. Dark thick and light-grey dashed lines represent mean bias and 95% confidence interval, respectively. Pink and green dashed lines represent the associations between participants’ clothing insulation and the difference between their measured and predicted Tcore during each trial.
Applsci 13 08561 g003
Applsci 13 08561 g004 550
Figure 4. Predictive capacity of the newly developed mobile phone application in pre-classifying individuals with an increased risk of experiencing elevated physiological heat strain, determined as core body temperature higher than the guidelines (≥38 °C) of the World Health Organization (WHO). Specificity and sensitivity correspond to the ability of the newly developed mobile phone application to correctly predict individuals that will or will not experience elevated core body temperature, respectively. Accuracy corresponds to the percentage of correct predictions made by the newly developed mobile phone application.
Figure 4. Predictive capacity of the newly developed mobile phone application in pre-classifying individuals with an increased risk of experiencing elevated physiological heat strain, determined as core body temperature higher than the guidelines (≥38 °C) of the World Health Organization (WHO). Specificity and sensitivity correspond to the ability of the newly developed mobile phone application to correctly predict individuals that will or will not experience elevated core body temperature, respectively. Accuracy corresponds to the percentage of correct predictions made by the newly developed mobile phone application.
Applsci 13 08561 g004
Table
Table 1. Participants’ characteristics.
Table 1. Participants’ characteristics.
nBody Mass
kg		Height
cm		Body Surface Area
m2		Age
y
Participants3777.1 ± 14.8175.2 ± 9.01.92 ± 0.2128.9 ± 5.7
Males2982.7 ± 12.3178.7 ± 6.92.01 ± 0.1630.0 ± 5.9
Females859.4 ± 4.0163.6 ± 4.41.64 ± 0.0725.8 ± 2.8
Hiking31 (26 M; 5 F)79.9 ± 14.6176.7 ± 8.71.96 ± 0.2129.0 ± 6.0
Guard duty17 (9 M; 8 F)73.0 ± 16.5172.3 ± 10.31.86 ± 0.2528.7 ± 6.6
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Ioannou, L.G.; Ciuha, U.; Fisher, J.T.; Tsoutsoubi, L.; Tobita, K.; Bonell, A.; Cotter, J.D.; Kenny, G.P.; Flouris, A.D.; Mekjavic, I.B. Novel Technological Advances to Protect People Who Exercise or Work in Thermally Stressful Conditions: A Transition to More Personalized Guidelines. Appl. Sci. 2023, 13, 8561. https://doi.org/10.3390/app13158561

AMA Style

Ioannou LG, Ciuha U, Fisher JT, Tsoutsoubi L, Tobita K, Bonell A, Cotter JD, Kenny GP, Flouris AD, Mekjavic IB. Novel Technological Advances to Protect People Who Exercise or Work in Thermally Stressful Conditions: A Transition to More Personalized Guidelines. Applied Sciences. 2023; 13(15):8561. https://doi.org/10.3390/app13158561

Chicago/Turabian Style

Ioannou, Leonidas G., Urša Ciuha, Jason T. Fisher, Lydia Tsoutsoubi, Kunihito Tobita, Ana Bonell, James D. Cotter, Glen P. Kenny, Andreas D. Flouris, and Igor B. Mekjavic. 2023. "Novel Technological Advances to Protect People Who Exercise or Work in Thermally Stressful Conditions: A Transition to More Personalized Guidelines" Applied Sciences 13, no. 15: 8561. https://doi.org/10.3390/app13158561

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop