Article Text

There is no ‘Swiss Army Knife’ of thermal indices: the importance of considering ‘why?’ and ‘for whom?’ when modelling heat stress in sport
  1. Andrew Grundstein1,
  2. Jennifer Vanos2
  1. 1 Department of Geography, University of Georgia, Athens, Georgia, USA
  2. 2 School of Sustainability, Arizona State University, Tempe, Arizona, USA
  1. Correspondence to Dr Andrew Grundstein, Geography, University of Georgia, Athens, GA 30602, USA; andrewg{at}uga.edu

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Combining extreme heat with exercise increases the risks of exertional heat illness, impairs performance and imposes thermoregulatory strain on athletes.1 What is the best way to monitor heat stress and strain in athletes? Wet bulb globe temperature (WBGT)—a direct environmental measure—coupled with activity modification, has long been used in sports, occupational safety and the military.2 Because of its longevity, it has well-known benefits (eg, simple to calculate/measure; integrates multiple weather variables) and limitations (eg, underestimation of stress of restricted evaporation; poor incorporation of clothing/adjustment factors).3 4 Practical decisions, such as cancelling an event or implementing countermeasures for heat, depend not only on the environmental conditions, but on the nature and length of the activity itself (eg, marathon vs beach volleyball), clothing and athlete anatomical characteristics.

In recent decades, additional heat metrics that are derived from human energy balance models (eg, the physiological equivalent temperature (PET) and the Universal Thermal Climate Index (UTCI)) have been introduced and increasingly applied, even for sporting events.5 These more comprehensive heat indices are designed to provide more meteorologically dynamic outputs to predict heat stress. However, they have their own limitations for sporting applications due to simplified and non-modifiable physiological/behavioural factors that are not sport-specific in their available formats.

Does our field of sport-related environmental physiology need to expand the arsenal of heat monitoring tools–both measurements and models–beyond indices that have been simplified for general applications? This editorial calls attention to the considerations and the potential misuse of thermal indices applied in outdoor sport settings.

From simple to complex: the devil is in the details

A simple thermal index with few inputs might seem useful for clinical purposes. However, it is critical to understand the underlying assumptions and the limits of applicability of a given model (ie, how, why, when and for whom/under what conditions a model can or should be applied). The bioclimate community, for instance, has focused on a population-level response to heat with generalised, mostly static physiological inputs (‘average human’), combined with advanced meteorological inputs (eg, fine-scale radiation data) to provide a thermal index value. In figure 1A, we present results of the thermal index values from four common indices within equivalent environments and observe different values due to variations in model design, purpose and physiological/behavioural assumptions (details in online supplemental file).6 7

Supplemental material

Figure 1

Comparisons of common bioclimate indices with a human heat balance (HHB) modelling approach. (A) Modelled bioclimate thermal index values (in full sun) for the Universal Thermal Climate Index (UTCI), physiological equivalent temperature (PET), wet bulb globe temperature (WBGT) and modified PET (mPET) using the given model assumptions (see online supplemental file) and standard sun-exposed environments as follows: windspeed=2.5 ms−1, RH=35%, Ta 26°C–32°C at 2°C intervals, Tmrt=Ta+8°C. (B) Application of a standard HHB modelling approach to the given environments in sun (see citations i–iii below). Dark blue lines show changing ratio between required and maximum possible evaporation (ie, Embedded Image or skin wettedness required (ωreq)) at one instant under the model/environmental assessments in (A), with 0.24* clo clothing insulation and four metabolic intensities and activity speeds for running from the compendium of physical activities (http://prevention.sph.sc.edu/tools/docs/documents_compendium.pdf). Activity speeds are converted to relative activity speeds based on windspeed. In this example, heat stress is compensable when Embedded Image <1.0. Thermal indices from (A) are modelled applying this same heat balance modelling approach, yet with the metabolic, intrinsic clothing insulation (in clo), and activity speed assumptions within the models applied, as shown to the right of the graph. Note that thermal index estimates were completed for illustrative purposes of this editorial and not the way these models were designed to be implemented. All input parameters for these thermal indices are included in online supplemental file. All modelling in (B) applies the following constants: human height (1.8 m), weight (75 kg), mean skin temperature (34°C), maximum skin wettedness=1.00. Rayman Pro was used to model PET and mPET; UTCI was modelled with both Rayman Pro and the online UTCI tool (http://www.utci.org/utcineu/utcineu.php). *0.24 clo is based on tank top, light shorts, athletic shoes from ISO.10–12

The exercise physiology community, on the other hand, has approached thermoregulation from an individual perspective, although with a tendency to use less sophisticated or variable weather measurements as inputs. They use heat balance models to determine rates of heat exchange over time (Wm−2) and include advanced inputs of clothing and metabolic rate, with the latter being the most critical factor predicting core temperature during activity.8

The above two approaches have similar goals in preventing heat illness and discomfort but have different starting points/inputs and different end points/end goals; thus, understanding the intended purpose of the model or index is critical to selecting the optimal method to accurately assess heat stress and strain (see online supplemental file). The UTCI and PET, for example, assume metabolism, activity speed and clothing for the general population and therefore estimate heat stress or thermal comfort for the average person rather than subpopulations like athletes.

In figure 1B, we illustrate how the PET, UTCI and WBGT would vary in a thermo-physiologically relevant way [via required vs maximum evaporative cooling (Ereq:Emax)] in equivalent environments by using the assumed static physiological inputs for each index (ie, metabolic level, activity speed, clothing). In comparison to a complex heat balance model at various running speeds, it is evident that the non-modifiable physiological assumptions within these indices (eg, PET assuming constant 80 W of activity) critically diminish the accuracy of the output if used for dynamic sports or occupational purposes. However, such assumptions were ‘made deliberately to define an index independent of individual behaviour’ (Höppe, p73).7 Thus, compared with a modifiable heat balance model, use of common thermal indices may dangerously underestimate the heat risk of a high intensity activity (eg, running, cycling, tennis, football).

Thinking ahead and future steps

Advancements in technology may allow sports scientists and event organisers to run human heat balance models in real time on mobile devices that use on-site meteorological data and human inputs that are changeable for different sports. Moving forward, there are at least three practical steps that are needed to bring human heat balance models to the field of activity modification and real-time heat stress assessment in sport.

  1. Heat balance models or thermal indices must be applied appropriately with active individuals in mind and allow for altering key physiological factors like metabolic output, clothing ensembles and activity speed (eg, cyclists vs runners).

  2. While the examples provided earlier in this editorial show differences in output among models, both laboratory and field-based studies in an athletic setting are needed to show the validity and reliability of these models in identifying dangerous conditions relevant to health outcomes.

  3. The most appropriate models should be developed so that they are easily conveyed to, and understood by, sport and exercise medicine clinicians such as athletic trainers (eg, Heat Stress Scale).9

These steps will advance heat safety for both recreational and elite athletes. It requires both technological advances and strong interdisciplinary collaborations among physiologists/sports medicine professionals and biometeorologists with expertise in both micrometeorological data collection/use and heat balance modelling. Thse advancements will bring reliable heat stress monitoring into the 21st century with thoughtfully applied evidence-based approaches.

Ethics statements

Acknowledgments

The authors would like to thank all reviewers for their helpful comments and Dr Ariane Middel for her critical insight on bioclimate indices.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Contributors AG conceived of the idea, JV and AG developed and wrote the editorial, JV developed the figure and supplemental table.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.