Validation of Core, Rectal and Skin Temperature Predictions of a Free Web-Based Predictive Heat Strain Software Based on the ISO 7933:2023 Standard in Recreational Athletes

Abstract

We previously developed the FAME Lab PHS software (PHS_FL), a free offline software to calculate the predicted heat strain for a group of individuals based on the ISO 7933. The objectives of this study were to: upgrade the PHS_FL from an offline (desktop-version) tool to a web-based platform, as well as assess its validity in recreational athletes in different forms of exercise and across various temperature recording methodologies and environmental conditions. The web PHS_FL was developed as browser-based software developed using HTML, CSS, and JavaScript, and included several updates from the previous offline version. Its validity was assessed in 83 healthy non-smoking males and females during rest, exercise, and post-exercise recovery in 165 trials (cycling: 97; running: 68). Trials were performed in an environmental chamber under varying environmental conditions: 19.1 to 40.6 °C air temperature, 30.0% to 60.0% relative humidity, 0.1 to 0.5 m/s wind speed, and 0 or 800 W/m² solar radiation. Comparison of actual vs. predicted core body temperature showed 0.85 Willmott’s Index of Agreement, 0.76 (P < 0.001) correlation coefficient, and 95% limits of agreement of 0.16 ± 0.83 °C (mean difference ± 95% limits). Results for rectal temperature showed 0.79 Willmott’s Index of Agreement, 0.68 (P < 0.001) correlation coefficient, and 95% limits of agreement of 0.18 ± 0.76 °C. Results for skin temperature showed 0.77 Willmott’s Index of Agreement, 0.75 (P < 0.001) correlation coefficient, and 95% limits of agreement of − 0.24 ± 2.28 °C. We conclude that the web PHS_FL provides acceptably accurate predictions of core body temperature and skin temperature to be used as indicators of physiological heat strain.