deliverable 3.2a. : Decision algorithm for the heat stressed worker : empirical analysis
Authors: Sarah Davey, Victoria Richmond, Katy Griggs and George Havenith
Elevated body core temperatures (Tc) of 1-3oC can induce fatigue and reduce cognitive ability, which in turn can have a detrimental effect on work productivity and worker’s health. In addition high body core temperatures can also cause increases in both absences from work and accidents within the workplace.
The proposed PROSPIE clothing system is designed to alert workers of any dangerous elevations in their Tc. Current reliable methods of measuring Tc are quite invasive and impractical for a clothing system designed for multiple uses. Therefore, a reliable method to measure Tc that uses only non-invasive measures is necessary for the production of the PROSPIE clothing system. Consequently, the main aim of this study was to develop decision algorithms that reliably predict Tc using only non-invasive measures that could potentially be incorporated into the PROSPIE clothing system.
Twenty-one participants (12 males, 9 females) of varying fitness and acclimation levels completed 103 tests (Age: 25.1 [±5.3] yrs, Body mass: 68.8 [±11.0] kg, Height: 179.9 [±6.6] cm, : 52.4 [±11.2] ml.kg.min-1, Body fat: 15.42 [±6.53] %). Each test involved participants undertaking a work/rest regime in one of six conditions. Each condition slightly differed in order to test the influence of: 1) solar radiation, 2) the type of protective clothing worn by the participants over a cotton undershirt (i.e. permeable vs. impermeable coveralls), and 3) ambient temperature and humidity, on Tc.
Several non-invasive measurements that have previously been shown to be associated with changes in Tc were recorded throughout each test. Regression analysis was used to establish the combination of non-invasive indirect measures that explain the highest variation in both mean skin temperature and Tc (measured using rectal temperature). Uniformity and the practical implication of any prediction error in the combinations were assessed by Bland and Altman charts. From this analysis, a total of 36 decision algorithms were developed. The decision algorithms were developed based on the PROSPIE three layer clothing system (i.e. layer 1= base clothing layer on the skin, layer 2= layer 1, plus the boundary layer between the base clothing layer and outer protective clothing layer, and layer 3 = layers 1 & 2, plus the surface temperature of the outer protective clothing layer).
This report provides the preliminary decision algorithms recommended for each layer of the PROSPIE clothing system that reliably predict changes in Tc in several types of environments that induce heat stress. It also provides recommendations on the minimum amount of sensors required to be incorporated into the PROSPIE clothing system based both on theoretical and practical implications.