Benchmarking Heat Index as an occupational exposure limit for heat stress.

While wet bulb globe temperature (WBGT) is the long-accepted index to represent the environmental contributions to heat stress, Heat Index (HI) is a commonly reported index and is used for heat stress guidance. The purpose of this article was to propose an HI-based heat stress exposure limit. The data came from previous progressive heat stress studies that identified the critical conditions between sustainable and unsustainable exposures. The experimental trials included five clothing ensembles at three levels each of relative humidity (rh) and metabolic rate (M). The critical Heat Index (HIcrit) was used to characterize the trial exposure. An analysis of variance (ANOVA) assessed the effects of M, clothing, and rh on HIcrit. After proposing a relationship between HIcrit and M to represent a benchmark exposure limit based on HI (called HIbel), the ability of the proposed relationship to discriminate between Sustainable and Unsustainable conditions was assessed using receiver operating characteristics curves (ROC curves). Based on the ANOVA results, the main effects of M, rh, and clothing on HIcrit were significant; the interaction between rh and clothing was not significant. There were differences in mean HIcrit among all the ensembles. For effects of relative humidity on HIcrit, the mean HIcrit at rh at 20% was 3 °C lower than the mean values for 50% and 70%. The benchmark exposure limit from the woven clothing data was HIbel [°C] = 49-0.026 M [W]. In terms of the ability of HIbel to discriminate, area under the ROC curve was 0.86, which was similar to WBGT-based exposure limits. Similar in purpose for WBGT-based exposure assessment, HI clothing adjustment values (HIcav) of 1.5 °C (particle barrier coveralls), 6 °C (water barrier coveralls), and 18.5 °C (vapor barrier coveralls) were supported. It should also be noted that the effects of the sun and lack of acclimatization were not included in this analysis; where the sun might reasonably increase the effects of the ambient HI by an additional 3.5 °C and being unacclimatized by 5.5 °C.