Graduation Year


Document Type




Degree Name

MS in Public Health (M.S.P.H.)

Degree Granting Department

Public Health

Major Professor

Thomas E. Bernard, Ph.D.

Committee Member

Rene R. Salazar, Ph.D.

Committee Member

Yougui Wu, Ph.D.


ROC Curve, Heart Rate, Core Temperature, Rectal Temperature


Introduction: Assessment of heat strain is an alternative approach to assessing heat stress exposures. Two common measures of heat strain are body core temperature (TC) and heart rate (HR). In this study TC was assessed by rectal temperature (Tre). Physiological Strain Index (PSI) was developed to combine both Tre and HR into one metric. Data collected from progressive heat stress trials were used to (1) demonstrate that PSI can distinguish between Sustainable and Unsustainable heat stress; (2) suggest values for PSI that demonstrate a sustainable level of heat stress; and (3) determine if clothing or metabolic rate were effect modifiers.

Methods: Two previous progressive heat stress studies included 494 trials with 988 pairs of Sustainable and Unsustainable exposures over a range of relative humidity (rh), metabolic rates (M) and clothing using 29 participants. To assess the discrimination ability of PSI, conditional logistic regression and logistic regression were used. The accuracy of PSI was assessed using Receiver Operating Characteristic curves (ROC).

Results: The present study found that primary (Tre, HR, and Tsk) and derived (PSI and ΔTre-sk) HSMs can accurately predict Unsustainable heat stress exposures based on AUCs that ranged from 0.73 to 0.86. Skin temperature had the highest AUC (0.86) with PSI in the mid-range (0.79).

The values of the HSMs associated with a predicted probability of 0.25 were considered as screening values (PSI < 2.6, ΔTre-sk > 1.9 °C, Tre < 37.5, HR < 109, and Tsk < 35.8). The value of using any one of these individual indicators is that they act as a screening tool to decide if an exposure assessment is needed.

Metabolic rate was found to be a confounder for all the HSMs except for RTsk. It was not statistically significant for HSMs derived models (PSI and ΔTre-sk). And its effect modification was not significant in any model.

Conclusions: Based on the ROC curve, PSI can accurately predict Unsustainable heat stress exposures (AUC 0.79). HR alone has a similar capacity to distinguish Unsustainable exposures (AUC 0.78) under relatively constant exposure (metabolic rate and environment) for an hour or so. Screening limits with high sensitivity, however, have low thresholds. This limits the utility of these heat strain metrics. To the extent that the observed strain is low, there is good evidence that the exposure is Sustainable.