A novel thermal comfort model for older adults – development and validation of the COMFA-OA model

Abstract

This study presents a population-specific adaptation of the COMFA model, customized to estimate thermal stress levels of older adults. We refer to this index as the COMFA thermal comfort – older adults (COMFA-OA) model. This model incorporates physiological adjustments to enhance accuracy in estimating thermal stress for the older demographic. Key modifications include empirical updates to sweat heat loss, core temperature, and metabolic rate equations to reflect age-related changes in heat regulation and metabolic efficiency. The model was validated using the ASHRAE Global Thermal Comfort Database II and a field experiment assessing thermal sensation of older adults during warm and hot seasons in Texas. Results show COMFA-OA (MAE = 4.03, RMSE = 4.14) outperforms traditional models like PET (MAE = 4.72, RMSE = 4.83), UTCI (MAE = 4.82, RMSE = 4.95), and COMFA (MAE = 4.17, RMSE = 4.29) in accuracy of predicting thermal sensation vote, demonstrating greater multinomial logit model fit (AIC: 4435.3) and achieving competitive computational efficiency (TOPS = 97.54) second only to original COMFA (TOPS = 331.21) and COMFAcourtyard (TOPS = 481.87). Sensitivity analysis identified air temperature, radiant temperature, age, weight, and height as primary contributors to thermal comfort variance in older adults. The COMFA-OA model offers a practical tool for environmental management and public health applications aimed at promoting climate resilience and thermal security for older populations that are vulnerable to heat conditions.