Research Letter
October 17, 2024
Effect of Electric Fans on Body Core Temperature in Older Adults Exposed to Extreme Indoor Heat
Fergus K. O’Connor, Robert D. Meade, Katie E. Wagar, et al
JAMA. Published online October 17, 2024.
doi:10.1001/jama.2024.19457
With rising global temperatures, sustainable and affordable cooling interventions are urgently needed for preventing heat-related mortality.1 Based on data in young adults and biophysical modeling,2,3 electric fans have been proposed as an effective cooling intervention for older adults (≥65 years) in temperatures as high as 38 °C.1 However, a recent reanalysis of these data suggested that fans provide minimal body cooling above 33 °C, especially for older adults due to age-related reductions in sweating.4 The current study evaluated whether electric fans limit core temperature increases in older adults exposed to conditions similar to those experienced in homes during deadly heat waves in North America.5
Methods
Following approval by the University of Ottawa Research Ethics Board, adults aged 65 to 85 years from Ottawa, Ontario, Canada, volunteered and provided written informed consent for the study, which was performed from December 2022 to March 2023, with final participant follow-up on April 10, 2023. Supplement 1 details the full protocol. This study followed the EQUATOR guidelines for the reporting of randomized trials (CONSORT).
On separate days, participants completed 3 randomized, 8-hour exposures to 36 °C and 45% relative humidity (separated by ≥6 days) seated in front of a fan delivering air speeds of 0 m/s (no fan control), 2 m/s (reflecting normal air speed provided by electric fans), or 4 m/s (reflecting air speed used in recent biophysical modeling).2,4 At the end of hours 1, 3, 5, and 7, participants performed 10 minutes of light stepping, simulating activities of daily living (~2.25 metabolic equivalents). Drinking water (~15-20 °C) was available ad libitum.
The primary outcome was peak core (rectal) temperature during the 8-hour exposure. A reduction of 0.2 °C or more was deemed the minimal clinically meaningful effect. A sample of 18 participants was required to detect this effect with more than 84% power. Of 15 prespecified secondary outcomes, 12 are reported here: core temperature and heart rate (3-lead electrocardiogram) area under the curve (index of cumulative strain) and end-exposure values for core temperature, heart rate, systolic and diastolic blood pressures (oscillometry), rate pressure product (heart rate × systolic pressure), fluid consumption, net fluid loss (% body mass change), percentage plasma volume change (estimated from hemoglobin and hematocrit), and thermal discomfort (anchors at −50 mm [extremely comfortable], 0 mm [neutral], and 50 mm [extremely uncomfortable]). Data were analyzed using linear mixed-effects models (baseline-adjusted; Bonferroni-corrected, 2-tailed α = .05) with R version 4.2.0 (R Foundation). The analyst (R.D.M.) was blinded to trial conditions.
Results
Of 23 individuals assessed for eligibility, 18 participated (8 females; median age, 72 years [IQR, 67-76]; Table 1). The mean (SD) peak core temperature was 38.3 (0.3) °C in control, 38.3 (0.2) °C with a fan at 2 m/s, and 38.3 (0.3) °C with a fan at 4 m/s. There were no significant differences in peak core temperature between conditions (fan at 2 m/s – control: −0.1 °C [95% CI, −0.2 °C to 0.0 °C]; fan at 4 m/s – control: 0 °C [95% CI, −0.1 to 0.1]; and fan at 4 m/s – 2 m/s: 0 °C [95% CI, −0.1 to 0.2]; all comparisons P ≥ .50; Table 2).
With a fan at 2 m/s compared with control, end-exposure, but not area under the curve, core temperatures (38.0 °C vs 38.1 °C) and heart rates (76 beats/min vs 81 beats/min) and end-exposure thermal discomfort (−1 mm vs −6 mm) were significantly lower (Table 2). Thermal discomfort was lower with a fan at 4 m/s compared with control (−9 mm vs −6 mm). There were no significant differences in blood pressure, rate pressure product, fluid consumption or loss, or plasma volume change between either fan condition and control, nor any outcomes between the 2 fan conditions.
Discussion
Electric fan use did not lower peak core temperature in older adults exposed to extreme indoor heat. Reductions in end-exposure core temperature and heart rate were observed, but they were small and of questionable clinical importance. Neither exceeded previous suggestions for clinical significance (≥0.3 °C and 5 beats/min, respectively).6 Consistent with recent modeling,4 these data do not support fans as an efficacious standalone cooling intervention for older adults in hot indoor environments (>33-35 °C). Limitations of this study included the small, homogenous sample; specific environmental conditions; and the controlled experimental setting.
