Heat stress vulnerability and critical environmental limits for older adults

Anthropogenic climate change has increased Earth’s average temperature by about 1.1 °C since the late 19th century, while the global population is rapidly aging, with U.S. adults aged 65+ projected to reach ~80 million by 2040. Adults aged 65+ experience disproportionate morbidity and mortality during heat waves. Although laboratory data demonstrate age-related declines in thermoregulatory function (e.g., reduced sweating and skin blood flow, increased cardiac strain), how these physiological changes translate to heat vulnerability in apparently healthy older adults remains unclear. The specific combinations of ambient dry-bulb temperature (Tdb) and humidity at which age disparities in heat balance emerge are not well defined. Critical environmental limits describe combinations of temperature and humidity above which heat balance cannot be maintained for a given metabolic heat load. Prior work from the authors’ laboratory developed a progressive heat stress protocol to identify these limits across environments and activity levels, emphasizing indoor exposures during activities of daily living (ADL). Recently, critical limits were published for healthy young adults during low-intensity activity approximating ADL, providing a best-case baseline for comparison with vulnerable populations. The present study aimed to: (1) document increased heat vulnerability of apparently healthy older adults based on thermal balance across indoor environmental conditions, and (2) establish and compare critical environmental limits for older adults during seated rest and minimal physical activity associated with ADL (MinAct). The authors also compared older adults’ MinAct limits to an updated young adult dataset. They hypothesized that older adults would reach uncompensable heat stress at lower combinations of temperature and humidity than young adults, and that these limits would be lower during MinAct than during Rest in older adults.

Laboratory studies have established biophysical and physiological limits to work in heat, largely in healthy young adults, and at higher metabolic rates relevant to industry and sport. Age-related declines in thermoregulatory function (attenuated sweating and skin blood flow, increased cardiovascular strain) are well documented. The concept of critical environmental limits (combinations of ambient temperature and humidity beyond which heat balance is not possible) has been developed and refined, including progressive heat stress protocols to identify these limits across environments, metabolic rates, and clothing ensembles. However, only one prior study examined critical environmental limits in adults over 60, comparing older unacclimated women with younger heat-acclimated and unacclimated women walking at 30% VO2max—an intensity more relevant to occupational or recreational contexts than typical conditions experienced by older adults during heat waves. The authors’ prior work defined critical limits for young adults at low-intensity activity approximating ADL, and documented core temperature dynamics across compensable vs. uncompensable heat stress. These foundations motivated the present focus on low metabolic rates and indoor settings relevant to older adults during extreme heat events.

Design and participants: Registered on ClinicalTrials.gov (NCT0428439), IRB-approved, and in accordance with the Declaration of Helsinki. One hundred twelve adults (54 young, 58 older) from Centre County, Pennsylvania were screened; 7 were excluded and 5 dropped out. Fifty-one young (23 ± 4 years; 22 men, 29 women) and 49 older adults (71 ± 6 years; 21 men, 28 women) completed testing. Participants were recruited without regard to body size, blood pressure, or blood biochemistry to enhance generalizability, and excluded for contraindications to heat or low-intensity activity, mobility limitations, abnormal ECG (older subjects cleared by a cardiologist), certain gastrointestinal disorders, tobacco or illicit drug use, pregnancy, or medications posing safety risks in heat. Menstrual status and contraceptive use were not controlled in young women. Clothing was standardized (thin short-sleeve cotton t-shirt, sports bra if applicable, shorts, socks, walking/running shoes). Participants, but not investigators, were blinded to the environmental condition. Each subject completed 2–4 experimental trials, totaling 248 trials across 100 participants. VO2max was measured by open-circuit spirometry during a maximal graded treadmill test. Hydration was verified via urine specific gravity ≤1.020 before trials.

Experimental protocol: Trials were conducted at least 72 h apart. Caffeine (12 h), alcohol, and vigorous exercise (24 h) were restricted pre-visit. Testing occurred in an environmental chamber during minimal physical activity (MinAct) for both age groups and during seated rest (Rest) for older adults only. MinAct was standardized as free pedaling on a recumbent cycle ergometer against zero resistance at 40–50 rpm, reflecting the metabolic demand of ADL. Rest trials involved seated rest in identical environments. Two trial types were performed:

• Tcrit trials: Tdb held constant at 34, 36, 38, or 40 °C while ambient water vapor pressure (Pa) increased stepwise by 1 mmHg every 5 min.
• Pcrit trials: Pa held constant at 12 or 16 mmHg while Tdb increased stepwise by 1 °C every 5 min. After a 30-min equilibration, stepwise increases proceeded until a clear, sustained rise in gastrointestinal core temperature (Tgi) from a steady state was observed, indicating the critical environmental limit. Typical trial duration was 90–120 min. Air movement was minimal (air velocity <2 m·s−1), and there was no radiative heat source.

Measurements: Gastrointestinal temperature telemetry capsules were ingested 1–2 h prior to testing; Tgi and heart rate were continuously recorded. VO2 and respiratory exchange ratio were measured at 5 and 60 min. Net metabolic heat production (Mnet) was calculated via partitional calorimetry. METs were determined from VO2 (ml·kg−1·min−1) assuming 3.5 ml·kg−1·min−1 at rest. Sweat rate and percent body mass loss were calculated from nude body mass change pre- to post-trial; no fluid was provided during trials.

Determination of critical limits: The critical Tdb or Pa was identified as the environmental value immediately preceding the Tgi inflection (transition from steady state to sustained increase), determined graphically by segmental line fitting; visual and segmental regression approaches showed excellent agreement (ICC=0.99). For a small number of 34 °C Pcrit trials where no inflection occurred within 120 min, individual isothermal lines were constructed from heat balance equations to estimate inflection points (<5 mmHg above the final completed stage).

Statistics: An a priori power analysis (effect size 1.2 from prior Pcrit data) indicated n=7 per group for power ≥0.8 at α=0.05. Subject characteristics were compared by independent t-tests. Mixed-effects models assessed effects of age, activity level (MinAct vs Rest), and environment on Mnet, VO2, METs, sweat rate, and body mass loss, and evaluated age and metabolic rate effects on psychrometric limits. Compensability curves (analogous to survival curves) were constructed for each condition, and differences tested via Gehan-Breslow-Wilcoxon. Psychrometric loci (mean and 95% CI) were plotted with second-order least squares polynomial fits. Significance was set at p=0.05.

• Sample: 51 young (23 ± 4 yrs) and 49 older (71 ± 6 yrs) adults completed MinAct trials; older adults also completed Rest trials. Groups were similar in height, BMI, body surface area, and body surface area-to-mass ratio (all p ≥ 0.30). VO2max was lower in older adults (28 ± 9 vs 49 ± 12 ml·kg−1·min−1; p < 0.001).
• Metabolic and sweat responses (MinAct): No age or environment effects on Mnet, VO2, or METs during MinAct (all p ≥ 0.06–0.07). Older adults had lower percent body mass loss and sweat rate (main effects: p < 0.0001 and p = 0.0003, respectively) with no environment or interaction effects.
• Rest vs MinAct (older adults): During Rest, Mnet, VO2, METs, sweat rate, and percent body mass loss were lower than during MinAct (all p < 0.01), with no environment or interaction effects.
• Compensability curves: During MinAct, compensability curves were significantly shifted leftward (lower compensable range) for older vs young in all environments (all p < 0.0001), indicating greater heat vulnerability in older adults. For older adults, MinAct vs Rest curves differed significantly across all environments (p ≤ 0.004) except at 34 °C Pcrit (p = 0.88), with Rest allowing a higher compensable range.
• Critical environmental limits (psychrometric loci): • Older vs young during MinAct: Critical limits were lower for older adults across all conditions (p < 0.0001). Examples (mean [95% CI]):
  • At Tdb 38 °C: Pcrit older 23.0 [20.6, 25.5] mmHg vs young 29.7 [27.8, 31.6] mmHg; rh older 46.1% [41.7, 50.6] vs young 59.5% [55.5, 63.6].
  • At Tdb 40 °C: Pcrit older 19.0 [16.3, 21.7] mmHg vs young 28.2 [26.6, 29.7] mmHg; rh older 34.0% [29.2, 38.8] vs young 51.0% [48.2, 53.8].
  • At Pa 16 mmHg: Tcrit older 40.3 [38.4, 42.2] °C vs young 46.4 [45.5, 47.3] °C; rh older 28.6% [25.8, 31.5] vs young 20.7% [19.9, 21.5].
  • At Pa 12 mmHg: Tcrit older 42.9 [40.9, 44.8] °C vs young 49.3 [48.2, 50.4] °C; rh older 18.8% [17.0, 20.5] vs young 13.8% [12.9, 14.7]. • Older MinAct vs Rest: Rest shifted critical loci upward/rightward (higher allowable temperature/humidity) relative to MinAct (p < 0.0001). Examples (older Rest means [95% CI]):
  • At Pa 16 mmHg: Tcrit 44.1 [41.8, 46.4] °C (vs MinAct 40.3 [38.4, 42.2] °C).
  • At Pa 12 mmHg: Tcrit 46.1 [44.1, 48.1] °C (vs MinAct 42.9 [40.9, 44.8] °C).
• Core temperature dynamics: Below critical limits, rates of Tgi change showed no effect of environment or age (Young MinAct 0.08 ± 0.13 °C·hr−1; Older MinAct 0.10 ± 0.12 °C·hr−1; Older Rest 0.07 ± 0.14 °C·hr−1). Above critical limits, rates were substantially higher (Young MinAct 0.70 ± 0.54 °C·hr−1; Older MinAct 0.62 ± 0.48 °C·hr−1; Older Rest 0.69 ± 0.50 °C·hr−1); slopes above were greater than below the inflection (p < 0.0001).

This study directly addresses how aging affects the capacity to maintain heat balance under indoor environmental stress at low metabolic rates relevant to daily life. Compensability analyses demonstrate that, even among apparently healthy individuals, older adults reach uncompensable heat stress at lower combinations of temperature and humidity than young adults during minimal activity, confirming heightened vulnerability. Within older adults, minimizing metabolic heat production (Rest vs MinAct) broadens the range of compensable conditions, underscoring the protective role of reducing activity during extreme heat. Psychrometric loci reveal that in hot-dry conditions the curve is steeper for older adults, consistent with lower maximal sweating capacity leading to earlier limits when evaporative cooling predominates; during Rest, reduced metabolic heat load shifts limits to more permissive environments. These findings integrate physiological mechanisms (age-related declines in sweating and skin blood flow) with biophysical constraints to define actionable environmental thresholds. The results provide an empirical foundation for developing evidence-based heat advisories, indoor environmental guidelines, and triage strategies for older adults during heat events.

The study quantifies heat stress vulnerability in older adults by establishing critical environmental limits for indoor settings during seated rest and minimal activity. Relative to young adults, older adults exhibit lower critical limits during minimal activity, indicating a narrower range of compensable environments. Within older adults, moving from minimal activity to rest shifts limits upward and rightward, expanding the compensable range. Above these critical thresholds, core temperature rises continuously, elevating risk for heat-related illness. These empirically derived limits can inform evidence-based guidance, risk communication, and interventions to protect adults aged 65+ during extreme heat. Future work should examine the influence of comorbidities, medications, acclimatization, regional climate adaptation, and potential sex differences (including menstrual and contraceptive status) on critical limits, and extend findings to outdoor settings with radiative and convective factors.

• Heat acclimatization was not controlled; however, testing spanned all seasons to mitigate seasonal bias.
• Participants were from central Pennsylvania; populations in hotter climates may have higher critical limits.
• Environmental chamber lacked radiative heat sources and had minimal air movement; findings may not generalize to outdoor settings or environments with significant radiation or airflow.
• Older adults were recruited without restrictions on medical history or medications (beyond safety contraindications); effects of comorbidities (e.g., cardiovascular disease, diabetes) and medications (e.g., diuretics, statins) on critical limits were not assessed.
• The study was not powered to detect sex differences; menstrual status and contraceptive use in young women were not controlled.
• Some inflection points at 34 °C Pcrit were estimated using heat balance equations when no inflection occurred within 120 min, though estimates were close to observed stages and within group ranges.