Global projections of heat exposure of older adults

The study investigates how concurrent global population aging and climate warming will interact to shape older adults’ exposure to heat by mid-century. Older adults are biologically more susceptible to heat due to reduced thermoregulation and prevalent comorbidities, and many face socioeconomic disadvantages (e.g., isolation, limited resources, inadequate housing) that hinder adaptation. As the share of people aged 60+ is set to rise to about 21% globally by 2050—with most living in low- and middle-income countries—the overlap with intensifying heat extremes could create vulnerability hotspots. The research aims to quantify current (circa 2020) and projected (circa 2050) chronic and acute heat exposures for the 69+ population worldwide, to identify regional disparities, and to inform health and adaptation planning under multiple socio-climatic scenarios.

Prior work has documented health risks from heat exposure and how they vary by socioeconomic and demographic factors, including age, gender, and race. Studies have examined contributions of climate change, population growth, and age structural change to heat exposure at national, regional, and global scales, often focusing on urban areas and using diverse exposure metrics (e.g., cooling degree days, hot days, unprecedented hot summers) and health outcomes (e.g., heat-related mortality). Exposure-response functions have been used to project morbidity and mortality, but relative risk estimates specifically for older adults remain comparatively rare. While some studies consider SSP-based futures, a comprehensive global assessment of future heat exposure specifically for older adults aligned with SSP scenarios has been lacking. This study fills that gap by combining age-stratified demographic projections with downscaled CMIP6 temperature projections to assess chronic and acute exposures among older adults globally.

Data: The analysis uses daily temperature data from NEX-GDDP-CMIP6 (0.25° grid) for 14 bias-corrected CMIP6 GCMs (excluding hot models), covering a historical baseline (1995–2014) and future projections (2041–2060) under SSP126, SSP245, SSP370, and SSP585. Age-stratified gridded population counts for 2020 at 1-km resolution come from WorldPop and Pezzulo et al. (aggregated into two groups: <69 and 69+). Downscaled total population projections (Gao) and country-level age-stratified projections (KC & Lutz) consistent with SSP narratives are used to derive future gridded age structures.

Age-stratified gridded projections: Starting from 2020 WorldPop grid-cell totals, grid-cell growth factors consistent with downscaled total population projections are computed and then adjusted to match country-level age-structured totals by SSP via country- and age-specific scaling. This yields 2050 gridded fractions for <69 and 69+ (assumes within-country homogeneous age-structure trends).

Heat exposure metrics: Three meteorological indicators are computed per grid cell: (1) Cooling Degree Days (CDD) with threshold 24°C (annual sum of positive daily mean temperature exceedances), representing chronic exposure; (2) TMAX95, the 20-year 95th percentile of daily maximum temperatures, representing acute intensity; and (3) #HD, the 20-year average annual count of days with daily maximum temperature ≥37.5°C, representing acute frequency.

Population exposure metrics: For each grid cell and scenario, exposures for older adults are computed as PDD = A69+ × N × CDD (Population Degree Days), PD95 = A69+ × N × TMAX95 (Population Degrees at 95th percentile), and PHD = A69+ × N × #HD (Population Hot Days). Exposures are aggregated to regions and globally. A decomposition attributes fractional changes in exposure to shifts in age structure (aging), population size (growth), and meteorology (climate change) using weighted sums over grid cells.

Validation and implementation: The age-stratified downscaling approach is benchmarked against existing subnational projections (USA, EU, UK, India, China), showing generally good consistency depending on region and scenario. Analyses are implemented in R (terra, raster, sf, tidyverse). Data and replication code are publicly available.

• Chronic exposure roughly doubles globally for older adults by mid-century across all warming scenarios.
• By 2050, more than 23% of the global 69+ population will live in climates where TMAX95 exceeds 37.5°C, up from 14% in 2020, exposing an additional 177–246 million older adults to dangerous acute heat; most will be in Asia and Africa.
• Under threshold benchmarks (≥30 hot days/year, TMAX95 ≥37.5°C, and ≥1200 CDDs/yr), the additional exposed 69+ population by 2050 is estimated at: 0.16–0.23 billion (#HDs), 0.18–0.25 billion (TMAX95), and 0.23–0.32 billion (CDDs), depending on scenario.
• Population Degree Days (PDDs) for 69+ more than quadruple from about 203 billion (2020) to 778–1008 billion (2050), largely driven by Asia (from ~150 to 585–768 billion).
• Global average hot days rise from ~10 to 19–21 days/year; TMAX95 rises from ~32°C to ~35°C. PD95 (population-weighted acute intensity) increases from ~15 to 35–45 billion, up to a three-fold increase.
• Europe sees the largest relative increase in TMAX95 among regions (from ~28°C to ~31°C; ~11% vs. ~9% global average under high warming), despite smaller absolute numbers than Asia.
• Asia will host the largest absolute number of older adults (69+ reaching ~588–748 million by 2050, up from ~239 million), dominating global exposure totals; Africa, Asia, and South America experience the largest increases in cumulative exposure (e.g., average elderly CDD growth of ~195–323 in Africa and 312–397 in Asia).
• Decomposition indicates climate change dominates exposure increases in temperate higher-income regions (Europe, North America), while population aging and growth dominate in warmer, lower-income regions (Africa, Asia, South America).

The findings demonstrate a substantial rise in both chronic and acute heat exposure among older adults worldwide by mid-century, directly addressing the research aim to quantify and locate future vulnerability hotspots. Increased exposures, particularly in Asia and Africa where adaptive capacity tends to be lower, imply heightened risks of heat-related morbidity and mortality among older people, who face physiological susceptibility and often socioeconomic barriers to adaptation. The results underscore the urgent need to integrate healthy aging considerations into climate adaptation and public health planning. Anticipated increases in hot days and extreme temperature intensities will escalate demand for cooling, with implications for energy systems and equity. Policymakers should consider targeted interventions—e.g., improving building thermal performance and passive/active cooling access, expanding early warning systems and public cooling centers—tailored to older adults’ specific needs and constraints. The work provides spatially resolved evidence to inform prioritization and resource allocation for healthcare and social services in regions where aging populations and heat exposure trends intersect.

This study provides the first comprehensive, SSP-consistent global assessment of mid-century heat exposure among older adults, revealing large increases in both chronic and acute exposures across all scenarios, with the largest absolute burdens in Asia and notable relative increases in Europe. The publicly available gridded datasets and exposure metrics can support risk assessments, mortality studies, and adaptation planning that account for demographic aging. Future research should quantify the costs and effectiveness of adaptation options specifically for older adults, integrate data on adaptive capacity (e.g., air conditioning access, urban green space cooling), and evaluate feedbacks involving mobility, time-use, appliance adoption, energy demand, and power system impacts under differential aging patterns and potential climate-induced migration.

Key limitations include uncertainty in the size and spatial distribution of future elderly populations and the assumption of geographically homogeneous within-country age-structure trends in the downscaling method. The lack of harmonized, high-resolution global age-stratified baseline data poses structural constraints. Exposures may be overestimated because autonomous and planned adaptation (e.g., increased cooling adoption, community interventions) is not fully captured; baseline assumptions about adaptation are uncertain. Additional dynamics such as changes in mobility, outdoor time, technology adoption and use, household energy demand, electric load profiles, and climate-induced migration could alter future exposures and are not comprehensively modeled.