Descomponiendo vulnerabilidades múltiples frente al calor extremo: una aproximación demográfica en México 

Climate change and population aging are simultaneously reshaping patterns of vulnerability worldwide. Older adults and people with disabilities are particularly susceptible to extreme heat due to physiological sensitivity, functional limitations, and reduced adaptive capacity. Yet, territorial differences in heat-related vulnerability are not determined by climate alone; they also reflect demographic composition and underlying health conditions. Existing studies have largely relied on regression models or composite indices, which are useful for identifying associations but provide limited insight into the relative contribution of different mechanisms. This study proposes a demographic approach to disentangling these sources of vulnerability and applies it to Mexico, a country characterized by rapid population aging, marked regional inequalities, and substantial climatic heterogeneity.  

We conceptualize exposure to extreme heat among people with severe disabilities as the result of the interaction of three components: population age structure, disability prevalence, and heat exposure. Demographic information comes from the 2020 Mexican Population and Housing Census, using the Washington Group framework to identify severe disability. Daily climatic conditions are measured through the Universal Thermal Climate Index (UTCI), derived from ERA5-HEAT data and aggregated to municipalities. Extreme heat is defined as days in 2020 exceeding the municipality-specific 95th percentile of the 2005–2019 distribution. These demographic and climatic components are combined into a synthetic exposure indicator representing the proportion of the population-year burden corresponding to individuals with severe disabilities under extreme heat conditions.  

Methodologically, the study extends demographic decomposition techniques to environmental vulnerability. We first illustrate the limitations of the classical Kitagawa decomposition when more than two interacting dimensions are involved and then apply the symmetric decomposition method proposed by Das Gupta. This approach attributes differences in exposure between each Mexican state and the national average to three additive components (age structure, disability prevalence, and heat exposure) while avoiding path dependence and preserving interpretability. Sensitivity analyses comparing municipal and state-level decompositions show strong agreement, supporting the robustness of state-level results.  

Results reveal substantial heterogeneity across Mexican states and contrasting vulnerability profiles. Age structure increases exposure in older populations, particularly in Mexico City, whereas younger age structures exert a protective effect in several states. Disability prevalence emerges as a major driver of elevated exposure in southern states, reflecting broader social and health inequalities. Climatic conditions also contribute significantly to territorial disparities, with states such as Morelos and San Luis Potosí experiencing markedly higher exposure due to extreme thermal conditions in 2020. Case studies illustrate distinct pathways to vulnerability: Mexico City combines population aging with favorable disability conditions, Oaxaca exhibits high disability-related vulnerability, and Nuevo León shows comparatively lower exposure across all components.  

By extending demographic decomposition methods to the analysis of climate-related risks, this study provides a transparent and interpretable framework for understanding multiple vulnerabilities. The approach highlights how demographic, health, and environmental factors interact to produce territorial inequalities and offers a flexible tool that can be applied to other contexts and environmental hazards. It also identifies regions where aging, disability, and heat converge, providing evidence relevant for health policy, care systems, and climate adaptation strategies.