If you’re interested in contributing a short What Am I Reading post, we’d love to hear from you! Email us at cache@colorado.edu. 

Written by Kelly Perry and Jenna Merenstein

When people think about brain health and aging, they might first think about the role of genetics or lifestyle factors. Both factors are important, though a growing body of research shows something else is also playing a major role in shaping how our brains age: the physical, social, and structural exposures we experience across our lives—from before birth through older age. Scientists refer to this as the exposome, and understanding it is key to addressing cognitive decline in a changing climate (Li et al 2025). In our recent Alzheimer’s & Dementia Perspective, we argued that understanding both healthy aging and Alzheimer’s disease and related dementias (ADRDs) requires an exposome- and equity-centered lens that considers the environmental and social conditions (including systemic inequities and systems-level resilience strategies) that shape brain health. We include the framework introduced in our paper in Figure 1. 

Figure 1. An equity- and justice-centered framework for linking the exposome and neurocognitive health across the life course (from Perry KE & Merenstein J, 2026) 

A recent Nature Medicine paper by Legaz et al. titled “The Exposome of Brain Aging across 34 Countries,” provides a compelling example of how this framework can be applied in practice. The authors examined 73 physical and social exposomal factors (e.g., air pollution, temperature, greenspace, gender equality, democracy, and access to drinking water) and their relation to neuroimaging measures of brain structure and brain function for 18,701 participants that varied in cognitive status (cognitively normal, mild cognitive impairment, Alzheimer’s disease, and frontotemporal lobar degeneration). Participants were recruited from 34 countries across Latin America, North America, Europe, Asia, Africa, and Oceania, thereby helping address the systemic underrepresentation of low- and middle-income countries (LMICs) in neuroimaging research.  

Legaz et al. demonstrated that aggregated exposome models explained significantly more variance in brain aging than any single exposure alone—up to 15.5 times more than individual factors. Furthermore, their results highlight the impact that structural inequities embedded within neighborhoods, institutions, and political systems have on brain aging: physical exposures (e.g., higher air pollution, reduced greenspace, and extreme temperatures) were strongly associated with accelerated structural brain aging. In contrast, social exposures (e.g., poverty, weaker rule of law, and decreased civic participation) were associated with accelerated functional brain aging. The latter was shown to be especially true for female participants in the study, where reduced rights-related factors and poor soil and water quality predicted accelerated aging in females more than males. Their findings underscore a central argument of our Perspective: that environmental neuroscience needs to overcome the siloed approaches that are traditional hallmarks of the discipline, e.g., focusing on a single exposure or a single high-income country cohort. 

Tools such as the Area Deprivation Index (ADI) and the Neighborhood Atlas are helpful in elucidating these structural, systemic inequities that adversely impact brain aging and increase the risk of ADRDs. While Legaz et al. used country-level indicators, neighborhood-level metrics such as ADI can provide finer resolution for studying how disadvantage shapes brain health within countries and across communities. ADI captures social dimensions such as income, education, employment, and housing quality, allowing researchers to better characterize cumulative disadvantage at the local scale. Similarly, the Neighborhood Atlas (which hosts the ADI) provides standardized geospatial measures of neighborhood deprivation that can be linked to neuroimaging, cognition, and dementia outcomes (Hunt et al 2020; Kim et al 2024). These tools help operationalize the “social exposome” by connecting place-based inequities to measurable differences in brain structure and function. 

Building better data in this field also means improving access to neuroimaging itself. As we discussed in our Perspective, one major barrier to exposome- and equity-informed neuroscience is that advanced MRI remains inaccessible in many under-resourced and rural settings. Emerging low-field MRI initiatives, including recent work at Cardiff University, are helping address this gap by developing lower-cost, portable systems capable of whole-brain imaging. These efforts are particularly important for LMICs, where the burden of environmental exposures is often highest but access to imaging infrastructure is lowest. Legaz et al.’s multi-country study underscores why this matters: without broader imaging access, we risk building evidence only from the least exposed populations.  

Protecting brain health equitably requires shifting toward prevention and systems change. Policies that promote cleaner air and water, safer housing, and equitable urban design can reduce harmful exposures across the life course. Investments in expanding access to green spaces, pollution control, and community infrastructure benefit physical health and cognitive resilience. Legaz et al.’s study provides strong empirical support for the notion that environmental and social conditions fundamentally shape brain aging outcomes. As climate change intensifies these exposures globally, building inclusive datasets (e.g., The Neighborhood Atlas), improving access to neuroimaging (e.g., investing in low-field scanners), and centering health equity in brain health research and policy will be essential for ensuring healthy cognitive aging for communities worldwide. 

If you’re interested in contributing a short What Am I Reading post, we’d love to hear from you! Email us at cache@colorado.edu.

Written by Jenna Tipaldo and Deborah Balk, CUNY Institute for Demographic Research

Researchers interested in studying the impacts of temperature on health face decisions on how to operationalize and measure exposures. This webinar “CAFE University: Considerations for temperature in public health studies” by Lauren Mock and Shreya Nalluri explores the myriad choices required when using temperature data to study health outcomes, from choice of metric and dataset to the study design.  

The webinar includes an overview of various temperature metrics including land surface temperature (LST), air temperature (AT), heat index, wet bulb globe temperature (WGBT), and universal thermal comfort index (UTCI). A fuller discussion  about choice of metrics can be found in this resource from the World Resources Institute (Engel et al. 2025), also referenced in the webinar. (It is useful to note that these are all measures of outdoor temperature. Measurement of indoor vs outdoor exposure is explored in another post on air pollution). 

Each metric is more suited for specific purposes and contexts than for others, and each has drawbacks. For example, air temperature is widely available but doesn’t include factors that influence how humans perceive temperature, such as humidity. Both WGBT and UTCI do consider humidity.  

Other dataset considerations include the availability of multiple metrics such as daily maximums, minimums, or averages. (All of the metrics discussed in the webinar provide daily temperature resolution, but some sensors have more fine-grain temporal data, such as hourly. See Table below.)  

Additionally, researchers may choose to use raw temperature values versus standardized values such as percentiles, which can account for acclimatization.  

Source: CAFE University 

Another aspect of operationalizing exposure is considering the timing of exposure relative to an outcome of interest – the webinar gives an example of defining the exposure period as the day of, day prior to, or three days prior to an event.  

Even so, a recent paper (Cruz et al., 2025) proposes a framing of heat as a chronic phenomenon and argues that chronic exposures pose different risks that are not captured by existing research that evaluates the health impacts of acute exposures to extreme heat. 

The webinar also explores the use of several gridded temperature datasets, including ERA5, gridMET, and PRISM, and how to link these with health data (e.g., calculating zonal statistics and population-weighting).  

The webinar provides strengths, limitations, and use cases for each dataset, which vary in their spatial and temporal resolution, as well as the measurements that can be calculated using each dataset. For example, gridMET and PRISM have high-resolution daily averages but do not have the additional measurement components to calculate UTCI or WBGT. ERA5 has higher resolution temporally with hourly data but is coarser spatially and provides global coverage with components to calculate more metrics, including UTCI and WBGT. Though, the ERA5-Land data is not well-suited for coastal studies.  

In general, gridded data are also limited by the data underlying them, a point that was touched on but not explored in depth in the webinar.  It was noted that “gridded products inherit uncertainty from models and stations,” and these limitations are important to keep in mind when choosing a dataset and measures.   

Regardless of which dataset is used, another consideration is how the resolution of the temperature data compares to the resolution of the health or socioeconomic data or outcome of interest. Using spatial resolution as an example, when a researcher wants to integrate temperature data to an administrative unit (say, county or municipality) associated with health records or survey respondents’ residence, they may care whether the temperature grid data in question is more or less coarse than those administrative units. At mismatched spatial scales, there may be implications for errors in measurement or attribution when averaging temperature into the administrative units. As an example, the figure below shows gridMET and ERA5 data for the same day (July 31, 2025) in the Philadelphia, PA and surrounding NJ area.  Averaging across finer resolution data would likely be more suitable for finer-scale (smaller) units such as administrative units that represent urban areas and seen in the center of the figure below.  

Finer scale data may also be more suitable when within administrative unit variation in temperature is high, or the analyst wants to capture that variation. For coarser administrative units, if additional information such as where population is concentrated is not available, use of the coarser temperature data may be as suitable as finer-grained data, noting of course that any sub-unit variation over these administrative areas is simply unobserved (Porter & Howell, 2016) or was reallocated with ancillary data (Zoraghein and Leyk, 2018, Uhl et al. 2018).  

In the context of gridded population data, Leyk et al. (2019) provide a review of products available and a guide for understanding underlying data used in each gridded product, including the population data (spatial coarseness and how change over time is handled), use of ancillary data (such as settlements, roads or land cover), and methods used to allocate population within a defined grid “cell size”. All of these impact the suitability of use in particular applications. Fitness-for-use guidelines for these global population gridded dataset may depend on spatial and temporal resolution, scale and setting (rural vs urban), and mechanism of interest as noted in Leyk et al. (2019) and the same principles can act as a lens when evaluating which climate data set is most suitable for a given analysis.  

Materials from a CACHE demonstration project “Heat, Disability in older adults and Care” from El Colegio de Mexico provides code and guidance for calculating UTCI using ERA5 data, producing municipality-level estimates of number of severe heat days:   https://agingclimatehealth.org/severe-heat-days-using-the-universal-thermal-comfort-index/  

References 

Cruz, M., Mach, K. J., Turek-Hankins, L. L., Ashad-Bishop, K. C., Bailey, Z. D., Evans, S. D., Fanning, A., Fernandez-Burgos, M., Gilbert, J., Howard, B., Mahabir, M., Marturano, J., Murphy, L. N., Muse, N., Pérodin, J., & Clement, A. C. (2025). Where heat does not come in waves: A framework for understanding and managing chronic heat. Environmental Research: Climate, 4(2), 023002. https://doi.org/10.1088/2752-5295/adc827 

Engel, R. E. Mackres, M. Palmieri and E. Anzilotti (2025). Beyond the Thermometer: 5 Heat Metrics That Drive Better Decision-Making, World Resources Institute Insights March 17, 2025 https://www.wri.org/insights/beyond-thermometer-measuring-heat  

Leyk, S., Gaughan, A. E., Adamo, S. B., De Sherbinin, A., Balk, D., Freire, S., Rose, A., Stevens, F. R., Blankespoor, B., Frye, C., Comenetz, J., Sorichetta, A., MacManus, K., Pistolesi, L., Levy, M., Tatem, A. J., & Pesaresi, M. (2019). The spatial allocation of population: A review of large-scale gridded population data products and their fitness for use. Earth System Science Data, 11(3), 1385–1409. https://doi.org/10.5194/essd-11-1385-2019 

Porter, J. R., & Howell, F. M. (2016). A spatial decomposition of county population growth in the United States: Population redistribution in the rural-to-urban continuum, 1980–2010. In Recapturing space: New middle-range theory in spatial demography (pp. 175-198). Cham: Springer International Publishing. 

Uhl, J. H., Zoraghein, H., Leyk, S., Balk, D., Corbane, C., Syrris, V., & Florczyk, A. J. (2020). Exposing the urban continuum: implications and cross-comparison from an interdisciplinary perspective. International Journal of Digital Earth, 13(1), 22–44. https://doi.org/10.1080/17538947.2018.1550120  

Zoraghein, H., & Leyk, S. (2018). Enhancing areal interpolation frameworks through dasymetric refinement to create consistent population estimates across censuses. International Journal of Geographical Information Science, 32(10), 1948–1976. https://doi.org/10.1080/13658816.2018.1472267  

If you’re interested in contributing a short What Am I Reading post, we’d love to hear from you! Email us at cache@colorado.edu.

Written by Kathryn Foster, Cornell University 

As sea-level rise worsens the impacts of hurricanes and storm surges, it also leads to more frequent tidal and seasonal floods in coastal areas, commonly referred to as sunny-day, blue sky, or nuisance flooding. The annual number of days with sunny-day flooding has more than doubled since 2000, with projections that it will triple by 2050, averaging 45-85 days a year nationwide (NOAA, 2025). Current research, such as Mueller et al. (2024), is beginning to document the many impacts of sunny-day flooding and other types of flooding on residential health.  

Mortality: Research shows that in Florida, a 20-mm (0.79 in.) increase in tidal flooding depth raises mortality rates by 0.46% to 0.60% among adults 65 and older. Sea-level rise could contribute to an additional 130 elderly deaths annually in Florida relative to 2019 (Mueller et al. 2024).  

Similarly, longer-lasting floods have greater impacts on mortality than other types of floods, such as flash floods (Lynch et al., 2025). An increase in frequency and severity of seasonal tidal flooding could increase the risk of mortality by blocking roads to medical services, such as regular doctor’s appointments, pharmacies, and hospitals. Other research supports these conclusions, finding that with anticipated intensified flooding, elderly populations will become more vulnerable to morbidity and mortality risks, largely due to mobility constraints among aging adults (Paavola, 2017; Hu et al., 2018; Sheahan et al., 2025). The mortality effects of flooding primarily affect those who require at least 8.85 minutes to reach the nearest hospital (Mueller et al. 2024).  

Infectious Diseases: Seasonal tidal flooding may also increase the incidence of waterborne and infectious diseases in the United States. As flooding increases, communities are exposed to standing water around their homes for longer periods. Seasonal floods are strongly associated with increased hospitalizations for Legionnaire’s disease, a type of pneumonia caused by Legionella bacteria that grows in warm, moist climates (Lynch and Shaman 2022). Enteric infections may also arise from drinking water contamination and sewerage disruption (Carr et al., 2024; Wright et al., 2018), and mosquito-borne diseases and infections may occur from exposure to polluted flood waters (Wright et al., 2018).  

Previous research on coastal storms finds that these events are associated with the spread of certain infectious diseases, such as E. Coli, Legionnaires’, Cryptosporidiosis, Paratyphoid fever, and Dengue in certain areas (Lynch and Shaman 2023; Zheng et al., 2017). These studies similarly posit that the spread of infectious disease will increase alongside predicted increases in coastal storms due to climate change. Although not yet explicitly studied, researchers hypothesize that the spread of infectious diseases associated with coastal storms will also be linked to seasonal tidal flooding (Lynch and Shaman 2022). This provides ground for future research to examine how increasing tidal floods relate to the rise in infectious diseases, like that of coastal storms.   

Other Impacts: Furthermore, other hazards, such as snakebites and wound infections, may occur with tidal flooding and prolonged inundation of residential areas (Wright et al., 2018). Flooding increases residents’ exposure to venomous snakes, as snakes try to enter homes to find dry land or as water snakes are found in inundated streets, leading to a potential health hazard if bitten (Ochoa et al., 2018). Furthermore, as residents document having to walk through flooded streets during seasonal tidal flooding, wounds may become infected by the floodwater, or unseen hazards in the water may cause wounds (Wright et al., 2018). While these hazards are beginning to be documented, little research has examined their relationship to tidal flooding; future research could fill this gap by examining the association between reported snake bites and wound infections and tidal flooding.  

From Research to Policy: These studies highlight that the costs of climate change go beyond heat impacts, emphasizing that the increased frequency and severity of seasonal tidal flooding may directly affect residents’ health. Mueller and colleagues (2024) suggest that communities design programs to improve transportation options for the elderly during the sunny-day flooding season. Furthermore, other researchers suggest that public health initiatives should inform clinicians and residents alike about the health risks of seasonal flooding (Lynch and Shaman, 2022). Alongside public-health initiatives, more effective and equitable preparations for flood risk should be put in place to better mitigate seasonal flooding-related health risks, especially for elderly adults in these communities (Lynch et al., 2025). Such initiatives could include increasing flood-risk awareness by hosting community classes on flood preparation, stockpiling essential medical supplies (Yodsubana and Nuntaboot, 2021), coordinating resource and information sharing with government agencies, healthcare providers, and community groups for elderly residents (Madani Hosseini et al., 2024), and constructing seawalls and elevating roads to prevent road closures (Mueller et al., 2024).   

References:

Hu, P., Zhang, Q., Shi, P., Chen, B., & Fang, J. (2018). Flood-induced mortality across the globe: Spatiotemporal pattern and influencing factors. Science of the Total Environment, 643, 171-182. 

Lynch, V. D., & Shaman, J. (2022). The effect of seasonal and extreme floods on hospitalizations for Legionnaires’ disease in the United States, 2000–2011. BMC Infectious Diseases, 22(1), 550. 

Lynch, V. D., & Shaman, J. (2023). Waterborne infectious diseases associated with exposure to tropical cyclonic storms, United States, 1996–2018. Emerging infectious diseases, 29(8), 1548. 

Lynch, V. D., Sullivan, J. A., Flores, A. B., Xie, X., Aggarwal, S., Nethery, R. C., … & Parks, R. M. (2025). Large floods drive changes in cause-specific mortality in the United States. Nature Medicine, 31(2), 663-671.  

Madani Hosseini, M., Zargoush, M., & Ghazalbash, S. (2024). Climate crisis risks to elderly health: strategies for effective promotion and response. Health Promotion International, 39(2), daae031. 

Mahmoudi, S., Moftakhari, H., Muñoz, D. F., Radfar, S., Sweet, W., & Moradkhani, H. (2025). Escalating high tide flooding along the Atlantic and Gulf Coast of the United States due to sea level rise. Earth’s Future, 13(9), e2024EF005328. 

Mueller, V., Hauer, M., & Sheriff, G. (2024). Sunny-day flooding and mortality risk in coastal Florida. Demography, 61(1), 209-230. 

NOAA Office for Coastal Management High tide flooding.. (n.d.). https://coast.noaa.gov/states/fast-facts/recurrent-tidal-flooding.html#:~:text=Here%20are%20some%20statistics%20on%20high%20tide,inches%20(21%20to%2024%20centimeters)%20since%201880  

Paavola, J. (2017). Health impacts of climate change and health and social inequalities in the UK. Environmental Health, 16 (Suppl 1), 113. 

Wright, L. D., D’Elia, C. F., & Nichols, C. R. (2018). Impacts of coastal waters and flooding on human health. In Tomorrow’s Coasts: Complex and Impermanent (pp. 151-166). Cham: Springer International Publishing. 

Yodsuban, P., & Nuntaboot, K. (2021). Community-based flood disaster management for older adults in southern of Thailand: A qualitative study. International journal of nursing sciences, 8(4), 409-417. 

Zheng, J., Han, W., Jiang, B., Ma, W., & Zhang, Y. (2017). Infectious diseases and tropical cyclones in Southeast China. International journal of environmental research and public health, 14(5), 494.