PROJECT SUMMARY / ABSTRACT
In recent decades, climate change has contributed to more frequent and extreme weather-related disasters
(EWRD), such as heatwaves, floods, hurricanes, storms and power outage (PO). The impact of these EWRDs
on human health has become a top public health priority. Research suggests that older adults, especially those
with low socioeconomic status (SES) and minority populations, are disproportionately vulnerable to disaster
hazards due to lack of access to the necessary resources for hazard mitigation or adaptation. What is now
needed is a much more comprehensive way to effectively address these disparities, by considering social and
contextual influences on both exposure and health responses to EWRDs. Currently, significant gaps remain in
our understanding of how all meteorological factors jointly affect health, and how health effects may differ during
transitional seasons. Major limitations on exposure assessment capacity, based on existing limited monitoring
sites in each state (particularly in rural areas), are also apparent. In addition, few large studies have attempted
to assess how the EWRDs-health may be modified by community and social contexts (e.g., greenness) in ways
that produce health disparities. To fill these gaps, the proposed study will test a central hypothesis that
vulnerable aging populations are particularly susceptible to the adverse health effects of extreme weather or
EWRDs. Specifically, we propose to: 1) Improve exposure assessment by generating high-resolution gridded
weather data; 2) Evaluate joint effects of multiple weather factors and disasters on cardio-respiratory diseases,
Alzheimer/dementia, injuries, and renal diseases in vulnerable older adults, as well as the modifying effects of
regional greenness and pandemic; and 3) Assess the impact of multiple community contextual factors in affecting
health during EWRDs by developing predictive models and vulnerability/resilience indices. Results will
serve as the basis for the development of effective communication strategies. HrGWD and weather simulations
will be created using a state-of-the-art, two-stage downscaling models based on unique Mesonet data. In addition
to utilizing NYS hospitalization and ED data, we will retrospectively follow-up readmission and other critical care
indicators in a unique 18-year dynamic cohort in NYS, while also evaluating US COVID-19 infection/death
rates after major EWRDs. We will use distributed lag non-linear models and interrupted time-series analysis to
evaluate the impacts of emergent EWRDs on the most common and fatal diseases among the aging population.
While causal influence analysis will be used to estimate the mediation effects from greenness and community
factors, a predictive model selected from over 300 factors at the community level will be developed to identify
vulnerability/resilience factors using machine-learning algorithms. Our multi-disciplinary and experienced
research team, access to numerous geocoded datasets, innovative data mining/analysis methods, culturally
appropriate communication materials planned for vulnerable older adults, and successful prior partnerships
with government agencies maximize the feasibility of this project and our probability of success.
