Weather extremes, natural disasters, and health outcomes among vulnerable older adults: New improvements on exposure assessment, disparity identification, and risk communication strategies - In recent decades, there have been more frequent extreme weather-related disasters (EWRDs), such as heatwaves, floods/hurricanes, power outages (POs), and recently endless Tornadoes. The impact of EWRDs/POs on human health has become a top public health priority. Research suggests that older adults, especially those with low socioeconomic status (SES), 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 assess and address these disparities, by considering social and contextual influences on both exposure and health responses to EWRDs/POs. Currently, significant gaps remain in our understanding of how all weather/contextual factors jointly affect health, and if health effects are worse in transitional months or within marginalized communities. Major limitations in exposure assessment capacity, based on limited monitoring sites in each state (particularly in rural areas), are also apparent. In addition, few large studies have attempted to assess how EWRDs-health associations may be modified by community contexts (e.g., PO, 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 EWRDs and POs. Specifically, we propose to 1) Estimate if POs and EWRDs and their impacts (short-/long-term) are stronger in older adults and underserved communities by generating high-resolution gridded weather data (HrGWD); 2) Evaluate if the EWRDs-health associations are modified by POs, greenness, soil, and social factors, and if the adverse effects of EWRDs/POs are greater during the COVID-19 pandemic period; and 3) Develop predictive models, vulnerability/resilience indices and maps, and other sustainable intervention tools using community engaged approaches. HrGWD and weather simulations will be created using 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, other critical care indicators, and mortality in a unique 20-year retrospective cohort in NYS. We will use distributed lag non-linear models and interrupted time-series analysis to evaluate the impacts of emergent PO/EWRDs on the most common and fatal diseases among the aging population. While additive and multiplicative interactions of PO, greenness and community factors on the EWRDs-health association will be evaluated, a predictive model selected from over 200 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 unique/ geocoded datasets, innovative data mining/analysis methods, community-engaged approaches to communicating our findings to vulnerable older adults, and successful prior partnerships with government agencies maximize the feasibility of this project and our probability of success.
