Background: Climate change disproportionately affects rural populations, yet Australian climate-health research focuses overwhelmingly on metropolitan cities. Whether rural and metropolitan areas experience climate-health impacts through different mechanistic pathways is unknown. Methods: We conducted an ecological time-series study linking daily climate data from the SILO database (2010-2023) to hospitalisation data across 15 Local Health Districts in New South Wales: annual diagnosis-specific rates (9 conditions) and quarterly total and mental health episodes (64 quarters). Rural-metropolitan stratified XGBoost models using climate-only features identified divergent predictors. Black Summer bushfire impacts were quantified using interrupted time series and cumulative Forest Fire Danger Index dose-response models. Sensitivity analyses included leave-one-out cross-validation, placebo tests, outlier exclusion, and positive/negative disease controls. Findings: Climate-only models explained substantially more variance in rural hospitalisation rates (R²=0.54-0.63 for chronic kidney disease, mental health, and potentially preventable hospitalisations) than in metropolitan areas (R² near zero). Rainfall and humidity dominated rural predictions (permutation importance 0.56 and 0.33); maximum temperature dominated metropolitan predictions. During Black Summer, interrupted time series identified +1,623 excess episodes per fire-affected district per quarter (p<0.001) and +61.7 excess mental health episodes (p<0.001), with mental health effects delayed by one quarter (p=0.029). Condition-specific dose-response showed positive associations for pneumonia (p=0.035), COPD (p=0.060), and CHD (p=0.093); diabetes was correctly null (p=0.65). The temperature-mental health relationship was significantly strengthening over time (p<0.001). Interpretation: Rural and metropolitan populations experience climate-health impacts through fundamentally different pathways. Rural communities are primarily affected by drought and humidity deficits — variables absent from current heatwave-focused response frameworks. Drought-health early warning systems for rural Australia are needed.
Building similarity graph...
Analyzing shared references across papers
Loading...
Hayden Farquhar (Mon,) studied this question.
www.synapsesocial.com/papers/69df2c2fe4eeef8a2a6b1355 — DOI: https://doi.org/10.5281/zenodo.19553046
Hayden Farquhar
Building similarity graph...
Analyzing shared references across papers
Loading...