Publication

To examine the associations between ambient temperature and hospitalizations for acute kidney injury (AKI) in Queensland, Australia, 1995-2016. Data were collected on a total of 34 379 hospitalizations for AKI from Queensland between 1 January 1995 and 31 December 2016. Meteorological data were downloaded from the Queensland Government's Department of Environment and Science. We assessed the temperature-AKI relationship using a time-stratified case-crossover design fitted with conditional quasi-Poisson regression model and time-varying distributed lag non-linear model. Stratified analyses were performed by age, sex, climate zone and socioeconomic group. Both cold and hot temperatures were associated with hospitalizations for AKI. There were stronger temperature-AKI associations among women than men. Cold effects were only positive in the > 70 years age group. Hot effects were stronger in the <= 59 years age group than in the >60 years age group. In different climate zone areas, cold effects decreased with increasing local mean temperatures, while hot effects increased. In different socio-economic status groups, hot effects were stronger in the poor areas than the affluent areas. From 1995 to 2016, the magnitude of associations between cold temperature and hospitalizations for AKI decreased, while the hot effect increased. The associations between hot temperature and hospitalizations for AKI become stronger, while the magnitude of cold effect decreased from 1995 to 2016. This trend may accelerate over the coming decades, which warrants further research. More attention is needed toward susceptible population including women, people > 70 years, and the people living in hot climate zones and in low socioeconomic status areas.