Establishment and validation of health vulnerability and adaptation indices under extreme weather events on the basis of the 2016 flood in Anhui province, China

Abstract

Climate change could intensify extreme weather events, such as flooding, which amplifies the public health threat of waterborne diseases. Thus, assessing health vulnerability and adaptation (V&A) could facilitate nationally effective responses to extreme climate events. However, related studies are still negligible, and the assessment urgently needs to be validated with actual health data after extreme weather events. We established health V&A indices through literature review and factor analyses. Then, we separately mapped the spatial distribution of flood exposure, social and public health sensitivity, and adaptive capacity before the 2016 flood in Anhui province and compared it with post-flood diarrhea risks by using the geographic information system method to assess health vulnerability. Finally, we validated the indices by exploring the relationship between health V&A indices and post-flood diarrhea risks by using the quantile regression model. Results revealed that health V&A can be framed and categorized as key components of exposure, sensitivity, and adaptive capacity. The remarkable differences in the spatial distribution of health vulnerability were generally consistent with the demographic sensitivity, geographic flooding exposure, and post-flood diarrhea risks in Anhui. In addition, health V&A indices exerted significant positive impacts on infectious diarrhea post-flooding at all quantiles and were significant across different percentiles. Moreover, the impacts of flood exposure on total infectious diarrhea were high and continuous, whereas the impacts of sensitivity were not obvious in the flood's early stage (Coeff = 0.643; p < 0.001) but high in the flood's middle (Coeff = 0.997; p < 0.001) and late stages (Coeff = 0.975; p < 0.001). However, the impacts of adaptive capacity were heterogeneous and high in the flood's early stage (Coeff = 0.665; p < 0.001 at the 25th percentile) and late stage (Coeff = 1.296; p < 0.001 at the 75th percentile) but were insignificant at the 50th percentile. This study contributed validated three-layered health V&A indices with 30 indicators and identified that the impacts of the key components on post-flood waterborne-disease risks are heterogeneous. For instance, local public health sensitivity and adaptive capacity are insufficient to reduce these risks in the long run. This study could be used to project population health risks after extreme weather events and thereby contributes to local government planning of health adaptation.