Association between hydrometeorological conditions and infectious diarrhea in mainland China: a spatiotemporal modeling study

Abstract

Infectious diarrhea imposes substantial disease burden in China, with local climate identified as a key risk factor. In this study, we aimed to explore the association between hydrometeorological conditions and the incidence of infectious diarrhea, and investigate the modification effect of urbanization, which remained unclear. Monthly data were collected from Jan 1, 2021 to Dec 31, 2022 from each city of mainland China on cases of infectious diarrhea other than cholera, dysentery, typhoid and paratyphoid (i.e., other infectious diarrhea, OID), temperature and precipitation. We used Palmer drought severity index (PDSI) to represent local hydrometeorological conditions. A spatiotemporal Bayesian hierarchical model combined with distributed lag nonlinear model was specified to explore the association between PDSI and the relative risk of OID. The effect size of hydrometeorological conditions at different urbanization levels were quantified through a linear interaction term. A total of 1,999,947 OID cases were included. There was a N-shaped cumulative association between PDSI and OID incidence over a maximum lag of 4 months. The effect of extreme dry conditions persisted over lag 1-3 months and peaked at the 2-month lag (RR=2.67, 95%CI: 2.44-2.91). By contrast, the relative risk of OID increased immediately under extreme wet conditions (RR=1.97, 95%CI: 1.68-2.32 at lag 0 month). Children and the older population were particularly susceptible to extreme dry and wet conditions, respectively. Areas with lower urbanization rate, healthcare resources and economic levels, and higher population density showed a higher risk of OID related to extreme dry conditions. While the modification effects of these urbanization characteristics were limited for extreme wet conditions. There was a nonlinear and lagged association between hydrometeorological conditions and the incidence of OID in mainland China, which may be modified in diverse patterns by urbanization indicators. Our findings will provide valuable reference for better prevention and control of OID.