Non-linear relationships and interactions of meteorological factors on mumps in Jinan, China.

Although vaccination is available, mumps remains a public health concern in many countries including China. Previous studies have indicated the impact of meteorological factors and mumps, but findings vary across different regions with limited evidence to inform local public health responses. We aim to examine the impacts of meteorological variables on mumps in Jinan, a temperate city of China, and explore the interactions of temperature with humidity or wind speed. Weekly meteorological data and notified cases of mumps in Jinan were collected for 2014-2018. Regression analyses using the generalized additive model were performed with considerations of multicollinearity, lag effects, school holidays, long-term trend, and seasonality. A stratification model was applied to investigate the interaction. We found a non-linear relationship between weekly mean temperature and the number of cases. Between 1.2 and 24.5 °C, the excess risk (ER) of mumps for a 1 °C increase in weekly mean temperature was 3.08% (95% CI 1.32 to 4.87%) at 0-week lag. The lagged effects could last for 3 weeks. There were interactions between mean temperature and relative humidity or wind speed. The effect of mean temperature was enhanced in days with low relative humidity or high wind speed. This study suggests that temperature is positively associated with mumps cases with thresholds in the temperate city of China, and the effect can be modified by relative humidity and wind speed and is independent of vaccine coverage. Findings could be integrated into current early warning systems of mumps in order to protect people's health from the risk of changing climate.