Background: Mumps remains a major global public health problem. Many studies have explored the relationship between meteorological factors and mumps, few have comprehensively explored such associations considering nonlinear relationship, delayed effects and collinearity in order to more accurately estimate them. This study aims to explore the relationship between meteorological factors and mumps in consideration of nonlinearity, delayed effects and collinearity. Methods: We collected daily reported mumps cases and meteorological data for Jining City, Shandong Province, China from 1 January 2007 to 31 December 2016. By building a Boosted regression tree model (BRT) for each day from lag 0 days to the maximum lag time, the optimal lag time was selected and the relationship between meteorological factors and mumps was explored for this lag time. Results: From 2007 to 2016, a total of 15,064 cases of mumps were reported in Jining, with a sex ratio of 2.11:1. Cases were most prevalent in 5–9-year-olds (42.15%) followed by 10–14-year-olds (24.72%). The optimal lag time identified was 10 days and the three meteorological factors that contributed the most to the risk of mumps were daily mean temperature, daily mean relative humidity and daily mean sunshine duration. Their relative contribution rates were 24.4%, 19.9% and 18.3%, respectively. The mean temperature and sunshine duration relationships approximated a U-shaped effect on the risk of mumps, with estimated thresholds of 5.5 °C and 9.5 h, respectively. The effect of relative humidity on mumps increased slightly and then decreased rapidly, with a threshold of 64%. Conclusions: Our study indicates that daily mean temperature, relative humidity and sunshine duration were three significant meteorological factors associated with the incidence of mumps in Jining, China. Understanding the shape of relationships and their thresholds are critical for establishing early warning systems which are important tools in the prevention and control of mumps.
- Boosted regression tree model
- Meteorological factors