Impact of meteorological factors on mumps and potential effect modifiers: An analysis of 10 cities in Guangxi, Southern China

• Published in: Environmental research 2018-10, Vol.166, p.577-587
• Main Authors: Yu, Guoqi, Yang, Rencong, Yu, Dongmei, Cai, Jiansheng, Tang, Jiexia, Zhai, Wenwen, Wei, Yi, Chen, Shiyi, Chen, Quanhui, Zhong, Ge, Qin, Jian
• Format: Article
• Language: English
• Subjects: CHINA; CLIMATES; CLIMATIC CHANGE; DEVELOPING COUNTRIES; Distributed lag non-linear model; ENVIRONMENTAL SCIENCES; INFECTIOUS DISEASES; Meteorological factor; METEOROLOGY; Modifier; Multi-city; Mumps; URBAN AREAS; WEATHER
• ISSN: 0013-9351; 1096-0953
• DOI: 10.1016/j.envres.2018.06.047

In the current context of global climate change, understanding the impact of climate on respiratory infectious diseases such as mumps and the potential modified factors is crucial, especially in developing countries. However, research on the climate-related incidence of mumps is rare, inconsistent and mainly limited to a single city or region. Daily mumps cases and meteorological variables of 10 cities in Guangxi, Southern China were collected for 2005–2017. Two-stage analyses were performed to assess the relationship between meteorological factors and mumps incidence during two time-periods: 2005–2012 and 2013–2017, separately. First, a Poisson regression model that allows over-dispersion was used to estimate the city-specific climate-related morbidity after controlling for temporal trends, day of week, and national statutory holidays. Then, we used a multivariate meta-analytical model to pool the city-specific effect estimates and conducted subgroup analyses. Multivariate meta-regression was applied to detect potential effect modifiers. Non-linear relationships were observed among mean temperature, wind speed, and mumps incidence in 2005–2012. The impact of high temperature on mumps incidence was short and rapid, whereas the impact of low temperature was long and slow. The total cumulative relative risk (RR) associated with hot temperature was 1.18 [95% Confidence Interval (CI): 0.93, 1.48], which was calculated by comparing the incidence of mumps above the 90th percentile of temperature with its incidence at the median temperature at lag of 0–30 days. Meanwhile, the RR associated with cold temperature was calculated to be 1.50 (95% CI: 1.08, 2.10) by comparing the incidence of mumps below the 10th percentile of temperature with its incidence at the median temperature. Similarly, the RRs associated with windless and windy conditions for the total population were 1.23 (95% CI: 1.04, 1.46) and 0.83 (95% CI: 0.67, 1.02), respectively. Effects based on extreme temperature and wind speed conditions were more prominent in males than in females. Compared with children and adults, adolescents (5–14 years old) were more sensitive to extreme weather conditions. Geographical latitude, Population density, GDP per capita, Number of health institutions, Highly educated population and Inoculation rate were considered the most likely associated modifiers. In addition, the correlation between meteorological factors and the incidence of mumps and modification of socioecon