In this study, we estimated the possible interactions of meteorological factors and PM on childhood asthma by using an 11-year time series data of 976, 350 outpatient visits for childhood asthma in Shanghai, China. Firstly, we found meteorological factors and PM had significantly lagged and non-linear effects on childhood asthma. Further, in interaction analysis, our results suggest significant antagonistic interaction between temperature and PM (RERI: -0.14, 95%CI: -0.15, -0.12 for PM2.5; RERI: -0.09, 95%CI: -0.10, -0.08 for PM10) but synergistic interaction between air pressure and PM (RERI: 0.16, 95%CI: 0.14, 0.18 for PM2.5; RERI: 0.17, 95%CI: 0.15, 0.19 for PM10) on childhood asthma.
In DLNM models, we found an approximate V-shape for the exposure-response curve of mean temperature on childhood asthma, indicating exposure to high and low temperatures was associated with an increased risk of childhood asthma. This finding is consistent with the majority of previous findings [11, 12, 18, 21, 29]. For instance, Xu et al. observed both hot (95th percentile of mean temperature) and cold (5th percentile of mean temperature) temperature increased the emergency department admissions for childhood asthma relative to the reference temperature of 24.0°C in Brisbane, Australia . Soneja et al. found the occurrence of extreme heat events was associated with higher risk of asthma hospitalization in Maryland, U.S.A. . A systematic review summarized available information about the relationship between temperature and childhood asthma, and called for special attention to extreme temperatures . Except for temperature, previous studies have reported that exposure to low relative humidity or high air pressure was associated with increased risk of childhood asthma [8, 19, 20]. In our study, relative humidity (< 82%) and air pressure (> 1003 hPa) were found to be positively connected with outpatient visits for childhood asthma in this study. In addition, we observed that exposure to PM2.5 and PM10 was positively associated with an increased risk of childhood asthma, which were align with prior findings in Chongqing (China) , Shanghai (China) , Seoul (Korea) , Québec (Canada) , and New York (USA) . Meteorological factors have been suggested to impact asthma through directly affecting airway hyperventilation or inflammation pathway and by indirectly affecting asthma through viral infections, allergens, bacterial activity or outdoor time [12, 29, 33]. The effects of PM on asthma are largely determined by the size and chemical composition, deposition in respiratory tract and immune response to the particles . Potential mechanisms of PM exposure on asthma exacerbation include oxidative stress , inflammatory response , mucosal barrier function disruption , and airway hyperresponsiveness . In general, align with the majority of previous findings, our study provides consistent evidence for associations of meteorological factors and PM with childhood asthma.
Even though a mounting body of evidence has suggested the independent impacts of meteorological factors or PM on childhood asthma, limited studies have explored their possible interactive effects to date [22, 39, 40]. In a multifactorial study of meteorological factors and ambient air pollutants (CO, NO2, SO2, PM2.5 and PM10) on asthma acute exacerbation in Taiwan, Yu et al. found ambient air pollutants showed different effects with or without meteorological factors, and suggested that meteorological factors should be simultaneously considered when identifying the impacts of ambient air pollutants on asthma . Mokoena et al. observed an antagonistic interaction between temperature and air quality index (AQI) on respiratory mortality (RERI = -0.235, 95%CI: -0.269, -0.163) in a time-series study conducted in Xi’an, China during 2014 and 2016 . The results of this study indicate antagonistic interaction between temperature and PM and synergistic interaction between air pressure and PM on the outpatient visits for childhood asthma. The antagonism between air temperature and PM might be related to adaptive behaviors such as the use of air conditioning and the reduction of outdoor activities under high temperature . It is well known that meteorological factors could influence the concentration of ambient air pollutants , and high air pressure has been reported to worsen air quality . This might partially explain the synergistic interaction of air pressure and PM on childhood asthma in this study. However, given to the limited knowledge of interactive effects of meteorological factors and ambient air pollution on childhood asthma, our findings should be interpreted with caution and need to be confirmed by further studies in different populations.
This study has several strengths. To our knowledge, this is the first study to explore the potential interactive effects of meteorological factors and PM on childhood asthma in Shanghai, China. Then, a large sample size of 976, 350 outpatient visits during a long-time span of 11 years was utilized in our study. In addition, consistent results in sensitivity analyses of interactions between temperature, air pressure and PM enhanced the reliability of our findings for the interactions between meteorological factors and particulate pollutants on childhood asthma. However, this study also has several limitations. Firstly, data of meteorological factors and air pollutants obtained from monitoring stations were used to reflect individual exposure levels, which might lead to measurement bias [44–45]. Secondly, information on some confounding factors, such as parental smoking, indoor air pollutants, aeroallergens and allergic constitution, was unable to acquire in the present study and might influence the risk assessment of meteorological factors and PM on childhood asthma. Thirdly, the generalizability of our study is limited considering the single city study design. Since dynamic changes in meteorological factors and ambient air pollutants simultaneously coexist in the real world, the assessment of their interactive effects on childhood asthma is imperative and strongly encouraged [16, 40].