Domestic coal use and exposure to ETS in the home were both associated with adverse respiratory effects in this population of Chinese adolescents. Coal burning was associated with increased wheezing, whereas living with smokers was associated with increased cough and phlegm production.
Coal burning produces high concentrations of particulate matter, SO
2, and other pollutants [
2,
3,
11]. Exposure to these pollutants may impair clearance mechanisms, and lead to airway inflammation [
21,
22]. Decreased pulmonary function has been associated with exposure to particulate matter and SO
2 in several air pollution studies during the past decades [
21]. Although residential coal burning has been linked to decreased pulmonary function and asthma among children [
23-
25], conflicting data exist. In two European studies, domestic coal burning has been associated with lower risk for childhood asthma and allergic diseases [
26,
27]. The findings in these two studies, however, may reflect some early life or other lifestyle factors related to coal use in Europe.
In our study, residential coal burning was predominantly associated with wheezing. Coal cooking was a stronger risk factor for wheezing than was coal heating. This may be explained by relatively low heating use in the Wuhan area, whereas cooking is a year around activity. The greater association with coal use for both cooking and heating may suggest an exposure-dependent relationship.
Although wheezing is often closely related to asthma, coal use was not positively associated with asthma diagnosis (data not shown) in this population. The majority of the diagnosed asthmatics (76.4%) lived in urban areas, where prevalence of coal use was lower than in non-urban areas. The diagnostic ascertainment of asthma most likely was greater in the urban than in the rural areas.
The harmful effects of ETS in children, primarily from living with smokers, have been widely studied [
14,
28-
31]. In general, evidence that ETS causes cough, phlegm, and wheezing has not been as strong for school-aged children as it has been for infants and preschool children [
28]. There are few data among Chinese populations where smoking behavior differs from Western populations.
In utero exposure, via maternal smoking, that is believed to contribute to adverse effects of ETS in children [
32,
33] is uncommon in China. Thus, it is of interest that in this group of middle school children, where maternal and personal smoking were low, exposure to ETS in the home was clearly associated with chronic cough and phlegm production, with and without colds.
Our results indicated an exposure dependent response to ETS; having two or more smokers in the household increased the odds of cough and phlegm production compared to having only one smoker in the household. We did not find strong evidence suggesting modifying effects by gender, although the effect of ETS on persistent cough without colds was more pronounced among boys than girls (data not shown). Exposure levels may be influenced by time-activity patterns that can differ by gender. Boys may be more likely to spend time in close proximity with their smoking fathers or male relatives than girls.
Mechanisms responsible for the respiratory effects of ETS have been proposed in the literature [
28]. In addition to decreased mucociliary clearance and goblet cell hypertrophy/hypersecretion, local and central nervous system components are thought be involved in cough and phlegm production [
28,
34]. Although exposure to ETS may affect childhood lung growth and result in lower pulmonary function [
14,
35], wheezing was not appreciably related to the presence of smokers in our study. Genetic susceptibility may influence the effects of ETS on bronchial obstruction. For example, parental atopy was found to modify the effects of ETS on bronchial obstruction and asthma considerably in a Norwegian birth cohort study [
36]. However, we were unable to examine potential interactions between family history and ETS in relation to atopic illness in our population because, consistent with previously published data on Chinese children [
37,
38], the prevalence of asthma (3.2%) and hay fever (1.8%) was very low.
In general, our findings agree with available data on Chinese children's respiratory health [
16,
23]. However, residential exposures in the current study were more selectively associated with the respiratory symptoms than in previous studies. This may reflect differences in the study settings. In the previous studies [
16,
23], for example, most of the children were younger in age than in the current study. Prevalence of symptoms and factors associated with childhood respiratory symptoms may differ between different age groups [
39]. It is also possible that using students rather than parents as a source of information on child's symptoms may contribute to the observed differences.
Exposure to indoor air pollutants is not only influenced by the source strength and other emission characteristics, but also by air exchange rates. A recent study showed that ventilation could modify effects between respiratory health outcomes and indoor air pollutants [
40]. In that study, the modifying effects were found most relevant when air exchange rates were low. Residences in Wuhan, however, were not energy-efficiently built [
16]. Air conditioning was uncommon, and most of the homes, both in urban and non-urban areas, relied on natural ventilation. In this study, we were unable to evaluate the effects of ventilation rates, because we did not collect detailed information on ventilation practices. We thought that children would not be able to give this information accurately.
The composition of pollutants produced by residential coal burning and smoking can be highly variable, but both exposures contribute substantially to inhalable and respirable particulate matter in indoor environments [
2,
3,
41]. Existing data suggests that coal burning and smoking may have synergistic effects on respiratory symptoms [
5]. In our data, we did not find consistent evidence of interaction between coal burning and ETS exposure.
Our outcome and exposure measures were determined by questionnaire alone, which is one of the major limitations of the study. Nonetheless, large epidemiological studies of respiratory health often rely on reports on recent symptom history because self-reported measures are cost efficient, practical and their repeatability is good [
42,
43]. Generally, respiratory symptoms have been reported consistently across populations [
43]. To improve the quality of our self-reported outcomes we included audiovisual presentation of wheezing symptoms [
15]. Because the temporal relationship between outcome(s) and exposure(s) can be difficult to determine in cross-sectional studies we focused on respiratory symptoms in the past 12 months to minimize recall bias. We did not use parents as source of information on child's symptoms. Some studies suggest that Chinese parents may deny or underreport child's symptoms or illnesses [
44,
45]. In addition, parents living in non-urban areas around Wuhan have lower educational level than parents living in urban areas [
16], and their literacy level may be lower than their children attending middle school. Therefore, adolescents' reports on their own symptoms and health status may be more accurate than their parents'. Because children were answering in school about exposures in their home, we were not able to acquire very detailed information on exposure characteristics. Given that questionnaires have limited ability to quantify exposures, the possibility of exposure misclassification cannot be excluded. However, serious differential misclassification either of the exposures or outcomes is unlikely because health hazards of indoor air pollutants were not widely known among Chinese school children at the time when the survey was conducted [
46].
Although urban air pollution has long been a major environmental concern in China, we do not believe that outdoor air pollution alone could explain the observed associations. Exposures to indoor air pollutants are likely to dominate the total exposure burden [
47], especially among children, who spend much of their time inside the home [
8]. In Chinese homes with coal stoves and smokers, not only levels of particulate matter, but also levels of many other air pollutants, including concentrations of SO
2, often exceed the levels outdoors [
2,
3]. In Wuhan, where coal stoves are not usually vented via flue, concentrations of respirable particulate matter (291 μg/m
3) and SO
2 (173 μg/m
3) can reach high levels indoors [
2]. Concentrations of these pollutants have been found to be lower in ambient air. For example, a study investigating long-term air pollution in Wuhan estimated that the annual means for PM
2.5, PM
10, and SO
2 in urban areas were 73 μg/m
3, 129 μg/m
3, and 73 μg/m
3, respectively [
48]. Because indoor air quality is influenced by infiltration of outdoor air, we cannot fully exclude possible confounding effects of ambient air pollution [
1]. However, the effects of living area, as measured by school districts, were taken into account in our models, providing some control for differing air pollution levels in the study area.
The major strength of this study is that the public school system ensured a large and representative sample of rural, suburban, and urban populations in the Wuhan area. Our study is one of the few studies that have examined effects of major indoor pollutants in relation to children's respiratory health in mainland China [
16,
23-
25].
1
This article has been 