Household environmental factors and children's respiratory health: comparison of two cross-sectional studies over 25 years in Wuhan, China.

BACKGROUND: Over the recent decades, residential lifestyle and household environment have changed substantially with rapid development of industrialization and urbanization in China. Whether the prevalence of respiratory diseases changed is still lack of evidence. The objective of this study is to assess potential changes in children's respiratory disease prevalence and associated household environmental factors in Wuhan over a 25-year time interval.
METHODS: Two cross-sectional studies in the Period 1 (1993 to 1996) and Period 2 (2017 to 2018) were compared in this research. Elementary school children in period 1 (N=2,517) and in period 2 (N=3,152) were recruited in Wuhan, China. The respiratory health condition, home environmental factors, and family socioeconomic status of each subject were acquired through questionnaire survey using the same protocols in both periods. We used the Chi-square test to analyze the difference of household environmental factors (focused on three indoor air quality determinants) and children's respiratory health condition between two periods. Logistic regression models were used to assess the impacts of household environmental determinants on children's respiratory diseases and symptoms between the two studies, by adjusting a set of covariates.
RESULTS: The three indoor air quality determinants have reduced substantially in prevalence from period 1 to period 2: environment tobacco smoke (ETS) from 86.6% to 45.9%, household coal use from 47.6% to 4.9%, and kitchen smoke from 58.9% to 7.3%. The prevalence of certain respiratory symptoms in children significantly decreased, such as cough with colds (51.1% to 41.6%) and phlegm with colds (22.3% to 17.7%). The prevalence of asthma was 2.5% and 2.4% and that of bronchitis was 27.1% and 29.8% in both periods. Coal use was a risk factor for asthma in period 1 (OR =2.34, 95% CI: 1.30-4.23), while it was not significantly associated with prevalence of asthma in period 2 (OR =0.60, 95% CI: 0.08-4.51).
CONCLUSIONS: Household indoor air quality determinants and respiratory health condition of children in Wuhan has been improved over the last 25 years. At present, kitchen smoke is an important factor affecting the prevalence of wheeze whatever child has a cold or not and reducing exposure to ETS could be beneficial to protect children to be less likely to develop bronchitis. 2021 Journal of Thoracic Disease. All rights reserved.

Introduction

Respiratory diseases have been a major health burden in children. Lower respiratory tract infections are one of the leading causes of death worldwide in children, especially in developing countries (1-3). A study of several respiratory diseases in one of the largest children’s teaching hospitals in China between 2009 and 2018 has reported that the top four most prevalent respiratory diseases were bronchitis with a prevalence of 27.6%, pneumonia (18.5%), pneumonia affecting other systems (18.4%), and asthma and status asthmaticus (10.7%). These four diseases accounted for 75.2% of all records of respiratory disease diagnoses (4). Children with respiratory diseases were more likely to be absent from school, suffer from sleeping disruptions and mental health disorders (5,6). Respiratory infection in early childhood may develop respiratory diseases in adult life, for instance, children with moderate to severe asthma tend to persist to adulthood (7).

Air pollution has been one of the recognized risk factors for respiratory health (8,9). Air pollutants originated from indoor sources, such as coal burning, environmental tobacco smoke (ETS), and kitchen smoke, often contribute to the total exposure more than pollutants from outdoors, especially in children who spent most of their time indoors (10-12). For example, children’s acute lower-respiratory-tract infections attributable to indoor air pollution accounted for 4.6% of all deaths and 3.1% of disability-adjusted life years in Europe (13). A review concluded that environmental tobacco smoke exposure is a major risk factor for acute and chronic respiratory illness and reduced lung function (14). Coal use for household cooking and space heating has been recognized to cause substantial respiratory diseases due to the incomplete combustion and poor ventilation (15). Epidemiologic studies have shown that exposure to kitchen smoke is associated with respiratory health (16,17). These indoor air pollution sources or determinants may have changed substantially in Chinese cities over the last decades. Such changes may have significant impact on children’s respiratory health conditions. On the other hand, socioeconomic, lifestyle, and parental health conditions may have also changed substantially with the rapid economic development and urbanization. These factors may affect children’s respiratory health as well (18,19).

Although urban living is associated with having a better standard of living in developing countries and has been generally considered as a beneficial factor for children’s health. However, a study in Malaysia found that children living in households or studying in schools in urban areas are more likely to suffer from respiratory illnesses compared with children living in homes or studying in schools in rural areas (20). Adverse factors for urban living include poor air quality especially in Chinese cities as Wuhan located (21). Moreover, it remains uncertain that whether the prevalence of respiratory symptoms and diseases in Chinese children has changed over the past decades. Remarkably, China with its rapid economic and social development offers a natural contrast scenario to see how children’s respiratory symptoms and diseases changes as society develops and whether the household environmental risk factors still play the same roles in their respiratory health.

This study was designed to leverage a previous study carried out in four Chinese cities in 1993 to 1996, which was the first study on the health effects of air pollution on Chinese children’s respiratory health (22). In order to examine the potential changes in respiratory disease prevalence in relation to changes in household air quality determinants, we conducted a follow-up cross-sectional survey in the same four Chinese cities in 2017 to 2018 using the same study protocols. In this study, we took Wuhan, located in central China as an example, to compare health outcomes and household exposure factors as well as socioeconomic and parental history of respiratory disease between period 1 (1993 to 1996) and period 2 (2017 to 2018).

We present the following article in accordance with the STROBE reporting checklist (available at https://dx.doi.org/10.21037/jtd-20-2170).

Methods

Study city and population

Wuhan is the capital city of Hubei province located in central China. It has a subtropical monsoon climate as located in the transitional zone of China’s geographical characteristics. Wuhan’s permanent resident population was 11.2 million in 2018 (23). As a representative of the major megalopolis in central China, Wuhan has developed rapidly in social and economic as GDP grew nearly four times in the last decade (2008 to 2018) (24,25).

A total of 2,517 children (male: 1,253, 49.8%; female: 1,264, 59.2%) participated in the period 1 study. There were 1,923 and 594 students chosen from an urban and a then suburban elementary school, respectively. Since the two schools which were selected 25 years ago did not exist now, six elementary schools located in the same districts where the period 1 schools located were chosen in the period 2 study. A total of 3,152 children were recruited from the six primary schools with a distribution of male 1,798 (57.0%) and female 1,354 (43.0%) child participants. Among them, one school in suburban area covered 1,167 children, five schools in urban area covered 1,985 children because of the small enrolment in each school. In every school, stratified sampling was conducted by each grade and children had an age of 9.81±1.60 and 8.88±1.38 in period 1 and period 2, respectively. This study was approved by the Committee on Ethics of Biomedicine Research, Duke Kunshan University, Jiangsu (No. FWA00021580). Parents gave written consent for the participation of themselves and their children in the study. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013).

Questionnaire survey

All the parents were asked to fill in a questionnaire concerning exposure to potential risk factors that might influence children’s respiratory health and the presence of symptoms of respiratory disease. The questionnaire of period 1 was adapted from American Thoracic Society Epidemiology Standardization Project (ATS-DLD-78-C) (26), which was composed of three parts concerning general family information including family member and lifestyle and household indoor environment, child’s respiratory health and their parents information such as educational levels and occupations. Except for the same questions as in period 1, some additional questions were added in period 2 according to the characteristics of the current social lifestyle.

The original values of the questionnaire responses were double inputted. In order to compare the influence of household environmental factors to children’s respiratory health between the two study periods, the questionnaire responses of identical questions in two periods were combined into a database according to a standardized code book and file structure. Since socioeconomic status was typically evaluated by parents’ education level and occupation, we classified parents’ education into low (high school or below) and high (technical college, university or above). Parental occupation was classified into ‘white collar’ and ‘blue collar’ basically according to the labor types and risk of occupational exposure (27). The former referred to businessman, educational or medical staff, civil servant, and military personnel, while the latter referred to farmer and factory worker in the production of chemicals, fertilizers, and paints. Parents’ history of respiratory diseases, such as ‘Has a doctor ever diagnosed asthma’ and ‘Has a doctor ever diagnosed bronchitis’ was also investigated.

Definitions of household risk factors and health outcomes

This study concerned about the common sources of household air hazards and typical childhood respiratory diseases. Based on the information both investigated in two questionnaire surveys, we focused on the health effects of three household environmental factors on children’s respiratory health, namely, indoor coal use, ETS, and kitchen smoke. Therefore, the data were preprocessed to be consistent and we assessed household environmental determinants using the questions in , such as kitchen smoke was determined by question ‘What is the ventilation for cooking in the kitchen?’ and answering other than ‘open window’ and ‘nothing’ is determined to be the presence of kitchen smoke. Respiratory symptoms and diseases are determined by the questions showed in . Respiratory symptoms were self-reported by children’s guardian and the history of respiratory diseases should ever be diagnosed by doctor.

Table 1

Questions and definitions of household environmental determinants

Household environmental determinants	Questions	Classification criteria
Coal use	Whether coal is the main fuel of heating your home? (yes/no)	Either answering yes to the two questions is determined to be the presence of coal use
	Whether coal is your main fuel for cooking? (yes/no)
ETS	Has father ever smoked? (yes/no)	Either parent answering yes is determined to be the presence of ETS
	Has mother ever smoked? (yes/no)
Kitchen smoke	What is the ventilation for cooking in the kitchen? (hood/fan/chimney/open window/nothing)	Answering other than ‘open window’ and ‘nothing’ is determined to be the presence of kitchen smoke

Table 2

Questions and definitions of respiratory symptoms and diseases in children

Respiratory health condition	Question
Symptoms
Cough with colds	Has this child ever coughed when having a cold? (yes/no)
Cough without colds	Has this child ever coughed when not having a cold? (yes/no)
Phlegm with colds	Has this child ever has brought up phlegm or mucus from the chest when having a cold? (yes/no)
Phlegm without colds	Has this child ever has brought up phlegm or mucus from the chest when not having a cold? (yes/no)
Wheeze with colds	Has this child’s chest ever sounded wheezy or whistling when having a cold? (yes/no)
Wheeze without colds	Has this child’s chest ever sounded wheezy or whistling when not having a cold? (yes/no)
Diseases
Asthma	Has a doctor ever diagnosed asthma in this child? (yes/no)
Bronchitis	Has a doctor ever diagnosed bronchitis in this child? (yes/no)

Statistical analysis

The continuous variables were described as mean ± SD, while the categorical variables were showed as numbers and proportions. To assess the changes of household environmental factors and children’s respiratory health conditions from period 1 to period 2, we calculated the percentage of the home environmental characteristics and the prevalence of children’s respiratory symptoms and diseases. We carried out chi-square tests to examine differences between the two periods. Logistic regression models were used for exploring the association of ETS, coal use and kitchen smoke on children’s respiratory symptoms and disease and also used to compare whether the role of the risk factors in two periods have changed, by adjusting the confounders such as residential area, gender, age, parental occupation, parental education, breast-feeding, sleeping in own room, sleeping in own bed, parental asthma, parental bronchitis. When assessing the health effect of ETS, coal use and kitchen smoke were also adjusted for in the ETS model. In the same way, the coal use model was adjusted for ETS and kitchen smoke, and the kitchen smoke model was adjusted for ETS and coal use. The data were analyzed using the statistical software SPSS 26.0 version. All statistical tests were two tailed, and statistical significance was declared when a P value was <0.05.

Results

The information of study subjects, family conditions and parental characteristics are shown in . The proportion of parents as blue-collar workers decreased after 25 years, although this still accounted for a significant proportion among occupation types in period 2 (48.1%for father and 42.1% for mother). There was a slight improvement in parental education level in period 2. The prevalence of paternal asthma and maternal asthma showed a decrease from period 1 to period 2, for example, 1.4% of mothers suffered from asthma in period 1 versus 0.5% in period 2. Also, the prevalence of parental bronchitis decreased from period 1 to period 2, particularly for paternal bronchitis (14.9% versus 6.1%). The proportion of exposure to coal use, ETS and kitchen smoke have all markedly decreased from period 1 to period 2 (P<0.001) (). The proportion of exposure to ETS fell by approximately half from 86.5% to 45.9%. Coal smoke exposure and kitchen smoke exposure have both reduced even more, from more than 40% to less than 10%.

Table 3

Characteristics of subjects, parental and household environmental factors in the two study periods

Characteristics	Period 1	Period 2	P for period difference
Total	2,517	3,152
Age (year, mean ± SD)	9.81±1.60	8.88±1.38	<0.001
District, n (%)			<0.001
Urban	1,923 (76.4)	1,985 (63.0)
Suburban	594 (23.6)	1,167 (37.0)
Gender, n (%)			<0.001
Male	1,253 (49.8)	1,798 (57.0)
Female	1,264 (50.2)	1,354 (43.0)
Environmental tobacco smoke (ETS), n (%)	2,083 (86.6)	1,382 (45.9)	<0.001
Coal use, n (%)	1,198 (47.6)	155 (4.9)	<0.001
Kitchen smoke, n (%)	1,479 (58.8)	231 (7.3)	<0.001
Sleep in own room, n (%)	713 (28.3)	1,573 (51.2)	<0.001
Sleep in own bed, n (%)	1,599 (63.6)	1,938 (63.4)	0.916
Breast feeding, n (%)	1,707 (67.8)	2,402 (82.7)	<0.001
Father occupation1, n (%)			<0.001
Blue collar2	1,490 (59.8)	1,439 (48.1)
White collar3	964 (38.7)	1,143 (38.2)
Mother occupation1, n (%)			<0.001
Blue collar2	1,470 (59.1)	1,195 (42.1)
White collar3	894 (36.0)	1,291 (45.4)
Father education, n (%)			<0.001
Less than high school	2,124 (85.2)	2,283 (76.1)
University and above	368 (14.8)	718 (23.9)
Mother education, n (%)			<0.001
Less than high school	2,331 (93.7)	2,364 (79.0)
University and above	156 (6.3)	628 (21.0)
Father diseases, n (%)
Asthma	38 (1.5)	23 (0.8)	0.010
Bronchitis	373 (14.9)	179 (6.1)	<0.001
Mother diseases, n (%)
Asthma	34 (1.4)	15 (0.5)	0.001
Bronchitis	220 (8.8)	137 (4.7)	<0.001

1, occupations not belonging to the below types or parents had multiple occupations are not shown in the chart. 2, farmer and worker who works in a factory or in the production of chemicals, fertilizers, and paints. 3, businessman, education or medical staff, civil servant, and military personnel.

The respiratory health conditions of children are shown in . Overall, children’s respiratory health has improved, particularly for the prevalence of wheeze with cold from period 1 to period 2 (13.5% vs. 6.5%). The improvement in respiratory health was more pronounced in urban children than in suburban children. Except for wheeze without colds and asthma, children’s respiratory symptoms and diseases in urban area has decreased significantly, such as cough with colds (from 55.0% to 40.5%), bronchitis (from 30.8% to 26.6%). In the suburb area, however, there was no significant decrease in those symptoms or diseases. And we even observed a significant increase in the prevalence of bronchitis, from 14.8% in period 1 to 35.4% in period 2 (). Interestingly, the prevalence of respiratory symptoms and diseases in the suburban area was lower than that in urban area in period 1, while it turned to be opposite for most symptoms and diseases in period 2 (Table S1).

Table 4

The prevalence of respiratory diseases and symptoms in two study periods

Variables	Total				Urban				Suburban
	Period 1, n (%)	Period 2, n (%)	P		Period 1, n (%)	Period 2, n (%)	P		Period 1, n (%)	Period 2, n (%)	P
Symptoms
Cough with colds	1287 (51.1)	1283 (41.6)	<0.001		1058 (55.0)	789 (40.5)	<0.001		229 (38.6)	494 (43.3)	0.059
Cough without colds	173 (6.9)	151 (4.9)	0.002		142 (7.4)	81 (4.2)	<0.001		31 (5.2)	70 (6.2)	0.429
Phlegm with colds	562 (22.3)	525 (17.7)	<0.001		475 (24.7)	326 (17.5)	<0.001		87 (14.7)	199 (18.2)	0.067
Phlegm without colds	120 (4.8)	104 (3.5)	0.020		102 (5.3)	62 (3.3)	0.003		18 (3.0)	42 (3.9)	0.387
Wheeze with colds	340 (13.5)	187 (6.5)	<0.001		290 (15.1)	122 (6.7)	<0.001		50 (8.4)	65 (6.0)	0.061
Wheeze without colds	49 (1.9)	48 (1.7)	0.444		43 (2.2)	27 (1.5)	0.097		6 (1.0)	21 (2.0)	0.143
Diseases
Asthma	64 (2.5)	69 (2.4)	0.710		56 (2.9)	48 (2.6)	0.611		8 (1.3)	21 (2.0)	0.362
Bronchitis	681 (27.1)	864 (29.8)	0.025		593 (30.8)	485 (26.6)	0.004		88 (14.8)	379 (35.4)	<0.001

shows the association on the basis of ORs and 95% CI between the coal use, EST, or kitchen smoke and children’s respiratory health, after adjusting for area (urban vs. suburban), gender, age, parental occupation, parental education, breast-feeding, sleep in own room, sleep in own bed, parental asthma, parental bronchitis and other two focused home environmental risk factors. Coal use was significantly associated with phlegm when having colds (OR =1.26, 95% CI: 1.02–1.56) and asthma (OR =2.34, 95% CI: 1.30–4.23) in period 1. There were significant associations of ETS exposure with cough (OR =1.30, 95% CI: 1.03–1.64) and phlegm (OR =1.35, 95% CI: 1.01–1.82) when having colds in period 1. Twenty-five years later, exposure to ETS was associated with cough without colds (OR =1.70, 95% CI: 1.070–2.70) and bronchitis (OR =1.35, 95% CI: 1.09–1.68). There was no significant association between kitchen smoke exposure and respiratory disease and symptoms in period 1. However, in period 2, children who exposed to kitchen smoke had increased risk of wheeze with colds (OR =2.05, 95% CI: 1.15–3.66) and wheeze without colds (OR =3.60, 95% CI: 1.22–10.65), while having a decrease risk for bronchitis (OR =0.51, 95% CI: 0.32–0.81).

Figure 1

The association between respiratory symptoms and diseases and household environmental determinants, showing odds ratios (ORs) in reference to the absence of each determinant. ORs were adjusted for residential area, gender, age, parental occupation, parental education, breast-feeding, sleep in own room, sleep in own bed, parental asthma, parental bronchitis, and the two co-determinants.

Discussion

The two cross-sectional surveys on children’s respiratory health, which were respectively conducted in 1993 to 1996 and 2017 to 2018 in Wuhan, found significant improvements in both household environmental conditions and children’s respiratory health. The proportions of children exposure to ETS (86.6% to 45.9%), coal use (47.6% to 4.9%) and kitchen smoke (58.9% to 7.3%) have greatly decreased, which used to be a great concern in household environment. As a whole, there was also a significant decrease in the prevalence of respiratory symptoms and diseases in children. The association between those three risk factors and respiratory diseases and symptoms have also changed, after adjusting for the covariates of gender, age, residential area, breast-feeding, sleeping in own room, sleeping in own bed, parental occupation, parental education, parental asthma, parental bronchitis and the other two home environmental risk factors.

The improvements in children’s respiratory health have been accompanied by the improvement in three household environmental risk factors pertinent to indoor air quality. Particularly, such improvements were more pronounced in the urban area. Some early studies have observed that children in urban areas have a higher prevalence of respiratory problems than children in suburban areas (28,29), which is consistent with our findings from period 1 survey. However, recent studies have found some conflicting results. The occurrence of acute respiratory tract infection in a group of Indian children living is rural areas was higher as compared to those living in urban areas (30). In period 2, we also found higher prevalence of respiratory diseases and symptoms in the suburban (closer to rural) area than in the urban area. The discrepancies of that change could be mainly attributed to the difference in air pollution in rural areas in the past 25 years (31). Nowadays in China, ambient air pollution in suburb areas tends to be higher than that in urban area (32) because industrial facilities are typically located in suburban areas and the implementation of environmental protection policies tends to be less stringent in suburban areas (33). Additionally, differences in health awareness and environmental related behaviors between urban and rural areas over two periods also lead to differences in the prevalence of respiratory symptoms and diseases. The condition of medical care can affect the consciousness and treatment behavior of parents to children’s respiratory health problems (34). Increased health awareness has led parents to pay more attention to their children’s symptoms and seek medical diagnosis, resulting in fewer missed detection (35). The Chinese government has been expanding health insurance coverage and improving health service utilization. In comparison with period 1, urban-rural disparities in children’s respiratory diseases and symptoms have been narrowed in period 2. It could be attributed to those suburban parents may pay more attention to the health of children and tend to seek medical consultation resulting in observation of increasing prevalence after decades. However, we cannot conclude which one was more important to the disparities among environmental factors and medical awareness. Thus, it is necessary to determine whether this is caused by past prevalence reporting bias due to lack of medical awareness or by differences in living environment between urban and rural areas in the future.

Among the respiratory symptoms and diseases, bronchitis of children in Wuhan requires special attention. The prevalence of bronchitis remained high, at 29.8% in period 2. Similarly, in Chongqing, one of the four cities in the original and follow-up Four Chinese City Study, the prevalence of children’s bronchitis in 2017 (23.6%) was significantly higher than that in 1993 (17.7%) (OR =1.89, 95% CI: 1.54–2.31) (36). To date, few studies focused on the prevalence of bronchitis in children. Bronchitis is a serious disease and is associated with worse lung function, more dyspnea, and increased risk for developing asthma and chronic obstructive pulmonary disease in later life (37-39).

Although we have observed a decrease in ambient concentrations of air pollutants in Wuhan in recent decades, the annual mean of fine particulate matter (PM2.5) in 2017 was still above 50 µg/m3, exceeding the World Health Organization (WHO) Air Quality Interim Target-1 (WHO 2005) of 35 µg/m3 (40,41). Outdoor air pollution alone cannot fully account for the health effects, especially when children stay indoors for a long time. In this study, we focus on three indoor air quality related household environmental determinants, namely coal use, EST, and kitchen smoke.

Indoor coal burning in Chinese homes is mainly used for cooking and heating (42). According to Chinese governmental policies, Wuhan is classified as a “heating-needless” city where no heating infrastructure is planned in building designs. Our research also found that the vast majority of indoor coal burning in Wuhan is used for cooking rather than heating. We found that burning coal in period 1 was a risk factor for cough with colds and asthma whose associations reach statistical significance. Coal burning releases sulfur dioxide, carbon monoxide, particulate matter and other toxic and harmful pollutants (43,44), which are harmful to respiratory health. However, we did not find significant respiratory health risks in children from burning coal in period 2. This may be attributed to the relatively low coal use rate (4.9%) and improved coal quality in period 2. Other cities in this series of studies, such as Lanzhou and Chongqing, have also seen significant declines in indoor coal use. In period 2, coal burning was no longer a prominent problem affecting children’s respiratory health in Lanzhou and Chongqing (45,46).

The association between ETS and respiratory issues such as chronic bronchitis is well known (47,48) and our finding in period 2 was consistent with this. Children are exposed to ETS mainly through living with smokers (49,50). Studies have found that ETS exposure can cause cough, wheezing and asthma attacks in children (51-53). The proportion of mothers in Wuhan who smoked was very low. It is also possible for children to live with family members who smoke, but this situation rarely occurred and it is impractical to assess the total true ETS exposure in our study. Hence, smoking status of fathers is the best indicator for home ETS exposure in our study subjects. We observed a significant decline in paternal smoking rates, which may attribute to increased health awareness and smoking bans (54,55). Both the reduction in amount of cigarettes and change in smoking site can affect children’s actual exposure to ETS.

In this study, we investigated the ventilation in kitchen as a surrogate of indoor kitchen smoke exposure. Although its effects were not significant on most of the outcomes assessed in this study, kitchen smoke has been observed to have a negative effect on wheezing symptoms in children. Indoor mechanical ventilation has also been shown to modify the effects of indoor air pollutants on respiratory health (56,57). However, kitchen may not be the main living area for elementary-school-age children. This may explain the null effects of kitchen ventilation (kitchen smoke exposure) in our children participants.

There are some limitations in this study. It is an inherent limitation of cross-sectional studies that two cross-sectional comparisons cannot accurately measure the influence of factors varying over time on the results. We hope to find out the corresponding exposure of children through questionnaires, although questionnaires cannot enable us to obtain the accurate exposure dose. Although we know coal use, ETS and kitchen smoke are important contributors to indoor air pollution, but we realize that indoor pollution sources are beyond those three factors. Emerging sources in Chinese households, such as decoration materials (58), indoor mildew (59), pet dander (60), were not considered in this comparison study. Since we considered the new pollution characteristics brought about by lifestyle changes or new concerns found in recent studies, we modified the questionnaire concerning these factors in period 2 including the proportion of the use of air purifiers and whether there was dampness. These can affect indoor air quality and further benefit to respiratory health (61-64). Unfortunately, the period 1 study did not collect data on these sources as they were nonexistent or not recognized, prohibiting us to consider these in our comparisons between the two periods, which was the main aim of the present study.

Conclusions

As a whole, children’s respiratory health symptoms have improved after 25 years from earlier 1990s in Wuhan. However, in the stratified analysis we found that this improvement was mainly due to the improvement in children living in urban areas of the city. In contrast, no significant improvements were observed for the prevalence of asthma and that of bronchitis. Household risk factors pertinent to indoor air quality, including coal use, ETS, and kitchen smoke, were significantly improved. These improvements may have contributed to the improvements in some of the respiratory health outcomes.

The article’s supplementary files as