Prevalence of COPD and its association with socioeconomic status in China: Findings from China Chronic Disease Risk Factor Surveillance 2007

Chronic Obstructive Pulmonary Disease (COPD) is a major cause of chronic morbidity and mortality throughout the world [1]. COPD mortality has been rising steadily over the past few decades and it is projected to be the third leading cause of death in the world by 2020 [2]. Cigarette smoking is the major risk factor for COPD; passive smoking, indoor air pollution, occupational dust exposure and genetic factors are recognized as potential factors contributing to the development of COPD [1, 3].

There have been many prevalence studies of COPD across countries and regions. A recent meta-analysis [4] comprehensively summarized all the 62 papers published from 28 different countries during 1990-2004 and estimated the pooled prevalence of COPD was 6.4%, 1.8% and 9.2% according to the definition of chronic bronchitis, emphysema and airflow obstruction respectively. Compared to the information available from developed countries, studies of COPD prevalence in China are limited. Three studies done in China were included in Halbert's review [4], one used physician diagnosed emphysema [5] and the other two [6, 7] used respiratory symptoms as the case definition. A study in Nanjing, east China, reported that the overall prevalence of physician-diagnosed COPD, using self-reported data from over 29,000 adults over the age of 35 years, was 5.9% [8]. Another study in Guangdong Province, using spirometry to define COPD, reported 9.4% prevalence based on 3286 subjects aged over 40 from two localities. The most recently published population based prevalence study was part of the worldwide Burden of Lung Disease (BOLD study) [9]. They used Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria to define COPD and recruited more than 20,000 subjects aged 40 years and older from 7 provinces and cities including urban and rural areas from China. The COPD prevalence was highest in Chongqing (13.7%) and lowest in Shanghai (3.9%) among urban areas, and highest in Guangdong (12.0%) and lowest in Liaoning (6.8%) and Shanxi (6.9%) among rural areas. There is no prevalence study carried out in a nationwide sample in China.

Many studies from developed countries suggested that socioeconomic status (SES), measured by income and educational level, is associated with lung function and COPD in terms of exacerbation, prevalence and mortality [10‐13]. This association may be partly explained by the greater proportion of smokers among people in lower socioeconomic groups, but smoking may not explain all of the association. In the Chinese population, COPD is prevalent not only among smokers but also among never smokers. This study aimed to clarify this association.

We used nationwide data from China Chronic Disease Risk Factor Surveillance in 2007 to estimate the prevalence of self-reported physician diagnosed COPD and to examine its association with SES.

To our knowledge, this is the first study conducted in a nationwide sample covering all 31 provinces, municipalities and autonomous regions of China to estimate the prevalence of COPD based on self-report. We found the prevalence of COPD among resident aged 15-69 years was 2.9%, with geographic variation from 2.2% in the east to 2.7% in the central and 3.7% in the west. Prevalence increased with age, and was significantly associated with smoking and rural areas, consistent with other studies [8, 9].

There have been a number of studies in China that reported prevalence of COPD either in certain provinces or selected combination of provinces which cannot provide information on variation of COPD prevalence. Zhong et al reported COPD prevalence in 11 provinces, and found the highest prevalence in Chongqing (west) and lowest in Shanghai (east) among urban areas and highest in Guangdong (east) and lowest in Liaoning (east) among rural areas [9]. Our findings agree with previous reports that the prevalence of COPD was highest in western areas and lowest in eastern areas. This can be explained by the different economic levels in these regions in China. Solid fuel use for cooking and heating in the household without adequate ventilation in west China might contribute to the high prevalence of COPD in the areas. Poorer nutrition, greater occupational exposure and lower health care standard may also play an important part in the significantly higher COPD prevalence in western China compared to central and eastern China. The geographic variation of COPD prevalence in China we found in the present study is consistent with other studies from Latin American and developed countries [15, 16].

We found a significant relationship between socioeconomic status reflected by educational level and household income and COPD prevalence in both urban and rural areas. Low household income was associated with high prevalence of COPD in urban areas. This is in accordance with previous studies [8, 9, 17]. In rural areas, we found no association between income and COPD after adjusting for confounding factors. This might add to the evidence that houshold income is not a good indicator for SES in rural China as self-reported income may be missclassified whether intentionally or unintentionally in rural Chinese pupulation.

Smoking has been well documented as a major cause of COPD, yet never smokers comprise a substantial proportion of individuals with COPD [18] and previous studies indicated that the prevalence of COPD in never smokers is high in China [19]. It is important to note that the dose response relationship between educational level and COPD were found in both never smokers and smokers, suggesting educational level might be a risk factor for COPD independent of smoking. Possible mechanism explaining the adverse effects of low SES on COPD among never smokers might be poor dietary habits (low in antioxidants and fresh fruit) [20, 21], poor housing conditions [22], more occupational dust exposure and indoor air pollution from biomass combustion in low SES group [23].

Educational level and household income are two imperfect indicators for socioeconomic status. Highest educational level is usually achieved in young adulthood and income earning generally begin after education is completed. The subjects included in our sample were aged 15-69 years and some of them have not achieved the highest educational level yet. Thus educational level may not reflect changes in social status that occur after schooling ends. The same association we found in the subgroup aged 30-69 years confirmed the dose response association between educational level and COPD. As features of asthma can overlap with COPD to some extent, we repeated the analysis excluding those with self reported asthma to avoid potential misclassification. Exclusion of the subgroup did not alter the magnitude of OR for the association.

We used self reported physician diagnosis as definition to COPD and the reporting relied on level of knowledge of the terms COPD, chronic bronchitis and emphysema in the population. In urban areas, COPD (known as 'Man Zu Fei' in Chinese), chronic bronchitis and emphysema are common terms, especially among the patients. Chronic bronchitis and emphysema are more widely used terms when the doctors announce a diagnosis of COPD in rural areas. The estimation of self-reported COPD prevalence is therefore reliable since it's less likely that the subjects would misreport whether they had physician diagnosed COPD.

The study population included in the analysis represents the largest study estimating COPD prevalence and addressing the relation between socioeconomic status and COPD in China. Subjects were randomly selected from all 31 provinces, autonomous regions and municipalities in China and covered a wide range of age group and variable socioeconomic status. We were able to examine the association in never smokers separately without the major confounding factor of active smoking.

One limitation of the study is its cross-sectional nature. We cannot infer temporal association between SES and COPD. It is difficult to know whether low income is the result of COPD or a risk factor, as the disease may lead to disability and job loss which directly cause decrease of the income. Yet it's unlikely that COPD would alter the educational attainment. Secondly, biomass exposure and occupational dust exposure were implicated as potential risk factors for COPD [3, 24, 25]. We were not able to adjust for these two factors. Indoor air pollution from biomass combustion is more common in rural areas and we have examined the relation in urban and rural separately. The evidence of adverse effect of occupational dust exposure on COPD is not consistent compared with smoking [26‐29], and we have adjusted the main risk factor like smoking and passive smoking. It was therefore unlikely that adding these two factors into the model would have yielded significantly different results. Thirdly, our study population has an upper limit of age of 70 years. The prevalence of COPD is therefore relatively underestimated due to the absence of elderly people who is at highest risk of COPD. The lack of confirmation of COPD diagnosis might also limit the interpretation of the finding due to the difficulty of performing spirometry in this large scale nationwide study. As this study is based on the national surveillance focusing on chronic diseases and risk factors, we did not collect data on respiratory symptoms and treatments which were both useful to confirm COPD diagnosis. Furthermore, due to general lower health care infrastructure and lack of medical knowledge among people in the west area, COPD patients had less frequent clinic visit compared to those in the central and east. The COPD prevalence might be underestimated in the west areas. The current result, however, showed the high prevalence of COPD and the burden it has caused in China. Information on childhood SES was not collected in this study, it would be worthwhile to examine the relationship between childhood SES and COPD because some previous studies have shown that childhood SES played an important role in the development of adult respiratory diseases [11, 22, 30]. A large study in the UK analyzed data from 1,204,110 people aged over 35 years old from a general practice database and found that risk of developing COPD was strongly associated with adult height (a marker of SES in early life) [31].

In conclusion, the prevalence of self-reported COPD is high among residents aged 15-69 years in China based on the nationwide sample from 31 provinces, autonomous regions and municipalities. SES was associated with COPD in both urban and rural areas and in both never smokers and smokers. Future research will be needed to better understand the relationship in order to design appropriate interventions to reduce the burden of COPD in China.