Published ahead of print on June 15, 2007, doi:10.1164/rccm.200612-1749OC
© 2007 American Thoracic Society doi: 10.1164/rccm.200612-1749OC
Prevalence of Chronic Obstructive Pulmonary Disease in ChinaA Large, Population-based Survey1 Guangzhou Institute of Respiratory Diseases, The First Affiliated Hospital, Guangzhou Medical College, Guangzhou, Guangdong, China; 2 Beijing Institute of Respiratory Medicine, Beijing Chaoyang Hospital, Capital University of Medical Sciences, Beijing, China; 3 The Third Hospital, Peking University, Beijing, China; 4 The Shenyang Military General Hospital, Shenyang, Liaoning, China; 5 The First Affiliated Hospital, China Medical University, Shenyang, Liaoning, China; 6 Ruijin Hospital, Shanghai Jiaotong University, Shanghai, China; 7 The General Hospital, Tianjin Medical University, Tianjin, China; 8 Xinqiao Hospital, The Third Military Medical University, Chongqing, China; 9 Xijing Hospital, The Fourth Military Medical University, Xi'an, Shanxi, China; 10 The First Affiliated Hospital of Jinan University, Guangzhou, Guangdong, China; 11 The First Municipal People Hospital of Shaoguan, Shaoguan, Guangdong, China; 12 The Second Hospital of Liwan District of Guangzhou, Guangzhou, Guangdong, China; and 13 Department of Epidemiology, Guangzhou Medical College, Guangzhou, Guangdong, China Correspondence and requests for reprints should be addressed to Pixin Ran, Ph.D., Guangzhou Institute of Respiratory Diseases, The First Affiliated Hospital, Guangzhou Medical College, 151 Yanjiang Road, Guangzhou, Guangdong, 510120, China. E-mail: pxran{at}gzhmc.edu.cn
Rationale: The prevalence of chronic obstructive pulmonary disease (COPD) in China is largely unknown. Objectives: To obtain the COPD prevalence in China through a large-population, spirometry-based, cross-sectional survey of COPD. Methods: Urban and rural population-based cluster samples were randomly selected from seven provinces/cities. All residents 40 years of age or older in the selected clusters were interviewed with a standardized questionnaire revised from the international BOLD (Burden of Obstructive Lung Diseases) study. Spirometry was performed on all eligible participants. Patients with airflow limitation (FEV1/FVC < 0.70) were further examined by post-bronchodilator spirometry, chest radiograph, and electrocardiogram. Post-bronchodilator FEV1/FVC of less than 70% was defined as the diagnostic criterion of COPD. Measurements and Main Results: Among 25,627 sampling subjects, 20,245 participants completed the questionnaire and spirometry (response rate, 79.0%). The overall prevalence of COPD was 8.2% (men, 12.4%; women, 5.1%). The prevalence of COPD was significantly higher in rural residents, elderly patients, smokers, in those with lower body mass index, less education, and poor ventilation in the kitchen, in those who were exposed to occupational dusts or biomass fuels, and in those with pulmonary problems in childhood and family history of pulmonary diseases. Among the patients who had COPD, 35.3% were asymptomatic; only 35.1% reported lifetime diagnosis of bronchitis, emphysema, or other COPD; and only 6.5% have been tested with spirometry. Conclusions: COPD is prevalent in individuals 40 years of age or older in China.
Key Words: chronic obstructive pulmonary disease • prevalence • epidemiology • cross-sectional studies • GOLD
Chronic obstructive pulmonary disease (COPD) is a disease state characterized by airflow limitation that is not fully reversible (1). As a major public health problem, COPD is the fourth leading cause of morbidity and mortality in the United States, with direct and indirect medical costs up to $24 billion in 1993 (2). In China, respiratory diseases (of which COPD is a major component) are the third leading cause of death in rural areas and the fourth leading cause of death in urban areas, accounting for 1 million deaths and over 5 million disabilities each year. According to an estimation by the World Health Organization (WHO), COPD ranks first among the burdens of diseases in China and is predicted to rank as the fifth burden of diseases in the world by 2020 (3). Earlier surveys have yielded varied global prevalence of COPD ranging from 0.23 to 18.3% because of disagreements on diagnostic criteria and epidemiologic study designs (4). In China, a previous study reported the prevalence of COPD in the northern and the central rural regions of China (Liaoning, Beijing, and Hubei province) to be 3% for individuals 15 years of age or older (5), in which only subjects with respiratory symptoms or smoking habits were recruited to receive lung function testing. Another study estimated that the prevalence of COPD was 5.9% in adults 35 years of age or older in Nanjing, China (6), based on self-reported physician diagnosis rather than spirometry. Thus, asymptomatic or never-smoking patients with COPD could have been overlooked, resulting in underestimation of the prevalence of this disease. To provide an overall prevalence of COPD in China, the present study, one of the key research projects funded by the 10th National Five-year Development Plan of China and a pertaining parallel study of Burden of Obstructive Lung Disease (BOLD), was conducted in seven provinces/cities of China between September 2002 and September 2004. To the best of our knowledge, this is the first large-scale, population-based epidemiologic study on COPD prevalence in China. Some results of this study have been previously orally reported in the form of an abstract at the 11th congress of the Asian Pacific Society of Respirology (7).
Study Design and Subjects A population-based, cross-sectional survey of COPD was conducted in seven provinces/cities in China: Beijing, Tianjin, and Liaoning (northern China); Shanghai (eastern China); Guangdong (southern China); and Shanxi and Chongqing (western China), covering a wide range of geographic areas in China and a total population of more than 230 million. In each of these provinces/cities, we used a multistage cluster sampling strategy, in which the first stage was the stratification of census tracts. Due to the large socioeconomic differences between rural and urban regions, an urban district and a rural county were randomly selected in each province/city. As the second step of the sampling process, we randomly selected an urban street or a rural township from each of the selected urban districts and rural counties. Finally, we used a randomized cluster sampling method to select urban communities or villages (as cluster units). The number of selected clusters depended on the size of population in the communities or villages. For example, in a 20-community urban district with the average number of subjects 40 years of age or older in each community estimated to be about 300, five communities were to be selected randomly to meet the designed sample size of this study, which is 1,450. (The calculation of this sample size can be found in the online supplement). In the selected sample clusters, we recruited all residents 40 years of age or older according to the latest census by local police stations where residential registry data were kept electronically. Those temporarily out of reach were given a home interview on a later occasion. The study protocol was approved by the institutional review board of each participating center.
Data Collection
Spirometry and diagnostic criteria
Questionnaire
Quality control
Statistical Analysis
Sample Demographics Of 25,627 subjects sampled from seven provinces/cities, 21,270 (83.0%) were interviewed. Among 21,270 interviewees, 315 were not eligible for spirometry, 532 failed to finish the spirometry, and 20,245 completed acceptable spirometry (grades A, B, or C) and questionnaires, yielding a response rate of 79.0% (20,245/25,627) (Figure 1) (the response rate by sites is shown in Table E1 of the online supplement). The reasons for nonresponse included refusals, contact failures, spirometry ineligibility, and failed attempts. Despite differences in age, sex, and urban/rural between responders and nonresponders, there was no significant difference in smoking status. The pattern of age and sex distribution in responders was similar to that of the whole population (Table 1). The pattern of age and sex distribution among responders, sample, and the sample frame are shown in Figure E1. The mean age of the total study population was 56.7 (SD, 11.7) years, with a range of 40 to 99 years. The percentages of current smokers, ex-smokers, and total smokers in the study population were 29.2, 9.2, and 38.4%, respectively. Smokers accounted for 74.3% of the male subjects and 11.3% of the female subjects. Those with more than 15 pack-years of smoking history added up to 25.9% of study subjects. The smoking rates of study population by urban/rural, sex, and age groups are shown in Table 2.
Prevalence of COPD The overall prevalence of COPD in China was 8.2% (95% CI, 7.9–8.6) according to GOLD diagnostic criteria (Table 3). Standardized to the world population in 2000–2005 and to the United States population in 2000, the prevalence rates of COPD were 7.8% (95% CI, 7.4–8.1) and 8.5% (95% CI, 8.1–8.9), respectively.
The crude prevalence of COPD was the highest in Chongqing (13.7%) and lowest in Shanghai (3.9%) among urban areas and was highest in Guangdong (12.0%) and lowest in Liaoning (6.8%) and Shangxi (6.9%) among rural areas. The COPD prevalence was significantly higher in rural (8.8%) areas compared with urban areas (7.8%) (P = 0.007) (Table 3). According to modified GOLD diagnostic criteria (prebronchodilator FEV1/FVC <70%), the overall prevalence of COPD in China was 11.5% (11.0–11.9%) (Table E2). The prevalence of modified restrictive disorder was 8.5% (Table E3).
Stratification Analysis on COPD Prevalence
Multivariate logistic regression analyses showed that smoking, pulmonary problems in childhood, family history of respiratory diseases, male sex, low education level, aging, lower BMI, poor ventilation in the kitchen, and exposure to biomass and occupational dusts/gases/fumes are associated with COPD. Adjusted ORs (95% CI) of these potential risk factors are shown in Table 4.
Prevalence of COPD According to GOLD Severity Strata The total prevalence of formerly proposed stage 0 defined by chronic cough or phlegm production with an FEV1/FVC of 0.70 or greater was 16.1% (95% CI, 15.6–16.6) and was higher in male patients than in female patients (20.0 vs. 13.2%; P < 0.001) (Table 5).
Respiratory Symptoms and Lifetime Diagnosis of COPD The prevalence of cough, phlegm, wheezing, and breathlessness respiratory symptoms in patients with COPD was 44.0%, 39.5%, 29.3%, and 48.6%, respectively (Table 6). A total of 64.7% of patients with COPD exhibited at least one of above respiratory symptoms, whereas more than one-third (35.3%) of patients with COPD were asymptomatic. The frequency of these respiratory symptoms increased with GOLD severity stages (all Ptrend < 0.001). Only 35.1% of patients with COPD had ever been "diagnosed" with emphysema, asthma, bronchitis, or COPD. Of all patients with COPD, only 6.5% had ever been tested by lung function tests (spirometry).
In this large-scale population and spirometry-based, cross-sectional survey in China, the prevalence of COPD (8.2%) in people 40 years of age or older indicated COPD as a more serious public health problem in the Chinese than expected in previous studies (5, 6). The finding was close to the WHO model–estimated prevalence (6.2%) in Chinese population aged 30 years or older (15) and consistent with an expected range of 4% to 10% from international review of COPD prevalence based on spirometry (4). Compared with studies using the same diagnostic criteria (GOLD criteria) in the same age groups (adults  40 yr of age), our prevalence was similar to those in Japan (10.9%) (16), Poland (10.7%) (17), and Mexico City (7.8%) but was lower than those in the other four cities of Latin America (12.1–19.7%) (18) and Salzburg of Austria (26.1%) (19). Prevalence of COPD in other age groups has also been described in Korea (17.2% among subjects above 45 yr of age according to GOLD criteria) (20), Finland (9.4% among adults 20–70 yr of age according to GOLD criteria) (21), and the United States (16.8% in residents 30–80 years of age according to modified GOLD criteria) (22). With regard to other studies, a Greek survey that was limited to smokers and used a different and more rigorous definition of COPD reported a COPD prevalence of 8.4% in people older than 35 years (23), and surveys that followed criteria of the European Respiratory Society (ERS) on spirometry (i.e., FEV1/FVC ratio in percentage predicted < 88% in men or < 89% in women) reported a 9.1% COPD prevalence in adults 40 to 69 years of age in Spain (24) and 13.3% among subjects 35 years of age in England (25). The Northern Ireland study showed a 14.4% prevalence of COPD in subjects 40 to 69 years of age according to British Thoracic Society (BTS) criteria (26). In India, COPD was diagnosed in 4.1% of adults 35 years of age and above according to chronic bronchitis (CB) criteria (cough and expectoration for at least 3 months in a year for 2 consecutive years or more) (27). A report from Italy indicated that the prevalence of airway obstruction for subjects over 46 years was 12.2% (ERS), 28.8% (GOLD), and 57% (ATS) (28). As reported from the Obstructive Lung Disease in Northern Sweden Studies, the prevalence rates of BTS-COPD were mild, 5.3%, moderate, 2.2%, and severe, 0.6% (GOLD-COPD: mild, 8.2%; moderate, 5.3%; severe, 0.7%; and very severe, 0.1%), for subjects over 45 years of age (29). The prevalence of COPD displays a wide range of variation due to differences in epidemiologic methodology, proportions of age and sex, response rate, and diagnostic criteria of COPD used, as described by Haibert (4, 30) and others (28, 29). Even if objective tools such as spirometry were performed in the same population, the prevalence of COPD may remain highly variable across ATS, ERS, GOLD, and BTS criteria (28, 29). Although the results of those studies cannot be directly compared with ours, they all have pinpointed COPD as a global threat to human health. Smoking is a well-documented risk factor that contributes substantially to COPD (5, 6, 14–32). As shown in our study, about two-thirds (61.4%) of patients with COPD, including 81.8% of male patients with COPD and 24.0% of female patients with COPD, were smokers; 13.2% of smokers had COPD, and the risk for COPD increased with the number of cigarettes consumed. In contrast to other sources of data (22, 24), this study revealed higher COPD prevalence in ex-smokers than in current smokers and a higher relative OR for COPD in ex-smokers after adjusting for index of smoking and other variables. This can be explained by the lifetime bias (ex-smokers possibly had lived longer than current smokers) and the fact that majority of Chinese smokers, with poor education in tobacco control and the harmfulness of smoking to health (data not shown), have not been determined to quit smoking unless afflicted with severe diseases or old age. About one-third (38.6%) of patients with COPD were nonsmokers, and the prevalence of COPD in nonsmokers, which was as high as 5.2%, suggested that factors other than smoking exposure might also be involved in COPD. As shown in the present study, the prevalence of COPD was significantly higher in those with lower BMI, less education, exposure to occupational dusts, and pulmonary problems in childhood, all of which are consistent with the findings of other studies (33–37). However, because some degree of recall bias might exist, more rigorous studies on the association between COPD and pulmonary problems in childhood, such as cohort studies, are needed for further evaluation and validation. The cause–effect relationship between BMI and COPD remains to be elucidated. Family history of respiratory diseases was also found to be associated with a higher prevalence of COPD in the present study, suggesting possible involvement of genetic or family-related environmental factors in the development of this disease. The prevalence of COPD in China varied with geographic area, gender, and age, which is consistent with other studies (18, 24). The association between indoor air pollution and COPD is a worldwide concern, especially in developing countries (36–38). The present study also showed a higher prevalence of COPD in subjects with exposure to biomass or with poor ventilation in the kitchen. In addition, there was a higher adjusted OR for COPD in rural areas than urban areas, which may be associated with more biomass exposure, lower socioeconomic status, lower education degree, lower health care standard, and poorer quality of cigarettes consumed in the countryside (data not shown). More than one-third of patients with COPD were asymptomatic in this population study, and nearly two thirds had never been diagnosed before this survey. Similarly, in a Japanese study, only 9.4% of cases with airflow limitation reported a previous diagnosis of COPD (16). This suggests that diagnosis of COPD based on symptoms may not be adequate. The fact that an even smaller percentage (6.5%) of diagnosed subjects had ever been examined with spirometry further indicates the need for the use of spirometry to be strongly encouraged in community clinics. On the other hand, although COPD diagnostic criteria of GOLD are widely accepted, they may yield more false-negative results among younger adults and more false-positive results among older adults (39, 40), which makes it necessary for asymptomatic patients be further investigated in a cohort study.
The present study showed higher prevalence of stage III (1.7%) and IV (0.4%) and lower prevalence of stage I (2.0%) and stage II (3.8%) of COPD in China than that in some of the Latin American cities (18). This could be attributed to the fact that due to lower health care standard and general lack of medical knowledge among people, most of the patients with COPD in China did not visit a doctor until they develop significant symptoms and exacerbations, especially in rural areas where major health care disparities exist. Hence, most diseases would have developed to a more advanced and severe stage by the time the diagnosis is made. The lower proportion of younger subjects in the responders may be another explanation. In addition, subjects with mild obstruction but short expiration times may not meet the FEV/FVC threshold, resulting in a "low" COPD prevalence and a "high" restrictive prevalence. However, in the present study, spirometry tests were performed according to ATS recommendation by well trained spirometric physicians. Subjects were encouraged to inhale as much air as they could and to continue to exhale the air at the end of the maneuver. FEV1 of more than 6 seconds and/or the 15 volume–time curve reached a plateau, which meant there was no change in volume ( One of the limitations in the present study was that data on smoking status were not obtained from all of the nonresponse population due to contact failure, and there were differences in age, sex, and urban/rural between responders and nonresponders. However, because the pattern of age and sex distribution in responders was similar to that of the whole population, the total response rate was relatively high. Moreover, contact failure, the major cause of nonresponse in this study, seems to be a random cause of nonresponse, so some possible selection bias should thus be negligible. Second, the predicted normative values of FEV1 were derived from ECCS equations and adjusted with conversion factors recommended by Zheng and Zhong (10) to minimize variations from ethnic differences. The adjusted conversion factors were obtained by comparing ECCS equations with Chinese equations, which were summarized from a nationwide normal lung function study (41), with 26 participating hospitals and institutions throughout the country and an enrollment of 4,773 Chinese healthy individuals ranging from 15 to 78 years of age. Although other prediction equations are available, larger ethnic differences can arise from these equations (10). Nevertheless, because there is no predicted value of FEV1 for adults over 78 years of age, the COPD diagnosis could not be further identified by the criteria of below fifth percentile of the predicted value, and the severity of COPD would also be affected if there is a systemic reason (e.g., diet or air pollution) responsible for poor lung function. However, according to diagnostic criteria of GOLD, the investigation on COPD prevalence must to be consistent with the study methods of BOLD and comparable with the results reported by BOLD study.
Third, the possibility of misclassifying some subjects with asthma and other diseases as patients with COPD could not be eliminated. In addition, a recall bias was inevitable, like in all cross-sectional studies. Due to the impracticality of testing with the bronchodilator for all subjects in this large-scale study, only those with airflow limitation received a dose of Ventolin. This would bias toward a lower prevalence because some people reportedly move from unobstructed to obstructed after a bronchodilator dose (because their FVC improves more significantly than FEV1 or due to some type of paradoxical response), as reported by Johannessen and colleagues (42). To evaluate the bias and to compare our study with other studies, the prevalence of COPD according to "modified GOLD criteria" (11.5%; see Table E2), the modified restrictive disorder (8.5%; see Table E3), and the changed lung function from prebronchodilator to post-bronchodilator for subjects with a low ratio of prebronchodilator FEV1/FVC (<70%) (see Figure E2S) are shown in the online supplement. As shown in Figure E2, for subjects with a low ratio of prebronchodilator FEV1/FVC (<70%), FEV1 and FVC increased by a mean of 50 ml and 40 ml, respectively, after a post-bronchodilator spirometric testing, in alignment with Johannessen and colleagues' findings (42). Also consistent with the findings (3.1%) of Johannessen and colleages (42) was our finding that 3.3% of population had a low ratio of FEV1/FVC (FEV1/FVC <70%) before bronchodilation and a normal ratio (FEV1/FVC In conclusion, this nationwide, population-based and spirometry-based, cross-sectional survey showed that the prevalence of COPD in China is 8.2% in people 40 years of age or older according to GOLD criteria and 11.5% according to modified GOLD criteria, which is higher than previously expected. Our results highlight COPD as a major public health problem in China and call for more research to be directed toward preventive measures and efforts.
The authors thank Professor Sonia Buist (Oregon Health and Science University, Portland, OR), Robert Crapo (LDS Hospital, Pulmonary Division, Salt Lake City, UT), and the BOLD committee for providing technical training and questionnaire. The authors thank all investigators and local administrations for their great assistance in field surveying in this study: Dr. Zeng Guangqiao, M.D. (Guangzhou Institute of Respiratory Disease, Guangzhou Medical College, China), and Jian Wang, M.D., Ph.D. (Division of Pulmonary and Critical Care Medicine, Johns Hopkins University), for assistance in linguistic considerations; Ms. Mei Jiang (Guangzhou Institute of Respiratory Disease, Guangzhou Medical College, China) for assistance in statistical considerations, and Prof. Qingyi Wei (Department of Epidemiology, The University Texas MD Anderson Cancer Center, Houston, TX) for scientific editing.
Supported by Chinese Central Government key research projects of the 10th national 5-year development plan grants 2001BA703B03(A) (P.R.) and in part by Guangdong key research project grant B30301 (P.R.). This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Originally Published in Press as DOI: 10.1164/rccm.200612-1749OC on June 15, 2007 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form December 2, 2006; accepted in final form June 13, 2007
Related articles in AJRCCM:
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
0.03 L) for at least 1 second. Thus, in our study, good quality control in spirometry was assured, and "high" prevalence of modified restrictive disorder was not observed in some of the areas with "low" prevalence of modified-GOLD COPD (see 
