Prevalence of chronic obstructive pulmonary disease and chronic bronchitis in eight countries: a systematic review and meta-analysis

Abstract Objective To estimate the prevalence of chronic obstructive pulmonary disease (COPD) and chronic bronchitis in eight countries in South Asia through a systematic review and meta-analysis. Methods We searched MEDLINE® Complete, Web of Science, Embase®, Scopus, CINAHL and reference lists of screened studies for research on the prevalence of COPD and chronic bronchitis in South Asian countries published between January 1990 and February 2021. We used standardized diagnostic criteria for definitions of COPD and chronic bronchitis. Two reviewers undertook study screening, full-text review, quality appraisal and data extraction. Findings Of 1529 studies retrieved, 43 met the inclusion criteria: 32 provided data from India; four from Bangladesh; three from Nepal; two from Pakistan; and two from both India and Sri Lanka. Twenty-six studies used standardized diagnostic definitions and 19 were included in the meta-analysis. The estimated pooled prevalence of COPD was 11.1% (95% confidence interval, CI: 7.4–14.8%), using the Global Initiative for Chronic Obstructive Lung Disease fixed criteria and 8.0% (95% CI: 5.6–10.4%) using the lower limit of normal criteria. The prevalence of COPD was highest in north India (19.4%) and Bangladesh (13.5%) and in men. The estimated pooled prevalence of chronic bronchitis was 5.0% (95% CI: 4.1–6.0%) in India and 3.6% (95% CI: 3.1–4.0%) in Pakistan. Conclusion Included countries have a high prevalence of COPD although it varied by geographical area and study characteristics. Future research in South Asia should use standardized diagnostic criteria to examine the contribution of setting-specific risk factors to inform prevention and control strategies.


Introduction
Chronic obstructive pulmonary disease (COPD) is a common, preventable and treatable disease, with a worldwide prevalence of 10.1% in people aged 40 years or older. 1,2 In 2019, COPD was the third leading cause of deaths globally, contributing to 3.23 million deaths, with most deaths (80%) occurring in low-and middle-income countries. 3,4 A systematic review on COPD showed that estimates of the number of cases of COPD in countries of the World Health Organization South-East Asia Region had increased from 44.5 million to 75.1 million between 1990 and 2010, a 68.8% increase. 5 According to the World Bank, South Asia comprises Afghanistan, Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan and Sri Lanka, and is home to a quarter of the global population. 6 The area is currently undergoing a demographic transition, because of ageing and increased life expectancy. 7 According to the 2017 Global Burden of Disease study, despite a lower prevalence of COPD in South Asia, the attributable morbidity and premature mortality due to chronic respiratory diseases was highest in South Asia with COPD being the most common cause of premature deaths among chronic respiratory diseases. 8 The area is also experiencing a change in the burden of risk factors with ambient air pollution becoming a greater risk due to rapid economic development. 8 While systematic reviews on the prevalence of COPD from Latin America 9 and sub-Saharan Africa 10 have been published previously, literature on current prevalence estimates of COPD and its common risk factors in South Asia is scarce. 11 Notably, few published data exist on the rural-urban, sex, and within and between country differences in the prevalence of COPD. Relevant and timely information on the prevalence of COPD in the area is crucial to inform, develop and implement context-appropriate policies and programmes for its prevention and control, in a setting where the burden is rising. 8 In COPD, airflow in and out of the lungs is limited due to chronic inflammation and narrowing of airways, which is a result of repeated and long-term exposure of the respiratory tract to noxious stimuli (tobacco smoke, indoor air pollution and repeated respiratory tract infections during childhood). 4 COPD presents as chronic bronchitis and emphysema. 4 However, variation in the underlying pathology and clinical presentation, and overlap with the symptoms of asthma and other chronic lung diseases often pose a challenge to making an accurate diagnosis of COPD. 12 Moreover, the gold standard diagnosis of COPD, which needs evaluation of lung air volumes using a spirometer, requires skilled personnel and quality assurance measures, which makes obtaining accurate data more challenging, especially in low-and middle-income countries where a skilled workforce is limited. 1,12 A 2012 systematic review reported limitations in estimates of the prevalence of COPD in India due to a lack of data, concerns about the quality of studies, inconsistencies in study settings and population characteristics. 11 The availability of more recently published studies provides an opportunity to conduct a systematic review to obtain up-to-date estimates of the prevalence of COPD. These data can be used to inform policy-makers when planning and implementing populationlevel risk mitigation strategies to control the rising burden of COPD in South Asia.

Methods Design
We conducted a systematic review and meta-analysis of peer-reviewed literature to estimate the prevalence of COPD and chronic bronchitis in South Asia, according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. 13 We used the World Bank's classification of countries of the South Asia area: Afghanistan, Bhutan, Bangladesh, India, Maldives, Nepal, Pakistan and Sri Lanka. 6 We used standardized diagnostic criteria for definitions of COPD and chronic bronchitis. We defined COPD as the presence of persistent airflow limitation according to the Global Initiative for Chronic Obstructive Lung Disease, i.e. a post-bronchodilator ratio of forced expiratory volume in one second (FEV1) to the forced vital capacity (FVC), FEV1/FVC < 0.70 (fixed criteria), 1 or post-bronchodilator FEV1/FVC below the lower limit of normal, i.e. the lower fifth centile of values from a reference population. 14 We defined chronic bronchitis as the presence of chronic cough and phlegm according to the criteria of the Medical Research Council in the United Kingdom of Great Britain and Northern Ireland, i.e. cough and sputum production on most days continuously for 3 months for more than 2 consecutive years. 15 We registered the protocol for this systematic review in the international prospective register of systematic reviews (CRD42020206189).

Data sources and searches
We searched MEDLINE® Complete, Web of Science, Embase®, Scopus and CINAHL databases using keywords: chronic obstructive pulmonary disease; COPD; obstructive airway disease; obstructive lung disease; airflow obstruction; chronic bronchitis; emphysema; prevalence; and other related terms. These keywords were combined with terms for individual countries and demonyms: Afghan*; Bangladesh*; Bhutan*; India*; Maldiv*; Nepal*; Pakistan*; Sri Lanka*; and associated global areas -South Asia*; Central Asia* (see data repository for search terms). 16 The study inclusion criteria were (i) community-based studies reporting the prevalence of COPD or chronic bronchitis; (ii) cross-sectional, cohort and case-control design; and (iii) publication date between 1 January 1990 and 28 February 2021. We had no language restrictions. We hand-searched the reference lists of screened studies for additional relevant citations.

Study selection
We used Covidence for management of systematic review citations. 17 Two authors independently screened the study titles and abstracts using the inclusion criteria to identify studies for full-text review. These authors undertook an independent full-text review of shortlisted articles, made the final decision to include or exclude studies from the review, independently assessed the methodological quality of included studies using the Joanna Briggs Institute checklist for prevalence studies, 18 and extracted data from all included studies into a structured data extraction sheet. The following data were extracted from the studies: authors' names; year of publication and data collection; study title and area; method of disease ascertainment; sample size; sampling technique; study participants; residence; sex; age group; prevalence of COPD and/or chronic bronchitis with their 95% confidence intervals (CIs); and associations between risk factors for and prevalence of COPD and/or chronic bronchitis.
We resolved inconsistencies between reviewers in screening, inclusion and exclusion of studies, quality appraisal and data extraction decisions through discussion. In cases of disagreement, the third reviewer made the final decision. Where necessary, we also contacted the authors of some publications for further information about the methods and data.

Data analysis
We included studies that used probability sampling techniques and standard definitions for disease ascertainment for the meta-analyses. We did not include studies conducted before 2000 in the meta-analyses. If more than one published study reported the prevalence from the same data set, we included the prevalence from the most recently published study. However, if the older studies provided more details on prevalence data compared with recently published studies, we included the older study data. In the qualitative summary we described characteristics of the studies and charted patterns of the prevalence data according to area, sex and criteria for COPD diagnosis.
We used Stata, version 16.1 (Stata Corp, College Station, United States of America) for all meta-analyses. We used the Stata metaprop command to estimate the pooled prevalence and generate forest plots of COPD (total, men and women) according to fixed criteria and lower limit of normal criteria. 19 We assessed statistical heterogeneity using the χ 2 test, percentage of variance due to heterogeneity using the I 2 test and estimated standard deviation of prevalence using the τ 2 test. Due to the considerable heterogeneity across studies, we used random-effects models to calculate the pooled prevalence estimates.

Included studies
Our search of electronic databases and reference lists yielded a total of 1529 studies. After removing 493 duplicates, we screened 1036 titles and abstracts for inclusion and reviewed the full texts of 127 studies. Of these studies, 43 met the inclusion criteria and were included in the qualitative summary and 19 were included in the meta-analysis: 11 studies on COPD with the most recent or detailed data on prevalence of COPD and eight on chronic bronchitis that reported the overall prevalence ( Fig. 1).
In India, the prevalence of COPD varied, with rural Srinagar in northern India having the highest overall prevalence reported in adults aged 40 years or older (19.3% according to the fixed criteria; 16.1% according to the lower limit of normal criteria) 21 and the lowest prevalence reported in south India (4.1% according to the fixed criteria; Table 1). 28 A higher prevalence of COPD was also reported in Bangladeshi men from rural Matlab and suburban Kamlapur (22.0% based on fixed criteria; 16.2% based on lower limit of normal criteria; Table 1). 25 COPD prevalence was found to be significantly higher in men than women in four studies, in Delhi 38 and Srinagar 21 according to the fixed criteria, and in Nepal 30 and rural Matlab, Bangladesh 25 according to both spirometry criteria. Only one study in Bangladesh compared rural versus non-rural samples and reported a significantly higher prevalence of COPD in rural dwellers (17.0% by fixed criteria and 12.5% by lower limit of normal criteria) than suburban dwellers (9.9% by fixed criteria and 8.0% by lower limit of normal criteria). 25
Common risk factors associated with COPD and chronic bronchitis were age, smoking, lower socioeconomic status, exposure to environmental tobacco smoke, exposure to biomass fuel, exposure to dust, history of tuberculosis and history of allergy and/or asthma. 20

Discussion
Here we report on the prevalence of COPD and chronic bronchitis in South Asia. A substantial regional variation was seen in the prevalence of COPD and chronic bronchitis, with higher prevalence estimates reported by studies in north India 21 and Bangladesh. 25 While tobacco smoking and indoor air pollution were the most common risk factors assessed for their association with COPD, no population-based studies were found in the area that determined  (. . .continued)

Systematic reviews
Prevalence of COPD, eight countries Prashant Jarhyan et al.
the association of COPD with other important risk factors such as ambient air pollution and occupational hazards. Within-country and betweencountry variations in the prevalence of COPD have been reported previously due to differences in the prevalence of risk factors, especially tobacco smoking. 63,64 The higher prevalence of COPD in north India (Kashmir) was mainly ascribed to tobacco smoking using traditional hookahs and higher exposure to indoor air pollution. 21 The high prevalence of COPD in Bangladesh was also attributed to the high prevalence of tobacco smoking, particularly among men. 25 Traditional norms of offering smoking products, low awareness of the harmful effects of smoking among low-income groups and people living in rural areas, suboptimal implementation of tobacco control measures and limited access to cessation services may account for high prevalence estimates in these areas. 65 Thus, these relevant regional, sociocultural and economic factors need to be considered while planning strategies to reduce smoking, decrease the COPD burden and improve population lung health. 65 Indoor air pollution is one of the main causes of COPD, especially in South Asian women. 66 Although many governments have scaled up access to cleaner cooking fuels, the reach and change to cleaner fuels is suboptimal. Furthermore, the effect of this transition on the COPD burden, especially among women in South Asia, remains unclear and warrants further evaluation. 67 More specific approaches are required to understand the role of various indoor air pollutants, such as particulate matter, nitrogen dioxide, carbon monoxide, sulfur oxides, polycyclic organic matter and formaldehyde which are produced by combustion of biomass fuels, in the development of COPD and chronic bronchitis. 68,69 Ambient air pollution is one of the main risk factors for COPD mortality and disability-adjusted life years lost, with the highest burden reported in South Asia. 8,70 In 2015, Bangladesh, India and Nepal had the highest burden of particulate matter 2.5 (PM 2.5 ) and Bangladesh, India and Pakistan had the highest increase in the ozone levels and the highest mortality due to ambient PM 2.5 observed. 71 Increased air pollution aggravates COPD symptoms, known as COPD exacerbations, and in- creases hospitalizations and mortality. 72 Although a few studies from large cities in India have shown the short-term effect of increased ambient air pollution and increased hospital visits due to respiratory problems, 73,74 the long-term effects of extended exposure to ambient air pollution and its effects on lung function, morbidity and mortality need to be studied. 69,75 Educating health-care providers and patients about the adverse health effects of ambient air pollution and simple measures that can be taken to reduce exposure, such as avoiding going out during periods of high pollution and wearing masks outdoors, is needed. 76 Few studies have assessed the association between COPD and respiratory infections in South Asia, such as lower respiratory tract infections and tuberculosis. 21,24,30 Given that many people in the area have been affected by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), 77 the long-term effects of this virus on the burden of COPD and other chronic respiratory diseases need to be assessed. Poor lung function associated with SARS-COV-2 and respiratory infections also requires monitoring to assess their long-term effects and the attributable risk for the development of COPD.
We also found that none of the included studies had assessed the presence of chronic bronchitis with airway obstruction in COPD. Chronic bronchitis in people with airway obstruction is associated with more severe disease, poor general health status, greater limitations on physical activity and higher mortality. 78 Therefore, COPD should be assessed together with the presence of chronic bronchitis in both clinical and research settings.
In our study, we tried to assess the current prevalence of COPD in South Asian countries. The strengths of our study include selection of recent studies, a comprehensive literature search using explicit definitions for COPD and chronic bronchitis according to international guidelines and a comprehensive quality assessment of studies selected for the review.
Our study has some limitations. First, because few studies in the area were available, our pooled prevalence estimates of COPD cannot be generalized to all South Asian countries. Only one study each from Bangladesh and India assessed the prevalence of COPD in urban areas and the only study from Nepal   80 while the ambient air pollution has increased in South Asia. 81 These limitations highlight the lack of research to determine the accurate burden of COPD in the area and the need for large population-based studies with rigorous methods to generate accurate prevalence estimates in various population subgroups as well as attributable risks of common risk factors, including outdoor air pollution, occupational exposure and infections. Such data are important to inform the development and implementation of context-relevant policies and programmes for reducing the increasing burden of COPD and its risk factors. Estimation of the prevalence of COPD is challenging for researchers in low-and middle-income countries. As a result, there is likely a large burden of undetected COPD. COPD is a complex disease with varied presentations and new phenotypes being identified. 82,83 Conducting post-bronchodilation spirometry, which is required for the confirmation of a diagnosis of COPD, is resource intensive and requires a high   Some studies have suggested screening strategies for early detection of COPD. 84,85 However, COPD screening needs further evaluation and feasibil-ity studies and should be supported by strengthening of public health infrastructure for the confirmation of diagnosis, early initiation of pharmacological and non-pharmacological treatment including pulmonary rehabilitation. 85 In conclusion, given the paucity of studies on the current burden of COPD and its risk factors in most South Asian countries, future research in these countries should ensure that standardized diagnostic criteria are used to examine the contribution of exposure to contextrelevant risk factors to inform COPD prevention and control policies.

Systematic reviews
Prevalence of COPD, eight countries Prashant Jarhyan et al.

Systematic reviews
Prevalence of COPD, eight countries Prashant Jarhyan et al.