Burden of chronic obstructive pulmonary disease among Indian adults: systematic review and meta‑analysis

Abstract

Background

Chronic Obstructive Pulmonary Disease (COPD) is a long-standing respiratory illness marked by ongoing airflow obstruction and inflammation. It continues to be a major contributor to global disease, and death, with low- and middle-income countries (LMICs) experiencing a disproportionate impact. India, as one of the largest LMICs, plays a significant role in global COPD-related mortality and disability-adjusted life years (DALYs). In India, COPD continues to be underrecognized owing to limited spirometry availability, inconsistent diagnostic approaches, and weak surveillance systems. Previous prevalence estimates are both outdated and methodologically inconsistent, while the COVID-19 pandemic may have further shifted disease trends. This systematic review and meta-analysis seeks to bridge these gaps by delivering current, standardized, and comprehensive prevalence data.

Objective

To estimate the pooled prevalence of spirometry-confirmed COPD among Indian adults and identify key demographic and environmental correlates through a systematic review and meta-analysis of observational studies.

Methods

This systematic review and meta-analysis aimed to determine the prevalence of spirometry-confirmed COPD among Indian adults. The study was registered in PROSPERO (CRD420251140678) and conducted in accordance with PRISMA guidelines. Literature searches were carried out in PubMed, EMBASE, Scopus, and Web of Science up to June 9, 2025, using relevant MeSH terms and keywords on COPD, prevalence, and India. Eligible studies included observational designs reporting spirometry-based COPD prevalence in adults; studies relying on non-spirometry diagnosis, qualitative designs, interventions, or non-English publications were excluded. Three reviewers independently screened records, extracted study and population data, and evaluated methodological quality using the Joanna Briggs Institute (JBI) checklist. Pooled prevalence was calculated using a random-effects model. Heterogeneity was assessed with I² and Cochran’s Q, complemented by Baujat and Galbraith plots. Subgroup and sensitivity analyses examined variations by diagnostic criteria, demographics, and exposures, while publication bias was tested using funnel plots, Egger’s and Begg’s methods, and trim-and-fill analysis.

Results

Twenty-three studies comprising 27,319 Indian adults were included. The pooled prevalence of COPD was 13% (95% CI: 9%–18%), with substantial heterogeneity (I² = 99.8%). Higher prevalence was observed among smokers (37%), elderly adults (≥ 60 years: 27%), males (16%), and biomass fuel users (8%). Studies using GOLD criteria reported a higher prevalence (15%) than those using FEV₁/FVC < LLN (10%). Hospital-based studies showed a greater prevalence (27%) than community-based ones (12%). Regional variation was notable, with North India reporting the highest prevalence (19%) and West India the lowest (7%). Sensitivity analyses confirmed the robustness of findings; publication bias was minimal and did not significantly affect pooled estimates.

Conclusion

COPD remains a significant and underrecognized public health challenge in India. As all included studies were appraised as good quality using the JBI tool, the evidence base is strong and supports reliable pooled estimates. Therefore, our conclusions emphasize the importance of routine spirometry-based screening, targeted interventions for high-risk groups, and integration of COPD surveillance into India’s NCD framework, while reinforcing gender-sensitive strategies and clean fuel initiatives as evidence-based measures to reduce disease burden and guide policy planning.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12890-026-04134-0.

Introduction

Chronic Obstructive Pulmonary Disease (COPD) is a progressive, non-reversible respiratory condition characterized by persistent airflow limitation and chronic inflammation of the airways. It encompasses chronic bronchitis and emphysema, and is primarily diagnosed through spirometry [1]. COPD is a primary global health concern, responsible for approximately 213 million cases and 3.5 million deaths in 2021—accounting for nearly 5% of all deaths worldwide. Notably, low- and middle-income countries (LMICs) bear the brunt of this burden, with 90% of COPD-related deaths among individuals under 70 years occurring in these regions [2]. Although age-standardized mortality rates for COPD have declined slightly, its overall impact—reflected in disability-adjusted life years (DALYs)—is increasing, particularly in low- and middle-income countries where widespread exposure to tobacco, air pollution, and occupational hazards persists [3].

As one of the most densely populated LMICs, India carries an outsized portion of the global COPD burden. Data from the Global Burden of Disease Studies (GBD: 2019 and 2021) indicate that India contributes substantially to worldwide COPD-related deaths and DALYs, consistently ranking among the highest-burden nations [3, 4]. The prevalence of COPD in India varies widely across regions, ranging from 0.9% [5] to over 50% [6], depending on diagnostic criteria, population characteristics, and environmental exposures. The National Program for Prevention and Control of Non-Communicable Diseases (NP-NCD) identifies COPD as a key focus within chronic respiratory diseases; however, monitoring and data reporting remain inconsistent. Additionally, the absence of uniform spirometry-based diagnostic practices in community and primary healthcare settings leads to frequent underdiagnosis and misclassification [7].

In India, the epidemiology of COPD is shaped by a multifaceted array of risk factors. Tobacco use, particularly among men, is a major driver, while non-smoking exposures—such as biomass fuel use, indoor pollution, and occupational dust—significantly affect women and rural communities. The presence of comorbid conditions like tuberculosis, cardiovascular disease, and diabetes adds complexity to diagnosis and management. Given the variability in study methodologies, diagnostic approaches, and reporting practices, a comprehensive synthesis of existing evidence is essential to guide effective public health strategies and resource distribution.

The previous systematic review in India was published in 2021, evaluating studies from 2011 to 2020 [8]. The COVID-19 pandemic impacted COPD epidemiology by limiting spirometry testing, lowering reported diagnoses, and changing exacerbation patterns due to infection-control restrictions and reduced healthcare access [9–12]. Additionally, COPD patients faced increased vulnerability and mortality during the pandemic, which may have influenced prevalence estimates and the overall recognition of the disease [13, 14]. To address these uncertainties, this study undertakes a systematic review and meta-analysis of observational research on COPD prevalence among Indian adults. By integrating data from varied regions and population groups, this updated study aims to produce pooled prevalence estimates, examine relevant demographic and clinical correlates, and investigate sources of heterogeneity. The outcomes will offer critical insights for health professionals, researchers, and policymakers to develop targeted strategies, enhance respiratory healthcare systems, and reinforce the prioritization of COPD within India’s non-communicable disease framework.

Methods & design

Study protocol and design

This systematic review and meta-analysis aimed to estimate the burden of COPD diagnosed with spirometry among Indian adults. The study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) [15] guidelines, incorporating an in-depth evaluation of various research assessing COPD prevalence. The review included primary research focusing on Indian adult participants and their reported COPD burden. Additionally, it sought to identify key factors influencing COPD prevalence in this population. The study protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) under registration number CRD420251140678.

Inclusion criteria

Studies were included in the review based on specific criteria. Eligible studies were observational in design. The population of interest was comprised of adult Indian participants, with studies assessing COPD prevalence, which was measured using a standardized diagnostic tool (Spirometry). Only primary research conducted in India was considered. Additionally, studies had to be published in English and could be from any publication year up to June 09, 2025, without restrictions on the start date.

Exclusion criteria

Studies conducted on patients with self-reported or non-spirometry evaluated COPD were excluded. Studies were excluded if they lacked full-text availability, provided inadequate information, or were opinion-based without supporting empirical data. Research outside the scope of the primary outcome was also excluded, including qualitative studies, case reports, case series, letters, unpublished work, study protocols, and conference posters. Prior systematic reviews and interventional studies were not included. Additionally, studies published in languages other than English were excluded from the review. (Supplementary Table S1).

Information sources and search strategy

A thorough literature search was conducted across four major databases: EMBASE, PubMed, Scopus, and Web of Science, covering research published until June 09, 2025. The search employed a structured approach, incorporating Medical Subject Headings (MeSH) and relevant keywords, such as COPD, Prevalence, and Epidemiology. Boolean operators (AND, OR) were used to refine searches. An example search query included COPD AND (Burden OR Prevalence OR Epidemiology) AND India. Further search refinements were performed to ensure the most relevant articles were included. A detailed search strategy is available in Supplementary Table S2. Reference lists were examined manually to ensure comprehensive coverage, and citations were reviewed. The literature search was carried out independently and double-blinded by three authors (MS, NK, and MS).

Study selection

All retrieved citations were imported into the Rayyan software in 2025, where duplicate records were systematically removed. Three independent researchers (MS, NK, and MS) screened the titles and abstracts of the remaining articles to identify those eligible for full-text review. Disagreements regarding inclusion were discussed and resolved by consensus. If necessary, the co-author (PH) provided additional input. Full-text articles were assessed based on predefined inclusion criteria. Any missing information required for the eligibility assessment was obtained through consultation with the full research team. Reasons for article exclusion were documented at each stage of selection, and any unresolved disagreements were addressed by a third reviewer (PH) (Fig. 1).

Fig. 1.

PRISMA Chart

Data extraction

Data extraction was conducted using a standardized Microsoft Excel spreadsheet to ensure consistency. It included study characteristics (first author’s name, year of publication, year of study, sample size, study location, study design, and study setting), population characteristics (mean age, type of cooking fuel usage), and COPD assessment details (diagnostic tools used, COPD prevalence, and severity levels). MS, NK, and MS performed the extraction process independently, with any discrepancies resolved through discussion and mutual agreement.

Quality assessment and risk of bias [16, 17]

The 9-item Joanna Briggs Institute (JBI) Critical Appraisal Tool for Prevalence Studies was used to assess the included studies' methodological quality and risk of bias. Based on their quality scores, studies were categorized as low quality (1–3 points; high risk of bias), moderate quality (4–6 points; moderate risk of bias), or high quality (7–9 points; low risk of bias). Three researchers conducted the quality assessment independently (MS, NK, and MS), ensuring objectivity and consistency in the evaluation process.

Statistical analysis

The pooled prevalence/burden of COPD was calculated using a random-effects model to accommodate inter-study variability. Cochran's Q test and the I² statistic were employed to assess heterogeneity among the included studies. Subgroup analyses were performed to explore prevalence differences based on COPD grade, diagnostic tool, study setting, sampling method, zones, gender, age group, exposure to cooking fuel, and study period. To assess the robustness of the results, heterogeneity and outlier analysis were conducted using the Baujat plot and the Galbraith plot, followed by sensitivity analyses conducted using leave-one-out analysis, identifying studies that contributed disproportionately to heterogeneity. Publication bias was examined using funnel plots, Egger’s and Begg’s regression tests. If publication bias was detected, trim and fill analysis was conducted to show the imputed studies with variation in the pooled effect size. Additionally, meta-regression analyses were performed to investigate the impact of covariates such as mean age and sample size on COPD prevalence. The findings were presented visually using bubble plots. All statistical analyses were conducted using STATA 17, with statistical significance at p < 0.05.

Results

Figure 1 documents the study selection process following the PRISMA guidelines. A total of 1223 studies were identified through searches in PubMed (302), Web of Science (225), Scopus (680), and Embase (16). After removing 825 duplicate records, 398 studies remained for screening. Titles and abstracts were reviewed, resulting in the exclusion of 353 studies that did not meet the inclusion criteria. Full-text retrieval was attempted for 45 studies; 43 of them were accessible further assessed for eligibility, 5 were excluded for the wrong study population, 4 studies were excluded for the wrong study outcome, and 11 studies were excluded for COPD measurement by a non-spirometry test. After applying all eligibility criteria, 23 studies were included in the final meta-analysis.

This meta-analysis included 23 studies published between 2009 and 2024, with a total of 27,319 Indian adults, among whom 2631 were diagnosed with COPD. In the case of multicenter studies or studies with different populations from which COPD was documented, these were segregated into separate studies for better documentation and analysis. In an observational study by Mukherjee et al. (2014), this analysis documented populations exposed to biomass fuel and LPG as separate studies. These two distinct groups are treated as individual studies which repot outcome of interest (prevalence of COPD). In both of these studies participants were recruited from community settings and reflected the general population, making them appropriate surrogates for estimating COPD prevalence in disease-free populations and thereby supporting our inclusion criteria. A multicentric study by Triest et al. (2019) documented results from Kashmir, Mumbai, and Pune separately; while Burney et al. (2020) reported studies from Mumbai, Pune, Kashmir, and Karnataka separately, with different data collection periods and sampling techniques. We have segregated these studies for better representation in this meta-analysis. Thus, the total number of studies was 28.

Included study sample sizes (total number of participants) ranged from 104 (Chandra et al., 2018 [18]) to 4171 participants (Chhabra et al., 2010 [19]). All of the studies were cross-sectional except Mukherjee et al. (Case-control study). Most studies (25/18 = 89.28%) were community-based, whereas (3/28) 10.72% were conducted in hospital or facility settings. Geographically, the studies covered all the Indian zones except the North-East. North (10/28 = 35.71%) Zone documented the highest representation, followed by the South (9/28 = 32.14%) zone. Central Zone (1/28 = 3.57%) had the least representation. At the same time, the West Zone (5/28) and the East Zone (3/28) represented 17.86% and 10.72% of the total studies.

At the state level, Karnataka and Maharashtra had the highest number of studies (5 studies each). Twenty studies used the diagnosis of COPD as per GOLD criteria (71.43%), with eight studies using FEV1/FVC < LLN criteria (28.57%). Eighteen studies reported COPD among females, and 15 studies reported COPD among males. The included studies were published over a 15-year period (2009–2024), with the highest number of studies published from 2009–2019 (21/28 studies 75.00%) (Table 1).

Table 1.

Characteristics of studies included in this meta‑analysis

Author (Last name) et al	Year of publication	Year of Study	Study State	Type of study	Study setting (Hospital-based/Community-based)	Sampling Technique	Study Population	Residence	Rural(%)	Urban (%)	Mean Age (Years)	SD Age (Years)	Total Sample size	Diagnostic tool/method used to assess COPD	No of COPD	Prevalence of COPD (%)
Mahesh et al. [20]	2009	2009	Karnataka	Cross-sectional	Community	Consecutive	All adults above 40 years	Rural	100	NR	43.09	17.47	900	Spirometry: GOLD	64	7.11
Chhabra et al. [19]	2010	2010	Delhi	Cross-sectional	Community	Stratified random	Permanent residents of Delhi above 18 years	Urban	NR	100	NR	NR	4171	Spirometry: GOLD	85	2.04
Johnson et al. [21]	2011	2007	Tamil Nadu	Cross-sectional	Community	Cluster random	Non-smoking women aged above 30 years	Rural	100	NR	NR	NR	900	Spirometry: GOLD	22	2.44
Ramadoss et al. [6]	2013	2011–2012	Kerala	Cross-sectional	Hospital/Facility	Consecutive	Adults aged > = 18 years	Not mentioned explicitly	NR	NR	NR	NR	550	Spirometry: GOLD	275	50.00
Parasuramalu et al. [22]	2014	2008	Karnataka	Cross-sectional	Community	Cluster random	Subjects above 35 yrs age	Rural	NR	NR	47.39	10.28	3120	Spirometry: GOLD	61	1.96
Mukherjee et al. [5]	2014	2014	West Bengal	Case-Control	Community	Purposive	Pre-menopausal non-smoking women using biomass as fuel in the age group 23–43 years	Rural	100	NR	39	NR	681	Spirometry: GOLD	31	4.55
Mukherjee et al. [5]	2014	2014	West Bengal	Case-Control	Community	Purposive	Pre-menopausal non-smoking women using LPG as fuel in the age group 23–43 years	Rural	100	NR	42	NR	438	Spirometry: GOLD	4	0.91
Banjare et al. [23]	2014	2011–2012	Odisha	Cross-sectional	Community	Multistage random sampling	Elderly aged 60 years and above	Rural	100	NR	NR	NR	310	Spirometry: GOLD	62	20.00
Singh et al. [24]	2014	2012–2013	Haryana	Cross-sectional	Community	Purposive	Females > = 35 years	Rural	100	NR	NR	NR	1027	Spirometry: GOLD	52	5.06
Pinto et al. [25]	2015	2014–2015	Maharashtra	Cross-sectional	Hospital/Facility	Consecutive	Adults aged > = 18 years	Not mentioned explicitly	NR	NR	NR	NR	326	Spirometry: GOLD	38	11.66
Gupta et al. [26]	2016	2015	Uttar Pradesh	Cross-sectional	Community	Consecutive	People above 60 years	Urban	NR	NR	66	NR	1493	Spirometry: GOLD	159	10.65
Mahishale et al. [27]	2016	2013–2014	Karnataka	Cross-sectional	Hospital/Facility	Consecutive	Women > 40 years age with > 10 years exposure to biomass fuel	Both	71	29	NR	NR	2868	Spirometry: GOLD	529	18.44
Koul et al. [28]	2016	2010	Kashmir	Cross-sectional	Community	Simple random	Adults aged > = 40 years	Urban	NR	100	NR	NR	757	Spirometry: GOLD	146	19.29
Sinha et al. [29]	2017	2013	Delhi	Cross-sectional	Community	Systematic random	Adults > 30 years	Urban	NR	100	46	13	1203	Spirometry: GOLD	122	10.14
Townend et al. [30]	2017	2016	Kashmir	Cross-sectional	Community	Simple random	Adults aged > = 40 years	Not mentioned explicitly	NR	NR	NR	NR	738	Spirometry: FEV1/FVC < LLN	118	15.99
Chandra et al. [18]	2018	2016–2017	Delhi	Cross-sectional	Community	Consecutive	Sewage workers > = 18 years	Not mentioned explicitly	NR	NR	50.71	8.43	104	Spirometry: GOLD	51	49.04
Mahesh et al. [31]	2018	2006–2010	Karnataka	Cross-sectional	Community	Simple random	Adults aged 35 to 80 years	Rural	100	NR	49.9	10.5	1085	Spirometry: GOLD	10	0.92
Kumar et al. [32]	2019	2017	Haryana	Cross-sectional	Community	Simple random	Elderly aged 60 years and above	Rural	100	NR	68.1	6.6	392	Spirometry: GOLD	168	42.86
Triest et al. [33]	2019	2019	Kashmir	Cross-sectional	Community	Simple random	Adults aged > = 40 years	Not mentioned explicitly	NR	NR	51.1	8.9	739	Spirometry: FEV1/FVC < LLN	121	16.37
Triest et al. [33]	2019	2019	Maharashtra (Mumbai)	Cross-sectional	Community	Simple random	Adults aged > = 40 years	Not mentioned explicitly	NR	NR	52.4	9.8	440	Spirometry: FEV1/FVC < LLN	30	6.82
Triest et al. [33]	2019	2018	Maharashtra (Pune)	Cross-sectional	Community	Simple random	Adults aged > = 40 years	Not mentioned explicitly	NR	NR	51.7	10.3	843	Spirometry: FEV1/FVC < LLN	52	6.17
Christopher et al. [34]	2020	2018	Tamil Nadu	Cross-sectional	Community	Cluster random	Adults aged > = 30 years	Rural	100	NR	51.5	12.1	787	Spirometry: GOLD	32	4.07
Burney et al. [35]	2020	2006–2008	Maharashtra (Mumbai)	Cross-sectional	Community	Cluster random	Adults aged > = 40 years	Not mentioned explicitly	NR	NR	NR	NR	440	Spirometry: FEV1/FVC < LLN	30	6.82
Burney et al. [35]	2020	2008–2009	Maharashtra (Pune)	Cross-sectional	Community	Simple random	Adults aged > = 40 years	Not mentioned explicitly	NR	NR	NR	NR	842	Spirometry: FEV1/FVC < LLN	52	6.18
Burney et al. [35]	2020	2010–2011	Kashmir	Cross-sectional	Community	Cluster random	Adults aged > = 40 years	Not mentioned explicitly	NR	NR	NR	NR	752	Spirometry: FEV1/FVC < LLN	124	16.49
Burney et al. [35]	2020	2010–2011	Karnataka	Cross-sectional	Community	Cluster random	Adults aged > = 40 years	Not mentioned explicitly	NR	NR	NR	NR	601	Spirometry: FEV1/FVC < LLN	48	7.99
Vaz et al. [36]	2023	2019	Delhi	Cross-sectional	Community	Simple random	Auto Rickshaw Drivers age 20–60 years	Urban	NR	100	39	7.54	409	Spirometry: GOLD	56	13.69
Pradeep et al. [37]	2024	2022	Tamil Nadu	Cross-sectional	Community	Simple random	Adults above 30 years	Not mentioned explicitly	NR	NR	37	1	403	Spirometry: GOLD	89	22.08

NR Not Reported, FEV₁ Forced Expiratory Volume in 1 s, FVC Forced Vital Capacity, GOLD Global Initiative for Chronic Obstructive Lung Disease, LLN Lower Limit of Normal

Quality assessment using the JBI critical appraisal tool indicated that all of the studies (100%) were good quality, and none were classified as moderate or poor. (Supplementary Table S3).

Overall, the pooled prevalence of COPD among adult Indians was 13% (95% CI: 9%−18%), with significant heterogeneity observed among studies (I² = 99.8%). The overall prediction interval was 1%−25%. Ramadoss et al. (2013) [6] reported the highest COPD prevalence (50%) in a hospital-based study in Kerala, whereas Mukherjee et al. (2014) [5] documented the lowest prevalence (0.91%) in a community-based study in West Bengal.

When focusing on severity of COPD, the pooled prevalence was: Grade 1/Mild COPD- 4% (95% CI: 1% – 8%), Grade 2/Moderate COPD- 5% (95% CI: 1% – 8%), Grade 3/Severe COPD- 2% (95% CI: 0% – 5%) and Garde 4/Very Severe COPD- 1% (95% CI: 0%−2%); again, demonstrating substantial heterogeneity across studies (I² = 98.9%, 98.8%, 98.8 and 91.3% respectively). Therefore, a random-effects model was utilized to estimate the pooled prevalence (Figs. 2, and 3) (Supplementary Table S4).

Fig. 2.

Forest plot showing the pooled prevalence of COPD in Indian adults using a random-effects model

Fig. 3.

Forest plot showing the pooled prevalence of COPD as per the grading of GOLD criteria among Indian adults using a random-effects model

Subgroup analysis

Subgroup analyses were conducted to explore the high heterogeneity (I² ≈ 99.8%) in reported COPD prevalence among Indian adults. One primary source of variability was the choice of diagnostic tools and methods of COPD. Studies conducted using Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria (15% (95%CI: 8%−21%)) of spirometry had a higher pooled prevalence than using FEV1/FVC < LLN criteria (10% (95%CI: 7%−14%)) (Fig. 4).

Fig. 4.

Subgroup Analysis of prevalence of COPD in Indian adults by Diagnostic Tool/method

In addition to methodological differences, study setting variations, sampling methods, geographic location, and study setting also impacted the reported prevalence. Hospital/facility-based studies (27%) had slightly higher COPD prevalence than community-based studies (12%). Studies conducted with random sampling (12%) documented a higher prevalence than the non-random sampling (17%) studies (Fig. 5).

Fig. 5.

Subgroup Analysis of prevalence of COPD in Indian adults by A) study setting and B) sampling methods

In India, the North (19%) zone had the highest prevalence, followed by the South (13%), Central (11%), and East (8%) zones. The lowest prevalence was documented in the West (7%) zone. (Fig. 6). Variations were more striking across each state. As per community-based studies, the highest pooled prevalence was documented by Odisha (20%) and the lowest by Tamil Nadu (2.4%). In view of the hospital/facility-based study, Kerala (50.0%) had the highest prevalence, and Maharashtra (11.7%) had the lowest prevalence (Fig. 7).

Fig. 6.

Subgroup Analysis of the prevalence of COPD in Indian adults by various zones of India

Fig. 7.

Graphical representation of the prevalence of COPD in Indian adults: A community and B) Hospital-based study

The pooled COPD prevalence was higher in adult males (16% (95% CI: 9–24%)) than in females (9% (95% CI: 5–13%)) (Fig. 8). Elderly Indian adults aged 60 years and above (27%) had higher COPD prevalence than young and middle-aged adults aged 18–59 years (17%) (Fig. 9). AS per exposure to cooking fuel, COPD prevalence was higher in Biomass fuel users (8%) than LPG/clean fuel users (1%) (Fig. 10). COPD prevalence was higher in urban (11% (95% CI: 6–17%)) than rural (9% (95% CI: 1–17%)) areas (Fig. 11). Lastly, subgroup analysis of year of study revealed a significant temporal trend with higher COPD prevalence in Post-COVID/2023 onwards (18%), followed by Pre-COVID/2009–2019 (13%), while the Intra-COVID (8%) period had lower prevalence (Fig. 12).

Fig. 8.

Pooled Prevalence of COPD as per gender

Fig. 9.

Subgroup Analysis of the prevalence of COPD in Indian adults by age group

Fig. 10.

Subgroup Analysis of the prevalence of COPD in Indian adults by exposure to cooking fuel

Fig. 11.

Subgroup Analysis of the prevalence of COPD in Indian adults by residence (rural vs urban)

Fig. 12.

Subgroup Analysis of the prevalence of COPD in Indian adults by study period

The pooled prevalence of COPD among Indian adult smokers was 37% [95%CI: 18%−56%]. Pradeep et al. (2024) documented the highest prevalence (77%) while Christopher et al. (2020) documented the lowest (12%) prevalence of COPD among Indian adult smokers (Fig. 13).

Fig. 13.

Pooled prevalence of COPD in Indian adult smokers

Heterogeneity, outliers, and sensitivity analysis

Sensitivity analyses were conducted to assess the robustness of the pooled prevalence estimate. The estimation of heterogeneity and outliers was documented in Fig. 14 using the Baujat and Galbraith plots. These plots identified key studies contributing to heterogeneity and potential outliers, aiding in the assessment of influence and consistency across pooled prevalence estimates in the meta-analysis. Figure 15 illustrates how omitting each study affects the overall prevalence estimate. The pooled prevalence remains stable (13%; 95% CI: 9%–25%), indicating robustness. Chhabra et al. (2010) [19], Parasuramalu et al. (2014) [22], Mahesh et al. (2018) [31], Ramadoss et al. (2013) [6], Mahishale et al. (2016) [27], and Chandra et al. (2018) [18] show notable influence on heterogeneity. However, their removal did not substantially alter the summary prevalence estimate, staying within the original confidence intervals (9%−25%) of the pooled estimate. Thus, no individual study significantly changes the pooled estimates, indicating that the results are stable and not dominated by any single study.

Fig. 14.

Analysis of heterogeneity and outliers using the Baujat Plot and the Galbraith Plot

Fig. 15.

Sensitivity analysis using leave-one-out analysis

Publication bias

Publication bias was assessed using Egger’s regression test, Begg’s test, funnel plots (including the Luis Furuya–Kanamori [LFK] index), and the trim-and-fill method. Egger’s test suggested publication bias and no small study effect: (β₁ = 12.50, SE = 1.55, z = 8.07, p < 0.001). Begg’s test also suggests publication bias (Kendall’s score 229.00, SE 50.60, Z 4.52, p < 0.001). The funnel plot appeared relatively asymmetrical, with an LFK index of 1.32 (|LFK index|= 1–2 indicates minimal asymmetry). In addition, the trim-and-fill analysis imputed missing studies and produced a slightly higher pooled prevalence estimate (17.0%; 95% CI: 12.4% to 21.6%) than the observed data, though it lay within the confidence intervals (9%−25%) of the overall pooled prevalence (Fig. 16).

Fig. 16.

Publication bias (A) DOI/LKF Plot, B Funnel plot

Meta-regression

Meta-regression analysis was performed to examine whether any specific study-level covariates explained the observed heterogeneity in COPD prevalence among Indian adults. Two covariates were assessed: mean age in years and total sample size. None of the covariates documented a significant influence (Fig. 17).

Fig. 17.

A Meta-regression of COPD Prevalence on Mean Age. B Meta-regression of COPD Prevalence on Total sample size

Discussion

This systematic review and meta-analysis aimed to estimate the pooled prevalence of Chronic Obstructive Pulmonary Disease (COPD) among Indian adults using spirometry-based diagnosis and to identify key demographic, clinical, and environmental correlates. To begin with, this study focused solely on cases diagnosed through spirometry, ensuring high diagnostic precision and minimizing the risk of misclassification. We analyzed 23 studies spanning various regions and settings across India, providing a comprehensive and nationally representative overview. A thorough quality assessment using the Joanna Briggs Institute checklist affirmed the strong methodological rigor of all included studies. Additionally, the analysis employed advanced statistical methods—such as subgroup evaluations, sensitivity checks, and meta-regression—to investigate and account for heterogeneity. Notably, the study reflects a post-COVID context, capturing potential shifts in respiratory disease patterns following the pandemic.

The primary finding was a pooled COPD prevalence of 13% (95% CI: 9%–18%) among Indian adults, indicating a substantial public health burden. The prevalence estimate observed in this study exceeds the 7.4% reported by Daniel et al. (2021) [8], who analyzed data from 2011 to 2020. Several factors may explain this difference: the inclusion of intra- and post-pandemic studies in our review, a broader geographic scope, and more stringent spirometry-based diagnostic criteria. Over the last three decades, COPD has become an important contributor and leading cause of disability adjusted life years (DALY) from 1990 (at 12th) to 2023 (7th) globally [38]. Our results are consistent with findings from Burney et al. (2020) [35], Vaz et al. (2023) [36]. Banjare et al. (2014) [23] and Mahishale et al. (2016) [27] highlighted higher COPD rates among rural and biomass-exposed populations. In contrast, studies like Chhabra et al. (2010) [19] and Parasuramalu et al. (2014) [22] reported lower prevalence, likely due to sampling, methodological differences, or underreporting. These findings emphasize that COPD remains significantly underdiagnosed and insufficiently addressed within India’s NCD framework, reinforcing the urgency of incorporating routine COPD screening and expanding spirometry services at the primary care level.

Subgroup analysis revealed notable variations in prevalence based on diagnostic criteria, study setting, gender, age, fuel exposure, and geographic zone. Studies employing the GOLD criteria reported a higher COPD prevalence (15%) compared to those using the FEV₁/FVC < LLN threshold (10%), aligning with international evidence that fixed-ratio cutoffs may overdiagnose COPD in older adults [39]. Prevalence was notably greater in hospital-based studies (27%) than in community-based ones (12%), likely reflecting selection bias and the inclusion of symptomatic individuals. Men exhibited a higher prevalence (16%) than women (9%), mainly attributable to greater tobacco use, although biomass fuel exposure remains a significant contributor among women [40, 41]. Older adults (≥ 60 years) showed substantially higher prevalence (27%) than those aged 18–59 years (17%), consistent with age-related declines in pulmonary function [42, 43].

Geographically, North India had the highest reported prevalence (19%). In comparison, West India had the lowest (7%), possibly due to variations in environmental risk factors, healthcare access, and diagnostic capacity supported by previous SRMA [8]. These patterns are corroborated by findings from Koul et al. (2016) [28] and Kumar et al. (2019) [32], though Mahesh et al. (2018) [31] reported lower rates in Karnataka, potentially reflecting differences in sampling methodology or the impact of localized health interventions. The regional differences in COPD prevalence across India reflect underlying socio-economic inequalities, with greater burden seen in poorer areas where risk exposures are concentrated. Northern and Central regions, characterized by poverty, dependence on biomass fuels, widespread tobacco use, and inadequate healthcare, demonstrate higher rates, whereas more affluent western states show comparatively lower prevalence. Socio-economic disadvantage intensifies vulnerability through indoor air pollution, occupational risks, and delayed detection, producing this gradient. These observations are consistent with earlier studies connecting COPD with poverty and environmental determinants [44–46].

Beyond prevalence, the review identified associated risk factors such as tobacco use and exposure to indoor air pollution due to cooking fuel usage. The pooled prevalence among smokers was 37% (95% CI: 18%–56%), with Pradeep et al. (2024) [37] reporting the highest (77%) and Christopher et al. (2020) [34] the lowest (12%). Biomass fuel users had significantly higher COPD prevalence (8%) than LPG users (1%), reinforcing the need for clean fuel transitions. These associations are well-documented in national surveys and cohort studies, including Mukherjee et al. (2014) [5] and Johnson et al. (2011) [21]. Males (16%) documented higher COPD prevalence than females (9%), which is in concordance with a global systematic review by Wachami et al. in 2024 (male 15.5%, female 8.8%) [47]. Previous SRMA by Daniel et al. (2021) reported relatively lower prevalence (male 11.4%, female 7.4%). Our study documented slightly higher prevalence in females but a larger burden in males over the COVID period. This might be due to the replacement of solid biomass fuel with clean fuel/LPG under the Pradhan Mantri Ujjwala Yojana (PMUY) scheme, contributing to relatively lower prevalence in females [48]. Occupational exposure was variably documented across the included studies, likely reflecting the absence of uniform assessment methods or potential underreporting. Given the complex interaction of multiple risk factors—particularly affecting women and rural communities—there is a clear need for interventions that are both gender-responsive and regionally adapted [49].

Heterogeneity across studies was substantial (I² = 99.8%), driven by methodological and regional variability. Baujat and Galbraith plots identified influential studies such as Ramadoss et al. (2013) [6] and Chandra et al. (2018) [19], though sensitivity analysis confirmed the stability of pooled estimates. Leave-one-out analysis showed no single study significantly altered the overall prevalence, affirming robustness. Publication bias was detected via Egger’s and Begg’s tests, with an LFK index of 1.32 indicating minor asymmetry. Trim-and-fill analysis adjusted the pooled prevalence to 17%, still within the original confidence interval, suggesting reliability. Meta-regression showed no significant influence of mean age or sample size on prevalence, indicating that other unmeasured factors may contribute to heterogeneity.

Strengths and limitations

The main strengths of this study lie in its reliance on spirometry for diagnosis, broad geographic representation, and rigorous methodological evaluation. Incorporating intra- and post-COVID data further enhances its relevance for public health planning. However, limitations include excluding non-English publications, possible publication bias, and the predominance of cross-sectional designs, which restrict causal interpretation. These were mitigated by applying strict inclusion criteria, performing sensitivity analyses, and employing meta-regression to assess the influence of covariates. In contrast to Daniel et al. (2021) [8], which lacked substantial intra- and post-pandemic evidence, our study provides a more precise and up-to-date estimate of COPD burden in India. Spirometry, the gold standard for COPD diagnosis, was widely restricted during the pandemic due to aerosol generation risks and infection control protocols. Consequently, many studies relied on symptom-based screening or self-reported diagnoses, that may have led to underestimation of true prevalence; which were excluded as per our eligibility criteria. Cumulative contributions of multiple risk factors on pooled prevalence of COPD was not calculated. The FEV₁/FVC ratio is widely used to diagnose COPD because it reflects airflow limitation, but it has important scientific rationale and methodological limitations. It is simple, reproducible, and predictive of outcomes, yet the fixed cut-off (0.70) may misclassify patients depending on age, sex, and ethnicity.

Policy implications and recommendations

This study highlights the urgent need to integrate COPD screening into India’s NP-NCD framework [7], expand spirometry access at primary care levels, and promote clean fuel adoption. Gender-sensitive interventions targeting biomass exposure among women and targeted programs for the elderly and high-risk populations are essential. Strengthening surveillance, standardizing diagnostic protocols, and investing in respiratory health infrastructure will be critical to reducing the COPD burden in India. Additionally, public awareness campaigns and training of frontline health workers in spirometry use can enhance early detection and management.

Conclusion

This systematic review and meta-analysis offer a detailed and current assessment of the chronic obstructive pulmonary disease (COPD) burden among Indian adults, based solely on spirometry-confirmed cases. The pooled prevalence was 13% (95% CI: 9–18%), with notably higher rates among smokers, elderly individuals, men, and those exposed to solid biomass fuels. Subgroup analyses highlighted considerable variation linked to diagnostic criteria, geographic regions, and study settings. Although heterogeneity was high, sensitivity checks and meta-regression affirmed the reliability of the results. The study is strengthened by its diagnostic precision, broad regional representation, and methodological rigor, though limitations include potential publication bias and the predominance of cross-sectional designs. These findings emphasize the need to incorporate routine COPD screening into India's non-communicable disease strategy, enhance access to spirometry at the primary care level, and accelerate clean energy adoption. Tailored, gender-responsive interventions and improved surveillance are vital for reducing COPD burden and informing future health policy.

Authors’ contributions

PH- Conceptualization, Methodology, Resources, Data Curation, Writing- Review and editing, Visualization, Supervision RK- Resources, Data Curation, Writing- Review and editing, SG- Resources, Data Curation, Writing- Review and editing, NK- Conceptualization, Methodology, Resources, Data Curation, Writing- Review and editing, MS- Conceptualization, Methodology, Resources, Data Curation, Writing- Review and editing, MS- Conceptualization, Methodology, Resources, Data Curation, Writing- Review and editing, BN- Methodology, Resources, Data Curation, Writing. MCP- Methodology, Resources, Data Curation, Writing. AM- Methodology, Resources, Data Curation, Writing. SR- Resources, Data Curation, Writing, Supervision.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.