Prevalence of chronic obstructive pulmonary disease in Indian nonsmokers: a systematic review & meta-analysis

Introduction

Chronic obstructive pulmonary disease (COPD) is caused by prolonged exposure to noxious stimuli and is characterized by a combination of respiratory symptoms and fixed spirometric airway obstruction. It affects about 10.3 % of the global population aged between 30 and 79 [1]. It has emerged as a major public health challenge, becoming the world’s third leading cause of death according to the World Health Organization [2]. In 2019, about 200 million people suffered from COPD worldwide, and 3.2 million people died due to it, with most of the deaths occurring in low- and middle-income countries (LMICs) [3]. Of the global COPD burden, about 55 million patients reside in India alone [4].

Smoking has long been recognized as the leading exposure responsible for the development of COPD. However, it has been found that about 40–56% of all COPD patients are nonsmokers [5]. Furthermore, among nonsmokers, women are disproportionately affected [6]. Some risk factors for COPD among nonsmokers include biomass fuel use for cooking and heating, environmental tobacco smoke exposure, and childhood respiratory infections [7]. Notably, these risk factors predominantly affect individuals of lower socio-economic status in developing countries. Biomass fuel has long been the dominant energy source for cooking in rural Indian households. The use of biomass fuel for cooking is associated with higher levels of indoor particulate matter (PM2.5 and PM10) compared with liquified petroleum gas (LPG) use [8]. The higher indoor air pollution with biomass fuel usage has also been linked with spirometric abnormalities and increased risk of developing COPD [8], 9]. Despite recent efforts of the Indian government to increase the adoption of cleaner fuels, the lingering effects of biomass fuel exposure will continue to affect the populace for the coming decades [10].

Further, evidence has shown that COPD may have different clinical and pathological characteristics among smokers and nonsmokers. For instance, smokers are more likely to suffer from emphysema, whereas nonsmokers are more likely to have a small airway disease phenotype [11]. Hence, it is concerning that most studies of therapeutic interventions in COPD have predominantly recruited only smokers with COPD.

Large cohorts from developed countries have estimated the prevalence of COPD among adult nonsmokers to be around 4.9–7.1 % [7], [12], [13], [14]. However, there are sparse data regarding the same from LMICs like India. With over 1.4 billion individuals, India has about one-sixth of the world’s population but contributes almost one-third of the disability-adjusted life years due to chronic respiratory diseases [4]. Existing studies have investigated the prevalence of COPD in India and neighboring countries but have not focussed on the nonsmoker population [4], 15]. Given that COPD is a heterogenous disease with distinct pathophysiological mechanisms in nonsmokers, the lack of systematic epidemiological data regarding COPD in nonsmokers from LMICs like India is a major impediment to investment in research and management of this neglected patient group. Hence, in this systematic review, we aim to establish the prevalence of COPD among nonsmokers in India.

Methods

This systematic review was prospectively registered in PROSPERO (CRD42022358489). It was conducted using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines.

Results

Characteristics of included studies

After removing duplicates, 1,441 results were obtained on the literature search. Following the title, abstract, and full-text screening, we finally included seven studies in the review (see PRISMA flow diagram, Figure 1) [21], [22], [23], [24], [25], [26], [27]. The characteristics of the included studies are summarized in Table 1.

Figure 1:

PRISMA diagram of the study selection process.

Table 1:

Characteristics of the studies included in the systematic review.

Author, year	Study location	Study population	Sample size	Age distribution	Diagnostic criteria for COPD	Prevalence of COPD in nonsmokers	Prevalence of COPD in biomass-exposed subjects
Johnson et al., 2011	Tamil Nadu (rural)	Female nonsmokers	900	Above 30 years	Symptoms and post-BD FEV1/FVC<0.7 without BDR	2.44 %	2.50 %
Dutta et al., 2013	West Bengal (rural)	Premenopausal female nonsmoker with biomass fuel exposure	244	22–41 years	Symptoms and FEV1/FVC<0.7	6.6 %	6.6 %
Parasuramalu et al., 2014	Karnataka (rural)	Nonsmokers and smokers of either sex	1,400 (1,053 nonsmokers)a	Above 35 years	FEV1/FVC<0.7b	3.04 %	–
Kalagouda et al., 2016	Karnataka (urban & rural)	Nonsmoker females with biomass fuel exposure>10 years	2,868	Above 40 years	Post-BD FEV1/FVC<0.7	18.4 %	18.4 %
Sinha et al., 2017	Delhi (urban)	Nonsmokers and smokers of either sex	1,203 (744 nonsmokers)	Above 30 years	Post-BD FEV1/FVC<0.7b	1.6 %	17.7 %
Christopher et al., 2019	Tamil Nadu (rural)	Nonsmokers and smokers of either sex	1,015 (899 nonsmokers)c	Above 30 years	Post-BD FEV1/FVC<0.7	3.3 %	3.6 %
Kumar et al., 2021	Haryana (rural)	Elderly nonsmokers and smokers of either sex	449 (195 nonsmokers)	Above 60 years	Post-BD FEV1/FVC<0.7	26.6 %	–

1. BDR, bronchodilator reversibility; COPD, chronic obstructive pulmonary disease; FEV1, forced expiratory volume in 1st second; FVC, forced vital capacity; GOLD, global initiative for obstructive lung disease. ^aThe number of nonsmokers was not reported and was derived using the reported prevalence of COPD in the overall population, smokers and nonsmokers. ^bIn this study, spirometry was only performed in those subjects with symptoms suggestive of COPD. ^cA total of 787 participants performed technically adequate spirometries, of whom 700 were nonsmokers.

Of the seven studies, three included only females [21], 22], 24]. There was a wide variation in the age distributions of the studies. Dutta et al. enrolled only young, premenopausal females aged 22–41 years [22]. In contrast, Kumar et al. included elderly individuals from a rural community aged 60 and above [27]. All other studies included middle-aged and elderly subjects with a lower age cutoff between 30 and 40 years without any upper age cutoff [21], [23], [24], [25], [26]. Geographically, four studies were conducted in South India [21], 23], 24], 26], two in North India [25], 27], and one in East India [22]. Two studies enrolled urban subjects [24], 25], while the remaining studies examined rural populations. All studies used spirometric definitions to diagnose COPD (Table 1).

Five studies provided an estimate of the prevalence of COPD among nonsmokers in the community, not limited by any specific exposures [21], 23], [25], [26], [27]. The remaining two studies reported the prevalence of COPD only among female nonsmokers exposed to biomass fumes [22], 24].

Study outcomes

Prevalence of COPD in nonsmokers in India

The prevalence of COPD among nonsmokers varied between 1.6 and 26.6 % in the included studies. As the mean age was not reported for the nonsmoker subgroup in a few included studies, we explored the association between the minimum age cutoffs and prevalence of COPD among nonsmokers in individual studies (Figure 2). There was a significant association between a higher minimum age cutoff and higher prevalence of COPD among included studies (coefficient=0.69, 95 % CI: 0.23–1.15; p=0.012; R²=74.7 %).

Figure 2:

Meta-regression analysis of the association between the lower age cutoff and the prevalence of nonsmoker COPD in the included studies.

Among included studies, those which exclusively enrolled biomass fuel-exposed subjects reported higher prevalences (Table 1), specifically 6.6 % by Dutta et al. and 18.4 % by Kalagouda et al. [22], 24]. The prevalence was highest (26.6 %) in the study by Kumar et al., which included only elderly subjects, of whom 97 % were exposed to biomass fuel [27]. Among the studies which enrolled both nonsmokers and smokers, the prevalence of COPD was generally higher among the smoker subgroups (Supplementary Table 3). However, the proportion of nonsmokers who constituted the overall COPD patients in these studies varied widely between 10 and 72 % (Supplementary Figure 1) [23], [25], [26], [27].

Prevalence of COPD among middle-aged and elderly nonsmokers in India

Among the four studies that reported the prevalence of COPD among middle-aged and elderly nonsmokers not selected based on biomass fuel exposure, the prevalence varied between 1.6 and 3.3 %. Upon meta-analysis, the pooled prevalence of COPD was 3 % (95 % CI, 2–3 %) with medium heterogeneity (I²=50.52 %, p=0.11; Figure 3).

Figure 3:

Forest plot of the meta-analysis of the prevalence of chronic obstructive pulmonary disease (COPD) among middle-aged and older nonsmokers in India.

Prevalence of COPD in biomass-exposed population in India

The prevalence of COPD, specifically in the biomass fuel-exposed population, was reported in five included studies and ranged from 2.5 to 18.4 %. The pooled prevalence of COPD among the biomass-exposed population was 10 % (95 % CI, 2–18 %) with high heterogeneity (I²=98.8 %, p<0.001; Figure 4).

Figure 4:

Forest plot of the meta-analysis of the prevalence of chronic obstructive pulmonary disease (COPD) among biomass fuel-exposed subjects of the included studies.

Discussion

We performed the first systematic review of the prevalence of COPD among nonsmokers from India. We included seven studies comprising 6,903 nonsmoking adults and found that the prevalence of COPD varied between 1.6 and 26.6 %. This wide variation in the prevalence resulted from considerable heterogeneity in the included studies regarding the age and sex distribution and biomass fuel exposure profile of the subjects. Among the studies which included middle-aged and elderly Indian nonsmokers not selected for biomass fuel exposure, we found the pooled prevalence of COPD to be 3 % with medium heterogeneity. In contrast, among those exposed to biomass fuel, we found the pooled prevalence to be 10 % with high heterogeneity.

Although a systematic review of COPD prevalence in India had been conducted by Jarhyan et al., it had not studied the nonsmoking population separately [15]. Jahyran et al. included all South Asian countries in the review. Among the studies from India, the pooled prevalence of COPD was 11.1 %. The authors determined that besides smoking, additional risk factors included exposures to biomass fuel, occupational dust and environmental tobacco smoke, and a previous diagnosis of tuberculosis or asthma. Additionally, the Global Burden of Disease study found that only one-fourth of the COPD burden in India is attributable to tobacco use. They observed that about half of the COPD cases in India are primarily caused by outdoor and indoor air pollution [4]. Various single-center observational studies from India have also shown that about 40–69 % of COPD patients were non-smokers [28], [29], [30]. In the present systematic review, the studies which reported on both smokers and nonsmokers found that the proportion of COPD belonging to the nonsmoker subgroup varied between 10 and 72 % [23], [25], [26], [27]. The wide range probably reflected differences in the proportion of smokers in the overall sample and the rates of other exposures like biomass fuel usage. Nonetheless, smokers had a higher prevalence rate of COPD than nonsmokers in all of these studies [23], [25], [26], [27].

It is hardly surprising that the bulk of COPD in India is unrelated to smoking because the Global Adult Tobacco Survey found that only 10.4 % of the adult Indian population uses a smoked form of tobacco [31]. Smokeless tobacco use is more prevalent in India; however, this is not an established risk factor for COPD. In contrast, about 55 % of the Indian population used biomass fuel for cooking in 2017, according to the Global Burden of Disease study [32]. In 2016, the Government of India launched the “Ujjwala Yojana” to provide LPG connections to families below the poverty line [33]. However, those who have already experienced prolonged biomass fuel exposure will be at risk for COPD for years to come. Our finding of the high prevalence of COPD in the biomass-exposed population re-emphasizes its important contribution to the burden of nonsmoker COPD.

The choice of including middle-aged and elderly nonsmokers in the meta-analysis was done as these are the age groups for which COPD is a public health menace. The definition of the middle-aged population is contentious; while the lower age cutoff varies between 35 and 45 years, the upper age cutoff varies between 60 and 65 years [17], 18], 34]. We increased the inclusivity of our meta-analysis by using a lower age cutoff of 44 years or below as the eligibility criterion. Hence, no study using even a restrictive definition of middle-age (e.g., 44–60 years) got excluded. As per the United Nations Statistics Department, about one-fourth of India’s population in 2021 was aged 45 years old or above, amounting to around 340.8 million people [35]. Assuming that 89.6 % of these individuals are nonsmokers [31], approximately 9.2 million middle-aged and elderly nonsmokers may be suffering from COPD in India. This number is likely to grow in the coming decades owing to India’s gradually ageing population, and the lifelong exposure to indoor air pollution from biomass fuel usage in a substantial proportion of the populace.

Our findings are also a significant addition to the meager data from LMICs on the prevalence of nonsmoker COPD. Our estimate that 3 % of middle-aged and older Indian nonsmokers have COPD is comparable to results from other LMICs. Waked et al. reported that 3.4 % of Lebanese nonsmokers had COPD using spirometric definitions [36]. Denguezli et al. found that 4.7 % of Tunisian nonsmokers had COPD, which accounted for 45 % of all COPD cases [37]. The risk factors for nonsmoking COPD, which remained consistent among these LMIC studies, included advanced age and biomass fuel usage [36], [37], [38]. In addition to our finding of higher COPD prevalence among biomass fuel-exposed subjects, we also observed a disproportionately higher prevalence of COPD in the Indian study that enrolled only elderly subjects [27].

Although we have stressed upon the role of biomass fuel exposure and air pollution in the development of nonsmoker COPD in LMICs, a significant confounding factor may involve the concept of a “false start” in childhood lung development [39]. While many COPD patients experience accelerated lung function decline due to noxious exposures, a sizable proportion may have normal rates of decline of lung function. In these patients, COPD develops because they have a lower value of peak lung function achieved in early adulthood due to impaired lung development. Underserved populations of LMICs are particularly susceptible to impaired lung growth due to various maternal (e.g., anemia, smoking etc.), perinatal (low birth weight or prematurity), or childhood (e.g., malnutrition, biomass fuel exposure, frequent respiratory infections etc.) factors [39], 40]. Hence, despite normal rates of lung function decline, they may develop COPD in later life. The present review is unable to determine what proportion of patients may have acquired COPD in this manner.

Conclusions and future directions

In conclusion, few studies have examined the prevalence of COPD among nonsmokers in India. While the estimated prevalence is 3 % among middle-aged and older nonsmokers in the community, it may be about 10 % among biomass fuel-exposed individuals. Given India’s sizeable nonsmoking population and its widespread usage of biomass fuel for cooking, this implies that more than 9 million nonsmoker adults in India already have COPD. Furthermore, India’s increasing outdoor air pollution levels and rising median age will further expose tens of millions of nonsmokers to developing COPD in the coming years.

Considering that nonsmokers with COPD represent a different phenotype associated with female preponderance, lesser emphysema, and higher small airway disease, it is concerning that measly research has been devoted to therapies for nonsmokers with COPD. Even lesser attention has been paid to the most vulnerable sufferers of this neglected condition; i.e., those residing in LMICs. Future studies need to examine not just the epidemiological footprint of COPD among Indian nonsmokers but also its pathophysiology, morbidity, economic impact, and unique management aspects in LMICs worldwide. Since the current estimate of nonsmokers suffering from COPD in LMICs is based on a few small studies and likely imprecise, large multinational studies can shed further light on the true burden of nonsmoker COPD. Such information would help prioritize resource-allocation for the research and management of this grossly underappreciated condition. Long-term cohorts in LMICs can help unravel patterns of spirometric declines and the natural history of nonsmoker COPD in association with impaired childhood lung development and environmental exposures like air pollution and biomass fuel usage. A better understanding of pathophysiology of nonsmoker COPD will enable clinical trials enrolling nonsmokers with COPD for tailored pharmacological and non-pharmacological interventions including exposure mitigation. Finally, economic evaluation of these management strategies would help optimize treatment of nonsmoker COPD in LMICs.