Introduction: Chronic cough, sputum, wheeze and breathlessness are early warning signs of smoking-related lung disease, but many smokers ignore them. Linking these symptoms to objective airflow obstruction can help identify smokers at risk of chronic obstructive pulmonary disease (COPD). Aim: To compare the prevalence of respiratory symptoms and spirometric airflow obstruction between adult smokers and non-smokers, and to describe the duration of smoking among smokers. Materials and methods: A comparative cross-sectional study was conducted on 100 current smokers and 100 never-smokers aged 18–60 years, matched in sex distribution (70 men, 30 women per group). Symptoms were recorded with a structured questionnaire. Spirometry followed ATS/ERS 2019 standards, and airflow obstruction was defined as FEV1/FVC below the lower limit of normal (LLN) of the GLI-2012 equations. Groups were compared with the chi-square test, and odds ratios (OR) with 95% confidence intervals (CI) were calculated. Results: Nearly half of smokers (48%) had smoked for more than 10 years. Smokers more often reported chronic cough (36% vs 10%; OR 5.06, 95% CI 2.34–10.94), sputum (31% vs 9%; OR 4.54), breathlessness (28% vs 8%; OR 4.47) and wheeze (22% vs 5%; OR 5.36); all p < 0.001. Mean FEV1/FVC was lower in smokers (74.5 ± 8.4% vs 82.1 ± 6.1%, p < 0.001), and 27% of smokers had FEV1/FVC below the LLN compared with 6% of non-smokers (OR 5.79, 95% CI 2.27–14.77; p < 0.001). Conclusion: Smokers had about five times the odds of each respiratory symptom and of spirometric airflow obstruction. About one in four smokers had obstruction. Smokers with respiratory symptoms should be offered spirometry and cessation support.
Tobacco smoking caused an estimated 7.69 million deaths worldwide in 2019 and remains the main preventable cause of chronic respiratory disease [1]. In India, about 100 million adults smoke, most commonly bidis, and the average age at starting daily use is in the late teens [2]. Many Indian smokers therefore build up decades of exposure before they seek medical care.
Respiratory symptoms are common in Indian smokers. The INSEARCH study of about 170,000 adults across India found that smoking was strongly associated with chronic bronchitis and respiratory symptoms [3]. An earlier multicentric Indian study also linked tobacco smoking to COPD, with prevalence rising with the amount smoked [4]. Despite this, smokers often accept chronic cough and breathlessness as normal "smoker's cough" and do not report them.
These symptoms have clear biological causes. Tobacco smoke causes goblet-cell hyperplasia, enlargement of the submucosal glands and shortening of cilia. Mucus builds up in the airways and is cleared poorly, causing cough and sputum. Inflammation and remodelling narrow the small airways and loss of alveolar attachments reduces elastic recoil, which together cause wheeze and breathlessness on exertion [5]. Chronic mucus hypersecretion is more than a nuisance: in a British birth cohort followed through adult life, its presence was associated with later development of COPD [6]. Fletcher and Peto showed that susceptible smokers lose FEV1 rapidly, and that stopping smoking slows this decline [7].
Symptoms alone cannot confirm airflow obstruction. Spirometry measures it objectively by the ratio of forced expiratory volume in one second to forced vital capacity (FEV1/FVC). Current ERS/ATS guidance recommends defining obstruction as an FEV1/FVC below the LLN, the fifth percentile of a healthy reference population [8]. The fixed cut-off of 0.70 used by GOLD under-diagnoses obstruction in younger adults and over-diagnoses it in older adults [8,9]. The GLI-2012 equations provide LLN values for South-East Asian populations [10].
In South Asia, the PURE substudy found that self-reported chronic wheeze, cough or sputum and dyspnoea rose steadily from non-smokers to light smokers, heavy cigarette smokers and heavy bidi smokers [11]. Respiratory symptoms in Indian non-smokers are also common, because of biomass smoke, outdoor air pollution and past tuberculosis [12]. Comparisons between smokers and non-smokers should therefore exclude these exposures. Few Indian studies have reported symptoms and LLN-based obstruction together, and fewer have included women.
This study therefore compared the prevalence of respiratory symptoms and of LLN-defined airflow obstruction between adult smokers and non-smokers in central India, and described the duration of smoking among smokers.
Study design and setting. A comparative cross-sectional study was carried out in the Department of Medical Physiology, [name of institution / medical college], Dewas, Madhya Pradesh, India, between [month year] and [month year].
Ethics. The Institutional Ethics Committee approved the protocol [approval number, date]. The study followed the ICMR National Ethical Guidelines (2017) and the Declaration of Helsinki [13,14]. Written informed consent was obtained from every participant in English or Hindi. Testing was free, and participants could withdraw at any time.
Participants. The study enrolled 200 adults aged 18–60 years: 100 current smokers and 100 never-smokers, each group with 70 men and 30 women. Participants were selected by [consecutive / convenience] sampling from [source population].
A current smoker had smoked at least 100 cigarettes or bidis in their lifetime, had smoked daily for at least one year, and still smoked. A never-smoker had smoked fewer than 100 cigarettes or bidis and did not currently smoke [2]. Smokers were grouped by duration of smoking as 1–5, 6–10, 11–15 and more than 15 years.
Exclusion criteria. To limit confounding from other causes of respiratory symptoms [12], we excluded people with known asthma, COPD, bronchiectasis, interstitial lung disease or lung cancer; present or past tuberculosis; a respiratory infection in the previous four weeks; cardiac disease; chest-wall deformity or neuromuscular disease; occupational exposure to dust, gases or fumes for at least one year; regular cooking with biomass fuel; contraindications to spirometry [15]; and current use of bronchodilators, corticosteroids or beta-blockers. Ex-smokers and exclusive smokeless-tobacco users were also excluded.
Symptom questionnaire. A trained investigator interviewed each participant with a structured proforma. Four symptoms were recorded as present or absent: chronic cough, sputum production, breathlessness and wheeze. Chronic cough and sputum were defined as present on most days for at least three months in a year. Participants could report more than one symptom.
Anthropometry. Height and weight were measured, and body mass index (BMI) was calculated and grouped by WHO categories.
Spirometry. Spirometry was performed on a computerised spirometer [make, model] calibrated daily with a 3-litre syringe, following ATS/ERS 2019 standards [15]. Participants avoided smoking for one hour before the test. At least three acceptable and repeatable forced expiratory manoeuvres were obtained, and the highest FEV1 and FVC were used. The FEV1/FVC ratio was compared with the LLN from the GLI-2012 equations [10]. Airflow obstruction was defined as FEV1/FVC below the LLN [8].
Statistical analysis. Data were analysed in IBM SPSS Statistics version [26]. Categorical variables are presented as number (%) and compared with Pearson's chi-square test. Odds ratios with 95% CI were calculated for each symptom and for airflow obstruction, with non-smokers as the reference group. The mean FEV1/FVC was compared with the unpaired t-test. A two-sided p < 0.05 was considered significant.
All 200 participants completed the questionnaire and acceptable spirometry. The groups had similar age (p = 0.97), sex (identical) and BMI (p = 0.94) distributions.
Table 1. Age, sex and BMI distribution of the study participants
|
Characteristic |
Smokers (n = 100), n (%) |
Non-smokers (n = 100), n (%) |
p-value |
|
Age group (years) |
|
|
0.97 |
|
18–30 |
25 (25) |
24 (24) |
|
|
31–40 |
32 (32) |
30 (30) |
|
|
41–50 |
28 (28) |
29 (29) |
|
|
51–60 |
15 (15) |
17 (17) |
|
|
Sex |
|
|
1.00 |
|
Male |
70 (70) |
70 (70) |
|
|
Female |
30 (30) |
30 (30) |
|
|
BMI (kg/m²) |
|
|
0.94 |
|
Underweight (< 18.5) |
8 (8) |
7 (7) |
|
|
Normal (18.5–24.9) |
53 (53) |
50 (50) |
|
|
Overweight (25.0–29.9) |
30 (30) |
32 (32) |
|
|
Obese (≥ 30.0) |
9 (9) |
11 (11) |
|
Chi-square test.
The two groups were well balanced, so differences in symptoms and spirometry are unlikely to be explained by age, sex or body build.
Table 2. Distribution of smokers by duration of smoking (n = 100)
|
Duration of smoking |
Number of smokers |
Percentage |
Cumulative percentage |
|
1–5 years |
18 |
18 |
18 |
|
6–10 years |
34 |
34 |
52 |
|
11–15 years |
26 |
26 |
78 |
|
More than 15 years |
22 |
22 |
100 |
The most common duration was 6–10 years (34%). Nearly half of smokers (48%) had smoked for more than 10 years, and 22% for more than 15 years. Most smokers therefore had substantial cumulative exposure.
Table 3. Prevalence of respiratory symptoms in smokers and non-smokers
|
Symptom |
Smokers (n = 100), n (%) |
Non-smokers (n = 100), n (%) |
χ² |
p-value |
OR (95% CI) |
|
Chronic cough |
36 (36) |
10 (10) |
19.09 |
< 0.001 |
5.06 (2.34–10.94) |
|
Sputum production |
31 (31) |
9 (9) |
15.12 |
< 0.001 |
4.54 (2.03–10.16) |
|
Breathlessness |
28 (28) |
8 (8) |
13.55 |
< 0.001 |
4.47 (1.92–10.40) |
|
Wheeze |
22 (22) |
5 (5) |
12.37 |
< 0.001 |
5.36 (1.94–14.80) |
Pearson chi-square test, df = 1. OR = odds ratio, non-smokers as reference. Participants could report more than one symptom.
Every symptom was significantly more common in smokers. Chronic cough was the most frequent (36%), followed by sputum (31%), breathlessness (28%) and wheeze (22%). Smokers had between 4.5 and 5.4 times the odds of each symptom compared with non-smokers. The predominance of cough and sputum points to mucus hypersecretion as the earliest clinical effect of smoking.
Table 4. FEV1/FVC ratio and airflow obstruction in smokers and non-smokers
|
Parameter |
Smokers (n = 100) |
Non-smokers (n = 100) |
Test statistic |
p-value |
OR (95% CI) |
|
FEV1/FVC (%), mean ± SD |
74.5 ± 8.4 |
82.1 ± 6.1 |
t = 7.32 |
< 0.001 |
— |
|
FEV1/FVC below LLN, n (%) |
27 (27) |
6 (6) |
χ² = 16.00 |
< 0.001 |
5.79 (2.27–14.77) |
|
FEV1/FVC at or above LLN, n (%) |
73 (73) |
94 (94) |
|
|
|
t = unpaired Student's t-test (df = 198); χ² = Pearson chi-square test (df = 1). LLN = lower limit of normal (GLI-2012).
Mean FEV1/FVC was 7.6 percentage points lower in smokers. More than one in four smokers (27%) had FEV1/FVC below the LLN, compared with 6% of non-smokers, giving smokers nearly six times the odds of airflow obstruction. Since 73% of smokers had a normal ratio but more than a third reported cough, many symptomatic smokers did not meet the spirometric definition of obstruction.
This study found that respiratory symptoms and spirometric airflow obstruction were each about five times more likely in smokers than in comparable non-smokers. Chronic cough was the most common symptom, and 27% of smokers had an FEV1/FVC below the LLN.
The symptom pattern agrees with large South Asian studies. In the PURE substudy, the adjusted prevalence of chronic wheeze, cough or sputum and dyspnoea increased steadily from non-smokers to heavy bidi smokers [11]. The INSEARCH study also found smoking to be a major risk factor for chronic bronchitis and respiratory symptoms in Indian adults [3]. The high frequency of cough and sputum fits with mucus hypersecretion caused by goblet-cell hyperplasia and impaired ciliary clearance [5]. These symptoms matter: chronic mucus hypersecretion predicted later COPD in a long-term British cohort [6].
Obstructive impairment was also the most common abnormal pattern among smokers in rural Maharashtra [16]. Our 27% prevalence of obstruction is lower than in a north Indian hospital study, where 76% of smokers and 28.7% of non-smokers had an obstructive pattern [17]. Differences in setting, age, smoking intensity and definition probably explain this. Hospital-based studies include more symptomatic patients, and many earlier Indian studies used the fixed 0.70 ratio rather than the LLN. We used the LLN, as recommended by the ERS/ATS, because it reduces misclassification across age groups [8]. Using GOLD's fixed ratio would likely have given different figures [9].
The gap between symptoms and spirometry deserves attention. Many smokers in our study reported cough or breathlessness but had a normal FEV1/FVC. Woodruff and colleagues showed that symptomatic smokers with preserved spirometry still had exacerbations, activity limitation and airway-wall thickening [18]. Regan and colleagues found emphysema and airway disease on CT in many smokers with normal spirometry [19]. A normal ratio does not mean a healthy lung, and symptomatic smokers need follow-up even if spirometry is normal.
The duration data show that nearly half the smokers had smoked for more than 10 years. Long-term cohorts show that continued smoking leads to COPD in a large share of smokers. In the Copenhagen City Heart Study, at least a quarter of continuous smokers developed clinically significant COPD over 25 years [20]. The Framingham Offspring cohort showed faster FEV1 decline in persistent smokers [21]. Quitting changes this course. In the Lung Health Study, sustained quitters had a much slower FEV1 decline than continuing smokers [22], and cough and sputum improve within months of cessation [23]. Spirometry can show smokers their own results, which may support quitting.
The low symptom and obstruction rates in our non-smokers reflect the exclusion of biomass, occupational and tuberculosis exposure, which cause much non-smoking COPD in India [12,24].
This study has limitations. Its cross-sectional design cannot show cause or change over time. Symptoms were self-reported and may be affected by recall bias. Smoking intensity, pack-years and product type were not analysed, and symptoms and obstruction were not compared across duration groups. Post-bronchodilator spirometry was not used to confirm persistent obstruction, so these figures cannot be equated with COPD. The study was single-centre and did not use multivariable adjustment.
Smokers had about five times the odds of chronic cough, sputum, breathlessness and wheeze, and nearly six times the odds of LLN-defined airflow obstruction, compared with non-smokers. About one in four smokers had obstruction, and many more had symptoms despite normal spirometry. Smokers who report respiratory symptoms should be offered spirometry, structured cessation support and follow-up, whatever their spirometry result.