Chronic obstructive pulmonary disease (COPD) is a leading cause of morbidity and mortality1 and the economic burden of COPD is increasing.2

Smoking is the most important causative factor of COPD. However, while more than 4000 chemical substances contained in a cigarette have adverse effects on various pulmonary diseases, the pathophysiological mechanisms underlying the association between smoking and COPD have not been fully elucidated.3

The effects of gender on COPD are well known, and women are widely considered more susceptible to the effects of smoking than men.4,5 The suggested reasons for the increased susceptibility of women to the effects of smoking include a smaller airway size compared with men6 and female sex hormone-mediated metabolism differences.7

In a Norwegian study, more women were current smokers. The authors hypothesized that females were more susceptible to the lung damaging effects of cigarette smoking.4 In other studies based on larger reductions in forced expiratory volume in 1 second (FEV1) and after adjustment for smoking intensity, female smokers were shown to develop COPD more often than male smokers.8,9

However, the effects of sex on the association between current smoker status and COPD in Korean adults are controversial. In a community-based cohort Korean study, current smoker status was significantly associated with COPD in men (risk ratio, RR: 2.28, 95% confidence interval, CI: 1.86–2.80) and in women (RR: 4.04, 95% CI: 2.84–5.75).10 However, in another population sample-based study in which a non-smoking group of Korean adults was used for comparison, the odds ratio (OR, 95% CI) for COPD in the current smoker group was 3.49 (2.44–5.00) in men and 3.45 (2.20–5.40) in women.11

In addition, one previous study showed the ratios of cotinine-verified to self-reported smoking rates were 2.36 for females and 1.12 for males in Korea,12 and another study showed the effects of female hidden smoking on the association between smoking and hypertension in Korean adults.13

Therefore, in this study, we investigated the effects of sex differences on the association between smoking and COPD and the effects of female hidden smoking on the association between smoking and COPD in Korean adults.

The main findings of this study were that ex-smoker and current smoker status were significantly associated with COPD in the overall population. In addition, sex effects were observed as well as an effect of female hidden smokers on the relationship between smoking and COPD in Korean adults.

The effects of sex on COPD are well established, and females are widely considered more susceptible to the effects of smoking than males.4,5 There are several hypotheses that explain sex-related variation in COPD prevalence. First, women may experience greater exposure to cigarette smoke due to possessing smaller airways than men.6 Second, female sex hormones may contribute to oxidative stress and, eventually, greater airway injury. During smoking, numerous smoking chemicals are metabolized in two phases. Phase I is mediated largely by cytochrome P450 (CYP) enzymes; these are responsible for detoxifying cigarette smoke into intermediate metabolites. Subsequently, these metabolites are conjugated by phase II enzymes and excreted. Because several intermediate metabolites are toxic, estradiol upregulates CYP enzymes without necessarily altering the expression of phase II enzymes. This causes the female lungs to be more susceptible to oxidant damage in response to cigarette smoke.7,16

Compared with other previous foreign studies, the results in previous Korean studies regarding the effects of sex on the association between current smoking status and COPD were controversial.4,8–11 In a community-based cohort Korean study, current smoking was significantly associated with COPD in males (RR: 2.28, 95% CI 1.86–2.80) and in females (RR: 4.04, 95% CI 2.84–5.75).10 However, in another population sample-based study, results showed that, compared with a non-smoking group of Korean adults, the OR (95% CI) for COPD in the current smoker group was 3.49 (2.44–5.00) in men and 3.45 (2.20–5.40) in women.11

Consistent with a previous study using similar data (2008–2016 KNHANES in this study and 2007–2015 KNHANES in the previous Korean study), we did not find sex differences in the association between COPD and current smoking.11 In our study, male self-reported and cotinine-verified current smoking status were correlated with COPD (OR: 2.43, 95% CI: 1.86–3.18 and OR: 1.65, 95% CI: 1.40–1.96, respectively) and female self-reported and cotinine-verified current smoking status also correlated with COPD (OR: 2.52, 95% CI: 1.66–3.83 and OR: 2.11, 95% CI: 1.49–2.98, respectively).

Interestingly, in the association between COPD and ex-smoking, this study and previous studies showed ex-smoking was associated with COPD in men and not in women.10,11

We assumed the different effects of sex differences between COPD and current smoking in two previous studies might be due to selection bias associated with the sampling of different groups.10,11 One study used a region-based sample, while the other used a nationally representative sample.10,11. In addition, a previous Korean study reported large geographic variation in the prevalence of COPD in Korea and emphasized the importance of nationally representative sampling.17,18

In a previous study, a significant number of female hidden smokers in Korea was reported compared with other countries.12 Similar results were observed in this study. One hypothesis suggests that self-reported smoking in Korean females underestimates the true prevalence as a result of Confucianism. The adoption of Confucianism can result in a patriarchal culture in which female smoking is stigmatized.19

In a previous Korean study, the ratios of cotinine-verified to self-reported smoking rates were 2.36 for women and 1.12 for men,12 indicating that women are more likely than men to be hidden smokers. Conversely, in a study from other countries, sex differences in underreporting the rate of past or present smoking was not observed.20,21

In this study, the ratios of cotinine-verified to self-reported smoking rates were 1.95 (632/325) for women and 1.07 (2452/2300) for men. These rates were similar to the findings in a previous Korean study.12 Sex differences in underreporting the rate of past or present smoking were observed in this study: a greater number of hidden smokers were women. The discrepancy in the underreporting rates between the sexes could lead to statistical issues. In addition, a larger effect of hidden smokers was found in the association between COPD and smoking in females. In male participants, the adjusted OR for self-reported current smoking associated with COPD was 2.43 (95% CI: 1.86–3.18) and 2.29 (95% CI: 1.76–2.99) based on SCS. In women, the adjusted OR for self-reported current smoking status associated with COPD was 2.52 (95% CI: 1.66–3.83) and 2.13 (95% CI: 1.51–3.02) based on SCS.

The results from this study showed the risk of COPD associated with current smoking was lower in Korean adults than in those of other nationalities.22,23 One hypothesis is that this result was due to different lung functions and genetic polymorphisms among nationalities.22 In this study, an association between ex-smoker and COPD was observed; however, the association was not significant in women. These results might be due to lower amount of smoking by women than men, which was similar to results reported in a previous Korean study.11,22

Light and social smokers are often not detected because many of these individuals are self-reported non-smokers.24,25 Recently, a report was published showing that light smoking was associated with COPD and amount of smoking was associated with progressively higher risk of COPD than never smoking.22

To better understand the association between COPD and smoking, a new variable, SCS, was used in this study to evaluate the effects of hidden smoking and other types of smoking. This new variable showed similar results compared with self-reported smoking status. Therefore, passive and light smoking may affect COPD development similar to active smoking.

Parental smoking was a major risk factor for second-hand smoke exposure in children; and maternal, not paternal, smoking was known as a risk factor for decreased lung function and COPD in children.26–28 However, because there are many hidden female smokers in Korea, maternal hidden smokers should be considered when determining the impact of maternal smoking on lung function in Korean children.

This study had several limitations. First, because this study was based on a survey, selection and recall biases may have existed. Second, because this study was cross-sectional in design, the causal relationship between smoking and COPD could not be confirmed. Finally, potential confounding factors, including amount and duration of smoking, diet patterns, and genetic variations affecting nicotine metabolism were not evaluated. Further prospective studies are needed to clarify the effects of hidden female smokers on COPD by country.

However, the notable strength of our study is the use of national and widely sampled data to assess sex-specific relationships between smoking status and COPD. The other strength is the evaluation of the effects of sex and hidden smokers on the relationship between smoking and COPD.

This large population-based cross-sectional study showed that sex differences might affect the association between COPD and smoking status. Depending on country, hidden female smokers should be considered in the study on the relationship between smoking and COPD.

Ju Suk Lee conceived and supervised all aspects of this study. Hye Sung Ock contributed to the concept of study and wrote the final version of this manuscript. Sang Won Hwang analyzed data and contributed to design of our study. Hae Jeong Lee, Cheol Hong Kim, and Sung Hoon Kim collected data. Tae Hong Kim and Jun Hwa Lee collected data and carried out statistical analysis.

The authors have no commercial relationships or potential conflicts of interest to declare.