Minimizing secondhand smoke (SHS) exposure has been a public health target globally. Although the prevalence of SHS exposure in Korea is decreasing with extension of the range of smoke-free public facilities since 2012, the prevalence is not yet low. The Korean National Health and Nutrition Examination Survey (KNHANES) VII-1 reported that the prevalence of SHS exposure indoors at home or at the workplace was 27.5% for male never-smokers and 20.0% for female never-smokers [1
]. However, because information on cigarette smoking is based on self-reported smoking questionnaires, actual smoking status could not be exactly assessed; respondents overreport never smoking because smoking is considered a socially undesirable behavior. Cotinine has been used as a biomarker for the validation of self-reported smoking status. A previous study demonstrated that self-reported SHS significantly underestimated the actual SHS as determined by cotinine verification [2
SHS exposure as well as active smoking are known risk factors for various cardiovascular diseases [3
]. Meanwhile, the relationship between active smoking and hypertension somewhat showed discordant results in previous studies [5
]. Recently, researchers on a number of studies have investigated the relationship between SHS exposure and hypertension, and most have found a positive relationship between the two [9
]. However, the studies’ data have been based on self-reported questionnaires, and most studies used relatively small sample sizes or were restricted to only women.
Therefore, the aim of this cross-sectional study was to evaluate the relationship between SHS exposure and hypertension among self-reported never-smokers verified by cotinine levels and to compare the relationship according to sex.
This study showed that SHS exposure in self-reported never-smokers verified by urinary cotinine was significantly associated with prevalence of hypertension. This association was proportional to the frequency and duration of SHS exposure. However, even the group with relatively low frequency and short duration of SHS exposure was significantly associated with hypertension compared to the group without SHS exposure. Furthermore, current SHS exposure only at the workplace as well as current SHS exposure that has been exposed to home SHS was associated with hypertension.
With the enactment of the Law for the National Health Promotion in 1995, Korea introduced some nonsmoking areas in public places, and since 2012, efforts have been made to minimize SHS exposure by expanding designated smoking cessation areas in public facilities and places. As a result, according to KNHANE VII-1 report, the rate of SHS exposure has been decreasing since 2012 [1
], and our study also showed that the rate of SHS exposure decreased from 28.3% in 2011 to 15.1% in 2016.
Previous studies on acute effect of active smoke on BP have established that smoking temporarily increases BP [19
]. The effect of chronic smoking on the rise in BP is also known to be mainly due to nicotine-mediated sympathetic activation [21
]. However, several studies reported tolerance to the effect of nicotine on BP and the possibility of effect of smoking on decreasing BP via vasodilating mediators [22
]. Due to these various biological properties of cigarette smoking, previous epidemiologic studies have shown inconsistent results on the relationship between chronic smoking and BP [5
]. The mechanism linking SHS exposure to hypertension is not fully elucidated but is thought to be similar to the mechanism between active smoking and blood pressure. Previous investigators have reported that SHS exposure causes structural and functional alterations on arterial walls including endothelial dysfunction, increased arterial stiffness, and progression of atherosclerosis [19
], and these changes could lead to higher BP. There have been epidemiologic studies on the relationship between chronic SHS exposure and hypertension in diverse population groups [9
], and most of them demonstrated that SHS exposure was significantly associated with hypertension [9
], which was consistent with our results. However, considering that individuals who were enrolled in our study had a low level of cotinine and SHS contains relatively higher concentrations of the toxic materials from side stream smoke, putative mechanism of linking of SHS exposure to BP could not exclude the effects of other oxidant gases and toxic chemicals other than nicotine in cigarette. Further pathophysiologic studies are needed to clarify the biologic actions.
However, there are several key differences between previous studies and our study. First, most of the studies were limited to females [9
], and the sample sizes were small except for one study [12
]. The results of the studies did show a positive association of SHS exposure with hypertension or BP, which is consistent with our results. A recent study in a large Chinese population also revealed a significant relationship between SHS exposure and hypertension, but the study’s investigation was limited to how husbands’ smoking affected their wives’ hypertension status [12
]. Although the remaining three studies included both males and females [13
], authors of only one study of 10,532 males and females investigated the association between SHS exposure and hypertension according to sex [13
]. Those authors found that SHS exposure was associated with hypertension in females only. In contrast, the subgroup analyses of our study showed significant associations in males as well as females, which was consistent with the main results of our study. The main reason for the difference of result in males between the two studies could be the different sample sizes, although the authors of the previous study explained that the lack of association with hypertension in males could have been because the participants were younger or because of sex-dependent biologic effects and small sample size. Second, all of the above studies were conducted based on self-reported data on SHS exposure, whereas, for our study, we assessed SHS exposure with minimizing misclassification of never-smokers using a biomarker as well as a self-reported questionnaire. Studies with only self-reported questionnaires cannot evade recall bias. For instance, in a recent Korean study derived from KNHANES IV, respondents who had self-reported SHS exposure had significantly underestimated their actual SHS exposure as verified by cotinine [2
]. Furthermore, the results of previous studies have reflected a misclassification between self-reported and biomarker-verified smoking status [25
]. In our previous studies, we reported that the proportion of cotinine-verified current-smokers among self-reported never-smokers is 1.7–1.8% [7
]. In short, the prevalence of self-reported never-smokers may be higher than that of actual never-smokers. Third, for the present study, we indirectly analyzed the relationship of home and workplace SHS exposure with hypertension. We found two significant results; SHS exposure only at workplace was associated with hypertension; irrespective of presence/absence of SHS exposure at workplace, SHS exposure at home was associated with hypertension. These findings suggest the importance of banning smoking at home as well as at the workplace. Fourth, the association between SHS exposure and hypertension increased with increasing daily time, week frequency and duration units of SHS exposure, although the ORs was not high. However, even the group with lower levels of frequency and duration of SHS exposure were also associated with hypertension, similar to the group with higher levels of frequency and duration of SHS exposure. This result suggests that it is necessary to avoid SHS smoke exposure to reduce the prevalence of hypertension.
Cotinine, a main metabolite of nicotine, has been known to be a useful biomarker in assessing actual smoking status with long half-life of about 17 h. Until now, although there is no clear cutoff point for urinary cotinine in terms of verifying current smoking, 50 ng/mL is generally used as a reference cutoff [17
]. In our previous study with 167,868 individuals, the urinary cotinine cutoff of 50 ng/mL provided sensitivity of 90%, specificity of 98.2%, positive predictability of 95.6%, and negative predictability of 95.9% for assessing smoking status according to self-reported questionnaire and cotinine verification [7
]. In particular, high specificity is an important indicator for minimizing the misclassification of true never-smokers. Recently, one study of 2889 never-smokers reported that higher serum cotinine levels, an objective marker of SHS exposure, were associated with higher systolic BP and the presence of hypertension [28
]. However, strictly speaking, self-reported never-smokers with high cotinine levels would be referred to as unobserved smokers rather than as exposed to SHS. As mentioned in our previous studies, unobserved smoking includes SHS and under-self-reporting of cigarette smoking [7
]. In particular, some smokers in Korea tend to make false statements of being never-smokers on self-reported smoking-related questionnaires because of the social recognition that the formerly Confucian culture is reluctant to accept female smokers and that there are potential disadvantages of male workers’ job life and promotion. Notably, the current study is the first to include cotinine testing as well as self-reported smoking and SHS exposure questionnaire to minimize the misclassification of SHS exposure.
As described in our previous studies [7
], the present study has several limitations as well. First, because the design of this study is cross-sectional, we cannot confirm the causal relationship between SHS exposure confirmed by self-reported questionnaire and cotinine and hypertension. Second, most of the study population was middle-aged people who resided in metropolitan areas; therefore, they might not be representative of the overall Korean population. Third, because the questionnaire items in this study did not completely separate the SHS exposure at home and workplace, our results would not be able to pinpoint the effect of SHS exposure at home and workplace on hypertension, respectively. However, based on the questionnaire presented in Figure 1
, we could divide the four groups of SHS exposure at home and workplace indirectly. Lastly, we did not have any information on the potential parameters that could affect nicotine metabolism such as genetic variations and drugs, and this study did not include dietary factors influencing hypertension. Nonetheless, our study is meaningful in that we are the first to conduct a large epidemiologic study to assess the association between SHS exposure and hypertension in never-smokers verified by self-reported questionnaire and cotinine.