Lifetime prevalence and correlates of smoking behavior in Iranian adults’ population; a cross-sectional study

Tobacco smoking is one of the main leading global causes of avoidable premature mortality [1] and still remains as one of the main global public health issues [2]. It is responsible for almost 5 million (~ 20%) of the worldwide annual deaths [3, 4]. It is expected that this number would increase to more than 8 million in 2030 [5], of which around 80% will occur in countries with low or middle-income level [6]. Its causal role in the development of adverse health related outcomes including several types of cancer [7‐9], chronic diseases [5, 10] as well as its high economic cost [4, 11] has been proven already.

Waterpipe smoking, also known as hookah, is the oldest form of tobacco smoking [12]. It is not only the most common method of tobacco use among youth in the Middle East, but also it is increasingly becoming an important global tobacco use method [13, 14]. It has been shown in a number of studies that the adverse health outcome of waterpipe smoking is equal to or worse than cigarettes smoking [15]. Evidence supports a positive association between waterpipe smoking and the risk of cancer [16], respiratory and cardiovascular disease [15, 17] and multiple sclerosis [18]. The common misunderstanding about the less dangerous nature of waterpipe smoking may be responsible for the increasing prevalence of waterpipe smoking [19].

There are substantial differences between and within geographical world regions regarding the prevalence of smoking [4]. The burden of tobacco use in Iran is considerable [20]. Based on the results of a meta-analysis published in 2013, almost 20% of Iranian male adults smoke [21]. Currently, few surveys simultaneously addressed the prevalence and modifiable correlates of waterpipe smoking, cigarette smoking and exposure to second-hand smoking in Iran. Tehran, the Capital of Iran, with a wide variety of ethnics and mass linguistic groups from all over Iran has provided the opportunity of more generalizable estimates of the smoking pattern and its correlates. Here using a population-based study we tried to investigate smoking pattern and its predictors in 15–50-year-old in Iranian population.

In this study, we used data of a population-based cross-sectional study that was conducted between 2013 and 2015 in Tehran, Iran. [18, 22, 23]. All residents of 22 municipality areas of Tehran aged 15–50 years constituted the study reference population. Using the standard method of random digit dialing (RDD) the study sample was selected. For generating the random dialing numbers, a 4-digit number was randomly added to the well-defined pre-codes of every 22 municipality areas of Tehran. The participants were selected proportionally to population size of each 22 areas of Tehran which had been determined in the last national Iranian census report. Using the Kish method [24], we selected one eligible individual in each of the selected household. The Kish method is a detailed process in order to randomly selecting one or more individuals when randomly calling to the households. In sum, it gives priority to the male and older residents and then, based on the participants’ gender and age, gives a rank to every resident of household. Using pre-specified specific tables, the Kish method randomly selects one or more individuals among the residents of each household [24]. Detailed recruitment flow-chart of the study participants has been shown in Fig. 1. As demonstrated, 1057 (70.0%) of 1510 contactable households, within which at least one eligible person existed, have been fully interviewed. Ten trained interviewers conducted the interviews. [18]. At the start of each interview, the participants were informed about the goals of the study and oral consent was obtained. Each interview on average took 15 min.

The detailed recruitment flow-chart of participants is shown in Fig. 1. Of the 1510 contactable households with eligible people, 1057 (70.0%) agreed to a full interview. The mean age was 31.3 and 51.5% of them were female. The majority of participants were less than 40 years old (Table 1).

Using population-based data in Iranian adults, we found exposure to second-hand smoking as the most prevalent smoking behavior (37%) followed by water-pipe (23.9%) and cigarette smoking (20.25%), respectively. Age, sex, history of depression along with lifetime alcohol intake were the important predictors of all three types of smoking habits. Moreover, cumulative life-event number predicted the cigarette and exposure to second-hand smoking. The lifetime drug abuse was only associated with cigarette smoking. The latest statistics provided by Drope et al. in Tobacco atlas demonstrated that 14.2% of males and 0.4% of females smoke cigarettes daily [31]. This is less than our finding (41.07% in females vs. 32.68% in males) in Tehran.

We identified exposure to second-hand smoking as the most prevalent smoking type in adults of Tehran. This finding indicated that current public places smoke-free policies may not be effective enough. Moreover, it emphasizes the need for strengthening health-related knowledge and also planning effective health education programs in family environments. In a cross-sectional study enrolled 5900 adults (15–75 years old) in Kerman, the prevalence of exposure to second-hand smoking was 27.5% (30.1% in females vs. 25.0% in males) [32] which is moderately less than our findings (41.07% in females vs. 32.68% in males). Our estimate is moderately less than one published systematic review by Zeng et al. in 2016 in which the prevalence of exposure to second-hand smoking in Chinese 15–59 years old was 47.1 [33].

In a cross-sectional study in Tehran which recruited 1830 participants, the prevalence of current waterpipe smoking in ≥15 years old was 17.6% which is similar to our estimates [34]. However, the lifetime prevalence of waterpipe smoking in current study was higher than recent estimates (12%) for the UK in the 11–16 years old age group [35]. The different recruited age group of the UK study potentially limits the comparability of findings. In California Cigarette Survey study by Smith et.al, 24.5% of young men (18–24) were ever waterpipe smokers [36]. In a systematic review conducted in 2008 in Middle Eastern countries, the minimum and maximum national reported prevalence of current waterpipe smoking among adults were 5 and 15%, respectively[13]. While, evidence shows that the prevalence of cigarette smoking is decreasing in developed countries, an observed increasing prevalence of the other forms of tobacco smoking such as waterpipe smoking is highly concerning [36, 37]. A most common public misunderstanding of the less harmful adverse health effects of waterpipe compared to cigarette smoking may be one of the possible reasons for this increasing prevalence [38]. Nonetheless, studies have shown that sweetened flavored tobacco products used in waterpipe, contain several toxicant agents, carcinogenic materials and heavy metals [39, 40]. Moreover, the carbon monoxide dose as well as the volume of inhaled smoke when smoking waterpipe are very much higher than those in cigarette because of exposure to the charcoal used for heating the waterpipe [15]. There is also evidence demonstrating that wastepipes have adverse short-term and long-term health effects including infectious, respiratory and cardiovascular disease, multiple sclerosis, cancer as well as mental health disorders leading to an increased risk of illegal substance use in young people risk [17, 41].

In a population-based cross-sectional study in Tehran by Fotouhi et al. which enrolled 3397 residents, the reported prevalence of smoking (11.9%) was less than our estimate (20.25%) [42]. This could suggest an increasing trend in cigarette smoking among 15–50 years old residents of Tehran. The lifetime prevalence of cigarette smoking in males in this study is higher than results of a published meta-analysis in 2014 which reported that nearly 25% of 15–65 years old males of northern Iran were ever smokers [43]. Our estimate is also more than a published meta-analysis in 2013 in Iran in which only 20% of Iranian male adults have reported history of cigarette smoking [21]. The prevalence of three types of smoking habits in males and females were higher than those reported in an Iranian national report by Nemati et.al study. The results of this national report showed that 10.9 and 2.4% of the Iranian adults were daily cigarette and waterpipe smoker, respectively [44]. Moreover, the estimated joint prevalence of cigarette and waterpipe smoking in the present study was substantially higher than those reported in the Iranian national report by Nemati et.al (5.6% vs. 0.3%) [44].

The estimated prevalence of cigarette smoking in the current study is comparable with finding of a meta-analyses among Iranian adults published in 2013 [21]. In recent meta-analysis the range of cigarette smoking was reported between 12.3 to 38.5% in men, and between 0.6 to 9.8% in women, which is comparable to our estimates [21].

Consistent with a study by Albisser et al. conducted in Switzerland among 204 young adults [45], our findings indicated that waterpipe smoking could predict both cigarette and exposure to second-hand smoking habits. Although exposure to second-hand and cigarette smoking were associated with each other, they could not predict engaging in the water pipe smoking.

Alcohol intake was highly correlated with both lifetime waterpipe and cigarette smoking (p for trend< 0.05). Consistently, Tamim et al., in a cross-sectional study recruited 1964 students from Lebanon Universities, found that heavy drinkers were more likely to be waterpipe and cigarette smoker [46]. It is well-acknowledged that cigarette smoking can predispose to alcohol and substance use. The possible mechanism of this association is still not clear. As expected, engaging in drug use activities was associated with increased likelihood of cigarette smoking [47]. However, the cross-sectional nature of this design limits the proper interpretation of the temporality between cigarette smoking and drug abuse. Depression was the other factor that predicted waterpipe and cigarette smoking habits but not exposure to second-hand smoking. This was similar to the other studies [48, 49], but not all of them [50]. The cumulative number of life time stressful events was associated with both cigarette and exposure to second-hand but not waterpipe smoking (p for trend< 0.05). Prior studies have similarly demonstrated that the higher level of stress could directly or indirectly elevate substance use among young people [51, 52]. Further examination is needed to understand the role of life time stressful events in the context of other forms of tobacco use.

There is a gender difference in the lifetime prevalence of waterpipe (35.3% in males vs. 13.1 in females) as well as the lifetime prevalence of cigarette smoking (35.9 in males vs. 5.5% in females) in this study. In general, other than exposure to second-hand smoking, the estimated lifetime prevalence was higher in males than females. This is in line with the results of other studies [44, 53]. However, exposure to second-hand smoking was more likely in Iranian females. The later finding shows that women are often affected by the others’ smoking behaviors. Contrary to our finding, Riachy et al. in a study, recruited 37,579 participants during 2003 to 2005 in Lebanon, showed that females consumed more waterpipe than males [54]. While the likelihood of cigarette smoking increased with age, reversely, the probability of being waterpipe smoker increased in young adults (20 to 29 years old). This could well demonstrate the age-related differences in pattern of smoking behaviors in Iranian population. Finally, increased years of education could decrease the probability of exposure to the second-hand smoking in the study population.

There are some advantages and limitations that should be considered. This study recruited a population-based data set to describe lifetime prevalence and related factors of waterpipe, cigarette and exposure to second-hand smoking among Iranian 15–50 years old adults. The response rate (70%) in this study was found to be satisfactory. Although the usefulness and efficacy of RDD sampling as well as its similarity with address-based sampling were previously demonstrated [55‐57], some possibility of selection bias may still have remained. This could affect the validity of study findings. Although we conducted a population-based study in a multiethnic city, the generalizability of the study finding to the whole Iranian population should be done after considering the rural vs. metropolitan differences in smoking behavior. The more stressful nature of metropolitan environments [58], increasing smoking habits in immigrate of the metropolitan areas [59], and finally negative attitude toward smoking habits in the rural areas may differentiate the prevalence of smoking habits in rural area compared to urban area. Considering the illegal nature of alcohol and illicit drugs consumption in Iran society, there was the possibility of the underreporting with respect to these sensitive behaviors. However, we reported one of the largest estimates of the three smoking behaviors as well as alcohol intake and drug abuse compared to the previous studies in Iran. This could alleviate the possibility of a major under reporting. Moreover, the cross-sectional nature of the study design, especially the temporality issue, precluded us from drawing causal conclusions with respect to the identified associations.

Lifetime prevalence of waterpipe, cigarette and exposure to second-hand smoking is moderately high among Iranian adults. Similarly, compared with previous studies, dual smoking behaviors are increasing in Iranian adults. Age and male sex, history of depression, life time stressful events, drug abuse, alcohol consumption as well as years of education were all found to have moderate correlations with smoking habits. There was an apparent age-related difference in pattern of waterpipe and cigarette smoking behavior in Iranian population. These findings highlight the need for further educational and preventive programs, especially for dual smoking in Iranian young adults, to stress the underlying potential adverse effects of waterpipe smoking. This could provide practical information for evaluating and reforming the tobacco control programs and policies in Iran.