The role of sociodemographic factors associated with waterpipe smoking among male adolescents in western Iran: A cross-sectional study

INTRODUCTION Waterpipe smoking (WPS) is an increasingly popular leisure activity among young people in Iran. The purpose of this study was to identify the role of sociodemographic factors associated with WPS among male adolescents in Iran. METHODS The study used a cross-sectional design. It included 730 high school male students (Grades 10–12) recruited through multistage random sampling conducted in 2017 in Hamadan city, western Iran. The self-administered questionnaires included information on demographic variables and behavioral risk factors related to WPS. Descriptive statistics and multinomial logistic regression modeling were conducted using SPSS. RESULTS The student mean age, and standard deviation (SD), and age at WPS initiation were 16.41 (0.84) and 13.31 (2.43) years, respectively. The percentages of never, former and current WPS were 37.3%, 36.4% and 26.3%, respectively. We found that ever cigarette smoking (OR=5.14, 95% CI: 2.56–10.32) and WPS family (OR=2.55, 95% CI: 1.40–4.64) were significantly associated with former WPS. Furthermore, being 18 years, studying in technical fields, reporting ever and current smoking of cigarettes and family usage of WP were significantly associated with current WPS. Friends with WPS (OR= 0.50, 95% CI: 0.34–0.72) however played a protective role on former WPS. CONCLUSIONS The results indicate that the prevalence of former and current WPS was high in Hamadan city. Thus, designing and implementing interventions for increasing students’, friends’ and family’s awareness regarding the harms of WPS and cigarette smoking are necessary to facilitate behavior change.


INTRODUCTION
Waterpipe (i.e. hookah, narghile, hubble-bubble) smoking (WPS) has been an old tradition in different parts of the world 1 . Smoke for consumption is produced by burning wood coal situated at the top of a perforated aluminum foil below which mostly flavored tobacco blend is placed 2 .There are several harms associated with smoking waterpipe such as cancers, cardiovascular disease, stroke and others 3 . However, smoking waterpipes continues to be a common form of tobacco use in various regions of the world, particularly the Middle East. Many people, particularly in the Middle East, think that WPS is less harmful than cigarette smoking or even harmless 4 , but studies have shown that WPS has serious, chronic, harmful effects, particularly on the respiratory and cardiovascular systems 5 .
It is estimated that 100 million people, particularly  10.1% for ever use and 4% for current use of WP among students in Grades 9-12. In a study in Iran, 9.7% of students smoked WP in the past month, of which 66.6% were male and 33.4% were female 7 . In another study in Iran, WPS prevalence was reported as 17.1% in male high school students 9 . WPS has a strong social aspect that makes it attractive and addictive to its users 10,11 . The results of a qualitative study showed that WPS is considered as enjoyable entertainment among friends, regardless of the health outcomes 12 .
Teenagers are especially attracted to the fragrance, to the pipe's nice appearance, easy accessibility, low cost, less stigma, and greater social acceptance 7 . There is generally a paucity of studies done on WPS in Iran, especially among high schools 13,14 . Increasing prevalence in Iran and associated negative sequelae warrant planning and implementing preventative approaches that are geared for adolescents in particular. It is generally well accepted that epidemiological studies identifying the determinants of risky behaviors such as WPS, should be undertaken before developing interventions 15 .Therefore, the purpose of this study was to estimate the prevalence of WPS and assess behavioral and social predictors of WPS among male adolescents in Iran.

Study participants
A total of 780 participants were recruited in the study. However, only 730 participants finished the study yielding a 93.5% completion rate. Multistage random sampling from 13 schools in two sectors of Hamadan city was employed with 60 students from Grades 10-12 participating in the study from each school. This study was approved by the Ethics Committee of Hamadan University of Medical Sciences (ID: IR.UMSHA.REC.1396.21). Informed consent, after due explanation, was obtained from all participants before the commencement of the study.

Measures
The instrument was a self-report and consisted of information on demographic variables and WPSrelated behaviors. The completion of the instrument took about 15-20 minutes. The instrument was divided into two parts: 1) Demographic factors including age, grade, major, father's and mother's occupation, father's and mother's education, whether having own room and living status; 2) WPS-related behaviors including being never, former (smoked waterpipe even one or two inhalations in your life) and current (smoked waterpipe at least once in the previous month) WP smoker, age at WPS initiation, computer games (at all, low, medium, high) internet usage (at all, low, medium, high) 15 , experienced cigarette smoking (yes/no), current cigarette smoking (yes/no), having friends who smoke WP (yes/no), and having family members who smoke WP (yes/no).

Data analysis
Descriptive statistics including frequency, percentage, mean and standard deviation were used to describe study participants. To determine prevalence and predictive factors, a series of analytical tests and multinomial logistic regression were done. Adjusted odds ratios (AORs), with associated 95% confidence intervals (CI) are reported. All statistical analyses were performed using version SPSS 22.0. P-values less than 0.05 were considered statistically significant.

RESULTS
In this study students' ages ranged from 15 to 18 years with a mean of 16 3 cigarette smoking. A total of 403 (55.2%) students' friends did not smoke WP. Table 2 presents behavioral risk factors relating to WPS among the students. Table 3 shows that being 17 years old (AOR=3.61, 95% CI: 1.53-8.52), 18 years of age (AOR=3.99, 95% CI: 1.28-12.40), and studying in the technical fields (AOR=2.70, 95% CI: 1.49-4.90), were statistically significant predictors for current WPS. There was a trend (with reference group: 15 years old that never smoked WP) for increased likelihood of current smoking WP with increasing age; odds were 3.61 times higher for 17 years old (p=0.003) and 3.99 times higher for 18 years old (p=0.017) than for the reference group. Studying in the technical fields had odds of current WPS that were 2.70 higher than students in human sciences (p=0.001). Table 4 shows that using Internet more than 7 hours per day (AOR=4.53, 95% CI: 1.82-11.29), ever cigarette smoking (AOR=9.92, 95% CI: 4.68-21.02), current cigarette smoking (AOR=8.41, 95% CI: 2.80-25.48) and family WPS (AOR=7.22, 95% CI: 3.75-13.87) were statistically significant variables in relation to current WPS. Students who used the Internet more than 7 hours a day were 4.53 times more likely to be current smokers of WP compared to those who did not use the Internet and had never smoked WP (p=0.001). There were also significant differences according to ever smoking cigarettes, currently smoking cigarettes and family member smoking WP. These were, respectively, 9.92, 8.41 and 7.22 times more likely to be currently smoking WP than never WPS (all p<0.001). Table 4 also shows that ever smoking cigarettes (AOR=5.14, 95% CI: 2.56-10.32), having WPS   Based on this analysis, there were significant differences according to ever smoking cigarettes and having a WPS family member. These were, respectively, 5.14 and 2.55 times more likely to be former smokers of WP compared to never WPS. The odds of former WPS were also negatively associated with having friends who smoked WP; students who had friends that smoked WP were 0.50 times less likely to be former smokers of WP than those whose friends did not smoke WP (p<0.001).

DISCUSSION
The In the present study, the age range of the first experience with WPS was between 6 and 18 years, with a mean age of 13.31 years. The results of this part of the study were consistent with those of similar studies. For example in Iran, the mean ages for initiation of WPS that were found ranged from 13.39 to 13.80 years 9,19,20 . The results of our study revealed that there is a need to plan and implement effective interventions to prevent tendencies for Reference category: Never WPS. WPS: waterpipe smoking, CI: confidence interval, AOR: adjusted odds ratio. a Categorical variables, b Reference group, *p<0.05, **p < 0.01, ***p < 0.001 substance abuse in adolescence. According to our findings, WPS was associated with the age of students. In this regard, increased age of students added to the chance of WPS. These findings are in congruence with other studies on this topic 9,18,[21][22][23][24] . It seems that early WPS at a younger age leads to increased addiction to WPS in older ages, resulting in students continuing WPS. Our observation that studying in technical fields resulted in higher odds of WPS than those in the human sciences might indicate differences in understanding of the health risks of hookah smoking and different social norms across disciplines. Similar findings have also been reported elsewhere 25 .
This study has shown that students who use Internet more than 7 hours a day were likely to be current WPS. Studies have shown that Internet addiction is related to tobacco smoking 26,27 . Those addicted to the Internet may conceivably be bored, tired, shy, depressed and suffer from other types of addiction, such as cigarette smoking 28 . Students that reported lifetime (ever) cigarette smoking were significantly more likely to also report former and current WPS than non-smokers of cigarettes. These findings are consistent with previous research 9, [29][30][31][32] . A possible explanation might be that WPS is being used by youth as an alternative to cigarette smoking among those who want to quit cigarette smoking; it could also be that WPS leads to other forms of tobacco smoking. Also, we found that current cigarette smokers were significantly more likely to be current WPS than others. The results of this part of the study were consistent with the findings of similar studies 24,33 . Perceiving WP as less harmful than cigarettes was positively associated with its use, suggesting that students may be uncertain about the harms associated with WPS relative to cigarette smoking. This misperception was more prevalent among young adults and in never smokers.
Although, previous studies 7, 10,31,34 have shown that having friends as users of WP is related to WPS, our study found that having friends as users of WP was inversely associated with former WPS. The reason for this might be that students who have used WP in the past have had unpleasant health experiences of the negative effects of WPS. The sharing of this unpleasant experience with friends who have not yet smoked hookah might have played a protective role in preventing initiation.
Results from our study show that social influence from family was significantly associated with students' former and current WPS. There are some studies with findings that are similar to our result 10,35,36 . Family members might own a WP for home use 37 and hold social gatherings at home to smoke WP 38 . In order to reduce WPS prevalence, public health interventions need not only to target adolescents but also their family members' risk perceptions, attitudes and behaviors regarding WPS.
This study has some limitations. First, the crosssectional design used in this study is restricted in establishing cause-effect relationships. Being cross-sectional it provides only a snap-shot in time and does not provide guidance on time sequence. Second, this study was restricted to students in Grades 10-12 in high schools of Hamadan, which limits its generalizability. Third, self-reported data could be prone to bias in reporting (i.e. under reporting or over reporting). Ideally, biochemical testing should have been done, but our results were not biochemically confirmed. Fourth, because of challenges getting permission to enter the girls' high schools, the current study was restricted to only male high schools. These findings may therefore not apply to all Iranian adolescents.

CONCLUSIONS
Our study shows that WPS prevalence is high among male high school students in Hamadan. There is a need to design interventional studies to increase students', friends' and parents' knowledge about the harms of WPS and cigarette smoking, to encourage behavior change in this regard, and to put in place restrictive policies for curtailing adolescents' access to WP and tobacco. In addition, our findings require public health campaigns to access and educate students, teachers and schools, to inform them of the growing understanding of the health effects of WPS.