Social Aspects of Hookah Smoking Among US Youth
- Israel Agaku, DMD, MPH, PhD,
- Satomi Odani, MPH,
- Brian Armour, PhD, and
- Rebecca Glover-Kudon, PhD, MSPH
- Office on Smoking and Health, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, Georgia
Dr Agaku conceptualized and designed the study and drafted the initial manuscript; Ms Odani helped conceptualize the study, assisted in the statistical analyses, and critically reviewed and revised the manuscript; Drs Armour and Glover-Kudon helped conceptualize the study and critically reviewed and revised the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
BACKGROUND: We investigated 3 social dimensions of youth hookah smoking: frequency, places smoked, and descriptive social norms.
METHODS: Data were from the 2016 National Youth Tobacco Survey of US sixth- to 12th-graders (n = 20 675). Hookah smoking frequency was classified as never, former, current occasional, and current frequent. Places where past 30-day hookah smoking occurred and students’ perceptions of their classmates’ hookah smoking prevalence were assessed. Descriptive and multivariable analyses were performed (P < .05).
RESULTS: Overall, 10.5% reported smoking hookah ≥1 time in their lifetime. Of these, 65.8% were former, 26.3% were current occasional, and 7.9% were current frequent smokers. Overall, 59.3% of students overestimated hookah smoking prevalence in their grade. Current occasional smoking was predicted by female sex (adjusted odds ratio [aOR] = 1.54) and peer hookah-smoking overestimation (aOR = 9.30). Current frequent smoking was most strongly predicted by living with a hookah smoker (aOR = 20.56), speaking a second language other than English (aOR = 2.17), and co-use of mentholated cigarettes (aOR = 19.94) or other flavored noncigarette tobacco products (aOR = 17.59). The top 3 places hookah was smoked were a friend’s house (47.7%), the respondent’s own house (31.8%), and another family member’s house (20.8%).
CONCLUSIONS: The home environment was the most common place for youth hookah smoking. Home-tailored interventions that encourage voluntary smoke-free rules and warn about the dangers of social smoking could help denormalize hookah smoking.
- aOR —
- adjusted odds ratio
- e-cigarette —
- electronic cigarette
- NYTS —
- National Youth Tobacco Survey
What’s Known on This Subject:
During 2016, ∼1 million US sixth- to 12th-graders reported past 30-day hookah smoking. Most youth who smoke hookah misperceived that hookah is not harmful. Hookah is marketed in attractive flavors, which may reinforce perceptions of lowered harm.
What This Study Adds:
Of ever hookah smokers, 65.8% were former, 26.3% were occasional, and 7.9% were frequent smokers; 3 in 5 students overestimated peers' hookah smoking prevalence. Various home environments comprised the top 3 places where hookah tobacco was smoked.
The design, physical characteristics, and use patterns of hookah encourage group consumption and social adhesion.1,2 Unlike other forms of tobacco, for which the product may be presented to the user with minimal effort, hookah use is planned and intentional. Use of hookah devices requires a multistep process for assembly, and users come to the device to smoke. Moreover, some hookah smoking devices are designed for multiple users1,2; therefore, hookah smoking can be reinforced by social networks. Of all smoked tobacco products, hookah bears the least physical resemblance to cigarettes; this, coupled with the fact that hookah smoke passes through a water column before reaching the smoker, might contribute to misperceptions of reduced harm.1–4 Furthermore, unlike cigarettes, hookah is marketed in multiple flavors other than menthol, including fruit, chocolate, and candy, which may enhance product appeal.3 Whereas cigarette smoking prevalence has continued to decline, hookah smoking prevalence among certain populations is increasing.5–8 Often located in close proximity to college campuses, hookah lounges and bars, which are increasing in number, are frequently marketed as places for socialization.9–11 In these settings, hookah smoking may be the purpose for gathering; it is not tangential to the main event but rather is the main event.
Three important components of the hookah-smoking social experience include the frequency of use (eg, “social smoking”), place of use (eg, hookah bars and/or lounges), and social networks associated with use (eg, family and peers).9–13 Some of these themes are evident in hookah advertisements, which often depict hookah being smoked among friends within social settings.10,11 Internet search patterns within the United States revealed that online demand for information on water pipes and/or hookah products is large and increasing, and common queries made in relation to hookah included “hookah lounge,” “hookah near me,” and “does hookah have tobacco in it?”14,15 During 2016, ∼1 million US middle and high school students reported past 30-day hookah smoking.7
Although researchers in several studies have examined past 30-day hookah smoking and its correlates among US adolescents,7,16,17 relatively little is known about some of the aforementioned social aspects of hookah smoking among adolescents. For example, no previous researchers have directly compared perceived prevalence of hookah use versus actual use or the places for recent hookah smoking among a national sample of US adolescents. Most of the extant research on social dimensions of hookah smoking is among young adults and college students.9,12,18–20 Teenagers and adolescents, however, differ from young adults in key respects, including disposable income, ability to enter adult-only facilities, and tobacco access patterns. Therefore, it is reasonable to expect differences in hookah-related social behaviors between teenagers and young adults. We investigated hookah smoking frequency, places smoked in, and hookah-related descriptive social norms among a national sample of US adolescents.
Data were from the 2016 National Youth Tobacco Survey (NYTS), a school-based, paper-and-pencil, cross-sectional survey of US students attending public and private schools in grades 6 to 12.21 A 3-stage cluster sampling design was used to select a nationally representative sample of 20 675 students in 2016 (response rate = 71.6%).
Frequency of Hookah Smoking and Harm Perceptions
Ever hookah smoking was assessed with the question, “Have you ever tried smoking tobacco in a hookah or water pipe, even 1 or 2 puffs?” Response options were “yes” or “no.” Current (any past 30-day) use was assessed with the question, “During the past 30 days, on how many days did you smoke tobacco in a hookah or water pipe?” Response options were “0 days,” “1 or 2 days,” “3 to 5 days,” “6 to 9 days,” “10 to 19 days,” “20 to 29 days,” and “all 30 days.” Flavored hookah products were those that tasted like “menthol (mint), alcohol (wine, cognac), candy, fruit, chocolate, or other sweets.” Hookah smoking frequency was classified as never (“not even 1 or 2 puffs”; n = 17 562), former (last smoked >30 days ago; n = 1395), current occasional (smoked on less than one-third of the past 30 days [1–9 days]; n = 564), and current frequent smoker (smoked on greater than or equal to one-third of the past 30 days [10–30 days]; n = 199). Among ever hookah smokers, duration since initiation was computed as current age minus age at first trying hookah.
Hookah harm perception was assessed with the question, “How much do you think people harm themselves when they smoke tobacco in a hookah or water pipe some days but not every day?” The 4 response options provided were dichotomized as not harmful (“no harm” or “little harm”) or harmful (“some harm” or “a lot of harm”).
Hookah-Related Descriptive Social Norms
Social learning among adolescents is heavily influenced by descriptive norms, which describe perceived typical patterns of behavior (eg, the misperception that “everyone else is doing it”).22–24 We measured the discordance between actual and perceived prevalence and assessed associations between overestimation of peer hookah use and respondents’ hookah-related behaviors. Perceived prevalence of peer hookah smoking was assessed with the question, “Out of every 10 students in your grade at school, how many do you think smoke tobacco in a hookah or water pipe?” Answers ranged from 0 to 10. On the basis of comparisons with actual prevalence (past 30-day use) rounded to the nearest 10%, all responses were dichotomized as either overestimating (perceived prevalence greater than actual prevalence) or not overestimating (perceived prevalence less than or equal to actual prevalence) grade-specific peer hookah use. For example, if a 12th-grade student perceived that 3 of 10 (30%) students in his or her grade level smoked hookah, whereas the actual prevalence was 7.2% (after rounding becomes 10%), then this student would be classified as overestimating the prevalence of hookah smoking (30% > 10%).
Respondents’ household members’ tobacco use status was classified as nonusers, hookah smokers (exclusively or concurrently with other tobacco products), and other nonhookah tobacco product users alone.
Place of Hookah Smoking
Place of recent hookah smoking was assessed with the question, “During the past 30 days, where did you smoke tobacco in a hookah or water pipe?” Response options were, “I did not smoke tobacco in a hookah or water pipe during the past 30 days,” “at my house,” “at a friend’s house,” “at a family member’s house other than my house,” “at a hookah bar,” “at a cafe or restaurant,” and “some other place not listed here.”
Other Tobacco-Related Measures
Current (any past 30-day) use was also assessed for cigarettes, roll-your-own tobacco, electronic cigarettes (e-cigarettes), cigars, pipes, smokeless tobacco, and bidis. Menthol cigarette smoking was a report of smoking mentholated cigarettes or predominantly menthol brands (Newport and/or Kool).25 Flavored noncigarette tobacco products were those that tasted like “menthol (mint), alcohol (wine, cognac), candy, fruit, chocolate, or other sweets.” All other noncigarette tobacco products not meeting this characterization were classified as nonflavored.
Regarding attempted tobacco purchase, participants were asked, “During the past 30 days, did anyone refuse to sell you any tobacco products because of your age?” Response options were, “I did not try to buy any tobacco products during the past 30 days,” “yes,” and “no.” We were interested in assessing whether past 30-day hookah smokers who tried (regardless of success) to purchase a tobacco product differed from those who did not try at all (and conceivably obtained their tobacco products entirely from social contacts).
Sociodemographic variables included sex, age, race and/or ethnicity, grade and/or school level, and second-language aptitude (assessed with the question, “Do you speak a language other than English at home?”). The latter was used as a measure of acculturation and/or assimilation, particularly because hookah use, although relatively new in the United States, has strong roots in other parts of the world, including the Eastern Mediterranean region, Asia, and parts of Africa.6 Despite this variable’s low geographic and linguistic specificity (eg, impossible to discriminate students from families originating in regions of the world with high hookah prevalence versus low prevalence or individuals with an acquired and/or natural versus learned and/or conscious second language), we used it to crudely explore underlying social constructs.
All data were weighted to account for the complex survey design. Descriptive analyses were performed, and prevalence estimates were compared by using χ2 tests at P < .05. Multinomial logistic regression analyses were performed to determine factors associated with hookah smoking frequency among all students; never hookah smokers were the reference group. Independent variables assessed were overestimation of peer hookah smoking, household member tobacco use, sex, school level, race and/or ethnicity, second-language aptitude, perceived hookah harmfulness, and current use of cigarettes and other tobacco products.
Five separate binary logistic regression models were used to identify factors associated with hookah smoking setting among past 30-day smokers (locations were not mutually exclusive). Hookah bars were combined with hookah cafes and/or restaurants because of small sample sizes; all other places were analyzed separately. Independent variables were overestimation of peer hookah smoking, household member tobacco use, sex, school level, race and/or ethnicity, second-language aptitude, years since first hookah use, past 30-day hookah smoking frequency, hookah flavor preferences, current use of cigarettes and other tobacco products, and past 30-day attempted tobacco purchase. Unless otherwise stated, all comparisons shown below are statistically significant at P < .05.
Prevalence of Ever and Current Use of Hookah
The overall prevalence of ever and current hookah smoking was 10.5% and 3.6%, respectively (Table 1). Of ever hookah smokers, 65.8% were former, 26.3% were current occasional, and 7.9% were current frequent smokers. In total, 28.1% of those who first tried hookah ≥2 years ago, 31.6% of those who first tried hookah 1 year ago, and 39.6% of those who first tried hookah <1 year ago reported any past 30-day hookah smoking.
Descriptive Social Norms and Hookah Smoking Frequency
Both perceived and actual prevalence of hookah smoking generally increased with increasing grade level (Supplemental Fig 1). Perceived hookah use prevalence was higher than actual hookah use prevalence by as much as 10-fold in the ninth grade and by at least fivefold in the twelfth grade. Overall, 59.3% of students overestimated the prevalence of hookah smoking within their grade (Table 1).
Overestimation of perceived peer hookah smoking strongly predicted current occasional hookah smoking (adjusted odds ratio [aOR] = 9.30), whereas having a household member who smoked hookah strongly predicted current frequent hookah smoking (aOR = 20.56; Table 2). Female respondents had lower odds than male respondents of being current frequent hookah smokers (aOR = 0.65) but had higher odds of reporting current occasional (aOR = 1.54) and former (aOR = 1.12) hookah smoking. Odds of hookah smoking were also higher among Hispanics and African Americans than whites and among those who spoke a second language other than English at home versus English-only speakers. Smoking mentholated cigarettes (aOR = 19.94) or using flavored varieties of other nonhookah tobacco products (aOR = 17.59) strongly predicted current frequent hookah smoking. The perception that nondaily hookah smoking is harmful was protective of hookah smoking.
Place of Hookah Smoking
Settings reported (not mutually exclusive) for past 30-day hookah smoking among all ages were at a friend’s house (47.7%), the respondent’s own house (31.8%), another family member’s house (20.8%), a hookah bar (12.0%), a cafe and/or restaurant (5.9%), and other places (22.9%). Similar percentages were observed among hookah smokers <18 years of age in relation to the place of hookah use, including the hookah bar (11.6%) and cafe and/or restaurant (4.8%).
The odds of visiting a hookah bar, cafe, and/or restaurant were 1.96 and 3.05 times higher among those who smoked hookah on 6 to 29 days and all 30 of the past 30 days than among those who only smoked 1 to 2 days, respectively (both P < .05; Table 3). Similarly, the odds of smoking hookah in one’s own house were 2.03 and 1.87 times higher among those who smoked hookah on 3 to 5 days and 6 to 29 days of the past 30 days than among those who only smoked 1 to 2 days, respectively. Those with a hookah smoker living in the household had 2.52 times higher odds of visiting a hookah bar, cafe, and/or restaurant and 3.40 times higher odds of smoking hookah in their own house than those with no tobacco product user in the household. Furthermore, those who indicated a preference for flavored hookah products had 1.88 times higher odds of visiting a hookah bar, cafe, and/or restaurant and 1.75 times higher odds of smoking hookah in their own house compared with those who smoked nonflavored hookah products. Compared with those who started hookah smoking <1 year ago, those who started ≥2 years ago had 2.47 times higher odds of smoking hookah in their own house.
The odds of smoking hookah in a friend’s house were higher among those indicating a preference for flavored over nonflavored hookah products (aOR = 1.54) and among those who used flavored nonhookah tobacco products than nonusers (aOR = 2.32). The odds of smoking hookah in a friend’s house were lower among non-Hispanic African Americans than among non-Hispanic whites (aOR = 0.44).
The odds of smoking hookah in a family member’s house were higher among high school students than among middle school students (aOR = 3.35) and among those who tried to buy tobacco products in the past 30 days than among those who made no attempt (aOR = 2.39). Correlates of hookah smoking in some other place not specified are shown in Table 3.
Approximately 3 in 5 students overestimated hookah smoking prevalence within their grade, and most current hookah smokers reported occasional use. Female respondents were less likely than male respondents to smoke hookah frequently but more likely to smoke occasionally, revealing different social patterns of hookah smoking between the sexes. The top 3 places hookah was recently smoked in were at a friend’s house, the respondent’s own house, and another family member’s house. Hookah devices are typically cumbersome, less portable, and more difficult to hide compared with other tobacco products, which reveals that youth hookah smoking in the home might be occurring with parents’ or caregivers’ awareness, permission, or involvement. This is consistent with the fact that youth with a household member who smoked hookah were more likely to smoke hookah frequently and in their own homes. Hookah tradition may be culturally prescribed and closely tied to identity formation, which is a particularly salient topic for adolescents.1,2 This might help explain the finding of higher odds of hookah smoking among youth who spoke a second language other than English at home compared with English-only speakers.
Relatively easy access to tobacco products within the homes of friends and relatives, or from fellow household members who smoke, could attenuate the impact of laws aimed at reducing youth consumption of tobacco products. These findings buttress the importance of the home as a critical environment for tailoring the content, messaging, and placement of hookah-related health communication interventions for adolescents as well as parents and/or caregivers. For example, mass media campaigns that provide education about the harmfulness of hookah smoking and the importance of adopting voluntary smoke-free home rules could help denormalize hookah smoking.26
As a communal experience, hookah smoking takes longer and prolongs social and physical exposure.1,2 Relative to other forms of tobacco, social pressure may be greater because the behavior is immediately observable in a group setting; refraining from hookah smoking may require a higher degree of refusal skills. The group experience may further reinforce misperceptions about harms (eg, that hookah smoking is a harmless hobby rather than a dangerous addiction) and social norms, particularly if hookah smoking occurs in conjunction with parties and celebratory gatherings.1,2 This group experience might also allow the price for a hookah bowl to be split among the group as a cost-avoidance strategy; in our study, there was no significant difference between hookah smokers who recently attempted versus did not attempt to buy a tobacco product in relation to the odds of smoking hookah at most settings assessed.
Of hookah smokers <18 years of age in our study, 11.6% reported past 30-day hookah smoking in a hookah bar, whereas 4.8% reported smoking hookah in a cafe and/or restaurant. Researchers measuring air pollution from hookah bars and lounges have found levels many times higher than those deemed by the Environmental Protection Agency to be hazardous for outdoor air.27,28 One hookah smoking session alone is estimated to generate as much smoke inhaled as 100 to 200 cigarettes.3 As of December 31, 2015, a total of 23 US states had no complete smoke-free laws in either bars or restaurants (including 7 states with no form of smoking restriction in both places).29 Even in states with comprehensive smoke-free policies, hookah establishments often take advantage of various exemptions in statewide smoke-free laws, such as those exempting “retail or wholesale tobacco shops and private smokers’ lounges” as well as owner-operated business exemptions.30–32 Including hookah prohibitions within the context of comprehensive smoke-free policies, restricting the sale of flavors that are attractive to youth, and raising the minimum age of buying tobacco products or accessing tobacco establishments (eg, hookah bars) to 21 years could benefit public health.33
Pediatric practitioners can play a key role in addressing some of the social underpinnings of hookah smoking among youth, including addressing parents’ and caregivers’ tobacco use behavior during child-wellness visits.34 Furthermore, when counseling adolescents, pediatric practitioners can reinforce the fact that hookah smoking among youth is far from ubiquitous and is harmful and addictive. Previous researchers show that hookah use is associated with various cancers, heart and lung disease, as well as other adverse health outcomes.35–39 In our study, ∼3 in 10 youth who first tried hookah ≥2 years before the study and 4 in 10 of those who first tried within the past year still smoke hookah currently. More so, hookah smoking was associated with current use of other tobacco products, particularly flavored varieties, which might further increase the likelihood of nicotine addiction.
There are several limitations to this study. First, given the cross-sectional design, alternative explanations might exist for some of the noted associations. For example, youth might smoke hookahs because they think “everyone is doing it”; alternatively, they may overestimate hookah smoking because they, and presumably the people they directly know, are doing it. Secondly, rounded estimates of actual and perceived prevalence were used in measuring overestimation of hookah smoking; this reduces precision and might not preserve the actual relationships. Thirdly, these findings may have limited generalizability to school-aged youth who were excluded from the NYTS sampling frame (eg, school dropouts). Finally, data were unavailable for socioeconomic factors, such as poverty status and parental education.
Youth hookah smoking is a social behavior and most commonly occurs at a friend’s house, the respondent’s own house, or at another relative’s house. Peer influence most strongly predicted occasional smoking, whereas household member influence most strongly predicted frequent hookah smoking. More than 3 in 5 youth overestimated the prevalence of hookah smoking within their grade. Mass media and programs that educate youth and their parents, caregivers, and health providers on the risks of hookah smoking and encourage the adoption of smoke-free home rules could help denormalize hookah use among youth.
- Accepted May 18, 2018.
- Address correspondence to Israel Agaku, DMD, MPH, PhD, Office on Smoking and Health, Centers for Disease Control and Prevention, 4770 Buford Highway, Mailstop F-79, Atlanta, GA 30341. E-mail:
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the US Centers for Disease Control and Prevention.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- Copyright © 2018 by the American Academy of Pediatrics