Exploring Demographic and Substance Use Correlates of Hookah Use in a Sample of Southern California Community College Students

Background and Purpose: Hookah smoking is a growing young adult phenomenon, particularly among college students. Many users feel that it is safer than other tobacco products, although its health threats are well documented. Little is known about hookah use rates in community colleges that are attended by nearly half of all US college students. This study examined hookah use in a diverse convenience sample of students attending two southern California community colleges. Methods: In fall 2011, a crosssectional, in-classroom survey was administered to 1,207 students. A series of fully adjusted multivariate logistic regressions were conducted to explore demographic, other substance use, and attitudinal correlates of lifetime and current hookah use. Results: Lifetime hookah use (56%) was higher than lifetime cigarette use (49%). Gender and personal socioeconomic status were not related to hookah use. Current use (10.8%) was associated with current use of alcohol, cigars, and cigarettes. Compared to African-Americans, Whites were 2.9 times more likely to be current users, and students who perceive hookah to be more socially acceptable were 21 times more likely to currently use. Conclusion: Since hookah use rates are high, colleges should offer health education programs to inform incoming students about the health risks of hookah and cessation programs. © 2015 Californian Journal of Health Promotion. All rights reserved.

Common correlates of hookah use at four-year schools included the concurrent use of other substances like alcohol, marijuana, cigars/cigarillos, and/or other illegal drugs (Braun, Glassman, Wohlwend, Whewell & Reindl, 2012;Jarrett et al., 2012, Rath et al., 2012).Although cigarette smokers may already be using hookah (Braun et al., 2012;Jarrett et al., 2012), users perceived hookah smoke to contain less nicotine, and therefor saw it as less addictive and harmful than cigarette smoke (Heinz, et al, 2013;Griffith & Ford, 2014;Noonan & Patrick, 2012).This belief may have increased its acceptance, including on social media outlets such as Facebook, with more than 5% of college students displaying profile pictures referencing its use (Brockman et al, 2012).Others found that over time, more hookah bars and cafés are located near college campuses (Noonan & Patirck, 2013;Holtzman, et al 2013) resulting in easy access to hookah in social settings that are appealing to students.
Many studies have shown that hookah users are more likely to be males (Jarrett et al., 2012;Primack et al., 2010).However, Barnett et al. (2013), found that on the University of Florida campus, males and females used hookah at similar rates; similarly Johnson et al. (2014) reported that there are less gender differences in use of hookah than other tobacco products.The use of hookah appeals to those of younger age, generally college freshmen or 18-19 year olds (Jarrett et al., 2012;Sutfin et al., 2011), with 23% of females initiating hookah during their freshman year (Fielder et al., 2012) and an overall decline in hookah use by age 29 and older (Johnston, 2014).Hookah use in younger persons may be related to cost as hookah is inexpensive and relatively easy to purchase online, increasing accessibility (Grekin & Ayna, 2012).Whites are much more likely to be users (Blosnich, Jarrett, & Horn, 2011;Primack, Sidani, J., Agarwal, Shadel, Donny, & Eissenberg, 2008) as are Arabs (Abughosh, Wu, Peters, Essien & Crutchley, 2011), while African-Americans are significantly less likely (Blosnich et al., 2011;Jarrett et al., 2012) to use hookah.

The Present Study
Understanding who is a hookah user in the collegiate setting is an important first step for intervening to curtail its growth in popularity.To date, however, what we know comes from students enrolled at four-year universities.However, according to the American Association of Community Colleges (2014) 45% of enrolled US students study at two-year community colleges.In general, these students tend to be more ethnically diverse and come from families of lesser economic means than those at four-year schools.Over half (56%) of all Hispanic undergraduates, 44% of all Asian-Pacific Islanders, 48% of all African-Americans, and 59% of all Native Americans study at community colleges.
The purpose of this study was to explore hookah use in an ethnically diverse group of community college students in southern California to see if user characteristics differ from those attending four-year universities, according to previous research.Specifically, we hypothesized that as in four-year colleges non-White students, and older students attending community colleges would be less likely to use hookah and that students who use tobacco and alcohol would be more likely to use also use hookah.We anticipated however that in community college students we would find higher levels of social acceptance, high rates of use in females, and given its low cost, high overall use rates compared to students in four-year colleges.

Participants and Procedures
In fall 2011, a cross-sectional in-classroom survey with 67 questions on hookah and other substance use behaviors and attitudes was administered to a convenience sample of adult (aged 18+) community college students enrolled in a required general education class in one of two participating community colleges in the Inland Empire region of southern California.An IRB trained instructor teaching at the college, but not teaching the classes in which the survey was conducted administered the survey.The survey took between fifteen and twenty minutes to complete and was voluntary and anonymous.A tear-off passive consent form with contact information was the front page of the survey and the Loma Linda University Institutional Review Board approved all aspects of the study.
Completed surveys (N=1,207) were scanned into the Scannable Office software and systematically verified by comparing scanned data to the original surveys.The resulting Access file was then exported into SPSS for analysis.

Dependent Variable Measures
Our dependent variables were lifetime and current hookah use.Lifetime ("ever") hookah use was asked three ways.If a student answered affirmatively to any of the three questions, he or she was considered an "ever" user.

Statistical Analysis
Once descriptive frequencies, means (SDs), and medians were computed, the relationship between hookah use (ever and current) and demographic variables was analyzed via fully adjusted multiple logistic regression models; the same was done for the relationship between hookah (ever and current) and other substance use (ever and/or current).Hookah belief variables were regressed on hookah use (ever and current) via bivariate multiple logistic regression models.Criterion for significance was p < 0.05.

Results
Relationships between the demographic variables and "ever" and current hookah use are described in Table 1.The following is a brief summary of the main results.

Hookah Use
A majority (55.5%) of the 1,207 students surveyed used hookah at least once and more than one-third (34.1%) in the past year (not tabled); 10.8% were current users.The average age of first use was 18.2 (SD = 4.0).
Students living with a boyfriend, girlfriend, and/or friends were twice as likely (AOR = 2.07, p < 0.05) as those living with parents/other family to be current users.With regard to students entertainment budget, those who spend relatively little, ($1-$20 per week), were about 35% less likely (AOR = 0.66, p < 0.05) to have "ever" used hookah than those who indicated they spent $0, but spending more was not associated with either "ever" or current use.Of note, being involved in organized sports, receiving financial aid, or working hours per week were not associated with "ever" and current use.Also, school (two schools were surveyed) was not significantly different for both "ever" and current use.Thus, the subsequent analyses presented here were conducted with all students combined.

Associations with Other Substance Use
Details on substance use frequencies and their relationship to hookah use can be found in Table 2.The following is a summary of the main findings.
Current use of two or more substances at the same time, which can include hookah, was the most significant correlate of current use, (AOR = 3.94, p < 0.001), followed by current alcohol use (AOR = 2.83, p < .001).Female binge drinking (35.9% of females; four plus drinks in one setting) was associated with current use (AOR = 1.76, p < 0.05) but male binge drinking (39.2%; five+ drinks) was not.Current cigarette use (AOR = 1.82, p < 0.05) and current cigar use (AOR=2.45,p < 0.01) were also associated.
Lifetime (ever) Hookah Use was associated with current (AOR = 1.89, p < .001)and "ever" alcohol use (AOR = 3.06, p < 0.001)."Ever" cigarette (AOR = 2.06, p < .001)and "ever" marijuana use (AOR = 2.12, p < .001)were correlated with "ever" hookah use.Of note, current marijuana use was not significantly associated with "ever" or current use of hookah, nor was current illegal drug use and current or "ever" smokeless tobacco use.

Hookah Beliefs
Hookah beliefs and correlates are summarized in Table 3. Students who perceived hookah as definitely less harmful and addictive than cigarettes were significantly more likely to be "ever" and current users than those who perceived it as definitely more harmful.The more a student believed hookah to be socially acceptable, the more likely he or she was to be an "ever" or current user.When asked approximately what percent of college students had "ever" used hookah, the median perceived answer was 70% (mean 60%).When asked what percent of close friends had "ever" used hookah, the median and mean were both 50%.Students answering with high percentages to these social acceptability questions had significantly higher odds of being "ever" and current users.Those who said 90-100% had used hookah were almost 25 times as likely (p < .001) to be an "ever" user and 11 times (p < .001) a current user than someone who answered that only 0-10% of their friends had used hookah.Mean = # 2.63 (SD = 1.4), between less and not sure.b Mean = # 2.44 (SD = 1.4) between less and not sure.c Median = # 4 "acceptable," Mean = # 3.29 (SD = 1.6) between "not sure" and "acceptable."d Median = 70%, Mean = 60% (SD = 2.5).As described in Table 4, when asked what came to mind (Top of Mind) when thinking about hookah, a majority of respondents (56%) associated hookah with tobacco use.For 15% of respondents, marijuana was their "top of mind" association, and for 6.7% it was alcohol.Those who associated hookah with alcohol were significantly more likely to be "ever" (OR = 3.85, p < .001)and current (OR = 3.07, p < .001)users.Conversely, those who associated hookah with marijuana were less likely to be "ever" (OR = 0.45, p < .001)and current (OR = 0.51, p < .05)users.This was not the case for "athletes" however.Their 'top of mind' responses had no bearing on actual use, although those who answered associating hookah with multiple categories of substances were two times more likely to be "ever" users (OR = 2.01, p < .05).

Discussion
This is the first known hookah use study in community college settings, adding an important perspective to the hookah discussion, since nearly half of all college students attend a community college (American Association of Community Colleges, 2014).It is also the first study conducted solely in the West with an ethnically diverse sample: 43% Latino, 25% White, 14% Multirace and 9% African-American.
We found that across our two schools the participating community college students have one of the highest hookah use rates in the U.S. with the majority (55.5%) reporting they have used hookah sometime in their lifetime.Only one study at a large public university in the Southeast reported a higher lifetime use rate (61%) (Noonan et al., 2011).Notably, more community college students have tried hookah than a cigarette (48.8%), also only found in the literature at that Southeastern public university (Noonan et al., 2011).
We found many similarities between community college and four-year college students.Community college student's current use (10.8%) and past year use (34.1%) are within the range of most four-year college study findings.As seen in four-year studies those who identified as White were most likely to use hookah and those identifying as African-Americans were least likely.In this study, Whites have almost three times higher odds of current use.Community college students who live with friends are more likely to be current users than those who live with parents/family.This is similar to four-year school findings that students living in fraternity/sorority or on-campus housing, in other words those who live among peers, are most likely to use (Jarrett et al., 2012)."Ever" and concurrent use of alcohol and cigarettes were also common correlates.Adolescent cigar use appears to be on the rise in the U.S. and current use is associated with hookah use in this study similar to some studies in four-year colleges (Abughosh et al., 2011;Eissenberg, Ward, Smith-Simone, & Maziak, 2008).Low perceived health and addiction risks and high perceived social acceptability and popularity are common beliefs in our participants as had been reported in four-year colleges.
We also found some distinct differences.Fouryear college studies have found the highest odds of use among 18-19 year olds, referring to this as "freshmen experimentation," but this study found no significant difference in current use between 18-19 and 20-24 year olds, although once age 25 is reached, the likelihood of hookah use goes down.This may be the case, as most community college students take longer to complete their studies with many working part time, thus possibly prolonging this experimentation effect relative to four-year college students.
We were surprised that at the community college a student's personal economic status was not associated with use.The only significant economic finding was that students who spend relatively little on weekly entertainment, compared to those who spend either nothing, or more that $20 per week, were less likely to "ever" use hookah, although no difference was found in current (or past year) use.This finding underscores the low cost, "easy access" of hookah that should be considered by policy makers, especially given the many misperceptions about health risk we found.Hookah and associated paraphernalia are extremely easy and inexpensive to purchase at local smoke shops or online.Indeed, home gettogethers in the garage or backyard, known as "kick-backs" in southern California, frequently involve hookah (Wilson, 2011).

Limitations
Our study, while pointing to some clear hookah risk patterns in this community college student population, has some limitations.First, we surveyed a convenience sample of students taking general education classes at two southern California Inland Empire community colleges.However, while our respondents were slightly younger and more female, our demographic participant profile aligned well with both school's published overall attendance for the school year we surveyed suggesting that our respondents are fairly representative of their peers.Secondly, we used a cross-sectional design; hence it was not possible to examine longitudinal associations among the variables of interest.Third, we relied on self-reported data, which may be subject to socially desirable responding and recall bias.

Conclusion
We found that more than half of our participating community college students had used hookah and that users believed it to be less harmful and addictive than cigarettes.Users felt that it was highly socially acceptable and males and females used hookah alike, regardless of income.These high use rates, coupled with the fact that monitoring studies found that hookah use across the US continues to increase (Johnston et al., 2014), point to an urgent need for hookah health education and cessation programs that clarify its negative health effects.While use diminishes when students are older, we should not wait for this natural decrease, as risks during use can be substantial.In addition, a recently published study found that with a brief online intervention collegiate hookah users who received information on its dangers were likely to wish to quit and actually quit (Lipkus, Eissenberg, Schwartz-Bloom, Prokhorov, & Levy 2011), suggesting that even low cost health education efforts can be effective.To combat increasing rates of hookah use colleges should implement these types of brief interventions to their incoming classes in order to educate their students about hookah risks and motivate them to not engage in this seemingly safe and fun pastime.
The questions and answers read as follows: 1. "Have you ever smoked hookah (waterpipe, shisha, narghile), even one or two puffs?"a) Yes, I have smoked hookah within the last year, b) Yes, but the last time I smoked hookah was over a year ago, c) No, I have never smoked hookah, and d) I do not know what hookah is.Options a) or b) indicated "ever" use.2. "How old were you when you first tried hookah?" with two boxes to write in an age as a continuous variable.Any age written in constituted "ever" use.And 3. "Within the last 30 days, on how many TOTAL days did you use hookah?"a) Never used, b) Have used, but not in the last 30 days, c) 1-2 days, d) 3-5 days, e) 6-9 days, f) 10-19 days, g) 20-29 days, h) Used daily.Current use was assessed by "Within the last 30 days, on how many TOTAL days did you use hookah?"Substance use related question were adapted from the National College Health Assessment (NCHA) II (American College Health Association, 2008).Other questions were identified from prior studies (see below) and some were developed based on qualitative data we collected in preparation for this research.The resulting survey was then pilot tested and minor revisions made as indicated, before it was used for the study.

Table 3 .
Hookah Belief Frequencies and Correlates to Hookah Use(N=1,207) a

Table 4 "
Top of Mind" Responses to Substances with Hookah Use (N=1,207)