Hookah Tobacco Smoking During the Transition to College: Prevalence of Other Substance Use and Predictors of Initiation

Abstract

Introduction:

The prevalence of hookah tobacco smoking is increasing, and the transition to college is a vulnerable time for initiation. Hookah use is associated with other forms of substance use, but most research has been cross-sectional, thus limiting our understanding of temporal patterns of use. The goals of this longitudinal study were to assess the prevalence of hookah use and initiation, as well as other forms of substance use among hookah users, and identify which forms of substance use predicted hookah initiation during the first 30 days of college.

Methods:

Incoming students ( N = 936, 50% female) reported on past 30-day substance use prior to the start of the Fall 2011 semester and again 30 days later ( n = 817). Substances included hookah, cigarettes, other forms of tobacco, alcohol, marijuana, and other illegal drugs.

Results:

Current prevalence of hookah use increased from 9.0% before college to 13.1% during the first month of college. At baseline and follow-up, current hookah users were more likely than nonusers to report current use of cigarettes, cigars/little cigars/clove cigarettes, smokeless tobacco, marijuana, and alcohol. Among pre-college hookah never users, 13.8% initiated hookah use in the first month of college. Alcohol (adjusted odds ratio [AOR] 1.11, 95% confidence interval [CI] 1.05, 1.17) and marijuana (AOR 1.30, 95% CI 1.03, 1.65) were the only substances predictive of hookah initiation.

Conclusions:

Findings indicate that hookah prevention and intervention is needed during the transition to college, and interventions may need to address comorbid alcohol, marijuana, and hookah use.

Implications:

To our knowledge this is the first longitudinal study examining predictors of hookah initiation among male and female incoming first-year college students. While hookah users were more likely than nonusers to use all other substances before and during the first month of college, pre-college marijuana and alcohol use were the only two predictors of hookah initiation during the first 30 days of college. Collectively, these findings provide additional support for the need for efficacious hookah prevention and intervention programs. The transition to college appears to be an ideal time to deliver prevention programs given the increased prevalence of hookah use during the first 30 days of college. In addition to prevention, former users may benefit from targeted relapse prevention as one-fifth of former hookah smokers resumed use during the first 30 days of college.

Introduction

Hookah, or waterpipe, tobacco smoking has been called an “emerging threat” ¹ to public health that “poses a serious potential health hazard.” ² Initial research suggests hookah has harmful health effects and potential to induce nicotine dependence. ^{3 , 4} Indeed, hookah users inhale nicotine, carbon monoxide, and large quantities of smoke. ⁵ A systematic review found that hookah smoking was associated with increased likelihood of lung cancer, respiratory illness, periodontal disease, and low birth weight. ⁶ Further, a meta-analysis found that hookah smoking may negatively affect lung function as much as cigarette smoking. ⁷ Hookah smoking may also engender a tobacco-friendly culture, which may increase risk for initiating other forms of tobacco use, such as cigarette smoking. ⁸ The combination of increasing prevalence, ⁹ potential health consequences, and continued lack of regulation of the hookah industry has ignited public health concern. ⁴

Increasing Prevalence of Hookah Use

Hookah is growing in popularity worldwide ⁹ and in the United States. Among nationally representative samples of American 18–34-year olds ¹⁰ and middle and high school students, ¹¹ 17% and 7%, respectively, reported lifetime hookah use. Studies examining patterns of use over time have shown a 40% increase in hookah use between 2005 and 2008 among California adults ¹² and a 61% increase in hookah use between 2007 and 2012 among Florida high school students. ¹³ The growing popularity of hookah has been attributed to a combination of factors, including novelty, low cost, easy access, commercial marketing, pleasant taste and smell of hookah tobacco flavors, social acceptance, misperceptions of safety, and lack of public health warnings or regulation. ^{14 , 15}

Hookah Use Among College Students

Hookah smoking has become especially popular among American college students. ^{16 , 17} Recent studies have found that the lifetime prevalence rate of hookah use was almost as high, ¹⁸ if not higher, ^19–21 than that of cigarettes. A review of seven studies sampling this population found the average prevalence rates to be 30% for lifetime use, 22% for past year use, and 15% for past month use. ¹⁷ Students usually smoke with friends in groups, ¹⁹ and the social aspect appears to be a key facilitator of hookah use in this population. ^{22 , 23} The most common motives for hookah use among college students include socializing, flavor/smell, relaxation, peer influences, and experimentation. ¹⁴

Hookah Use During the Transition to College

Emerging adulthood, the transitional period (ages 18–25) between adolescence and young adulthood, ²⁴ is an important time for establishing health behaviors. The transition to college is a developmental context marked by identity exploration and experimentation. ²⁴ Consistent with this, prior research has found that many emerging adults initiate hookah tobacco smoking, ^{25 , 26} as well as other forms of substance use, such as marijuana use ²⁷ or nonmedical use of prescription stimulants, ²⁸ during the first year of college. The first year of college may be a particularly vulnerable time for hookah use, given permissive social norms about substance use in college environments, the highly social nature of hookah use, ²³ media portrayals of hookah as exotic and trendy, ¹ the popularity of hookah lounges in college towns, ¹⁸ the ability of students under age 21 to get into hookah lounges but not traditional bars, ⁴ and decreased parental monitoring. One study specifically examining hookah initiation found that 23% of female college students who had never tried hookah before college reported hookah use during the first year of college ²⁶ ; lifetime prevalence of hookah smoking increased from 29% at college entry to 45% at 1-year follow-up. ²⁵

Comorbid Substance Use

Hookah smoking is associated with other forms of substance use. Hookah use is more likely among youth who use other forms of tobacco, including cigarettes, ^{18 , 22 , 29 , 30} cigars and cigarillos, ³⁰ and bidis and smokeless tobacco. ³¹ In addition, hookah use is more likely among youth who use alcohol, marijuana, and other illegal drugs. ^{18 , 22 , 29 , 32}

Most research on comorbid substance use has been cross-sectional, preventing understanding of temporal patterns of use. The few extant longitudinal studies have primarily examined whether hookah use predicts cigarette smoking initiation. Two studies found that hookah use predicted cigarette uptake among adolescents, ^{33 , 34} and a third found that pre-college hookah use predicted initiation/resumption of cigarette smoking during the first year of college among females. ³⁵ Poly-tobacco use increases exposure to nicotine and may increase risk for nicotine dependence.

Predictors of Hookah Use

Despite the prevalence and consequences of hookah use, little is known about predictors because the majority of extant research has been cross-sectional. Therefore, the temporal patterns of the initiation of hookah use in relation to other forms of substance use are not well understood. To date, only three longitudinal studies have evaluated predictors of hookah initiation among US college students. One study found that positive attitudes and favorable normative beliefs were associated with increased odds of hookah initiation, whereas negative attitudes were associated with decreased odds. ³⁶ The other two used a sample of first-year college women and found that alcohol and marijuana use were the strongest risk factors for hookah initiation. ²⁶ Furthermore, controlling for impulsivity and sensation-seeking, pre-college marijuana use and binge drinking, but not cigarette smoking, predicted hookah initiation during the first year of college. ³⁵ Additional research is needed, including studies with male college students. Identification of predictors of hookah use will enable targeting of public health resources to individuals who would benefit most from prevention and intervention.

The Current Study

Few studies have employed longitudinal designs to investigate predictors of hookah use during the transition to college. Therefore, the current study examined other forms of substance use as predictors of initiation of hookah use during the transition to college to inform future prevention and intervention efforts. Our objectives were to (1) assess the prevalence of hookah use and initiation, (2) assess the prevalence of other forms of substance use (cigarettes, other tobacco products, alcohol, marijuana, other illegal drugs) among hookah users, and (3) identify predictors of hookah initiation during the transition to college. Based on prior research, ^{25 , 33–35} we hypothesized that cigarette, alcohol, and marijuana use would predict initiation of hookah tobacco smoking during the transition to college. We advance the literature by using a random sample and a longitudinal design to examine temporal patterns of hookah and other forms of substance use.

Methods

Participants and Procedure

A random sample of 1200 eligible incoming college students were invited to participate in a randomized controlled trial of an Internet-delivered alcohol education intervention. ³⁷ Students were eligible if they were incoming first-year undergraduates at a large, Mid-Atlantic state university and had a US mailing address. In total, 936 students (78%) provided informed consent to participate and completed the baseline assessment before the start of the Fall 2011 semester. Subsequently, 817 participants (87%) completed the follow-up survey that was administered 30 days after the start of the semester. Participants were entered in a drawing to win one of ten $50 gift cards for completing the follow-up survey. Additional information about the parent study is reported elsewhere. ³⁷ All procedures were approved by the institutional review board at the host site.

Participants were 49.7% female, and 94.7% were 18 years old, with an average age of 18.1 ( SD = 0.3). Ethnicity was 85.2% white, 7.5% Hispanic, 6.5% Asian, 6.0% Black or African American, and 2.8% other (participants could select more than one race/ethnicity). Twenty percent of participants were student-athletes, and 99% lived on campus.

Measures

Tobacco, Marijuana, and Other Drug Use

An 18-item measure assessed use of a wide range of substances. Participants were asked to select how many times they used these substances during the past 30 days: cigarettes; tobacco from a water pipe (hookah); cigars, little cigars, or clove cigarettes; smokeless tobacco; marijuana (pot, weed, hashish, hash oil); cocaine (crack, rock, freebase); methamphetamine (crystal meth, ice, crank); other amphetamines (diet pills, bennies); sedatives (downers, ludes); hallucinogens (LSD, PCP); opiates (heroin, smack); inhalants (glue, solvents, gas); MDMA (Ecstasy); other club drugs (GHB, ketamine, rohypnol); and other illegal drugs. Given low prevalence rates, we combined all of these substances besides tobacco and marijuana into an “other illegal drugs” category. Response options were: 0 = never used, 1 = have used, but not in the last 30 days, 2 = 1–2 days, 3 = 3–5 days, 4 = 6–9 days, 5 = 10–19 days, 7 = 20–29 days, 8 = used daily. Based on their self-reported hookah use, participants were classified as lifetime users or current (past 30 days) users at each assessment.

Alcohol Use

We assessed current alcohol use with a single item. Participants indicated on how many days (0–30) they drank alcohol in the past 30 days. We did not assess lifetime alcohol use.

Data Analysis

We calculated descriptive statistics for study variables. We used chi-square tests to compare the prevalence of lifetime and current substance use among hookah users and nonusers. Alcohol use was dichotomized (did not/did drink alcohol in the past 30 days) for the chi-square analyses. We used multivariate binary logistic regression to examine predictors (assessed at baseline) of hookah initiation at 1-month follow-up among the subset of participants who reported no lifetime hookah use at baseline. Even though the intervention did not target hookah use, we controlled for intervention condition given that half of participants participated in an Internet-delivered alcohol education intervention as part of the parent study. We controlled for gender (coded 0 = male, 1 = female) and white race (coded 0 = non-white, 1 = white) because men and whites are more likely to smoke hookah compared to women and non-whites, respectively. ^{11 , 17 , 38} Alcohol was treated as a continuous variable (range: 0–30 days of use in the past 30 days). We recoded the variables for other forms of substance use (predictors) such that 0 = never used or have used, but not in the last 30 days; 1 = 1–2 days (of use in the past 30 days); 2 = 3–5 days; 3 = 6–9 days; 4 = 10–19 days; 5 = 20–29 days; and 6 = used daily. We report adjusted odds ratios and 95% confidence intervals for the logistic regression results. Alpha was set ( a priori ) at .05. Analyses at baseline include all participants, whereas analyses at 1-month follow-up were restricted to the 817 complete cases.

Results

Chi-squared tests showed that completion of the 1-month follow-up survey was more likely among women (90.5%) compared to men (84.1%), χ ² (1, N = 936) = 8.80, P = .003; whites (89.7%) compared to non-whites (73.4%), χ ² (1, N = 936) = 28.44, P < .0001; lifetime hookah nonusers (88.5%) compared to lifetime hookah users (83.3%), χ ² (1, N = 936) = 4.09, P = .04; and lifetime smokeless tobacco nonusers (88.3%) compared with lifetime smokeless tobacco users (76.5%), χ ² (1, N = 936) = 9.22, P = .002. Also, participants who completed the follow-up survey reported significantly fewer days of alcohol use in the past month at baseline ( M = 3.12, SD = 4.47) than those who did not ( M = 4.62, SD = 6.80), Satterthwaite t (133.2) = 2.34, P = .02. However, there were no significant differences by age, athlete status, current use of alcohol, or lifetime use of cigarettes, cigars/little cigars/clove cigarettes, marijuana, or other illegal drugs (all P s > .05).

Prevalence, Frequency, and Initiation of Hookah Use

Lifetime hookah use was reported by 215 students (23.0%) at baseline and 209 (25.6%) at 1-month follow-up. At baseline, white students (24.5%) were more likely than non-white students (14.4%) to report lifetime hookah use, χ ² (1, N = 936) = 6.79, P = .009. However, there were no differences in lifetime hookah use by gender, age, or athlete status, (all P s > .05).

Current hookah use was reported by 84 students (9.0%) at baseline and 107 (13.1%) at 1-month follow-up. At baseline, white students (10.0%) were more likely than non-white students (2.9%), χ ² (1, N = 936) = 7.43, P < .01, and males (10.8%) were more likely than females (7.1%), χ ² (1, N = 936) = 3.99, P < .05, to report current hookah use. There were no differences in current hookah use by age or athlete status (all P s > .05). At baseline, the frequency of current hookah use was relatively low, with 77.4% of current users reporting 1–2 days, 14.3% 3–5 days, and 8.3% 6–9 days of use in the past 30 days. At follow-up, a greater number of participants reported more frequent hookah use, with 63.6% of current users reporting 1–2 days, 21.5% 3–5 days, 6.5% 6–9 days, 6.5% 10–19 days, 0.9% 20–29 days, and 0.9% daily use in the past 30 days.

Among the 638 baseline lifetime never hookah users who completed the 1-month follow-up, 88 (13.8%) initiated hookah use during the first month of college. Among the 110 students who reported lifetime, but not current, hookah use at baseline and completed the 1-month follow-up, 23 (20.9%) resumed hookah use during the first month of college.

Prevalence of Other Substance Use

Table 1 displays the lifetime prevalence of other forms of substance use among lifetime hookah users and nonusers at baseline and 1-month follow-up. Table 2 displays the current (past 30 days) prevalence of other forms of substance use among current hookah users and nonusers at baseline and 1-month follow-up. With the exception of current use of other illegal drugs at baseline, all forms of substance use were significantly more likely among lifetime as well as current hookah users compared to nonusers at baseline and 1-month follow-up.

Table 1.

Lifetime Prevalence of Other Substance Use by Lifetime Hookah Use Status

Substance	Baseline ( N = 936)					1-month follow-up ( N = 817)
	Hookah users ( n = 215)		Hookah nonusers ( n = 721)		χ 2 ( df = 1)	Hookah users ( n = 209)		Hookah nonusers ( n = 608)		χ 2 ( df = 1)
	n	%	n	%		n	%	n	%
Cigarettes	103	47.9	59	8.2	182.60*	97	46.4	42	6.9	171.91*
Cigars, little cigars, or clove cigarettes	136	63.3	143	19.8	149.25*	130	62.2	82	13.5	192.08*
Smokeless tobacco	56	26.1	25	3.5	106.81*	46	22.0	21	3.5	71.14*
Marijuana	135	62.8	137	19.0	154.05*	132	63.2	77	12.7	208.30*
Alcohol	185	86.1	399	55.3	66.55*	200	95.7	371	61.0	88.86*
Other illegal drugs	25	11.6	17	2.4	33.21*	31	14.8	3	0.5	80.18*

Alcohol use reflects any current use; all other substance use reflects lifetime use.

* P < .0001.

Table 2.

Current (Past 30 Days) Prevalence of Other Substance Use by Current Hookah Use Status

Substance	Baseline ( N = 936)					1-month follow-up ( N = 817)
	Hookah users ( n = 84)		Hookah nonusers ( n = 852)		χ 2 ( df = 1)	Hookah users ( n = 107)		Hookah nonusers ( n = 710)		χ 2 ( df = 1)
	n	%	n	%		n	%	n	%
Cigarettes	28	33.3	56	6.6	67.03**	35	32.7	48	6.8	68.60**
Cigars, little cigars, or clove cigarettes	48	57.1	105	12.3	112.32**	34	31.8	51	7.2	60.33**
Smokeless tobacco	7	8.3	26	3.1	6.27*	14	13.1	21	3.0	23.25**
Marijuana	43	51.2	107	12.6	84.80**	38	35.5	66	9.3	57.54**
Alcohol	79	94.1	505	59.3	39.41**	102	95.3	469	66.1	37.86**
Other illegal drugs	2	2.4	18	2.1	0.03	11	10.3	6	1.0	40.63**

* P < .05; ** P < .0001.

Predictors of Hookah Initiation

We examined predictors of hookah initiation during the first month of college among the 638 baseline lifetime never hookah users who completed the 1-month follow-up. Controlling for gender, white race, and intervention condition, current marijuana use and current alcohol use predicted hookah initiation ( Table 3 ), model likelihood ratio χ ² ( df = 9) = 57.93, P < .0001. Use of cigarettes, cigars/little cigars/clove cigarettes, smokeless tobacco, and other illegal drugs did not predict hookah initiation.

Table 3.

Other Forms of Substance Use at Baseline as Predictors of Hookah Initiation at 1-Month Follow-up

Predictor	B ( SE )	Wald χ 2	AOR	95% CI
Intercept	−2.42 (0.43)	31.95**	—	—
Intervention condition (reference group: control condition)	0.19 (0.25)	0.59	1.21	0.75, 1.96
Gender (reference group: male)	−0.46 (0.26)	3.09	0.63	0.38, 1.05
White race (reference group: non-white)	0.21 (0.41)	0.62	1.23	0.55, 2.75
Cigarettes	0.23 (0.18)	1.57	1.26	0.88, 1.80
Cigars, little cigars, or clove cigarettes	0.19 (0.19)	1.02	1.21	0.84, 1.75
Smokeless tobacco	0.11 (0.18)	0.38	1.12	0.79, 1.59
Marijuana	0.27 (0.12)	5.02*	1.30	1.03, 1.65
Alcohol	0.10 (0.03)	13.48**	1.11	1.05, 1.17
Other illegal drugs	0.96 (0.66)	2.09	2.61	0.71, 9.59

AOR = adjusted odds ratio; B = regression estimate; CI = confidence interval; SE = standard error. N = 638. All predictors were assessed at baseline. Alcohol use reflects current (past 30 days) use; all other substance use reflects lifetime use.

* P < .05; ** P < .001.

Discussion

Our results corroborate previous research showing that a significant minority of college students report trying hookah prior to college entry. ²⁶ The lifetime prevalence of 23.0% at baseline is lower than the average of 30% found by Grekin and Ayna ¹⁷ in their review; however, their rate included all college students, whereas our sample consisted entirely of first-year students, who would be expected to have a lower prevalence given that approximately one-third of college student hookah users do not try hookah for the first time until age 19 to 21. ³⁹ In our sample, 9% of students reported hookah use in the month before college, which is also lower than the average current prevalence of 15% reported in Grekin and Ayna, ¹⁷ but again may be related to our sample comprising all first-year students.

Over the 1-month follow-up period, the lifetime prevalence of hookah use increased from 23.0% to 25.6%, and the current prevalence increased from 9.0% to 13.1%. The increase in hookah use was likely a combination of prior hookah users resuming use as well as never users trying hookah for the first time. We found that 13.8% of baseline never hookah users initiated hookah use in the first month of college, which is comparable to rates reported in previous studies, such as 13% of first- and second-year undergraduates and graduate students initiating over 7 months ³⁶ and 23% of first-year college women initiating in the first year after college entry. ²⁵ Taken together, these data suggest that the transition to college is a vulnerable time for hookah initiation and use. The combination of proximity to and attractiveness of hookah lounges as venues in which underage students can socialize, ⁴ along with peers who may be inviting or encouraging students to try it, ²³ may contribute to hookah use. Of note, the host site has two hookah lounges in close proximity to campus.

With few exceptions, hookah users were more likely than nonusers to use all other forms of substance use examined in this study. For example, at follow-up, 96% of ever hookah users reported current alcohol use compared to 61% of hookah nonusers, and over 60% of ever hookah users reported lifetime use of both marijuana and cigars/little cigars/clove cigarettes, compared to approximately 13% for hookah nonusers. Notably, only 46% of ever hookah users reported lifetime cigarette use at follow-up, which is consistent with prior research ^{19 , 38} suggesting that many hookah users might have otherwise been naïve to tobacco if not for trying hookah. At follow-up, current hookah users almost universally reported current alcohol use, and but only 36% reported current marijuana use; approximately one-third of current hookah users reported current use of cigarettes and cigars/little cigars/clove cigarettes, but only 13% reported current use of smokeless tobacco. Although other forms of tobacco were not predictive of hookah initiation, these rates suggest the importance of considering poly-product tobacco use in future research. ⁴⁰

Consistent with our hypothesis, alcohol and marijuana use predicted initiation of hookah tobacco smoking. However, contrary to expectations, cigarette use was not associated with hookah initiation, and neither were other forms of tobacco use or other illegal drugs. This pattern of findings is consistent with prior research sampling college women, in which marijuana and alcohol use, but not cigarette use, predicted hookah initiation during the first year of college. ³⁵ The link between marijuana, alcohol, and hookah use may be explained by factors such as similar motives (eg, social enhancement, relaxation), similar social contexts (eg, hookah lounges, friend’s homes), and a general propensity to use substances. Another potential link between hookah and marijuana smoking is a similar route of administration, as 45% of college student lifetime hookah users reported using the waterpipe apparatus to smoke marijuana. ⁴¹ Additional research is needed to clarify exactly how marijuana and alcohol use may increase risk for hookah initiation. Given the strong association between cigarette and hookah use, ^{18 , 22 , 30} it was surprising that cigarette use did not predict hookah initiation. However, as noted previously, lifetime and current cigarette use was reported by only 46% and 33%, respectively, of hookah users at follow-up, which could mean that many students view hookah and cigarettes differently (eg, a false perception that hookah is safer than cigarettes). The lack of a relationship may also stem from the two forms of tobacco being used for different purposes or in different settings.

Limitations and Directions for Future Research

This study has five primary limitations. First, the study was conducted at one university, and these findings may not generalize to students at other colleges or to non-college attending emerging adults. The predictor analyses were limited to students who completed the 1-month follow-up survey, who were disproportionately female and white and were less likely to report baseline use of hookah and smokeless tobacco compared to students who were lost to attrition. Second, the sample consisted of students who were participating in a randomized controlled trial of an online alcohol intervention. Although the enrollment rate for the larger study was 78% of a random sample of incoming students, those who agreed to enroll may differ in some systematic way from those who declined to enroll. However, sample demographics were relatively similar to the overall campus demographics (eg, this sample was 49.7% female and 85.2% white vs. 46.2% female and 80% white overall at the host site). Third, we are not able to infer cause and effect, specifically as to whether alcohol or marijuana use increased the likelihood of hookah use. Fourth, some of our measures were not optimally sensitive, particularly our alcohol use measure, which assessed frequency, but not quantity, and our tobacco use measure, which combined cigars, little cigars, and clove cigarettes rather than asking about each separately. Fifth, we do not have biological or collateral verification of substance use behaviors. Nonetheless, self-report is generally considered valid, ⁴² and our good response and retention rates impart confidence in these findings. Additional strengths of our study include the random sample and longitudinal design. To our knowledge this is the first longitudinal study examining predictors of hookah initiation among male and female incoming first-year college students.

Future research is needed with nationally representative samples. Daily diary studies spanning the first month of college would be able to capture more precise temporal associations than our method allowed. In addition, research identifying theory-based constructs associated with hookah initiation is needed. ³⁶ For example, perceived descriptive and injunctive norms for hookah use may explain why some individuals initiate hookah use given the importance of the social context of hookah use among college students. ²³ Once the theory-based mechanisms of hookah initiation are understood, hookah prevention programs that address these constructs can be developed and evaluated.

Continued research is needed to inform public health policy related to hookah tobacco smoking and determine the extent to which regulation is warranted. Currently, hookah products are unregulated in the United States, which permits a lack of health warnings, misleading claims on packaging, marketing of flavored tobacco, and exemption from bans on indoor smoking that apply to cigarettes. ⁴ However, the Food and Drug Administration has proposed a deeming rule, which would give the Food and Drug Administration authority to regulate emerging tobacco products, including hookah and e-cigarettes. ⁴³ While currently pending and not yet final, this proposed rule seeks to prevent sales of hookah to underage youth, restrict hookah-related advertising, and include a health warning about nicotine on hookah tobacco products. These policy changes are expected to delay hookah smoking by emerging adults by delaying or preventing hookah initiation, decreasing awareness of hookah via marketing, and increasing perceived health risks associated with hookah use.

Conclusions and Implications

The transition to college is a formative period associated with hookah initiation and a high relapse rate for former hookah users. While hookah users were more likely than nonusers to use all other substances before and during the first month of college, pre-college marijuana and alcohol use were the only two predictors of hookah initiation during the first 30 days of college. Collectively, these findings provide additional support for the need for efficacious hookah prevention and intervention programs, and suggest that programs target alcohol and marijuana given their association with increased risk of hookah initiation. The transition to college appears to be an ideal time to deliver prevention programs given the increased prevalence of hookah use during the first 30 days of college. In addition to prevention, former users may benefit from targeted relapse prevention as one-fifth of former hookah smokers resumed use during the first 30 days of college.

Funding

This work was supported by the National Center for Advancing Translational Sciences at the National Institutes of Health (UL1 TR000127 and KL2 TR000126 to Lawrence Sinoway). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The funding source had no role in the study design, data collection, data analysis or interpretation, writing of the manuscript, or the decision to submit the manuscript for publication.

Declaration of Interests

None declared.