Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Dec 1.
Published in final edited form as: Drug Alcohol Depend. 2020 Oct 12;217:108346. doi: 10.1016/j.drugalcdep.2020.108346

Predictors of Nicotine Dependence Among Adolescent Waterpipe And Cigarette Smokers: A 6-year Longitudinal Analysis

Mohammad Ebrahimi Kalan a, Raed Bahelah b, Zoran Bursac c, Ziyad Ben Taleb d, Joseph R DiFranza e, Malak Tleis f, Rima Nakkash f, Rime Jebai a, Mohammad Masudul Alam g, Miguel Ángel Cano a, Matthew T Sutherland h, Kristopher Fennie i, Taghrid Asfar j,k,l, Thomas Eissenberg l,m, Kenneth D Ward l,n, Wasim Maziak a,l
PMCID: PMC7861130  NIHMSID: NIHMS1637905  PMID: 33075692

Abstract

Objective:

Identifying the factors associated with nicotine dependence (ND) is essential to prevent initiation and continued use, and to promote cessation among youth. This study aims to document the predictors of the appearance of initial ND symptoms and full ND syndrome among adolescent waterpipe (WP) and cigarette smokers.

Methods:

A 6-year longitudinal study was conducted among 8th and 9th graders from 38 schools in Lebanon. The analysis sample included exclusive-WP (n=228) and exclusive-cigarette smokers (n=139). Weighted Cox proportional hazards models were used to characterizing predictors of initial ND symptoms and full ND syndrome.

Results:

Predictors of experiencing initial ND symptoms among WP smokers included low maternal educational level, having a sibling who smoked WP, low physical activity, high body mass index (BMI), smoking initiation at a younger age. For cigarette smokers these were being male, younger, having lower BMI, having a sibling who smoked cigarettes, living in a crowded household, and smoking daily. Among WP smokers, predictors of developing full ND syndrome include being younger, believing that WP smokers have more friends, depression, high levels of impulsivity, and initiating smoking at a younger age. For cigarette smokers, predictors of full ND syndrome were being younger and initiating smoking at a younger age.

Conclusion:

Smoking cessation and prevention interventions targeting youth should address modifiable, and tobacco use-specific factors that influence the development of ND among young WP and cigarette smokers. They also need to start at a younger age to target those most vulnerable to developing life-long addiction to tobacco products.

Keywords: Nicotine Dependence, Waterpipe smoking, Cigarette Smoking, Adolescent

1. Introduction

Every year, an estimated 8 million people die of tobacco-caused diseases worldwide(WHO, 2019). Initiating tobacco use during adolescence doubles the rate of premature death.(Thomson et al., 2020) Waterpipe (WP; hookah, shisha, narghile)–a centuries-old tobacco use method– exposes users to nicotine(Aboaziza and Eissenberg, 2014) and other toxicants similar to those present in cigarettes.(Primack et al., 2016) WP tobacco smoking is becoming widespread globally especially in the Eastern Mediterranean Region (EMR)(Jawad et al., 2018; WHO, 2015) where approximately 10.3% of adolescents report past-month (current) WP use (Jawad et al., 2018).

Nicotine is a highly addictive psychomotor stimulant associated with continued use and dependence on tobacco products, and understanding factors influencing nicotine dependence (ND) in tobacco users is critical to design effective prevention and cessation interventions. Yet, studies addressing ND have just begun investigating the diversity of tobacco products used by youth and the importance of understanding product-specific features and contextual factors leading to ND(DiFranza et al., 2000, 2002a; Hu et al., 2008; Sharapova et al., 2020; Ward, K. D. et al., 2015). For example, WP users are exposed to significant amounts of nicotine during the typically long smoking sessions (average 1 hour) and experience ND symptoms (e.g., abstinence-induced withdrawal and craving that are relieved by subsequent smoking). Moreover, WP social aspects, commonly reinforced by WP cafés can provide important cues for ND in WP smokers (Bahelah et al., 2018)). Identifying these unique features of ND and their predictors in WP smokers is important to develop tailored interventions to address the rise of WP smoking among youth.

Milestones in the trajectory of ND development, such as the appearance of first ND symptoms and full syndrome of ND syndrome can help elucidate factors that need to be addressed at each stage of adolescents’ tobacco use trajectory for effective intervention. This is particularly important for WP smoking given its unique use features that can influence ND development such as its time-consuming preparation and consumption, intermittent use, and strong sensory and social cues(Akl et al., 2013; Maziak et al., 2019; Maziak et al., 2015). Our earlier studies based on a cohort of adolescents in Lebanon, a country with the highest WP use among adolescents(Jawad et al., 2018), showed that the initial ND symptoms and full syndrome of ND manifest more rapidly among WP smokers compared to cigarette smokers(Ebrahimi Kalan et al., 2020). We also found that 50% of both WP and cigarette smokers developed the full ND syndrome within 15 and 22 months after smoking initiation, respectively(Ebrahimi Kalan et al., 2020). As for contextual factors, our studies(Bahelah et al., 2017; Bahelah et al., 2018) showed that having at least a family member who smokes WP, and more importantly, not resisting WP use while in a restaurant were associated with a higher risk of ND. Building on such a unique cohort, we aim to identify and contrast the factors that predict initial ND symptoms and full-blown ND syndrome in adolescents WP and cigarette smokers. The findings of this study will be effective for developing WP-specific policies to prevent ND among adolescents and design smoking cessation interventions for those already hooked on nicotine.

2. Methods

2.1. Study design and sample

Data were drawn from the Waterpipe Dependence in Lebanese Youth (WDLY) study, an ongoing prospective study of 647 Lebanese 8th and 9th grades adolescent smokers and non-smokers recruited from 38 public and private schools. A brief in-class, self-administered recruitment baseline survey about students’ smoking status was administrated to determine eligibility. To compare cigarette and WP smokers in terms of ND patterns, students were eligible to participate if they were either current (use in the last 30 days) cigarette or WP smokers, but not both. Susceptible non-smokers (defined as the likelihood of cigarette/WP smoking initiation in next year) to smoke WP or cigarettes were also included. More details about the study design and procedures can be found elsewhere (Bahelah, Raed et al., 2016a; Ebrahimi Kalan et al., 2020).

The current longitudinal analysis used data from 8 waves and was restricted to participants who reported being current exclusive users of either WP (WP-only) or cigarettes (cigarette-only) during the study with a retention rate of 72.3% at wave 8 (i.e., n=179 were lost to follow-up during waves 2–8). The first 6 data collection waves were conducted between May 2015 and December 2017 with each wave separated by 6 months. The 7th and 8th waves were conducted in 2019 and 2020, respectively with a 1-year interval between waves. This was done because data collection every 6 months became challenging given political instability in Lebanon. Also, because changes in ND trajectories tend to stabilize as adolescents approach early adulthood(Hu et al., 2012). At each wave, participant age assessed and when a participant reached 18 years old, informed consent was obtained without the need for parental/guardian consent as was required for those younger than 18. The Institutional Review Boards of Florida International University and the American University of Beirut approved this study.

2.2. Measures

2.2.1. Predictors

The selection of ND predictors was guided by a review of the literature on adolescent WP (Auf et al., 2012; Bahelah et al., 2016b; Jaber et al., 2015; Maziak et al., 2005; Neergaard et al., 2007) and cigarette smoking(Hu et al., 2006; Kleinjan et al., 2012) as well as theories of ND in this population(Aboaziza and Eissenberg, 2014; DiFranza et al., 2007; DiFranza and Ursprung, 2008; Maziak et al., 2005). A description of these predictor variables follows (the predictor variable levels used as reference categories in regression models are underlined).

  1. Sociodemographic characteristics included age (years), gender (female/male), school type (public/private), body mass index (BMI= weight/height^2), regular physical activity defined as performing the physical activity at least once a week (yes/no), parental education (< 12 years of education vs ≥12), and crowding index (defined as the number of co-residents in a dwelling, excluding infants, divided by the number of rooms in the dwelling, excluding the kitchen and bathrooms(Bejjani et al., 2012; Melki et al., 2004). Crowding index is an indirect measure of socioeconomic status (SES) that is widely used in studies in the EMR and a higher crowding score indicates lower SES(Bejjani et al., 2012).

  2. Indicators of smoking in the social environment such as parental WP/cigarette smoking (mother and/or father; yes/no) and having ≥ 1 siblings/friends who smoke WP/cigarettes; yes/no.

  3. Beliefs about smoking included: WP/cigarette smokers look more attractive, WP/cigarette smokers have more friends, WP/cigarette smoking makes a person lose weight, and WP/cigarette smoking is harmful to health (response choices for all items: agree vs. disagree or don’t know). Due to a low number of participants answered “don’t know” and to simplify interpretation of results, we collapsed the responses “don’t know” with “disagree”.

  4. Smoking patterns included age of initiating WP/cigarette smoking, past-month WP/cigarette smoking frequency (daily vs non-daily), quantity (number of WP heads/bowls and amount of cigarettes smoked in the past 30 days), intention to quit WP/cigarettes (yes/no), and any attempt to quitting WP/cigarettes in the past 6 months (yes/no).

  5. Psychological indicators included perceived stress (15 items on four-point Likert scale “Not at all (0) to A whole lot (3)” with a possible score of 0–45)(Racicot et al., 2013), depressive symptoms (6 items on a four-point Likert scale “Never (0) to Often (3)” with a total score of 0–18)(Brunet et al., 2014), impulsivity (7 items on a five-point Likert scale “Not at all true (0) to Very true (4)” with 0–28 total score) (DiFranza et al., 2007), novelty-seeking (9 items on a five-point Likert scale “Not at all true (0) to Very true (4)” with 0–36 total score) (DiFranza et al., 2007), and self-esteem (10 items on a four-point Likert scale “Strongly Disagree (0) to Strongly Agree (3)” with a total score of 0–30) (Waters et al., 2006). The internal consistency of these scales, as measured by Cronbach’s alpha in our previous study, ranged from 0.63 to 0.81, indicating acceptable internal consistency(Bahelah et al., 2016b).

2.2.2. Outcomes

a). Initial ND symptom

The time interval between first WP/cigarette use and report of experiencing initial ND symptoms was assessed by the Hooked on Nicotine Checklist (HONC).(DiFranza et al., 2002a) HONC is a 10-item measure based on the Autonomy Theory of Tobacco Dependence which posits that the appearance of a single symptom of dependence (initial ND symptom) signals a loss of autonomy over tobacco use.(DiFranza et al., 2000) HONC was validated among adolescent cigarette smokers in previous work (O’Loughlin et al., 2002) and WP smokers in the WDLY study(Bahelah et al., 2016a).

b). Full ND syndrome

The WHO’s International Classification of Diseases, 10th Version Criteria for Tobacco Dependence (ICD-10) criteria for ND was assessed using 19 dichotomous (yes/no) items (DiFranza et al., 2007; WHO, 1993) across 6 criteria of ND and attainment of ≥ 3 criteria over a 12-month period is the standard threshold for diagnosis of full ND syndrome(DiFranza et al., 2007). The ICD-10 has been previously validated among adolescent cigarette smokers(DiFranza et al., 2007) and WP smokers in the WDLY study(Bahelah et al., 2016a).

2.3. Data preparation

Data were prepared for analysis in 3 steps. First, to calculate the time-to-event (i.e., number of months) from the first WP/cigarette puff to the appearance of initial ND symptoms and development of full ND syndrome, the date of the first puff was subtracted from the date when the initial ND symptom emerged or full ND syndrome developed(Ebrahimi Kalan et al., 2020). Second, those participants who had achieved these outcomes prior to Wave 1 were excluded from the current analysis. This study sample included participants who were at risk of experiencing an initial ND symptom and developing full ND syndrome at bassline and during the follow up (Figure 1). Third, a dynamic cohort and analytical approach allow us to use available data to increase the power of the study and represent all cohort members. Therefore, as shown in Supplemental Table 1, adolescent WP-only (n=79) and cigarette-only (n=49) smokers who were lost to follow-up were included in the analyses. As discussed below, the type of analysis (i.e., time-varying inverse probability weights for Cox regression) that we applied in this study produce marginal estimates but also adjust for selection bias resulting from lost to follow-up(Kohl et al., 2015; Robins et al., 2000). Third, time-varying predictors were measured in all 8 waves. In line with previous longitudinal studies,(Moahmed et al., 2014; O’Loughlin et al., 2009; Racicot et al., 2013) missing values for time-varying predictors (i.e., physical activity, BMI, smoking by parents, siblings, and friends, beliefs about smoking, and psychological indicators) were imputed using the “first observation carried backward” and “last observation carried forward” approaches.

Figure 1. Study sample flowchart.

Figure 1.

Open in a new tab

In the first model, we excluded those WP (n=116) and cigarette (n=29) smokers who endorsed ND symptoms at the baseline. In the second model, we excluded those WP (n=76) and cigarette (n=29) smokers who had already attained ICD10 criteria at the baseline. Note, the retention rate at wave 8 was 72.3% (i.e., n=179 were lost to follow up at wave 8); among adolescents who were lost to follow up, we excluded those who were either non-smokers or dual users at the time of loss to follow up (n=51). However, adolescents WP (n=79) and cigarette (n=49) smokers who were lost to follow up were included in the analysis (i.e., whether they were diagnosed as dependent or not at the end of the study) to increase the power. E-cigarettes module was added to the WDLY study at wave 8 and current e-cigarette users were excluded from current analysis.

2.4. Data analysis

Data analysis was performed in 3 steps. First, summary statistics were computed (categorical variables: frequencies/percentages; continuous variables: mean ± standard deviation [SD]). Chi-square or Fisher’s exact test (categorical variables) and Student’s T-test or Mann-Whitney U/Kruskal-Wallis tests (continuous variables) were used to assess for significant differences in baseline characteristics between adolescent WP and cigarette smokers. Second, to analyze data containing time-varying predictor variables, we applied the Counting Process technique in SAS(Allison, 2010; Andersen and Gill, 1982; Powell and Bagnell, 2012). Specifically, we constructed a new dataset containing multiple records for each individual, with each record corresponding to a time interval during which all predictors remained constant(Allison, 2010; Andersen and Gill, 1982; Powell and Bagnell, 2012). Third, the Cox Proportional Hazard Regression (CPHR) model is based on the assumption of the proportionality of hazards, meaning that the hazard ratio (HR) of each predictor is the same at all study times(Kleinbaum and Klein, 2010). However, this assumption was violated in our sample, hence, the SAS Macro PHSREG(Kohl et al., 2015) was used to apply weighted CPHR to test the unadjusted and adjusted HR’s (aHRs) and 95% confidence intervals of outcomes across levels of each predictor(Hosmer et al., 2002). A multivariable weighted CPHR was performed to identify independent predictors of experiencing initial ND symptoms and developing full ND syndrome for each smoking mode. SPSS v.26 and SAS/STATv14.2 for Windows were used for all analyses and statistical significance was set at p<0.05. Fourth, we checked for potential effect modification by gender, BMI, and crowding index (a proxy for SES) in univariate models controlling for the age of participants (see Supplemental Table 2). Lastly, frequency and quantity of WP(Bahelah et al., 2016b; Maziak et al., 2005; Maziak et al., 2004; Robinson et al., 2017) and cigarette(Kleinjan et al., 2012; Lam et al., 2014; O’Loughlin et al., 2003) use can be either risk factors for the development of ND or behavioral manifestations of established ND. Therefore, multivariable models were run both with and without frequency and quantity of use variables included. Associations of individual, environmental, and psychological variables with outcomes did not differ substantially in models that contained versus those that excluded frequency and quantity variables. As such, these variables were retained in the final multivariable models.

3. Results

3.1. Descriptive statistics

Of the 367 adolescents included in this study, 228 (62%) were current WP-only smokers and 139 (38%) were current cigarette-only smokers. Compared to cigarette smokers, a higher percentage of WP smokers were females (WP: 61.4%; cigarettes: 18%; p<0.001). WP smokers were also younger than cigarette smokers (WP:13.9±1.1years; cigarettes:15.0±1.2;p<.001). Table 1 shows the baseline characteristics of adolescents WP and cigarette smokers. Of the 647 participants, 9.1%(n=61) were dual users with a mean age of 14.1 years, males (85.2%) and enrolled in private schools (68.9%) at baseline (data not shown).

Table 1.

Baseline characteristics of adolescents waterpipe and cigarette smokers (n=367)

Study characteristics Total (n=367) Waterpipe smokers (n=228) Cigarette smokers (n=139) p-value
Individual characteristics Gender, n(%)
 Male 202 (55.0) 88 (38.6) 114 (82.0) <.001
 Female 165(45.0) 140 (61.4) 25 (18.0)
School type, n (%)
 Private 182 (49.6) 94 (41.2) 88 (63.3) <.001
 Public 185 (50.4) 134 (58.8) 51 (36.7)
Physical activity (at least once/week) (yes), n (%) 270 (73.6) 163 (71.5) 107 (77.0) .273
Father’s years of education (< 12 years/ illiterate) a 219 (59.7) 141 (61.8) 78 (56.1) .324
Mother’s years of education (< 12 years/ illiterate) a 185 (50.4) 122 (53.5) 63 (45.3) .133
Age, years M±SD 14.3±1.2 13.9± 1.1 15.0± 1.0 .001
BMI (weight/height^2) M±SD 21.7±4.3 21.2±4.1 22.4±4.5 <.001
Crowding index, M±SD 1.6±0.8 1.4±0.6 1.9±0.9 <.001
Indicators of smoking in social environmental Parent smokes cigarette (yes), n (%) 243 (66.2) 151 (66.2) 92 (66.2) 1.00
Parent smokes WP (yes), n (%) 171 (46.6) 122 (53.5) 49 (35.3) .001
≥1 sibling smoke cigarette, n (%) 96 (26.2) 57 (25.0) 39 (28.1) .542
≥1 sibling smoke WP, n (%) 144 (39.2) 109 (47.8) 35 (25.2) <.001
≥1 friend smoke cigarette, n (%) 226 (61.6) 106 (46.5) 120 (86.3) <.001
≥1 friend smoke WP, n (%) 311 (84.7) 189 (82.9) 122 (87.8) .233
WP smokers looks attractive (agree)b, n (%) 60 (16.3) 47 (20.6) 13 (9.4) .005
Beliefs about smoking Cigarette smokers looks attractive (agree)b, n (%) 52 (14.2) 24 (10.5) 28 (20.1) .013
WP smokers have more friends (agree)b, n (%) 97 (26.4) 68 (29.8) 29 (20.9) .067
Cigarette smokers have more friends (agree)b, n (%) 59 (16.1) 35 (15.4) 24 (17.3) .662
WP smoking makes a person lose weight (agree)b, n (%) 48 (13.1) 27 11.8) 21 (15.1) .425
Cigarette smoking makes a person lose weight (agree)b, n (%) 55 (15.0) 24 (10.5) 31 (22.3) .004
WP smoking is harmful to health (agree)b, n (%) 357 (97.3) 221 (96.9) 136 (97.8) .748
Cigarette smoking is harmful to health (agree)b, n (%) 359 (97.8) 221 (96.9) 138 (99.3) .268
Smoking patterns Age of initiation, years, M±SD — 13.3±1.9 13.9±1.9 .007
Frequency of use (daily)c n (%) — 22 (9.6) 80 (57.6) —
No of WPs/cigarettes smoked in the past month, M±SD — 12.3±17.7 249.9±23.8 —
Intention to quit (yes), n (%) — 84 (36.8) 35 (25.2) —
Made quit attempt (yes), n (%) — 60 (26.3) 51(36.7) —
Psychosocial indicators, M±SD Stress, M±SD 5.9±5.6 7.2±5.9 3.9±4.3 <.001
Depression, M±SD 6.1±4.2 6.9±4.2 4.9±4.1 <.001
Distractibility, M±SD 6.6±4.5 7.5±4.4 5.1±4.4 <.001
Novelty seeking, M±SD 12.3±7.1 13.1±6.5 11.0±7.7 .005
Impulsivity, M±SD 8.3±6.1 9.3±5.7 6.6±6.4 <.001
Self-esteem, M±SD 19.2±4.6 20.7±4.1 16.7±4.4 <.001
Open in a new tab
a

Compared to ≥12 years of education

b

Compared to Disagree/or don’t

c

Compared to non-daily use

3.2. Predictors of Initial Symptoms and the Full Syndrome of ND among WP smokers

Between waves 1 and 8, of the 193 WP smokers who were at risk of experiencing initial ND symptoms, 43.5% (n=84) did so and of the 228 WP smokers who were at risk of developing Full Syndrome of ND, 12.3% (n=28) did so. As shown in Table 2, predictors of experiencing initial ND symptoms among WP smokers were having a mother with <12 years of education and ≥ 1 sibling who smoke WP, and high BMI, whereas regular physical activity, being older, and smoking initiation at an older age were protective. There was an interaction effect between gender and school type, showing that for experiencing initial ND symptoms, the estimated marginal mean was higher for females at private schools. However, these two variables were not significant predictors in our multivariable models, therefore, there was no effect modification in the final model. Predictors of experiencing full ND syndrome were believing that WP smokers have more friends, greater depressive symptomatology, and being more impulsive, while older age and smoking initiation at an older age were protective factors.

Table 2.

Weighted Cox regression of the association between predictors and experiencing initial ND symptoms and developing full syndrome of ND among WP smokers, WDLY Study, 2015–2020

Study characteristics Initial ND symptoms (n=193) Full syndrome of ND (n=228)
Unadjusted HRs (95%CI) Adjusted HRs (95%CI) Unadjusted HRs (95%CI) Adjusted HRs (95%CI)
Individual characteristics Gender (Female vs Male)¥ 1.22 [0.95–1.57]# 1.04 [0.58–1.84] 0.68 [0.42–1.11]# 0.50 [0.19–1.32]
School (Public vs. Private)¥ 1.11 [0.86–1.44] 1.21 [0.68–2.15] 1.62 [1.06– 2.65]* 1.51 [0.66–3.48]
Physical activity (at least once/week) (yes vs no) 0.63 [0.49– 0.82]* 0.69 [0.50–0.95]* 0.97 [0.46–2.05] —
Father’s years of education (< 12 years/ illiterate vs ≥12 years)¥ 1.30 [1.02–1.68]* 0.97 [0.55–1.72] 0.74 [0.45–1.22] —
Mother’s years of education (< 12 years/ illiterate vs ≥12 years)¥ 2.110 1.62–2.74]* 1.99 [1.09–3.63]* 0.91 [0.56–1.48] —
Age, years 0.68 [0.64–0.78]* 0.68 [0.59–0.79]* 0.77 [0.67–0.89]* 0.76 [0.62–0.96]*
BMI (weight/height^2) 1.02 [0.99–1.05]# 1.06 [1.01–1.12]* 1.05 [1.01–1.09]* 1.07 [0.97–1.18]
Crowding index¥ 0.80 [0.67–0.96]* 0.72 [0.43–1.20] 1.16 [0.74–1.84] —
Indicators of smoking in social environmental Parents smokes cigarette (yes vs no) 0.95 [0.74–1.23] — 0.98 [0.58–1.62] —
Parents smokes WP (yes vs no) 0.72 [0.46–1.10]# 0.77 [0.50–1.18] 0.92 [0.56–1.50] —
>1 sibling smoke cigarette 0.99 [0.76–1.30] — 0.53 [0.27–1.05] —
>1 sibling smoke WP 1.43 [1.12–1.83]* 1.43 [1.07–2.24]* 0.71 [0.43–1.17]# 0.65 [0.27–1.54]
>1 friend smoke cigarette 0.52 [0.40–1.00] — 0.34 [0.19–0.60]* 0.25 [1.00–0.73]
>1 friend smoke WP 0.92 [0.58–1.44] — 1.04 [0.40–2.49] —
Beliefs about smoking WP smokers looks attractive (Agree vs Disagree/or don’t know) 2.03 [1.44–2.85]* 1.23 [0.77–1.96] 1.39 [0.71–2.73] —
Cigarette smokers looks attractive (Agree vs Disagree/or don’t know) 0.85 [0.530–1.38] — 1.00 [0.45–2.18] —
WP smokers have more friends (Agree vs Disagree/or don’t know) 1.39 [1.05–1.85]* 0.98 [0.64–1.49] 2.44 [1.47–4.05]* 2.76 [1.05–7.30]*
Cigarette smokers have more friends (Agree vs Disagree/or don’t know) 1.62 [1.14–2.31] — 0.48 [0.19–1.19] —
WP smoking makes a person lose weight (Agree vs Disagree/or don’t know) 0.91 [0.56–1.50] — 0.53 [0.21–1.34] —
Cigarette smoking makes a person lose weight (Agree vs Disagree/or don’t know) 0.37 [0.14–1.01] — 0.90 [0.41–1.97] —
WP smoking is harmful to health (Agree vs Disagree/or don’t know) 0.53 [0.33–0.86]* 0.65 [0.33–1.26] 2.10 [0.29–15.13] —
Cigarette smoking is harmful to health (Agree vs Disagree/or don’t know) 1.72 [0.54–5.45] — 1.67 [0.22–12.35] —
Smoking patterns Age of initiation 0.87 [0.84–0.91]* 0.84 [0.79–0.89]* 0.90 [0.82–0.98]* 0.88 [0.78–0.98]*
Frequency of use (daily vs non-daily) 1.03 [0.80–1.34]# 0.95 [0.57–1.59] 0.67 [0.38–1.20]* 0.85 [0.26–2.82]
No of WPs smoked in the past month 1.05 [1.01–1.09]* 1.00 [0.99–1.02] 1.06 [1.01–1.12]* 1.04 [1.02–1.08]*
Intention to quit (yes vs no) 1.1 2 [0.71–1.78] — 1.89 [1.15–3.09]* 1.68 [0.83–3.41]
Having quit attempt (yes vs no) 0.85 [0.54–1.35] — 1.70 [1.03–2.82]* 1.12 [0.36–3.47]
Psychosocial indicators Stress 1.05 [0.99–1.12] — 1.08 [1.04–1.12]* 0.99 [0.91–1.06]
Depression 1.03 [0.98–1.09] — 1.15 [1.09–1.21]* 1.13 [1.02–1.25]*
Distractibility 0.93 [0.87–1.01] — 1.10 [1.05–1.15]* 1.02 [0.93–1.12]
Novelty seeking 1.05 [1.04–1.07]* 1.02 [0.99–1.04] 1.06 [1.03–1.09]* 0.99 [0.93–1.07]
Impulsivity 1.07 [1.05–1.09]* 1.03 [1.01–1.07]* 1.08 [1.05–1.11]* 1.06 [1.02–1.11]*
Self-esteem 1.06 [1.03–1.10]* 0.98 [0.94–1.02] 1.05 [0.98–1.12] —
Open in a new tab

HR = Hazard Ratio; CI = Confidence Interval. BMI=Body Mass Index.

Sign (*) indicate p-value <0.05.

Sign (#) indicate p-value <0.25 (selected variables for multivariable analysis).

Sign (¥) indicates time-invariant predictors.

Note, for continuous variables, the HR approximates the risk change for every one-unit increase in the age, BMI, crowding index, age of initiation, number of WP/cigarettes smoked, and psychological indicators.

3.3. Predictors of Initial Symptoms and the Full Syndrome of ND among cigarette smokers

Between waves 1 and 8, of the 134 cigarette smokers who were at risk of experiencing initial ND symptoms, 51.5% (n=69) did so and among 139 cigarette smokers who were at risk of developing ICD-10 ND, 17.3% (n=24) did so. As shown in Table 3, predictors of experiencing initial ND symptoms among cigarette smokers were having ≥1 sibling who smokes cigarettes and living in a crowded household (lower SES), whereas being female, older age, smoking initiation at an older age, and lower BMI were protective factors. Being at an older age and initiating smoking at an older age were protective factors of experiencing full ND syndrome. In other words, smoking initiation at a younger age was a risk factor of experiencing full ND syndrome.

Table 3.

Weighted Cox regression of the association between predictors and experiencing initial ND symptoms and developing full syndrome of ND among cigarette smokers. WDLY Study. 2015–2020

Study characteristics Initial ND symptoms (n=134) Full syndrome of ND (n=139)
Unadjusted HRs (95%CI) Adjusted HRs (95%CI) Unadjusted HRs (95%CI) Adjusted HRs (95%CI)
Individual characteristics Gender (Female vs Male) ¥ 0.49 [0.23–1.02]# 0.43 [0.19–0.93]* 0.18 [0.02–1.32]# 0.50 [0.07–3.71]
School (Public vs. Private) ¥ 1.39 [0.82–2.33] 1.67 [0.88–3.13] 1.16 [0.49–2.71] 1.17 [0.46–2.99]
Physical activity (at least once/week) (yes vs no) 0.94 [0.57–1.57]# 0.97 [0.63–1.53] 1.74 [0.65–4.66] —
Father’s years of education (< 12 years/ illiterate vs ≥12 years) ¥ 1.22 [0.76–1.97] — 1.37 [0.87–2.17] —
Mother’s years of education (< 12 years/ illiterate vs ≥12 years) ¥ 1.00 [0.62–1.61] — 1.96 [1.24–3.09]* 1.13 [0.38–3.36]
Age, years 0.85 [0.81–0.91]* 0.76 [0.70–0.83]* 0.81 [0.66– 0.99]* 0.75 [0.65–0.87]*
BMI (weight/height^2) 0.98 [0.95–1.00]# 0.95 [0.93–0.97]* 0.94 [0.85–1.04] —
Crowding index ¥ 1.44 [1.09–1.91]* 1.40 [1.01–1.93]* 1.37 [0.89–2.10]# 1.50 [0.89–2.52]
Indicators of smoking in social environmental Parents smokes cigarette (yes vs no) 1.84 [1.08–3.16]* 1.47 [0.87–2.53] 1.86 [0.74–4.69]# 1.75 [0.74–4.15]
Parents smokes WP (yes vs no) 0.69 [0.43–1.13] — 2.17 [0.93–5.08]# 1.69 [0.61–4.69]
≥1 sibling smoke cigarette 1.73 [1.07–2.78]* 1.86 [1.19–2.93]* 0.78 [0.33–1.83] —
≥1 sibling smoke WP 0.99 [0.58–1.67] — 0.81 [0.32–2.06] —
≥1 friend smoke cigarette 1.13 [0.49–2.61] — 1.14 [0.26–4.87] —
≥1 friend smoke WP 0.64 [0.28–1.49] — 1.53 [0.20–11.33] —
Beliefs about smoking WP smokers looks attractive (Agree vs Disagree/or don’t know) 1.27 [0.46–3.50] — 2.99 [0.89–10.08]# 1.63 [0.56–4.78]
Cigarette smokers looks attractive (Agree vs Disagree/or don’t know) 1.21 [0.65–2.25] — 0.55 [0.16–1.85] —
WP smokers have more friends (Agree vs Disagree/or don’t know) 1.98 [1.12–3.51]* 1.14 [0.68–1.93] 2.81 [1.16–6.79]* 1.08 [0.48–2.40]
Cigarette smokers have more friends (Agree vs Disagree/or don’t know) 1.22 [0.58–2.55] — 1.78 [0.61–5.23] —
WP smoking makes a person lose weight (Agree vs Disagree/or don’t know) 1.13 [0.58–2.22] — 1.77 [0.66–4.76] —
Cigarette smoking makes a person lose weight (Agree vs Disagree/or don’t know) 1.35 [0.78–2.34]# 1.03 [0.61–1.73] 0.59 [0.17–1.97] —
WP smoking is harmful to health (Agree vs Disagree/or don’t know) 0.66 [0.21–2.09] — 0.34 [0.08–1.47] —
Cigarette smoking is harmful to health (Agree vs Disagree/or don’t know) 1.12 [0.27–4.57]# 0.80 [0.36–1.73] 0.86 [0.11–6.39] —
Smoking patterns Age of initiation 0.85 0.81–0.91]* 0.79 [0.74–0.84]* 0.81 [0.66– 0.99]* 0.76 [0.67–0.86]*
Frequency of use (daily vs non-daily) 2.68 [1.44–5.01]* 2.19 [1.16–4.13]* 2.62 [0.89–7.68]# 2.62 [1.75–9.15]*
No of cigarettes smoked in the past month 1.04 [1.02–1.08]* 1.02 [1.01–1.04]* 1.01 [0.97–1.03]# 1.02 [1.01–1.03]*
Intention to quit (yes vs no) 1.02 [0.59–1.76] — 1.75 [1.09–2.80]* 1.45 [0.41–5.09]
Having quit attempt (yes vs no) 0.97 [0.59–1.59] — 1.69 [1.08–2.64]* 1.20 [0.41–3.49]
Psychosocial indicators Stress 1.02 [0.96–1.09] — 1.07 [0.98–1.18]# 1.04 [0.97–1.11]
Depression 1.00 [0.94–1.06] — 0.96 [0.85–1.09] —
Distractibility 0.99 [0.89–1.10] — 1.03 [0.93–1.14] —
Novelty seeking 1.00 [0.97–1.03] — 1.03 [0.98–1.08]# 1.02 [0.96–1.07]
Impulsivity 0.98 [0.94–1.02] — 1.02 [0.96–1.08] —
Self-esteem 1.10 [1.01–1.18]* 1.06 [0.98–1.13] 1.19 [1.09–1.30]* 1.10 [1.00–1.18]
Open in a new tab

HR = Hazard Ratio; CI = Confidence Interval. BMI=Body Mass Index.

Sign (*) indicate p-value <0.05.

Sign (#) indicate p-value <0.25 (selected variables for multivariable analysis).

Sign (¥) indicates time-invariant predictors.

Note, for continuous variables, the HR approximates the risk change for every one-unit increase in the age, BMI, crowding index, age of initiation, number of WP/cigarettes smoked, and psychological indicators. .

4. Discussion

Adolescents are at a higher risk for ND because of nicotine’s irreversible and profound effect on their developing brain(Yuan et al., 2015). Generally, it is well-established that a younger age of exposure to addictive substances is strongly associated with lifelong addiction(Bonnie et al., 2015). Specifically for tobacco, early symptoms, and full ND syndrome can appear within days to weeks of the onset of WP(Bahelah et al., 2016a; Ebrahimi Kalan et al., 2020) and cigarette smoking(DiFranza et al., 2000; Ebrahimi Kalan et al., 2020; Gervais et al., 2006). While nicotine effect on the developing brain is likely to be universal, nuances related to the vehicle of delivery of nicotine and its context-specific factors and cues will likely shape the development and composition of ND among youth(Aboaziza and Eissenberg, 2014; CDC, 2010, 2012; Maziak et al., 2005). In fact, ND predicts smoking consistency and quantity across teenage years into young adulthood, therefore, understanding these nuances is important to develop tailored approaches to reduce tobacco use and addiction among youth and young adults (Bonnie et al., 2015; Allem and Unger, 2016). We already have extensive knowledge about the development of ND among young cigarette smokers and its predictors, but the same is not true for the most popular tobacco use method among adolescents in the EMR, the WP(Maziak et al., 2005; Maziak et al., 2015). Our cohort in the EMR is uniquely set to address this topic being the only such cohort that follows adolescent WP and cigarette smokers as they develop ND. In this study, for both tobacco use methods, we observed some commonalities in predictors of ND milestones (e.g. age of initiation, sibling tobacco use), which allows us to apply some of the experience of targeting these predictors among cigarette smokers. At the same time, there were substantial differences in the predictors of these ND milestones between the two tobacco use methods (e.g., the role of physical activity, BMI), which will be important to inform the development of tailored interventions targeting these popular tobacco use methods in the EMR.

Our findings show that the initiation of tobacco at a younger age is associated with a greater risk of experiencing initial ND symptoms and developing full ND syndrome for both WP and cigarette smokers. This study is the first to show that early initiation of WP is an independent risk factor for experiencing initial symptoms and developing the full syndrome of ND among adolescents; a finding that is already known for cigarette smoking(Breslau et al., 1993). Another shared predictor of the development of initial ND symptoms between WP and cigarette smokers is having at least one sibling who smokes the same product. This could be an important driver of WP smoking in the EMR context where family attitude and norms are important(Ali and Jawad, 2017), and where WP is rooted in this culture and is more tolerated than cigarettes(Akl et al., 2015; Hammal et al., 2008). Therefore, existing strategies to reduce cigarette smoking among youth such as those based on resisting peer pressure (in this case siblings) can be effective for both products(Akl et al., 2013; CDC, 2012). Special attention perhaps in the EMR context is the focus on communicating WP harmful and addictive properties to family members as a means to protect their children(Akl et al., 2013; Roohafza et al., 2015; Tobacco-Free Kids, 2009 ). Evidence shows, for example, that family discussion regarding the dangers of WP smoking (e.g., life-long dependency) significantly reduced the likelihood of being a current WP smoker among youth(Alzyoud et al., 2013). As universally effective for both tobacco methods perhaps, are upstream strategies to limit youth access to tobacco products (e.g., taxation, age restrictions) (Jawad et al., 2015).

Apart from commonalities, this study highlighted for the first time differences in the risk factors for experiencing ND symptoms between adolescent WP and cigarette smokers. For example, in contrast to cigarette smoking used often as a means for weight control among youth (girls in particular), a higher BMI and low physical activity levels were risks of developing initial ND symptoms among WP smokers. Population-based studies from Syria indicated that WP smokers, compared to never-users, had higher BMI, translating into 6 extra kg on average, and were 3 times more likely to be obese(Ward, Kenneth D. et al., 2015). This can be explained with the social context and prolonged sessions of WP smoking (an hour or more) and its association with the WP café/restaurant setting where food is served and consumed around the WP(Aboaziza and Eissenberg, 2014; Baalbaki et al., 2019). In fact, our earlier results from this cohort show that inability to resist WP smoking in these venues was the strongest predictor of ND symptoms progression(Bahelah et al., 2019). This was further supported by our findings showing that regular physical activity was inversely predicting the development of initial ND symptoms among WP but not cigarette smokers. Accordingly, strategies to promote a healthy diet and physical activities among youth, combined with upstream restrictions on underage access to WP venues and clean indoor air policies particularly in restaurants, bars, and cafes can be effective to reduce WP smoking among youth in the EMR.

Another unique aspect of the development of ND among WP compared to cigarette smokers in our cohort was that psychosocial factors (impulsivity and depression) were strong predictors of developing full ND syndrome in WP smokers, but not in cigarette smokers. Differences in who uses and why these tobacco methods are used, may shed some light on these findings. In the EMR context, and unlike cigarettes, WP is looked at as a way to spend good time in the company of friends and family, and as a mood enhancer, while cigarette smoking is looked at as a mundane addiction(Hammal et al., 2008). In fact, having positive beliefs about WP in this study was an independent predictor of developing full ND syndrome in WP smokers, but not in cigarettes. Evidence shows that positive WP outcome expectancies (e.g. connecting WP smoking with relaxation and socialization) are associated with ND, less motivation to quit, and reduced ability to quit WP tobacco smoking (Barnett et al., 2017; Aboaziza and Eissenberg, 2014; Maziak et al., 2015). The positive attitude towards WP in the EMR context is intertwined with its social acceptability compared to cigarettes, being considered as part of the local culture(Ebrahimi Kalan and Ben Taleb, 2018; Maziak et al., 2015). The effect of such norms is most noticeable for females, where WP smoking is way more tolerated than cigarette smoking in the EMR (Abdulrashid et al., 2018; Hammal et al., 2008). This explains the high prevalence of WP smoking among females in our sample compared to males (61.4% vs 38.6%, respectively) with the opposite is true for cigarette smoking (18% vs 82%, respectively). Being a female was a protective factor against experiencing initial ND symptoms for cigarettes but not WP in this study. As such, our findings indicate that at the individual level, adolescent WP smokers with depressive symptoms represent an important subgroup in need of targeted smoking cessation interventions in the EMR (Dierker et al., 2015). At the population level, strategies to increase awareness about the harmful effects of WP smoking and de-normalize its use, are warranted to counterbalance the “cultural endorsement” of this tobacco use method.

This study has some limitations. First, the longitudinal nature of the study makes possible errors in recalling dates of ND milestones. We tried to minimize such possibilities using techniques that improve event recall (e.g., personal landmarks, bounded recall, decomposition, and a visual aid) (Bahelah et al., 2017; DiFranza et al., 2002b). Third, our findings may not be generalizable to adolescent WP and cigarette smokers in other countries and there is a need for additional research elsewhere as context-specific factors will likely be important (e.g. tobacco control policies). Fourth, although we included those exclusive WP and cigarette smokers who were lost to follow-up in the main analysis, a statistical approach and dynamic characteristic of this cohort study made it possible to use available data to increase the power of the study and represent all cohort members who were exclusive WP and exclusive cigarette smokers. Finally, data were collected only from students who attended public or private schools, so our findings may be less generalizable to youths who are home-schooled or have dropped out of school. However, almost 97% of Lebanese youths aged 11–18 years were enrolled in a public or private school in 2015(Chaaban and El Khoury, 2015). Also, the findings from this study can be informative for other countries in the MER due to the shared culture and factors related to WP smoking among youth in this region(Maziak et al., 2014, 2015).

Despite these limitations and given the commonalities in the EMR cultural context that is driving the huge epidemic of WP smoking among adolescents, our findings are important to guide tailored approaches to reduce tobacco use among adolescents in the EMR.

5. Conclusion

We previously hypothesized that ND in WP smokers is shaped by factors unique to its social context and cues beyond the addictive effects of nicotine. In this study, we further advance our knowledge about WP-specific nuances affecting the development of ND among youth. Taken together, our findings indicate that focusing on communicating WP harmful and addictive properties, de-normalizing its use, promoting a healthy diet and physical activity, and targeting atrisk children for cessation intervention as promising strategies to reduce WP smoking among youth. At the population level, our findings indicate the importance of limiting underage access to tobacco products and venues, and clean indoor air policies particularly in cafés and restaurants serving the WP.

Supplementary Material

1

Highlights.

  • No studies have documented the predictors of ND among adolescent WP and cigarette smokers.

  • For both tobacco use methods, we observed some commonalities in predictors of ND milestones (e.g. age of initiation, sibling tobacco use).

  • There were substantial differences in the predictors of these ND milestones between the two tobacco use methods (e.g., the role of physical activity, BMI).

  • Smoking cessation interventions targeting youth should address modifiable and tobacco use-specific factors.

Acknowledgements

The authors are grateful to the Lebanese Ministry of Education, school administrators, and students who participated in this longitudinal study.

Role of Funding Source:

This study was supported by the National Institute of Health Forgery International Center (NIH FIC) under award number R01TW010654. MEK is supported by the National Institute on Drug Abuse at the National Institutes of Health (NIDA NIH) under award R01DA042477 and NIH FIC under award number R01TW010654. ZB supported by NIH NIMHD U54. RJ is supported by NIH FIC under award number R01TW010654. MTS is supported by NIDA (R01DA041353, K01DA037819) and NIMHD (U54MD012393, subproject 5378). TE is supported by the NIDA NIH under award number U54DA036105 and the Center for Tobacco Products of the US Food and Drug Administration (FDA). WM is supported by the NIH FIC under award R01TW010654 and the NIDA NIH under award R01DA042477. The content is solely the responsibility of the authors and does not necessarily represent the views of the NIH or the FDA.

Footnotes

Conflict of Interest:

Dr. Eissenberg is a paid consultant in litigation against the tobacco industry and also the electronic cigarette industry and is named on a patent for a device that measures the puffing behavior of electronic cigarette users. All other authors reported no conflict of interest.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Abdulrashid OA, Balbaid O, Ibrahim A, Shah HBU, 2018. Factors contributing to the upsurge of water-pipe tobacco smoking among Saudi females in selected Jeddah cafés and restaurants: A mixed method study. J Family Community Med 25(1), 13–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aboaziza E, Eissenberg T, 2015. Waterpipe tobacco smoking: what is the evidence that it supports nicotine/tobacco dependence? Tob Control 24 Suppl 1(Suppl 1), i44–i53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Akl EA, Jawad M, Lam WY, Co CN, Obeid R, Irani J, 2013. Motives, beliefs and attitudes towards waterpipe tobacco smoking: a systematic review. Harm Reduct J 10, 12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Akl EA, Ward KD, Bteddini D, Khaliel R, Alexander AC, Lotfi T, Alaouie H, Afifi RA, 2015. The allure of the waterpipe: a narrative review of factors affecting the epidemic rise in waterpipe smoking among young persons globally. Tob Control 24(Suppl 1), i13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Ali M, Jawad M, 2017. Health Effects of Waterpipe Tobacco Use: Getting the Public Health Message Just Right. Tobacco Use Insights 10,1179173X17696055. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Allison PD, 2010. Survival analysis using SAS: a practical guide. SAS Institute. [Google Scholar]
  7. Alzyoud S, Weglicki LS, Kheirallah KA, Haddad L, Alhawamdeh KA, 2013. Waterpipe smoking among middle and high school Jordanian students: patterns and predictors. Int J Environ Res Public Health 10(12), 7068–7082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Allem JP, Unger JB, 2016. Emerging adulthood themes and hookah use among college students in Southern California. Addictive Behaviors 61, 16–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Andersen PK, Gill RD, 1982. Cox’s Regression Model for Counting Processes: A Large Sample Study. Ann Statist 10(4), 1100–1120. [Google Scholar]
  10. Auf RA, Radwan GN, Loffredo CA, El Setouhy M, Israel E, Mohamed MK, 2012. Assessment of tobacco dependence in waterpipe smokers in Egypt. INT J TUBERC LUNG D 16(1), 132–137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Baalbaki R, Itani L, El Kebbi L, Dehni R, Abbas N, Farsakouri R, Awad D, Tannir H, Kreidieh D, El Masri D, El Ghoch M, 2019. Association Between Smoking Hookahs (Shishas) and Higher Risk of Obesity: A Systematic Review of Population-Based Studies. J Cardiovasc Dev Dis 6(2), 23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Bahelah R, DiFranza JR, Fouad FM, Ward KD, Eissenberg T, Maziak W, 2016a. Early symptoms of nicotine dependence among adolescent waterpipe smokers. Tob control 25(e2), e127–e134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Bahelah R, DiFranza JR, Ward KD, Fouad FM, Eissenberg T, Ben Taleb Z, Jaber R, Osibogun O, Maziak W, 2016b. Correlates of nicotine dependence among adolescent waterpipe smokers. Drug Alcohol Depend 168, 230–238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Bahelah R, DiFranza JR, Ward KD, Eissenberg T, Fouad FM, Taleb ZB, Jaber R, Maziak W, 2017. Waterpipe smoking patterns and symptoms of nicotine dependence: The Waterpipe Dependence in Lebanese Youth Study. Addict Behav 74, 127–133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Bahelah R, Ward KD, Ben Taleb Z, DiFranza JR, Eissenberg T, Jaber R, Maziak W, 2019. Determinants of progression of nicotine dependence symptoms in adolescent waterpipe smokers. Tob Control 28(3), 254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Barnett TE, Lorenzo FE, Soule EK, 2017. Hookah Smoking Outcome Expectations Among Young Adults. Substance use & misuse 52(1), 63–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Bejjani N, El Bcheraoui C, Adib SM, 2012. The social context of tobacco products use among adolescents in Lebanon (MedSPAD-Lebanon). Clin Epidemiol Glob Health 2(1), 15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Bonnie RJ, Stratton K, Kwan LY, editors., DC: National Academies Press, 2015. Committee on the Public Health Implications of Raising the Minimum Age for Purchasing Tobacco Products. 1–399. https://www.ncbi.nlm.nih.gov/books/NBK310412/ doi: 10.17226/18997 [DOI] [PubMed] [Google Scholar]
  19. Breslau N, Fenn N, Peterson EL, 1993. Early smoking initiation and nicotine dependence in a cohort of young adults. Drug Alcohol Depend 33(2), 129–137. [DOI] [PubMed] [Google Scholar]
  20. Brunet J, Sabiston CM, Chaiton M, Low NCP, Contreras G, Barnett TA, O’Loughlin JL, 2014. Measurement invariance of the depressive symptoms scale during adolescence. BMC Psychiatry 14, 95–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Centers for Disease Control and Prevention (CDC), 2010; National Center for Chronic Disease Prevention and Health Promotion (US); Office on Smoking and Health (US),How Tobacco Smoke Causes Disease: The Biology and Behavioral Basis for Smoking-Attributable Disease: A Report of the Surgeon General, Nicotine Addiction: Past and Present; Available from: https://www.ncbi.nlm.nih.gov/books/NBK53018/. [PubMed] [Google Scholar]
  22. Centers for Disease Control and Prevention (CDC), 2012. Reports of the Surgeon General, National Center for Chronic Disease, Prevention, Health Promotion Office on, Smoking, Health, Preventing Tobacco Use Among Youth and Young Adults: A Report of the Surgeon General. Centers for Disease Control and Prevention (US), Atlanta (GA). [PubMed] [Google Scholar]
  23. Chaaban J, Khoury El., 2015. Spot light on youth in Lebanon. United Nations Development Programme; (Accessed 09/22/2020) https://www.undp.org/content/dam/lebanon/docs/Poverty/Publications/Spotlight%20on%20Youth%20Final%20(23-06-2016).pdf [Google Scholar]
  24. Dierker L, Rose J, Selya A, Piasecki TM, Hedeker D, Mermelstein R, 2015. Depression and nicotine dependence from adolescence to young adulthood. Addict behav 41, 124–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. DiFranza JR, Rigotti NA, McNeill AD, Ockene JK, Savageau JA, St Cyr D, Coleman M, 2000. Initial symptoms of nicotine dependence in adolescents. Tob Control 9(3), 313319. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. DiFranza JR, Savageau JA, Fletcher K, et al. , 2002a. Measuring the loss of autonomy over nicotine use in adolescents: The dandy (development and assessment of nicotine dependence in youths) study. Arch Pediatr Adolesc Med 156(4), 397–403. [DOI] [PubMed] [Google Scholar]
  27. DiFranza JR, Savageau JA, Rigotti NA, Fletcher K, Ockene JK, McNeill AD, Coleman M, Wood C, 2002b. Development of symptoms of tobacco dependence in youths: 30 month follow up data from the DANDY study. Tob Control 11(3), 228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. DiFranza JR, Savageau JA, Fletcher K, Pbert L, O’Loughlin J, McNeill AD, Ockene JK, Friedman K, Hazelton J, Wood C, Dussault G, Wellman RJ, 2007. Susceptibility to nicotine dependence: the Development and Assessment of Nicotine Dependence in Youth 2 study. Pediatrics 120(4), e974–983. [DOI] [PubMed] [Google Scholar]
  29. DiFranza JR, Ursprung WWSA, 2008. The latency to the onset of nicotine withdrawal: A test of the sensitization-homeostasis theory. Addict Behav 33(9), 1148–1153. [DOI] [PubMed] [Google Scholar]
  30. Ebrahimi Kalan M, Behaleh R, DiFranza JR, Bursac Z, Ben Taleb Z, Tleis M, Asfar T, Nakkash R, Ward KD, Eissenberg T, Maziak W, 2020. Natural Course of Nicotine Dependence Among Adolescent Waterpipe and Cigarette Smokers. J Adolesc Health. DOI: 10.1016/j.jadohealth.2020.04.030 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Ebrahimi Kalan M, Ben Taleb Z, 2018. Waterpipe tobacco smoking: A reality or hidden iceberg for Iranian women. Health Promot Perspect 8(4), 252–254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Gervais A, O’Loughlin J, Meshefedjian G, Bancej C, Tremblay M, 2006. Milestones in the natural course of onset of cigarette use among adolescents. CMAJ 175(3), 255–261. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Hammal F, Mock J, Ward KD, Eissenberg T, Maziak W, 2008. A pleasure among friends: how narghile (waterpipe) smoking differs from cigarette smoking in Syria. Tob Control 17(2), e3. [DOI] [PubMed] [Google Scholar]
  34. Hosmer DW, Lemeshow S, May S, 2002. Applied survival analysis: regression modeling of time to event data. Wiley New York, NY. [Google Scholar]
  35. Hu MC, Davies M, Kandel DB, 2006. Epidemiology and correlates of daily smoking and nicotine dependence among young adults in the United States. Am J Public Health 96(2), 299–308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Hu MC, Griesler PC, Schaffran C, Wall MM, Kandel DB, 2012. Trajectories of criteria of nicotine dependence from adolescence to early adulthood. Drug Alcohol Depend 125(3), 283–289. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Hu MC, Muthén B, Schaffran C, Griesler PC, Kandel DB, 2008. Developmental trajectories of criteria of nicotine dependence in adolescence. Drug Alcohol Depend 98(12), 94–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Jaber R, Madhivanan P, Khader Y, Mzayek F, Ward KD, Maziak W, 2015. Predictors of waterpipe smoking progression among youth in Irbid, Jordan: A longitudinal study (2008–2011). Drug Alcohol Depend 153, 265–270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Jawad M, Charide R, Waziry R, Darzi A, Ballout RA, Akl EA, 2018. The prevalence and trends of waterpipe tobacco smoking: A systematic review. PLoS ONE 13(2), e0192191. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Jawad M, El Kadi L, Mugharbil S, Nakkash R, 2015. Waterpipe tobacco smoking legislation and policy enactment: a global analysis. Tob Control 24(Suppl 1), i60–i65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Kleinbaum DG, Klein M, 2010. Survival analysis. Springer. [Google Scholar]
  42. Kleinjan M, Vitaro F, Wanner B, Brug J, Van Den Eijnden RJ, Engels RC, 2012. Predicting nicotine dependence profiles among adolescent smokers: the roles of personal and social-environmental factors in a longitudinal framework. BMC Public Health 12(1), 196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Kohl M, Plischke M, Leffondré K, Heinze G, 2015. PSHREG: A SAS macro for proportional and nonproportional subdistribution hazards regression. Computer methods and programs in biomedicine 118(2), 218–233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Lam E, Giovino GA, Shin M, Lee KA, Rolle I, Asma S, 2014. Relationship between frequency and intensity of cigarette smoking and TTFC/C among students of the GYTS in select countries, 2007–2009. J Sch Health 84(9), 549–558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Maziak W, Eissenberg T, Ward KD, 2005. Patterns of waterpipe use and dependence: implications for intervention development. Pharmacol Biochem Behav 80(1), 173–179. [DOI] [PubMed] [Google Scholar]
  46. Maziak W, Nakkash R, Bahelah R, Husseini A, Fanous N, Eissenberg T, 2014. Tobacco in the Arab world: old and new epidemics amidst policy paralysis. Health Policy Plan 29(6), 784–794. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Maziak W, Osibogun O, Asfar T, 2019. Waterpipe smoking: the pressing need for risk communication. Expert Rev Respir Med, 1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Maziak W, Taleb ZB, Bahelah R, Islam F, Jaber R, Auf R, Salloum RG, 2015. The global epidemiology of waterpipe smoking. Tob Control 24 Suppl 1, i3–i12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Maziak W, Ward KD, Eissenberg T, 2004. Factors related to frequency of narghile (waterpipe) use: the first insights on tobacco dependence in narghile users. Drug Alcohol Depend 76(1), 101–106. [DOI] [PubMed] [Google Scholar]
  50. Melki IS, Beydoun HA, Khogali M, Tamim H, Yunis KA, National Collaborative Perinatal Neonatal, N., 2004. Household crowding index: a correlate of socioeconomic status and inter-pregnancy spacing in an urban setting. J Epidemiol Community Health 58(6), 476–480. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Moahmed TA, El Gayar N, Atiya AF, 2014. Forward and Backward Forecasting Ensembles for the Estimation of Time Series Missing Data, Springer Berlin Heidelberg, pp. 93–104. [Google Scholar]
  52. Neergaard J, Singh P, Job J, Montgomery S, 2007. Waterpipe smoking and nicotine exposure: a review of the current evidence. Nicotine Tob Res 9(10), 987–994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. O’Loughlin J, Karp I, Koulis T, Paradis G, Difranza J, 2009. Determinants of first puff and daily cigarette smoking in adolescents. Am J Epidemiol 170(5), 585–597. [DOI] [PubMed] [Google Scholar]
  54. O’Loughlin J, Tarasuk J, Difranza J, Paradis G, 2002. Reliability of selected measures of nicotine dependence among adolescents. Ann Epidemiol 12(5), 353–362. [DOI] [PubMed] [Google Scholar]
  55. O’Loughlin JO, DiFranza J, Tyndale RF, Meshefedjian G, McMillan-Davey E, Clarke P, Hanley J, Paradis G, 2003. Nicotine-dependence symptoms are associated with smoking frequency in adolescents. Am J Prev Med 25. [DOI] [PubMed] [Google Scholar]
  56. Powell TM, Bagnell ME, 2012. Your “survival” guide to using time-dependent covariates, Proceedings of the SAS Global Forum; pp. 22–25. [Google Scholar]
  57. Primack BA, Carroll MV, Weiss PM, Shihadeh AL, Shensa A, Farley ST, Fine MJ, Eissenberg T, Nayak S, 2016. Systematic Review and Meta-Analysis of Inhaled Toxicants from Waterpipe and Cigarette Smoking. Public Health Rep (Washington, D.C. : 1974) 131(1), 76–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Racicot S, McGrath JJ, Karp I, O’Loughlin J, 2013. Predictors of nicotine dependence symptoms among never-smoking adolescents: A longitudinal analysis from the Nicotine Dependence in Teens Study. Drug Alcohol Depend 130(1–3), 38–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Robins JM, Hernán MA, Brumback B, 2000. Marginal structural models and causal inference in epidemiology. Epidemiology 11(5), 550–560. [DOI] [PubMed] [Google Scholar]
  60. Robinson JN, Wang B, Jackson KJ, Donaldson EA, Ryant CA, 2017. Characteristics of Hookah Tobacco Smoking Sessions and Correlates of Use Frequency Among US Adults: Findings From Wave 1 of the Population Assessment of Tobacco and Health (PATH) Study. Nicotine Tob Res 20(6), 731–740. [DOI] [PubMed] [Google Scholar]
  61. Roohafza H, Kasaei Z, Heidari K, Omidi R, Alinia T, Naji M, Jaberifar M, Sadeghi M, 2015. Better view on attitudes and perceived parental reactions behind waterpipe smoking among Iranian students. J Res Med Sci 20(11), 1032–1038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Sharapova S, Reyes-Guzman C, Singh T, Phillips E, Marynak KL, Agaku I, 2020. Age of tobacco use initiation and association with current use and nicotine dependence among US middle and high school students, 2014–2016. Tob Control 29(1), 49. [DOI] [PubMed] [Google Scholar]
  63. Thomson B, Rojas NA, Lacey B, Burrett JA, Varona-Pérez P, Martínez MC, Lorenzo-Vázquez E, Constantén SB, Morales Rigau JM, Hernández López OJ, Martínez Morales MÁ, Alomá IA, Estupiñan FA, González MD, Muñoz NR, Asencio MC, Emberson J, Peto R, Lewington S, Herrera AD, 2020. Association of childhood smoking and adult mortality: prospective study of 120 000 Cuban adults. The Lancet Global Health 8(6), e850–e857. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Tobacco-Free Kids, 2009. Campaign for Tobacco-Free Kids. How parents can protect their kids from becoming addicted smokers; (Accessed 06/20/2020) https://www.amadorgov.org/home/showdocument?id=6650 [Google Scholar]
  65. Ward KD, Ahn S, Mzayek F, Al Ali R, Rastam S, Asfar T, Fouad F, Maziak W, 2015. The relationship between waterpipe smoking and body weight: population-based findings from Syria. Nicotine Tob Res 17(1), 34–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Ward KD, Siddiqi K, Ahluwalia JS, Alexander AC, Asfar T, 2015. Waterpipe tobacco smoking: The critical need for cessation treatment. Drug Alcohol Depend 153, 14–21. [DOI] [PubMed] [Google Scholar]
  67. Waters K, Harris K, Hall S, Nazir N, Waigandt A, 2006. Characteristics of social smoking among college students. J Am Coll Health 55(3), 133–139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. World Health Organization (WHO), 1993. The ICD-10 Classification of Mental and Behavioural Disorders: Diagnostic criteria for research, in: Health, W., Organization, W.H. (Eds.). Geneva, Switzerland. [Google Scholar]
  69. World Health Organization (WHO), 2015. World Health Organization Study Group on Tobacco Product Regulation. Advisory note: waterpipe tobacco smoking: health effects, research needs and recommended actions by regulators, 2nd ed. [Google Scholar]
  70. World Health Organization (WHO), 2019. WHO REPORT ON THE GLOBAL TOBACCO EPIDEMIC, 2019. Offer help to quit tobacco use; (Accessed 06/22/2020) https://apps.who.int/iris/bitstream/handle/10665/326043/9789241516204-eng.pdf?ua=1. [Google Scholar]
  71. Yuan M, Cross SJ, Loughlin SE, Leslie FM, 2015. Nicotine and the adolescent brain. J Physiol 593(16), 3397–3412. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1637905-supplement-1.docx (17.1KB, docx)

RESOURCES