TABLE 2

Multinomial Logistic Regression Analyses of Factors Associated With Frequency of Hookah Smoking Among US Middle and High School Students, NYTS, 2016 (n = 20 675)

CharacteristicFormer Versus Never,a aOR (95% CI)Occasional Current Versus Never,a aOR (95% CI)Frequent Current Versus Never,a aOR (95% CI)
Overestimation of peer hookah useb
 No (referent)
 Yes3.41 (3.39–3.43)*9.30 (9.18–9.42)*6.54 (6.42–6.66)*
 Indeterminate3.44 (3.40–3.49)*7.55 (7.38–7.73)*7.74 (7.50–7.99)*
Tobacco use by household member
 None (referent)
 Hookah (alone or with other products)3.29 (3.26–3.32)*7.81 (7.73–7.89)*20.56 (20.27–20.86)*
 Nonhookah onlyc1.58 (1.57–1.58)*1.09 (1.09–1.10)*1.54 (1.52–1.56)*
Sex
 Male (referent)
 Female1.12 (1.12–1.13)*1.54 (1.53–1.55)*0.65 (0.65–0.66)*
 Indeterminate1.13 (1.10–1.16)*0.46 (0.44–0.48)*1.41 (1.35–1.46)*
School
 Middle (referent)
 High2.66 (2.64–2.67)*1.61 (1.60–1.62)*1.25 (1.24–1.27)*
 Indeterminate1.99 (1.93–2.04)*1.02 (0.97–1.07)4.59 (4.40–4.78)*
Race and/or ethnicity
 White, non-Hispanic (referent)
 African American, non-Hispanic1.01 (1.00–1.01)1.49 (1.48–1.51)*1.39 (1.37–1.41)*
 Asian American, non-Hispanic0.76 (0.75–0.77)*0.47 (0.46–0.48)*0.73 (0.70–0.76)*
 Hispanic1.46 (1.46–1.47)*1.67 (1.66–1.69)*2.15 (2.12–2.17)*
 Other, non-Hispanic1.23 (1.22–1.24)*1.25 (1.24–1.27)*2.07 (2.03–2.10)*
 Indeterminate1.03 (1.02–1.04)*1.77 (1.75–1.80)*0.66 (0.64–0.68)*
Do you speak a language other than English at home?
 No (referent)
 Yes1.36 (1.35–1.36)*1.68 (1.67–1.69)*2.17 (2.15–2.19)*
 Indeterminate1.52 (1.50–1.54)*1.69 (1.65–1.72)*1.26 (1.22–1.29)*
Perceived harm of hookah
 No and/or little harm (referent)
 Some and/or a lot of harm0.33 (0.33–0.33)*0.21 (0.21–0.22)*0.15 (0.15–0.16)*
 Indeterminate0.31 (0.31–0.31)*0.21 (0.21–0.21)*0.51 (0.50–0.52)*
Cigarette smoking status (past 30 d)d
 Noncurrent and/or never smoking (referent)
 Nonmenthol cigarette2.51 (2.49–2.52)*4.52 (4.48–4.56)*9.84 (9.71–9.98)*
 Menthol cigarette2.95 (2.92–2.97)*6.20 (6.14–6.25)*19.94 (19.69–20.20)*
 Indeterminate2.20 (2.17–2.22)*3.88 (3.83–3.93)*15.18 (14.95–15.40)*
Other tobacco product use status (past 30 d)e
 Noncurrent and/or never use (referent)
 Nonflavored and/or menthol4.45 (4.43–4.48)*10.20 (10.11–10.28)*15.26 (14.99–15.54)*
 Flavored4.43 (4.41–4.45)*14.83 (14.72–14.94)*17.59 (17.30–17.89)*
 Indeterminate2.85 (2.83–2.87)*8.52 (8.42–8.61)*27.33 (26.83–27.84)*
  • Multinomial logistic regression analyses were controlled for all factors listed. Hookah smoking frequency was classified as never smoker (not even 1 or 2 puffs; n = 17 562), former smoker (last smoked hookah >30 d ago; n = 1395), current occasional smoker (smoked hookah on less than one-third of the past 30 d [1–9 d]; n = 564), and current frequent smoker (smoked hookah on greater than or equal to one-third of the past 30 d [10–30 d]; n = 199). Indeterminate responses include missing or “not sure” responses. CI, confidence interval.

  • a Never hookah smokers were the reference category within the multinomial logistic regression analyses.

  • b Respondents were asked how many students in their grade they thought used hookah out of 10 (perceived prevalence). A difference score (δ) was computed to measure the gap between perceived prevalence and grade-specific actual prevalence. Respondents were then dichotomized as either overestimating (δ >0) or not overestimating (δ ≤0) peer hookah use.

  • c Assessed nonhookah product use by household members included e-cigarettes, cigarettes, bidis, cigars, pipes, and smokeless tobacco.

  • d Menthol cigarette smoking was a report of smoking mentholated cigarettes or predominantly menthol brands (Newport and/or Kool).

  • e Assessed other noncigarette tobacco products included e-cigarettes, bidis, cigars, pipes, roll-your-own tobacco, and smokeless tobacco Flavored noncigarette tobacco products were those that tasted like “menthol (mint), alcohol (wine, cognac), candy, fruit, chocolate, or other sweets.”

  • * P < .05.