Dataset and study population

Data come from the 2014 NYTS; a school-based, self-administered survey among a large, nationally representative sample of middle (6-8 grades) and high (9-12 grades) school US adolescents from public and private schools¹⁵. In 2014, the NYTS inquired about ever and current (past 30 days) use of nine tobacco products (cigarettes; cigars, cigarillo, or little cigars; chew tobacco, snuff, or dip; electronic cigarettes; waterpipe; pipe; bidis; snus; and dissolvable tobacco). Information about student socio-demographics, smoking behaviors, attitudes, and exposure to tobacco ads were collected¹⁶. A stratified, three-stage cluster sampling design was used by the NYTS to obtain a representative sample of US middle and high school adolescents. The sampling design includes primary sampling units (PSUs) within each stratum, schools within each selected PSU, and in the final stage, classes within each selected school^15,16. More details about NYTS are available elsewhere¹⁶.

The sample for this study included adolescents who never smoked cigarettes, not even a puff or two. This was assessed by the survey question: ‘Have you ever tried cigarette smoking, even one or two puffs?’. Adolescents who answered ‘No’ to this question were considered never smoked cigarettes and were included in the analysis (N= 16,709; weighted %= 76.39). Since the data are publicly available, this study deemed exempt from IRB approval.

Outcome variables

Two outcomes, curiosity about cigarettes and susceptibility to cigarette smoking, were studied. Each of these outcomes independently predicts experimentation with cigarettes¹². Susceptibility to cigarette smoking was assessed by a composite index using three survey questions: ‘Do you think you will smoke a cigarette in the next year?’, ‘Do you think that you will try a cigarette soon?’, and ‘If one of your best friends were to offer you a cigarette, would you smoke it?’. Each question has four response options: ‘Definitely yes’, ‘Probably yes’, ‘Probably not’, and ‘Definitely not’. Students who answered ‘Definitely not’ to all three questions were considered not susceptible to cigarette smoking and those who gave a response other than ‘Definitely not’ to any of the three questions were considered susceptible to cigarette smoking¹³. Curiosity was assessed by the survey question: ‘Have you ever been curious about smoking a cigarette?’, with four response options: ‘Definitely yes’, ‘Probably yes’, ‘Probably not’, and ‘Definitely not’. Students who answered ‘Definitely not’ were considered not curious, and curious if answered otherwise^14,17.

Explanatory variable

The primary independent variable is current waterpipe smoking, defined as smoking waterpipe at least once in the past 30 days¹³. Eight hundred (3.33%) participants have missing data on current waterpipe smoking and were excluded from the analysis. Thus, the final sample included 1,349 (6.17%) current waterpipe smokers.

Covariates

Selection of covariates was guided by the socio-environmental framework of smoking initiation among adolescents, and by reviewing the literature for factors related to curiosity and susceptibility to cigarette smoking or smoking initiation among youth. Covariates included were socio-demographics, current use of other tobacco/nicotine products (electronic cigarettes, cigar/cigarillo/little cigar, smokeless tobacco), exposure to secondhand smoke/vape, exposure to pro-tobacco ads, receptivity to tobacco marketing, and beliefs about smoking-related harm^13,15,17,18.

Current use of tobacco/nicotine was defined as the use of a tobacco/nicotine product on at least one day during the past 30 days. Smokeless tobacco was defined as the use of chewing tobacco, snuff, dip, snus, or dissolvable tobacco¹⁹. To assess exposure to secondhand smoke (from a waterpipe) or vape (from electronic cigarettes), students were asked if anyone who lives with them now uses either of these products.

Exposure to pro-tobacco ads in convenient stores was assessed by the question: ‘When you go to a convenience store, supermarket, or gas station, how often do you see ads or promotions for cigarettes or other tobacco products?’; on the internet by the question: ‘When you are using the internet, how often do you see ads or promotions for cigarettes or other tobacco products?’; in magazines/newspapers by the question ‘When you read newspapers or magazines, how often do you see ads or promotions for cigarettes or other tobacco products?’; and in movies by the question ‘When you watch TV or go to the movies, how often do you see actors and actresses using cigarettes or other tobacco products?’. For all these variables, the exposure was classified into never, rarely/sometimes, and most of the time/always. Receptivity to tobacco marketing was assessed by the question: ‘How likely is it that you would ever use or wear something – such as a T-shirt, hat or sunglasses – with a tobacco brand name, logo, or picture on it?’. Students who answered ‘Very likely’ or ‘Somewhat likely’ were classified as having high receptivity and those who answered ‘Somewhat unlikely’ or ‘Very unlikely’ were classified as having low receptivity¹⁴.

Two questions were used to assess students’ beliefs about smoking-related harm: ‘How much do you think people harm themselves when they smoke cigarettes some days but not every day?’, and ‘How strongly do you agree with the statement “All tobacco products are dangerous”?’. Responses to the first question were classified into no harm, little/some harm, and a lot of harm. The second question has four responses: strongly agree, agree, disagree, and strongly disagree. Agree and strongly agree were grouped under the category ‘agree’, while disagree and strongly disagree were grouped under the category ‘disagree’.

Statistical analysis

Descriptive statistics for the sample’s characteristics were calculated and presented by curiosity and susceptibility status using percentages and 95% confidence intervals (CIs). All percentages were weighted using sampling weights to account for the complex survey design and non-response²⁰. In the bivariate analysis, percentages were considered statistically different when their corresponding CIs were not overlapping²¹. Participants with missing values in the outcome variables were excluded: 64 (0.33%) were missing curiosity status, while 54 (0.25%) were missing susceptibility status. Missing values in the covariates were excluded as they were overall low and ranged from 0.61% in age groups, to 3.33% in current use of smokeless tobacco products. Consequently, the percentages throughout this paper are for those with complete data. Separate multivariable logistic regression models were used to examine the factors associated with each outcome (curiosity, susceptibility) and all factors were entered simultaneously into these models. Hosmer-Lemeshow test was used to evaluate the model’s goodness-of-fit, and a p-value >0.05 indicates a good model fit²². Adjusted odds ratios (aOR) with 95% CIs were reported and standard errors were estimated using Taylor linearization method. All analyses were performed using svy command in Stata 13 (StataCorp, College Station, Texas) and the level of significance was set at p<0.05.

Sample characteristics

The sample for this study consists of 16,709 US youths who never smoked a cigarette, not even a puff or two. Among them 27.61% (95% CI: 26.47-28.78) were susceptible to cigarettes and 30.05% (95% CI: 28.91-31.21) were curious about cigarette smoking. Table 1 presents the sample characteristics. There were no significant differences in the percentage of boys and girls who were curious or susceptible to cigarette smoking. Youth age groups, those aged 17 years or more were less susceptible to cigarettes (24.16%, 95% CI: 21.51-27.03). Hispanic youth were more susceptible to cigarette smoking than all other races/ethnicities (Table 1).

Relationship between curiosity and susceptibility to cigarettes

Among the whole sample, 1736 youths (9.95%, 95% CI: 9.32-10.58) were not curious but susceptible to cigarettes, compared with 3060 youths (17.84%, 95% CI: 16.85- 18.82) who were both curious and susceptible to cigarettes.

Bivariate analysis

Susceptibility to cigarette smoking was higher among current waterpipe smokers and among individuals who were living with a waterpipe smoker than those who were not waterpipe smokers or living with someone who smokes waterpipe. Similar findings were found for electronic cigarettes, cigar/cigarillo/little cigar, or smokeless tobacco. Students exposed to pro-tobacco ads at convenient stores, internet, newspapers/magazines, or movies, and those highly receptive to tobacco marketing were more susceptible to cigarette smoking. Students who agree that all tobacco products are dangerous or believe people harm themselves a lot when they smoke somedays were less susceptible to cigarette smoking.

A higher percentage of current users of waterpipe or electronic cigarettes, but not cigar/cigarillo/little cigar or smokeless tobacco, were curious about cigarettes than non-users. Students living with someone who uses waterpipe, cigarettes, or electronic cigarettes were more curious about cigarettes than those who did not. Compared to those never exposed to pro-tobacco ads and those with low receptivity to tobacco marketing, students exposed to any kind of pro-tobacco ads assessed in this study (convenient store, internet, newspaper/magazines, movies) and those with high receptivity to tobacco marketing were more curious about cigarettes. Finally, students who agree that all tobacco products are dangerous or believe people harm themselves when they smoke somedays were less curious about cigarette smoking (Table 1).

Multivariable analysis

Table 2 presents results from multivariable logistic regression models assessing the association between current waterpipe smoking, and curiosity and susceptibility to cigarette smoking.

1)Currentwaterpipeuse and susceptibility to cigarette smoking

Among students who reported never smoked cigarettes, current waterpipe smokers were 1.5 times more susceptible to cigarette smoking than non-current waterpipe smokers; controlling for socio-demographics, use of other combustible and non-combustible tobacco products, receptivity to tobacco marketing, beliefs and attitudes towards tobacco use, and exposure to pro-tobacco ads (aOR=1.52, 95% CI: 1.01-2.26, p=0.04). Students living with someone who smokes waterpipe (aOR=1.51, 95% CI: 1.06-2.14, p=0.02), or cigarettes (aOR=1.24, 95% CI: 1.08-1.42, p<0.002) were more likely to be susceptible to cigarette smoking.

2)Use of other tobacco products, exposure to tobacco ads and susceptibility to cigarette smoking

Electronic cigarette users were 3.6 times more likely than non-users to be susceptible to cigarette smoking (aOR=3.6, 95% CI: 2.76-4.86, p<0.0001). Those reported exposure to pro-tobacco ads rarely/sometimes in newspapers/magazines (aOR=1.31, 95% CI: 1.08-1.57, p=0.005), or the internet (aOR=1.24, 95% CI: 1.01-1.54, p=0.04), and those with high receptivity to tobacco marketing (aOR=2.81, 95% CI: 2.40-3.30, p<0.0001) were more likely to be susceptible to cigarette smoking. Finally, the belief that people harm themselves a lot when they smoke somedays was protective against susceptibility to cigarettes (aOR=0.29, 95% CI: 0.16-0.54, p<0.0001).

3)Currentwaterpipeuse and curiosity about cigarette smoking

Current waterpipe smoking was not associated with curiosity about cigarettes (aOR=1.19, 95% CI: 0.83-1.70, p=0.33). A borderline association was observed between curiosity about cigarettes and living with someone who smokes waterpipe (aOR=1.37, 95% CI: 0.97-1.93, p=0.06) or cigarettes (aOR=1.15, 95% CI: 0.98-1.34, p=0.07).

4)Use of other tobacco products, exposure to tobacco ads and curiosity about cigarette smoking

Electronic cigarette users were more curious about cigarettes (aOR=2.18, 95% CI: 1.58-2.99, p<0.0001) than non-users, while smokeless tobacco users were less curious about cigarettes (aOR=0.44, 95% CI: 0.24-0.78, p=0.006). Higher odds of curiosity about cigarettes were associated with exposure to pro-tobacco ads in newspaper/magazines rarely/sometimes (aOR=1.23, 95% CI: 1.03-1.46, p=0.02), or most of the times/always (aOR=1.30, 95% CI: 1.03-1.65, p=0.03), exposure to pro-tobacco ads rarely/sometimes on the internet (aOR=1.28, 95% CI: 1.05-1.55, p=0.01), and high receptivity to tobacco marketing (aOR=2.18, 95% CI: 1.83-2.60, p<0.0001). The belief that all tobacco products are dangerous (aOR=0.80, 95% CI: 0.65-0.98, p=0.03), and that people harm themselves a lot when they smoke somedays (aOR=0.63, 95% CI: 0.42-0.942, p=0.02) were protective against curiosity about cigarettes. (Table 2).