Abstract
Introduction
There is very little insight into the psychosocial characteristics of young adults susceptible to waterpipe tobacco use and their reactions to messages about harms of waterpipe tobacco smoking (WTS). We investigated how young adults who were or were not susceptible to WTS differed on various characteristics and their reactions to messages about WTS harms.
Methods
Young adults ages 18 to 30 who had never used waterpipe tobacco were recruited through an online crowdsourcing site. Participants were stratified on susceptibility status (susceptible or not) and randomized to receive messages about harms and addictiveness of WTS or a control condition that received no messages. Participants’ perceptions of risk and worry, their attitudes toward, and willingness/curiosity to try WTS were assessed.
Results
Compared to nonsusceptible participants, susceptible participants perceived themselves to be at lower risk and worried less about harms and addictiveness of WTS, had more positive attitudes toward use, and expressed a greater willingness and curiosity to try it. Among susceptible participants, messages decreased willingness/curiosity to try WTS; messages had no effect on nonsusceptible participants. The message effects among susceptible participants were explained by more negative attitudes and less ambivalence toward WTS.
Conclusions
Susceptible young adults’ psychosocial characteristics place them at high risk for future uptake of WTS. Brief public health messages about harm and addiction may deter susceptible young adults’ willingness to try WTS and prevent WTS initiation and progression.
Implications
Findings suggest that in order to curb the initiation of WTS among susceptible young adults, interventions should target risk appraisals and attitudes toward WTS.
Introduction
A waterpipe is a nicotine delivery device in which tobacco smoke passes through water before it is inhaled. Epidemiological and laboratory data suggest that waterpipe smoke is no less dangerous than cigarette smoke.1,2 For example, waterpipe tobacco smoking (WTS) is related to cancer, poorer pulmonary function, and heart disease.1,3–6 Despite the health risks and addictive potential,7 WTS continues to spread in the United States among young adults.8–10 For example, in the 2012–2013 National Adult Tobacco Survey, prevalence of WTS was 18.2% among 18–24 year olds.11 Young adults are a particularly important group to target before the establishment of WTS sets in. Indeed, young adulthood remains a critical time in the development of tobacco use risk-taking behaviors,12 possibly owing to greater experimentation with alcohol and other substances and the lack of effective antismoking campaign messages for this audience,13 especially concerning WTS.
Efforts to curb the rise of WTS would benefit from interventions targeting young adults who do not use waterpipe tobacco yet and are susceptible. That is, persons who do not adamantly oppose WTS and may be open to WTS in the future. Indeed, in a prospective study that involved 964 university students who responded to a four-item susceptibility measure modified after the measure of Pierce and colleagues,14–16 nearly 27% were found susceptible to WTS.17 Of importance, those found susceptible had 2.5 the odds of trying waterpipe the following year compared to those who were not susceptible.17 Thus, the proportion of susceptible young adults at risk of WTS uptake is not trivial, merits investigating, and requires intervention.
Intervention processes and contents to derail susceptible individuals from WTS are unknown, partly due to the very limited understanding of what psychosocial characteristics (eg, knowledge, attitudes, perceived risk, social acceptability, efficacy beliefs) potentiate WTS among these individuals versus those who are not susceptible. Two studies provide some insight. Barnett and colleagues18 assessed WTS susceptibility among students at one university (N = 852), using the question: “Do you intend to smoke tobacco from a hookah sometime in the rest of your life?” Students were classified as susceptible if they responded with anything other than “definitely no.” Capturing the constructs from the Theory of Reasoned Action,19 they found that greater susceptibility (ie, WTS intentions) was associated with positive attitudes and the normative belief that “hookah is socially acceptable”; lower susceptibility was associated with beliefs that hookah is addicting and harmful. Further, Nuzzo and colleagues (using the same sample) reported the relationship between factual knowledge of the harms of WTS and susceptibility.20 While susceptibility was unrelated to overall knowledge, it was associated with responding with a “don’t know” to several of the knowledge items (eg, “which has more nicotine?”). Overall, the extant literature suggests that individuals susceptible to WTS possess more positive and less negative attitudes toward the product and lack information about harms and addictiveness. As addressed in this paper, confidence in these results would be strengthened by generalizing findings to populations not limited to college students, using a more robust multi-item susceptibility measure,17 and conducting controlled experiments to isolate constructs that potentiate WTS use between individuals who are and are not susceptible.
Individuals susceptible to WTS may lack information about its harms and addictiveness. Limited knowledge may translate into lower perceived risk and worry about harms and addiction—constructs deemed important in predicting intentions and behavior change.21 For example, among college and noncollege WTS users, brief public health messages about harms of waterpipe have been shown to influence perceptions of risk and worry about harms of WTS and increase intentions to quit.22,23 Among susceptible individuals, these messages should have similar effects. Further, messages should lower willingness and curiosity to try WTS in the future. Whether in this population such messages influence overall risk appraisals (eg, perceptions of risk and worry) and attitudes toward WTS remains to be seen. We tested if the above-hypothesized effects were supported.
In sum, this study consists of young adults who have never used waterpipe tobacco and are or are not susceptible to WTS, to examine two critical areas in need of data. First, we examine, utilizing the Theory of Reasoned Action and work on risk appraisals, which constructs (ie, attitudes, norms, perceived risk, and worry) differentiate between young adults susceptible or not to WTS. Second, we compare the extent to which a brief public health messaging intervention on harms and addictiveness of WTS versus no intervention influences willingness/curiosity to try WTS in the future and their psychosocial determinants. The main hypotheses are as follows:
Hypothesis 1
Compared to individuals not susceptible to WTS, susceptible individuals will: (1) have lower perceived risk and worry about the harms of waterpipe and of becoming addicted (ie, risk appraisals); (2) have a more positive global attitude towards WTS; (3) perceive greater social acceptance of WTS; and (4) express a stronger willingness/curiosity to try WTS in the future.
Hypothesis 2
Interactions will occur between susceptibility status (susceptible or not) and receipt of messages about harms and addictive potential of WTS (no/yes). Among nonsusceptible individuals, we expect the intervention will not influence any outcome. Among the susceptible, we expect the intervention will result in heightened perceptions of risk and worry and a lower willingness to try WTS compared to the control.
Methods
Setting and Sample
Participants were recruited in December 2015 through Amazon Mechanical Turk (AMT), a crowdsourcing Internet marketplace designed to efficiently gather data from a large group of respondents.24 Studies demonstrate AMT’s validity for behavioral and consumer product research, including studies of tobacco use23,25,26 and other behaviors.27 Using AMT, researchers collect data using human intelligence tasks, which interested AMT members can choose to complete. The study task involved completing an online survey with exposure to WTS messages based on the experimental conditions described below.
Following a brief description of the study, AMT members residing in the United States who were interested in participating reviewed a complete study description with a link to the online consent form and eligibility screener. Individuals between 18 and 30 years of age who reported never using waterpipe tobacco were eligible to participate. Age and waterpipe tobacco use eligibility criteria were assessed using screening questions derived from population-based surveys11 and previous research on waterpipe tobacco use.17,22,23 To ensure approximately equal numbers of susceptible and nonsusceptible participants randomized to the experimental conditions, recruitment was stratified into two tasks that were identical but with eligibility based on susceptibility status (susceptible and nonsusceptible, respectively), in addition to the age and waterpipe tobacco use criteria.25 Thus, after answering screening items about age and waterpipe tobacco use, potential participants completed a four-item measure of susceptibility to WTS.17 These four questions were: Do you think that you will smoke tobacco from a waterpipe soon? Do you think that you will smoke tobacco from a waterpipe in the next year? Do you think that in the future you might experiment with waterpipe tobacco smoking? If one of your best friends asked you to smoke tobacco from a waterpipe, would you? Response options included: Definitely yes; Probably Yes; Definitely no; Probably No. Participants were considered susceptible if they gave a response other than Definitely No to any item. Participants responding Definitely No to all items were considered to be nonsusceptible.
Eligible, consenting AMT members proceeded to the online experiment. Participants completing all study procedures were provided a small monetary credit through AMT. The Georgetown University Institutional Review Board approved the study protocol.
Procedures
Prior to randomization participants answered initial questions about demographics and tobacco use behaviors. Then, based on an algorithm within the online survey, within each susceptibility stratum (susceptible or not) participants were randomized in approximately equal numbers to either the control or experimental condition. Participants randomized to the control condition completed measures only and did not view any messages on WTS. Participants in the experimental condition viewed online messages about the harms and addictive potential of WTS. Messages were presented within the online survey as slides that participants self-advanced. Participants viewed a total of six slides that, other than the title and reference slides, described: (1) potential health harms of WTS and levels of harmful chemicals contained in waterpipe tobacco smoke relative to cigarettes, and (2) nicotine levels from WTS that can lead to addiction and described research evidence that users indicate they are hooked on WTS. Messages were used in a prior trial conducted using AMT involving young adult waterpipe users,23 and shown to increase knowledge of harms, perceived risk and worry, and desire to quit WTS. Participants were instructed to read the slides carefully and allowed to view the content for as long as they wished.
Measures
Demographics
Demographic characteristics assessed included age, gender, race/ethnicity, whether participants were current college/university students, educational attainment, and employment status.
Other Tobacco Product Use
Cigarette smoking was assessed using epidemiological survey items, with current smokers defined as those who had smoked ≥100 lifetime cigarettes and now smoke every day or some days.11 Among current smokers, a single item was used to assess the number of cigarettes smoked per day. Participants also reported past month use of large cigars, little cigars/cigarillos, smokeless tobacco, and electronic cigarettes based on a single item for each product.11 We considered noncigarette tobacco use to include use of cigars, cigarillos, smokeless tobacco, and electronic cigarettes.
Outcome Variables
Perceived Relative Harm and Addictiveness.
To assess degree of harm and addictiveness of WTS relative to cigarettes, we posed two questions: “Compared to regular cigarettes how harmful do you think waterpipe tobacco use is?” (1 = much less harmful to 5 = much more harmful); and “Compared to regular cigarettes, how addictive do you think waterpipe tobacco use is?” (1 = much less addictive to 5 = much more addictive). The two items were summed and averaged (r = 0.58, p <.001).
Risk Appraisals (ie, Worry and Perceived Risk).
Perceived risk of WTS was assessed by asking “What do you think is your chance of getting a serious smoking-related disease, such as cancer, lung disease, or heart disease, if you were to smoke waterpipe tobacco?” (1 = no chance to 7 = certain to happen). Worry about harm was assessed by asking “How worried would you be about getting a serious smoking-related disease, such as cancer, lung disease, or heart disease, if you were to smoke waterpipe tobacco?” (1 = not at all to 7 = very). Perceived risk of becoming addicted was assessed by “What do you think is your chance of becoming addicted to nicotine in tobacco from waterpipe if you were to smoke it?” (1 = no chance to 7 = certain to happen). Worry about becoming addicted was measured with “How worried would you be about becoming addicted to nicotine in waterpipe if you were to smoke it?” (1 = not at all to 7 = very). A factor analysis revealed that all four items loaded on the same factor (alpha = 0.82). Hence, the four items were summed and averaged.
Attitudes Toward WTS.
A global evaluation of WTS was assessed using four bipolar items: positive/negative, like/dislike, good/bad, desirable/undesirable (scored 3 to −3).28 Items were summed and averaged (alpha = 0.94). A negative score represents a stronger negative evaluation of WTS.
We also assessed felt attitudinal ambivalence towards WTS using three items that captured having mixed feelings, felting torn, and conflicted about WTS. Response anchors were from 1 = strongly disagree to 7 = strongly agree. Items were summed and averaged (alpha = 0.93).
Social Acceptance of WTS.
This was assessed by, “How socially acceptable is tobacco waterpipe smoking?” (1 = not acceptable to 7 = acceptable).
Willingness/Curiosity to Try Waterpipe.
We created a measure that captured the essence of what is means to be susceptible, that is, a willingness to try tobacco, that included an assessment of curiosity because adding this concept to existing susceptibility measures improves predicting tobacco use.29 Thus, the willingness/curiosity measure was composed of four items: (1) “How tempted are you to try waterpipe tobacco smoking within the next year, just for the experience?” (1 = not at all to 7 = very tempted); (2) “How curious are you about trying waterpipe tobacco smoking” (1 = not at all to 7 = very curious); (3) “Do you see yourself smoking tobacco from a waterpipe within the next year, just to see how it’s like?” (1 = definitely not to 7 = definitely yes); and (4) “If you were with friends and they offered you to smoke waterpipe within the next year, how likely is it that you would try it, even a puff?” (1 = no chance to 7 = certain to happen). All items loaded on a single factor (alpha = 0.94); they were summed and averaged.
Evaluation of the Messages.
Participants who received the messages were asked the extent to which they agreed the information on the slides were relevant and believable (1 = strongly disagree to 7 = strongly agree). Further, to examine engagement, they were asked how much mental time and energy they placed on reading the information (1 = very little to 7 = a lot).
Statistical Analysis
Bivariate analyses confirmed that no participant characteristics differed at p <.05 between the experimental and control conditions, thus none were included in multivariable analyses. However, given the prevalence of cigarette smoking in the sample and the potential influence on susceptibility to WTS,17 cigarette smoking status was included as a covariate in all models. Analysis of covariance was used to determine whether outcomes differed by susceptibility status, study condition, and their interaction. Least-square means were examined for significant main and interaction effects using Tukey’s adjustment for multiple comparisons.
Mediation analyses were conducted to examine whether observed differences by study condition in willingness/curiosity to use waterpipe tobacco in the future could be explained by risk appraisals, social acceptability, attitudes, and ambivalence toward WTS. These analyses estimated the direct and indirect effects of experimental condition on curiosity/willingness to use waterpipe tobacco among those who were susceptible to WTS, following recommended methods.30 Indirect effects were estimated using a bias-corrected bootstrapping approach with 1000 resamples to address non-normality in the product of coefficients.31 Asymmetric 95% confidence intervals (CIs) around estimates that do not include zero indicate statistically significant indirect effects.31 Mediation analyses were conducted using Mplus 7.1 (Los Angeles, CA); all other analyses were conducted using SAS 9.3 (Cary, NC).
Results
Sample Characteristics
Overall, 4129 AMT members provided a valid response to eligibility screening questions, 508 (12%) of whom were eligible and completed the study. Table 1 displays the sample characteristics. On average, participants took 7.0 mins to complete the study procedures (standard deviation (SD) = 6.4, Median 5.5); time to complete procedures was significantly greater for participants in the experimental condition (mean (M) = 8.1, SD = 7.0, Median 6.6) than participants in the control condition (M = 6.0, SD = 5.6, Median 4.7, p < .001). Participants averaged 25.3 (SD = 3.2) years of age, about half were male (49.9%), most were White (80.1%), and most completed at least some college education (86.8%); about 41% were students. With respect to other tobacco products, close to 20% were current cigarette smokers (M = 7.1 cigarettes/day, SD = 6.7), while about 12% used another tobacco product. We obtained roughly an equal number of susceptibility and nonsusceptible participants. There was between 120 and 140 participants per cell.
Table 1.
Sample Characteristics
| Full sample (N = 508) | Control condition (N = 267) | Education condition (N = 241) | p | |
|---|---|---|---|---|
| Demographics | ||||
| Age (M, SD) | 25.3 (3.2) | 25.4 (3.2) | 25.2 (3.1) | .600 |
| Gender | .826 | |||
| Male | 253 (49.9) | 132 (49.4) | 121 (50.4) | |
| Female | 254 (50.1) | 135 (50.6) | 119 (46.6) | |
| Race | .971 | |||
| Black/African American | 37 (7.3) | 19 (7.1) | 31 (12.9) | |
| White | 407 (80.1) | 215 (80.5) | 192 (79.7) | |
| Other | 64 (12.6) | 33 (12.4) | 31 (12.9) | |
| Hispanic ethnicity | .956 | |||
| Yes | 46 (9.1) | 24 (9.0) | 22 (9.1) | |
| No | 462 (90.9) | 243 (91.0) | 219 (90.9) | |
| Current student | 206 (40.8) | 108 (40.9) | 98 (40.7) | .955 |
| Nonstudent | 299 (59.2) | 156 (59.1) | 143 (59.3) | |
| Education | .750 | |||
| Less than college education | 67 (13.2) | 34 (6.7) | 33 (6.5) | |
| Some college education or higher | 441 (86.8) | 233 (87.3) | 208 (86.3) | |
| Employment | .138 | |||
| Not full-time employed | 281 (55.3) | 156 (58.4) | 125 (51.9) | |
| Full-time employed | 227 (44.7) | 111 (41.6) | 116 (48.1) | |
| Cigarette smoking | .083 | |||
| Nonsmoker | 349 (69.0) | 191 (71.5) | 158 (66.1) | |
| Former smoker/experimenter | 57 (11.3) | 33 (12.4) | 24 (10.0) | |
| Current smoker | 100 (19.7) | 43 (16.1) | 57 (23.9) | |
| Cigarettes/day—current smokers (M, SD) | 7.4 (6.1) | 6.8 (5.8) | 7.8 (6.3) | .386 |
| Past month noncigarette tobacco use | .074 | |||
| Yes | 62 (12.2) | 26 (9.7) | 36 (14.9) | |
| No | 446 (87.8) | 241 (90.3) | 205 (85.1) | |
| Susceptibility to waterpipe tobacco use | .552 | |||
| Susceptible | 248 (48.8) | 127 (47.6) | 121 (50.2) | |
| Not susceptible | 260 (51.2) | 140 (52.4) | 120 (49.8) |
Data display N and % unless otherwise indicated. Noncigarette tobacco use includes use of cigars, cigarillos, smokeless tobacco, and electronic cigarettes. Some totals do not add to the sample n or n within condition due to sporadic missing data (<1% of cases for any variable). M = mean, SD = standard deviation.
There were no differences in participant characteristics across experimental conditions, indicating successful randomization. However, compared to nonsusceptible participants, those who were susceptible were more likely to smoke cigarettes (33.7% vs. 6.5%, p < .001) and use other combustible and noncombustible tobacco products (19.4% vs. 5.4%, p < .001). Given that current (35.0%) and former (24.6%) smokers were also more likely than nonsmokers (3.7%) to use other tobacco products (p < .001), we covaried smoking status in multivariable analyses as noted above.
Evaluation of Messages
Participants who received the messages found the messages to be somewhat relevant (M = 4.1, SE = 0.12), believable (M = 5.9, SE = 0.07), and spent a fair amount of mental energy reviewing the messages (M = 5.6, SE = 08). There were no differences by susceptibility status expect for relevance. Participants who were susceptible viewed the information as more relevant than nonsusceptible participants (M = 4.7, SE = 0.19 vs. M = 3.8, SE = 0.21, p < .001).
Psychosocial Characteristics by Susceptibility Status
Item means between participants who were susceptible and not susceptible to waterpipe on the psychosocial characteristics are presented in Table 2. Consistent with the first set of hypotheses, compared to the not susceptible, susceptible participants: (1) viewed WTS are less harmful and addictive than cigarettes; (2) perceived themselves to be at lower risk and worried less about harms of and becoming addicted to waterpipe; (3) had a more positive, albeit still negative, overall attitude towards WTS; (4) felt more ambivalent; (5) deemed the product more socially acceptable; and (6) expressed greater willingness/curiosity to try waterpipe.
Table 2.
Results for Main Effects for Education Condition and Susceptibility on Study Outcomes
| Psychosocial variable | Education condition | Susceptibility | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| F | p | η 2 | Control M (SE) | Education M (SE) | F | p | η 2 | Susceptible M (SE) | Not susceptible M (SE) | |
| Relative harm and addictiveness of waterpipe vs. cigarettes | 122.8 | <.001 | 0.20 | 2.7 (0.05) | 3.4 (0.05) | 5.4 | .020 | 0.01 | 3.0 (0.04) | 3.1 (0.05) |
| Risk appraisals (perceived risk and worry) | 16.3 | <.001 | 0.03 | 4.6 (0.08) | 5.0 (0.09) | 32.9 | <.001 | 0.06 | 4.5 (0.08) | 5.1 (0.09) |
| Global attitude | 19.3 | <.001 | 0.04 | −1.0 (0.10) | −1.5 (0.11) | 46.3 | <.001 | 0.08 | −0.8 (0.10) | −1.7 (0.11) |
| Ambivalence | 11.0 | .001 | 0.02 | 2.7 (0.09) | 2.4 (0.09) | 87.5 | <.001 | 0.15 | 3.1 (0.09) | 2.0 (0.10) |
| Socially acceptable | 0.00 | .960 | <0.01 | 4.1 (0.12) | 4.1 (0.13) | 20.1 | <.001 | 0.04 | 4.4 (0.12) | 3.7 (0.14) |
| Curiosity/willingness | 3.6 | .057 | <0.01 | 2.5 (0.08) | 2.3 (0.09) | 178.3 | <.001 | 0.26 | 3.1 (0.08) | 1.6 (0.09) |
Effects of Messages
We predicted that among nonsusceptible participants, messages about harm and addiction would have little to no effects on the psychosocial outcomes; messages would have effects on susceptible participants. Table 2 presents the main effects comparing the control versus experimental condition on the psychosocial characteristics. Table 3 displays results of the interaction between experimental condition and susceptibility status.
Table 3.
Results for Interaction Effects Between Education Condition and Susceptibility on Study Outcomes
| Psychosocial variable | Control | Education | |||||
|---|---|---|---|---|---|---|---|
| F | p | η 2 | Susceptible M (SE) | Not susceptible M (SE) | Susceptible M (SE) | Not susceptible M (SE) | |
| Relative harm and addictiveness of waterpipe vs. cigarettes | 0.28 | .597 | <0.01 | 2.6 (0.06) | 2.8 (0.07) | 3.3 (0.06) | 3.4 (0.07) |
| Risk appraisals (perceived risk and worry) | 0.59 | .442 | <0.01 | 4.2 (0.11) | 4.9 (0.12) | 4.7 (0.11) | 5.3 (0.11) |
| Global attitude | 0.11 | .107 | <0.01 | −0.41 (0.13) | −1.5 (0.14) | −1.2 (0.14) | −1.9 (0.14) |
| Ambivalence | 0.33 | .569 | <0.01 | 3.3 (0.11) | 2.1 (0.13) | 2.9 (0.12) | 1.8 (0.13) |
| Socially acceptable | 0.09 | .763 | <0.01 | 4.4 (0.16) | 3.7 (0.17) | 4.4 (0.16) | 3.7 (0.17) |
| Curiosity/willingness | 4.5 | .035 | 0.01 | 3.3 (0.10) | 1.6 (0.11) | 2.9 (0.11) | 1.6 (0.12) |
Cigarette smoking status (nonsmoker, experimenter/former smoker, current smoker) included as a covariate in all models.
Providing educational messages versus the lack thereof had several main effects. Compared to participants who did not receive any messages, those who did were: (1) less likely to view WTS as less harmful and addictive than cigarettes; (2) rated themselves at higher risk and reported more worry about the harms and of becoming addicted; and (3) reported a stronger negative attitude towards WTS (see Table 2). There were no differences for willingness/curiosity to try waterpipe.
The interaction between messaging condition and susceptibility status revealed only one significant finding. Susceptible participants who received the educational messages reported lower willingness/curiosity to try waterpipe compared to susceptible participants who did not receive any messages. Among nonsusceptible participants, messages produced no significant effects. Given these findings, we explored which psychosocial characteristics may explain the results via mediational analysis.
Mediational Analysis
Results of the mediation analysis among susceptible participants are shown in Table 4. The direct effect of study condition among susceptible participants was no longer significant. However, there were statistically significant indirect effects; the observed differences in curiosity/willingness to try WTS among susceptible participants based on study condition were explained by more negative attitudes (β = −0.33, 95% CI = −0.60, −0.16) and lower ambivalence (β = −0.06, 95% CI = −0.18, −0.01) toward WTS among susceptible participants in the experimental condition versus susceptible participants in the control condition.
Table 4.
Mediation Analysis of Intervention Effects on Willingness/Curiosity to Try WTS in the Future Among Individuals Susceptible to WTS
| Independent variable | Mediating variable | Β (95% CI) | |
|---|---|---|---|
| Direct effect | Education vs. control | — | 0.08 (−0.27, 0.41) |
| Indirect effects | Education vs. control | General perceptions | 0.01 (−0.15, 0.18) |
| Risk appraisals | −0.04 (−0.14, 0.03) | ||
| Attitudes | −0.33 (−0.60, −0.16) | ||
| Ambivalence | −0.06 (−0.18, −0.01) | ||
| Social acceptability | 0.00 (−0.02, 0.03) |
Regression coefficients and 95% CIs are displayed. CIs that do not include 0 are statistically significant at p <0.05 and are in bold text. Cigarette smoking status (nonsmoker, experimenter/former smoker, current smoker) included as a covariate. CI = confidence interval.
Discussion
In this study, we addressed two issues that have received very little attention in the waterpipe tobacco literature. The first issue is characterizing young adults who are susceptible to WTS outside of college settings—importantly, our results were not moderated by level of education or being in college or not. Second, to the best of our knowledge, we examined for the first time effects of a risk communication intervention to modify risk appraisals, attitudes, and beliefs that may result in WTS among susceptible young adults. Results point to several psychosocial characteristics that place susceptible versus nonsusceptible young adults at higher risk for WTS uptake, including: underappreciating the harms and addictive potential of waterpipe (ie, risk appraisals); less negative evaluations of waterpipe (ie global attitude); and importantly, a stronger willingness/curiosity to try WTS. Viewing WTS as safer than cigarettes, and more positive and less negative attitudes toward WTS have been linked empirically with WTS.8,32,33 These findings point to crucial factors to target with public health efforts to curb WTS uptake among susceptible young adults.
In this spirit, we examined how messages about harm and addiction of WTS influence at the very minimum risk appraisals. The decision to use messages about harm and addiction is based on findings that college students who are susceptible more often report not knowing facts about product harms than students who are not susceptible.20 As predicted, messages were found effective in heightening perceptions of risk and worry (ie, risk appraisals) as well as producing more negative evaluations of WTS. Importantly, these messages lowered willingness/curiosity to try WTS among participants deemed susceptible only. This effect was mediated by more negative attitudes and less ambivalence toward waterpipe tobacco, the latter interpreted to mean participants felt less conflicted about not trying WTS.
Study findings have potential implications for public health efforts to prevent WTS among young adults, a group where the prevalence of waterpipe tobacco use is at its highest at the population level. For example, under the Family Smoking Prevention and Tobacco Control Act, the Food and Drug Administration (FDA) is authorized to regulate tobacco products and is engaged in public education about the potential harms and addictiveness of tobacco product use.34 Ongoing public education efforts include the FDA’s “The Real Cost” mass media campaign targeting youths who are susceptible to cigarette smoking with persuasive smoking prevention messaging conveying the potential harms and addictiveness of cigarettes. The FDA’s enacted new “deeming rule” brings waterpipe tobacco and other unregulated products under the agency’s regulatory authority.35 The rule now positions the FDA to engage in public health messaging about the potential harms and addictiveness of WTS. Our data suggest such messaging targeting susceptible young adults and conveying the harmful and addictive nature of WTS may have an impact for preventing initiation in this vulnerable group.
There are several study limitations. First, results of AMT users may not be representative of the general population. Indeed, our sample was mostly White and highly educated. The methods used to recruit an appropriate sample of susceptible and nonsusceptible participants also required us to screen a larger number of potential participants and led to a somewhat lower eligibility rate than similar studies conducted using AMT.23,25 Second, our risk perception items did not specify frequency of waterpipe use or length of time (eg, lifetime risk) for occurrence of being harmed or of becoming addicted. Third, we did not assess uptake of WTS prospectively. One can question whether such a brief intervention would have any long-term effects. What may be required to buffer against uptake is delivering booster messages. Fourth, the messages only focused on harm and addiction. In order to obtain at least moderate effect sizes, interventions will likely need to target several constructs, as suggested herein. However, to find that such a minimal intervention can have any effect, admittedly small, is itself promising and suggests that these messages could be a useful adjunct to larger multicomponent messaging interventions. Indeed, the present effects closely mirror those found in a prior study with waterpipe users that employed the identical messages.23 As such, these risk messages may have significant effects across populations and WTS status (ie, current users, susceptible nonusers).
Despite these limitations, findings reinforce which variables may need to be targeted to reduce uptake of WTS among susceptible young adults. Among these, we present the first evidence to suggest that brief risk messages about harm and addictiveness can reduce willingness/curiosity to try WTS, a concept akin to and incorporated in susceptibility measures.29 Future experiments should investigate whether these risk messages, with or without targeting other constructs (eg, social acceptance, attitudinal ambivalence), deters the uptake of WTS in this high-risk population.
Funding
Preparation of this publication was supported in part by the National Institutes of Health (NIH) and the Food and Drug Administration (FDA) Center for Tobacco Products (CTP) under NIH grant number (Mays, PI) K07CA172217. This work was also supported in part by the Georgetown Lombardi Comprehensive Cancer Center Support grant number P30CA051008. The study sponsors had no role in the study design; in the collection, analysis, and interpretation data; in the writing of the report; and in the decision to submit the paper for publication. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the FDA.
Declaration of Interests
None declared.
References
- 1. Akl EA, Gaddam S, Gunukula SK, et al. The effects of waterpipe tobacco smoking on health outcomes: a systematic review. Int J Epidemiol. 2010;39(3):834–857. [DOI] [PubMed] [Google Scholar]
- 2. Waziry R, Jawad M, Ballout RA, Al Akel M, Akl EA. The effects of waterpipe tobacco smoking on health outcomes: an updated systematic review and meta-analysis. Int J Epidemiol. 2016. [DOI] [PubMed] [Google Scholar]
- 3. Cobb C, Ward KD, Maziak W, Shihadeh AL, Eissenberg T. Waterpipe tobacco smoking: an emerging health crisis in the United States. Am J Health Behav. 2010;34(3):275–285. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Knishkowy B, Amitai Y. Water-pipe (narghile) smoking: an emerging health risk behavior. Pediatrics. 2005;116(1):e113–e119. [DOI] [PubMed] [Google Scholar]
- 5. Raad D, Gaddam S, Schunemann HJ, et al. Effects of water-pipe smoking on lung function: a systematic review and meta-analysis. Chest. 2011;139(4):764–774. [DOI] [PubMed] [Google Scholar]
- 6. Sibai AM, Tohme RA, Almedawar MM, et al. Lifetime cumulative exposure to waterpipe smoking is associated with coronary artery disease. Atherosclerosis. 2014;234(2):454–460. [DOI] [PubMed] [Google Scholar]
- 7. Aboaziza E, Eissenberg T. Waterpipe tobacco smoking: what is the evidence that it supports nicotine/tobacco dependence? Tob Control. 2015;24(suppl 1):i44–i53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Grekin ER, Ayna D. Waterpipe smoking among college students in the United States: a review of the literature. J Am Coll Health. 2012;60(3):244–249. [DOI] [PubMed] [Google Scholar]
- 9. Primack BA, Shensa A, Kim KH, et al. Waterpipe smoking among U.S. university students. Nicotine Tob Res. 2013;15(1):29–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Soule EK, Lipato T, Eissenberg T. Waterpipe tobacco smoking: a new smoking epidemic among the young? Curr Pulmonol Rep. 2015;4(4):163–172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Agaku IT, King BA, Husten CG, et al. ; Centers for Disease Control and Prevention (CDC) Tobacco product use among adults–United States, 2012-2013. MMWR Morb Mortal Wkly Rep. 2014;63(25):542–547. [PMC free article] [PubMed] [Google Scholar]
- 12. Terry-McElrath YM, O’Malley PM. Trends and timing of cigarette smoking uptake among US young adults: survival analysis using annual national cohorts from 1976 to 2005. Addiction. 2015;110(7):1171–1181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Tercyak KP, Rodriguez D, Audrain-McGovern J. High school seniors’ smoking initiation and progression 1 year after graduation. Am J Public Health. 2007;97(8):1397–1398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Pierce JP. N Measures Guide for Youth Tobacco Research: Susceptibility to Smoking. 2012; cancercontrol cancer.gov/brp/tcrb/susceptibility.html. Accessed June 12, 2014.
- 15. Pierce JP, Choi WS, Gilpin EA, Farkas AJ, Merritt RK. Validation of susceptibility as a predictor of which adolescents take up smoking in the United States. Health Psychol. 1996;15(5):355–361. [DOI] [PubMed] [Google Scholar]
- 16. Pierce JP, Farkas AJ, Evans N, Gilpin E. An improved surveillance measure for adolescent smoking? Tob Control. 1995;4:S47–S56. [Google Scholar]
- 17. Lipkus IM, Reboussin BA, Wolfson M, Sutfin EL. Assessing and predicting susceptibility to waterpipe tobacco use among college students. Nicotine Tob Res. 2015;17(9):1120–1125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Barnett TE, Shensa A, Kim KH, et al. The predictive utility of attitudes toward hookah tobacco smoking. Am J Health Behav. 2013;37(4):433–439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Ajzen I, Fishbein M. Understanding Attitudes and Predicting Social Behavior. 3rd ed. Englewood Cliffs, NJ: Prentice-Hall; 1980. [Google Scholar]
- 20. Nuzzo E, Shensa A, Kim KH, et al. Associations between hookah tobacco smoking knowledge and hookah smoking behavior among US college students. Health Educ Res. 2013;28(1):92–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Sheeran P, Harris PR, Epton T. Does heightening risk appraisals change people’s intentions and behavior? A meta-analysis of experimental studies. Psychol Bull. 2014;140(2):511–543. [DOI] [PubMed] [Google Scholar]
- 22. Lipkus IM, Eissenberg T, Schwartz-Bloom RD, Prokhorov AV, Levy J. Affecting perceptions of harm and addiction among college waterpipe tobacco smokers. Nicotine Tob Res. 2011;13(7):599–610. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Mays D, Tercyak KP, Lipkus IM. The effects of brief waterpipe tobacco use harm and addiction education messages among young adult waterpipe tobacco users. Nicotine Tob Res. 2016;18(5):777–784. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Brabham DC, Ribisl KM, Kirchner TR, Bernhardt JM. Crowdsourcing applications for public health. Am J Prev Med. 2014;46(2):179–187. [DOI] [PubMed] [Google Scholar]
- 25. Mays D, Moran MB, Levy DT, Niaura RS. The impact of health warning labels for Swedish Snus Advertisements on young adults’ snus perceptions and behavioral intentions. Nicotine Tob Res. 2016;18(5):1371–1375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Hall MG, Ribisl KM, Brewer NT. Smokers’ and nonsmokers’ beliefs about harmful tobacco constituents: implications for FDA communication efforts. Nicotine Tob Res. 2014;16(3):343–350. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Mays D, Tercyak KP. Framing indoor tanning warning messages to reduce skin cancer risks among young women: implications for research and policy. Am J Public Health. 2015;105(8):e70–e76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Crites SL, Fabrigar LR, Petty RE. Measuring the affective and cognitive properties of attitudes: conceptual and methodological issues. Pers Soc Psychol Bull. 1994;20(6):619–634. [Google Scholar]
- 29. Strong DR, Hartman SJ, Nodora J, et al. Predictive validity of the expanded susceptibility to smoke index. Nicotine Tob Res. 2015;17(7):862–869. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Hayes AF, Preacher KJ. Statistical mediation analysis with a multicategorical independent variable. Br J Math Stat Psychol. 2014;67(3):451–470. [DOI] [PubMed] [Google Scholar]
- 31. Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods. 2008;40(3):879–891. [DOI] [PubMed] [Google Scholar]
- 32. Akl EA, Jawad M, Lam WY, et al. Motives, beliefs and attitudes towards waterpipe tobacco smoking: a systematic review. Harm Reduct J. 2013;10:12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Akl EA, Ward KD, Bteddini D, et al. The allure of the waterpipe: a narrative review of factors affecting the epidemic rise in waterpipe smoking among young persons globally. Tob Control. 2015;24(suppl 1):i13–i21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Husten CG, Deyton LR. Understanding the Tobacco Control Act: efforts by the US Food and Drug Administration to make tobacco-related morbidity and mortality part of the USA’s past, not its future. Lancet. 2013;381(9877):1570–1580. [DOI] [PubMed] [Google Scholar]
- 35. FDA Administration. Deeming Tobacco Products To Be Subject to the Federal Food, Drug, and Cosmetic Act, as Amended by the Family Smoking Prevention and Tobacco Control Act; Restrictions on the Sale and Distribution of Tobacco Products and Required Warning Statements for Tobacco Products 2016. http://federalregister.gov/a/2016–10685. Accessed October 5, 2016.