The effect of e-cigarette warning labels on college

Addictive Behaviors 76 (2018) 106–112

Contents lists available at ScienceDirect

Addictive Behaviors journal homepage: www.elsevier.com/locate/addictbeh

The effect of e-cigarette warning labels on college students' perception of ecigarettes and intention to use e-cigarettes

MARK

Hsiao-Yun Lee1, Hsien-Chang Lin⁎,1, Dong-Chul Seo1, David K. Lohrmann1 Department of Applied Health Science, School of Public Health, Indiana University, Bloomington, IN 47405, USA

H I G H L I G H T S warning label proposed by FDA was more effective than it created by companies. • The warning label reduced e-cigarette use intention by increasing risk perception. • FDA • Companies' own label had lower readability and comprehensibility.

A R T I C L E I N F O

A B S T R A C T

Keywords: E-cigarette Warning label Theory of Planned Behavior Structural equation modeling

Objective: This study examined the effect of two e-cigarette warning labels on college students' perceived advantages and risks of e-cigarette use, as well as students' intentions to use e-cigarettes. The company-produced ecigarette warning label carries abundant information with small font size while the governmental warning label has only two sentences presented in large font size. The effect of both labels have not yet been examined and verified. Methods: Data were collected in October 2015 from college students at a Midwestern university. A pretestposttest design was employed with 338 students exposed to the warning label proposed by the FDA and 328 students exposed to the label created by e-cigarette companies. Structural equation modeling analysis was implemented to examine the effect of warning labels with the analytical model grounded in the Theory of Planned Behavior. Results: Findings showed that college students' perceived advantages of e-cigarette use were positively related to their intentions to use e-cigarettes, while perceived risks were negatively associated with their intentions. When comparing two labels, the governmental label was found to reduce college students' intentions to use e-cigarettes via increasing perceived risks of e-cigarette use (β = 0.10, p < 0.05), however, not via decreasing perceived advantages of e-cigarette use. The warning label currently used by e-cigarette companies showed no influence on beliefs about or intentions to use e-cigarettes. Conclusions: The warning label proposed by the FDA is more effective than that created by e-cigarette companies, however, has room for improvement to make a greater impact on e-cigarette use intention.

1. Introduction The electronic nicotine delivery system (also known as the e-cigarette) entered the United States (US) in 2006; (Dockrell, Morrison, Bauld, & McNeill, 2013) however, it has not yet been fully regulated by the government. Because of this lack of regulation, e-cigarettes were heavily advertised in media, leading to the high awareness of e-cigarette (Kim, Arnold, & Makarenko, 2014; King, Patel, Nguyen, & Dube, 2015). According to the results of previous studies, it is estimated that > 70% of US adults were aware of e-cigarettes by 2012 (Adkison,

⁎

1

Corresponding author. E-mail address: [email protected] (H.-C. Lin). Current address: 1025 E. 7th Street, SPH 116, Bloomington, IN 47405-7129, USA.

http://dx.doi.org/10.1016/j.addbeh.2017.07.033 Received 20 March 2017; Received in revised form 20 July 2017; Accepted 26 July 2017 Available online 28 July 2017 0306-4603/ © 2017 Elsevier Ltd. All rights reserved.

O'Connor, Bansal-Travers, et al., 2013), while about 14% used e-cigarettes between 2012 and 2013 (Agaku, King, Husten, et al., 2014). The emergence of the e-cigarette has aroused a rigorous debate over its health effects. Supporters believe that the e-cigarette is a safer alternative to smoking (Barbeau, Burda, & Siegel, 2013; Dawkins, Turner, Roberts, & Soar, 2013), while opponents are concerned that e-cigarettes will re-normalize smoking behavior (Fairchild, Bayer, & Colgrove, 2014; Hajek, Etter, Benowitz, Eissenberg, & McRobbie, 2014). Even though the long-term health effects of e-cigarette use have not yet been observed, giving an early warning to the public about the potential risks


H.-Y. Lee et al.

proposal, the warning label should be addressed as “WARNING: This product contains nicotine derived from tobacco. Nicotine is an addictive chemical.” and has to cover at least 30% of the package with contrasting colors and maximum font size (as shown in Fig. 1b) (US Food and Drug Administration, 2014). Nonetheless, this regulation has not yet been passed, and the anticipated effect of this label on consumers' e-cigarette related beliefs and behaviors is unknown. Although tobacco companies claimed that their warning labels is a way to communicate with consumers honestly regarding possible adverse health effects (Richtel, 2014), without examining the effect, their influence on consumers' beliefs and behaviors of e-cigarette use remains unknown. On the other hand, although the e-cigarette warning label created by FDA was based on current mandatory tobacco warning label design, there is no direct evidence showing this label to be as well effective on e-cigarettes. This study aimed to investigate the effect of e-cigarette warning labels on college students' beliefs about and intention to use e-cigarettes, given that college students have a high prevalence rate of using non-traditional tobacco products (Johnston, O'Malley, Bachman, Schulenberg, & Miech, 2014) and previous studies show the highest rate of ever-use e-cigarettes among this age group (Baeza-Loya, Viswanath, Carter, et al., 2014; Trumbo & Harper, 2013). The purpose of this study was to examine the effect of the two warning labels, generated by FDA and e-cigarette companies, on college students' perceived advantage and risk of e-cigarette use as well as their intention to use e-cigarettes. To the best of our knowledge, this was the first study examining whether or not e-cigarette warning labels have impacts on consumers' beliefs and behaviors. The findings of this study not only provided evidences showing the effect of e-cigarette warning labels, but also provided insightful information for future warning label design.

a. Warning label created by e-cigarette companies

b. Warning label proposed by the FDA

Fig. 1. Warning labels produced by e-cigarette companies and proposed by the FDA shown in real size. a. Warning label created by e-cigarette companies. b. Warning label proposed by the FDA.

2. Methods 2.1. Conceptual framework

of e-cigarette use beforehand is still necessary. Warning labels on cigarette packages have been used to decrease smoking rates (Borland, Wilson, Fong, et al., 2009; Hammond et al., 2007; Moodie, MacKintosh, & Hammond, 2010). Previous studies found that smokers' desire to consume cigarettes decreased after warning labels were launched (Devlin, Anderson, Hastings, & MacFadyen, 2005; Emery, Romer, Sheerin, Jamieson, & Peters, 2014). For non-smokers, warning labels could reduce their likelihood of initiating cigarette use (Singh, Owusu-Dabo, Britton, Munafo, & Jones, 2014). Because e-cigarettes are not fully regulated by the (US Food and Drug Administration (FDA), 2009), manufacturers are not obligated to include warning labels on packages. Nevertheless, some e-cigarette companies created their own warning label and place it on packages. One commonly used e-cigarette warning label example is shown in Fig. 1a. The content of this warning label reads: “This product is not a smoking cessation product and has not been tested as such. This product is intended for use by persons of legal age or older, and not by children, women who are pregnant or breastfeeding, or persons with or at risk of heart disease, high blood pressure, diabetes, or taking medicine for depression or asthma. Nicotine is addictive and habit forming, and it is very toxic by inhalation, in contact with the skin, or if swallowed. Nicotine can increase your heart rate and blood pressure and cause dizziness, nausea, and stomach pain. Inhalation of this product may aggravate existing respiratory conditions. Ingestion of the non-vaporized concentrated ingredients in the cartridges can be poisonous.” (Nu Mark LLC, n.d.) Although the company-produced warning label carries abundant information, the effect on users' perception and behavior has not been examined. Moreover, the lengthy text and small font size may lead to doubtful validity. In 2014, the FDA proposed a requirement for tobacco products, including e-cigarettes, to place a warning label on the packages (US Food and Drug Administration, 2014). Based on this

Constructs of the Theory of Planned Behavior were used for the structural equation modeling (SEM) in this study. The key concept is that subjects' behavioral intention is influenced by subjective evaluation of the risks and benefits of the expected outcome (Ajzen, 1985). Therefore, we hypothesized that college students' intention to use ecigarettes was influenced by their subjective evaluation of the risks and relative advantages of e-cigarette use, which could be swayed by the warning label. Regarding determinants influencing warning label effect, previous studies have demonstrated that label design, such as readability, had impact on warning label effect (O'Hegarty, Pederson, Yenokyan, Nelson, & Wortley, 2007; Wogalter, Conzola, & SmithJackson, 2002). Furthermore, the effect of the health message was associated with readers' comprehension and level of believability. Hence, we further hypothesized that the effect of warning labels is influenced by label design, including readability, comprehensibility, and believability of the content. Consumers' intention to use cigarettes was not only influenced by information they acquired from the warning labels but also predisposed by their knowledge and using experiences. Previous tobacco studies confirmed the association between knowledge about smoking and smoking intention (Romer & Jamieson, 2001; Tyc, Hadley, Allen, et al., 2004). Sutfin, McCoy, Morrell, Hoeppner, & Wolfson, 2013 reported similar findings for e-cigarette use, although they did not specify the direction of the relationship between knowledge and e-cigarette use. Accordingly, we hypothesized that college students' intention to use e-cigarette is influenced by their knowledge about e-cigarettes. Lastly, demographic variables such as gender, age, and race (white/non-white) were included as covariates. 2.2. Research design and experiment administration This study used pretest-posttest design to examine the effect of e107


H.-Y. Lee et al.

to inhale, (2) the main components of e-cigarettes are nicotine and propylene glycol (or glycerol), (3) e-cigarettes contain toxic ingredients (e.g., tars, carbon monoxide) that are responsible for many of tobaccoattributable diseases, (4) e-cigarette use could lead to development of nicotine dependence, (5) e-cigarettes have been approved by the FDA for smoking cessation, (6) e-cigarettes are under FDA regulation, and (7) the long-term health effects of e-cigarette use are known. Seven true and false questions were created. Subjects scored 1 point for each correct answer. Subjects scored 0 if they answered incorrectly or indicated that they do not know. The higher sum scores subjects had, the more knowledge they possessed about e-cigarettes. The effect of each warning label was assessed by one question: In your opinion, how effective is this warning label? A 5-point Likert scale was used, ranging from “not effective at all” (1) to “very effective” (5). Warning label design was evaluated by its readability, comprehensibility, and believability. Subjects were asked to respond to statements: “The warning label is easy to read”, “The information on the warning label is easy to understand,” and “The information on the warning label is believable,” adapted from previous studies (Emery et al., 2014; Moodie et al., 2010). A 5-point Likert scale was used. The higher scores, the higher level of agreement regarding the statement. Lastly, subjects' experience of using e-cigarettes was probed by asking if they have ever tried e-cigarettes, even just once. Moreover, their demographic characteristics comprised with gender (male/female), age, race (white/non-white) were surveyed.

cigarette warning labels on college students' intention to use e-cigarettes, mediated by warning label design, their perceived advantages and risks of e-cigarette use. Because this study examined two warning labels, two questionnaires were designed accordingly. One questionnaire demonstrated the warning label proposed by the FDA while the other demonstrated the warning label currently used by e-cigarette companies. Both versions had warning labels presented in real size with black text on a white background (Fig. 1). Regarding the process, subjects were first asked to answer questions related to their perceived advantage and risk of e-cigarette use. After that, they were asked to read the content of the warning label. Subsequently, subjects were asked again about their perception of the advantages and risks of ecigarette use together with their intention to use e-cigarettes. Additionally, they were asked to evaluate the effect of the warning label and its design. 2.3. Data and study groups This study employed a subset of data from a survey conducted in September and October 2015 from undergraduate college students aged 18 to 25 years at a Midwestern university. Detailed information about the survey has been addressed in another study (Lee, Lin, Seo, & Lohrmann, 2017). Subjects expressing an intention to try e-cigarettes in the next 12 months were randomly assigned to be exposed to either the warning label proposed by the FDA (Group 1) or the label used by e-cigarette companies (Group 2). Consequently, a total of 666 subjects were retained for analysis, with 338 subjects in Group 1 and 328 subjects in Group 2. The study was granted exemption from IRB review at the authors' institution.

2.5. Statistical analysis Descriptive analysis was performed to confirm the comparability between the two groups. After that, structural equation modeling (SEM) analysis was performed to examine the effect of warning labels on college students' intention to use e-cigarettes. The proposed model of SEM is shown in Fig. 2. In terms of the measurement model, the warning label design was determined by three indicators: readability, comprehensibility, and believability. Regarding the structural model, subjects' intention to use e-cigarettes was influenced by (1) knowledge about e-cigarette, (2) experience of trying e-cigarettes, and (3) the effect of warning label mediated by label design, subjects' perceived advantage, and risk of e-cigarette use. Moreover, subjects' perceived advantage and perceived risk of e-cigarette use were swayed by their perception in the pre-test. The error terms of perceived advantage and perceived risk were correlated because perceived advantage and perceived risk are both related to subjects' belief and have a trade-off relationship. Lastly, an invariance test was performed to compare across two groups. All analyses were performed by STATA 14.0.

2.4. Variables and measurement Subjects' intention to use e-cigarettes was probed by questions adapted from previous studies (Bunnell, Agaku, Arrazola, et al., 2015; Choi, Gilpin, Farkas, & Pierce, 2001; Pierce, Choi, Gilpin, Farkas, & Merritt, 1996; Wakefield, Kloska, O'Malley, et al., 2004). These questions were, “Do you think you will use e-cigarettes in the next 12 months?” and “If one of your best friends were to offer you an ecigarette, would you use it?”. A 5-point Likert scale was used. A higher sum score illustrated a higher intention to use e-cigarettes. Four positive statements were adapted from previous studies to measure perceived advantage of e-cigarette use (Choi & Forster, 2014; Tan & Bigman, 2014). Subjects were asked to respond to the same statements before and after label exposure. These statements were: (1) “E-cigarettes are less harmful than traditional cigarettes”, (2) “E-cigarettes can help reduce tobacco consumption”, (3) “E-cigarettes can help quit smoking”, and (4) “E-cigarettes are less addictive than traditional cigarettes”. A 5-point Likert scale was employed. The higher the sum score, the more a subject perceived e-cigarette use to have advantages. Subjects' perceived risk of e-cigarette use was measured by asking how likely they think it is to experience negative health consequences from continual use of e-cigarettes. Seven negative health consequences adapted from previous studies were used (Hine, Honan, Marks, & Brettschneider, 2007; Pokhrel, Little, Fagan, Muranaka, & Herzog, 2014). Subjects were asked to respond to the same possible health consequences before and after their exposure to the warning label. These items were: (1) damage overall health, (2) damage lungs, (3) develop lung cancer, (4) develop heart disease, (5) develop mouth/teeth problem, (6) develop nicotine addiction, and (7) die prematurely. A 10-point semantic differential scale was employed. The higher the sum scores subjects had, the greater risk they perceived from e-cigarette use. To examine subjects' knowledge about e-cigarettes, seven questions that address confirmed understandings about e-cigarettes were chosen. The statements were: (1) instead of smoke, e-cigarettes generate vapor

3. Results 3.1. Characteristics of the sample Descriptive results of the study sample are shown in Table 1. In total, 44.2% were male, 73.7% were non-Hispanic white, and the average age was 19.91 (SD: 1.55). Regarding e-cigarette using experience, 70.4% indicated that they have ever tried e-cigarettes. On average, subjects scored 3.13 in knowledge about e-cigarettes (range: 0–7; SD: 1.57). Before their exposure to the warning label, their mean perceived advantage of e-cigarette use was 13.49 (range: 4–20; SD: 2.80), while the mean perceived risk of e-cigarette use was 43.17 (range: 7–70; SD: 14.77). After the exposure, the mean perceived advantage decreased to 12.31 (SD: 3.68), the mean perceived risk increased to 45.45 (SD: 15.28) whereas subjects' intention to use e-cigarettes scored 5 (range: 2–10; SD: 2.33). Regarding warning label effect and design, all of results from independent t-tests showed that the FDA label was significantly superior than the label created by e-cigarette companies. 108


H.-Y. Lee et al.

Perceived advantage before warning label exposure

ε ε Warning Label Effect

Male

ε

Perceived advantage after warning label exposure Label Design

Intention Perceived risk after warning label exposure

Readability

ε

Comprehensibility

Believability

ε

Age

Have ever tried e-cigarettes

ε Perceived risk before warning label exposure

ε

White

Knowledge

Fig. 2. Conceptual model of the effect of warning label on college students' intention to use e-cigarettes mediated by perceived advantage and perceived risk.

use after exposure as well as their knowledge about e-cigarettes (γ = 0.13, p < 0.01) and use experiences (γ = 0.26, p < 0.01).

3.2. Structural equation modeling 3.2.1. Effect of the warning label proposed by FDA Results of SEM are shown in Table 2. The results showed that warning label effect had influence on subjects' intention to use e-cigarettes, mediated by label design (γ = 0.49, p < 0.01). Label design was a latent variable with three indicators: readability, comprehensibility, and believability. The standardized factor loadings of indicators were 0.94, 093, and 0.82, respectively. All factor loadings were significant (all ps < 0.01). Regarding subjects' belief about e-cigarette use, subjects' perceived advantage of e-cigarette use after exposure was related to their perception before exposure (γ = 0.57, p < 0.01), however, was not related to the label design. In contrast, subjects' perceived risk of e-cigarette use after exposure was associated with both their perceived risk of e-cigarette use before exposure (γ = 0.67, p < 0.01) and the label design (β = 0.10, p < 0.05). Lastly, subjects' intention to use e-cigarettes was related to their perceived advantage (β = 0.24, p < 0.01) and risk (β = −0.16, p < 0.01) of e-cigarette

3.2.2. Effect of warning label used by e-cigarette companies Results of SEM are shown in Table 2. The results showed that warning label effect was not able to influence subjects' intention to use e-cigarettes through label design, although the relationship between label effect and label design was significant (γ = 0.17, p < 0.01). The standardized factor loadings of indicators in determining the label design were 0.63, 0.83, and 0.72, respectively (all ps < 0.01). The label design, however, was not found to be correlated with subjects' perceived advantage and risk of e-cigarette use after exposure. Besides, subjects' perceived advantage and risk of e-cigarette use after exposure were related to their perceived advantage (γ = 0.62, p < 0.01) and risk (γ = 0.69, p < 0.01) of e-cigarette use before exposure. Lastly, subjects' intention to use e-cigarettes was associated with their perceived advantage (β = 0.23 p < 0.01) and risk (β = −0.16, p < 0.01) of e-cigarette use after exposure, together with their

Table 1 Descriptive statistics of two study groups by warning labels. Variable

1. Warning label effect and design Warning label effect (range: 1–5) Label design (range: 1–5) Readability Comprehensibility Believability 2. Before and after label exposure Perceived advantage before warning label exposure (range: 4–20) Perceived advantage after warning label exposure (range: 4–20) Perceived risk before warning label exposure (range: 7–70) Perceived risk after warning label exposure (range: 7–70) 3. Outcome variable Intention (range: 2–10) 4. Control variables Male White Age Have ever tried e-cigarettes Knowledge (range: 0–7)

Group 1: FDA labela

Group 2: E-cigarette company labelb

Total

N = 338 (50.1%)

N = 328 (49.9%)

N = 666 (100%)

3.19 (0.99)

2.84 (0.92)

3.03 (0.97)

< 0.001

4.49 (0.71) 4.44 (0.74) 4.39 (0.79)

3.18 (1.15) 3.59 (1.00) 3.78 (0.90)

3.85 (1.15) 4.03 (0.97) 4.10 (0.89)

< 0.001 < 0.001 < 0.001

13.29 12.27 43.61 45.67

13.69 12.44 42.73 45.43

13.49 12.31 43.17 45.45

0.06 0.68 0.45 0.81

(2.76) (3.57) (14.45) (15.08)

(2.83) (3.82) (15.10) (15.53)

p-Value

(2.80) (3.68) (14.77) (15.28)

5.02 (2.19)

5.01 (2.28)

5.00 (2.33)

0.80

155 (45.9%) 251 (74.3%) 19.87 (1.53) 233 (68.9%) 3.09 (1.57)

139 (42.4%) 240 (73.2%) 19.95 (1.58) 236 (72.0%) 3.17 (1.56)

294 (44.2%) 491 (73.7%) 19.91 (1.55) 469 (70.4%) 3.13 (1.57)

0.35 0.75 0.50 0.39 0.76

Notes: Results are either N (%) or means (SD). a E-cigarette warning label was proposed by the FDA. b E-cigarette warning label was created by e-cigarette companies.

109


H.-Y. Lee et al.

perceived advantage was positively related to their intention to use ecigarettes. Our findings revealed that current e-cigarette warning labels only focus on increasing consumers' risk perception rather than decreasing their perceived advantage. Based on findings of previous studies (Etter, 2013; Etter & Bullen, 2014; Hajek, Foulds, JLe, Sweanor, & Yach, 2013), a greater percentage of the public believes e-cigarette to be a safer alternative to cigarette smoking. This perception may make people mistakenly believe the e-cigarette use is safe. Accordingly, to make a label more effective, other than addressing potential risks, authorities should also focus on weakening the linkage between advantage perception and e-cigarette use. With respect to knowledge, our findings are not consistent with previous tobacco studies (Romer & Jamieson, 2001; Tyc et al., 2004), college students' knowledge about e-cigarettes may be positively related to their use intention. This may be because the negative health impacts of e-cigarette use have not yet been confirmed. The only certain point is that e-cigarettes are less harmful than traditional cigarettes, leading to more positive opinions about e-cigarettes and a higher intention to use them. Lastly, college students who have ever tried e-cigarettes had higher use intention. Based on the findings from previous studies (Agaku et al., 2014; Baeza-Loya et al., 2014), young adults have the highest rate of ever-tried e-cigarettes, indicating a very high percent of young adult e-cigarette experimenters who may progress to regular e-cigarette users. This finding may arise another concern regarding the progression of e-cigarette use and addiction. Therefore, not only are more studies needed to investigate the possible negative health outcomes of e-cigarette use, but more interventions are required to educate college students that a safer alternative does not equate to absolute safety. Additionally, more attention should be paid on college students' e-cigarette using behavior and long-term monitoring is needed.

Table 2 Results of structural equation modeling of the e-cigarette warning label effect on college students' intention to use e-cigarettes. Dependent variables and predictors

Measurement model Label design Readability Comprehensibility Believability Structural model Label design Warning label effect Perceived advantage Perceived advantage before exposure Label design Perceived risk Perceived risk before exposure Label design Intention Perceived advantage Perceived risk Male White Age Have ever tried ecigarettes Knowledge

Group 1: FDA label (N = 338)

Group 2: E-cigarette company label (N = 328)

Standardized regression coefficient

SEa

Standardized regression coefficient

SEa

0.94⁎⁎ 0.93⁎⁎ 0.82⁎⁎

0.01 0.13 0.02

0.63⁎⁎ 0.83⁎⁎ 0.72⁎⁎

0.03 0.03 0.03

0.49⁎⁎

0.03

0.17⁎⁎

0.05

0.57⁎⁎

0.04

0.62⁎⁎

0.03

−0.03

0.06

−0.01

0.05

0.67

⁎⁎

⁎⁎

0.03

0.69

0.03

0.10⁎

0.05

0.06

0.04

0.24⁎⁎ −0.16⁎⁎ 0.09 −0.01 0.12⁎ 0.26⁎⁎

0.06 0.06 0.05 0.05 0.05 0.05

0.23⁎⁎ −0.16⁎⁎ 0.14⁎⁎ −0.12⁎ 0.09 0.20⁎⁎

0.06 0.06 0.05 0.05 0.05 0.05

0.13⁎⁎

0.05

0.14⁎⁎

0.05

a

Standard error. p < 0.05. ⁎⁎ p < 0.01. ⁎

4.2. Comparison of warning label effect Compared with the warning label proposed by the FDA, the warning label created by e-cigarette companies had no impact on college student's intention to use e-cigarettes. This finding proved that volume of information is not proportionally related to label effect. To the contrary, presenting too much information in a label may decrease its effect because consumers may be reluctant to read the content if the label is too lengthy and/or the font size is too small. To further compare the predictors of warning label effect between two groups, results from additional analyses (upon request of authors) showed that the label created by companies had significantly lower readability and comprehensibility. Part of our results confirmed what has been found in previous studies: that label design influences the effect of health messages (O'Hegarty et al., 2007; Wogalter et al., 2002). Clearly presenting information could increase the label's readability and effectiveness. Rather than showing all information in one label, selecting important and trustful information could enhance the labels' comprehensibility as well as its effectiveness. Further studies are needed to ensure the effectiveness of selected information. The results of invariance test showed that the FDA label had better effects than the company label. In particular, the FDA label performed better in decreasing college students' intention to use e-cigarettes via increasing their risk perception of e-cigarette use. In contrast, the warning label currently used by companies showed no effect on use intention. Consequently, to decrease prevalence of e-cigarette use, policy makers may take actions not only to mandate the use of warning labels but also regulate the content and presentation of a label. Meanwhile, government should continue seeking a more effective warning label while the current label has much room for improvement. Further studies are also needed to investigate the mechanism causing the effect difference between labels. This study had some limitations. First, a self-administered questionnaire was applied, which may contain inaccurate information due

knowledge about e-cigarettes (γ = 0.14, p < 0.01) and experiences of using them (γ = 0.20, p < 0.01). 3.2.3. Invariance test to compare coefficient between two groups Results of the invariance test used to compare coefficient between two aforementioned groups. The relationship between waning label effect and label design was found significantly higher in FDA group (p < 0.001, χ2 = 106.36, df = 1). Moreover, FDA group showed better readability (p < 0.001, χ2 = 39.06, df = 1) and comprehensibility (p < 0.001, χ2 = 28.61, df = 1) than the counterpart. 4. Discussion This empirical study helped to fill the literature gap by investigating the effect of the aforementioned e-cigarette labels. As far as is known, this study is the first to examine and compare the effect of the warning labels proposed by the FDA and created by e-cigarette companies. The findings not only provide direct preliminary evidences about the effect of warning labels, but also could be references for future governmental studies that could be conducted to verify these findings. 4.1. Factors influencing intention to use e-cigarettes Consistent with findings from previous tobacco studies (Devlin et al., 2005; Emery et al., 2014), e-cigarette warning labels influenced college students' intention to use e-cigarettes, mediated by the label design and their perceptions of e-cigarette use. To be more specific, a well-designed warning label may decrease college students' intention to use e-cigarettes by increasing their risk perception of e-cigarette use. In contrast, warning labels were found to have no impact on college students' perceived advantage of e-cigarette use, even though their 110


H.-Y. Lee et al.

243–253. Barbeau, A. M., Burda, J., & Siegel, M. (2013). Perceived efficacy of e-cigarettes versus nicotine replacement therapy among successful e-cigarette users: a qualitative approach. Addiction Science & Clinical Practice, 5, 8AA. http://dx.doi.org/10.1186/ 1940-0640-8-5. Borland, R., Wilson, N., Fong, G. T., et al. (2009). Impact of graphic and text warnings on cigarette packs: Findings from four countries over five years. Tobacco Control, 18, 358–364. http://dx.doi.org/10.1136/tc.2008.028043. Bunnell, R. E., Agaku, I. T., Arrazola, R. A., et al. (2015). Intentions to smoke cigarettes among never-smoking US middle and high school electronic cigarette users, National Youth Tobacco Survey, 2011–2013. Nicotine & Tobacco Research, 17, 228–235. http:// dx.doi.org/10.1093/ntr/ntu166. Choi, K., & Forster, J. L. (2014). Beliefs and experimentation with electronic cigarettes: A prospective analysis among young adults. American Journal of Preventive Medicine, 46(2), 175–178. http://dx.doi.org/10.1016/j.amepre.2013.10.007. Choi, W. S., Gilpin, E. A., Farkas, A. J., & Pierce, J. P. (2001). Determining the probability of future smoking among adolescents. Addiction, 96, 313–323. http://dx.doi.org/10. 1046/j.1360-0443.2001.96231315.x. Dawkins, L., Turner, J., Roberts, A., & Soar, K. (2013). “Vaping” profiles and preferences: An online survey of electronic cigarette users. Addiction, 108, 1115–1125. http://dx. doi.org/10.1111/add.12150. Devlin, E., Anderson, S., Hastings, G., & MacFadyen, L. (2005). Targeting smokers via tobacco product labelling: Opportunities and challenges for Pan European health promotion. Health Promotion International, 20(1), 41–49. http://dx.doi.org/10.1093/ heapro/dah506. Dockrell, M., Morrison, R., Bauld, L., & McNeill, A. (2013). E-cigarettes: Prevalence and attitudes in Great Britain. Nicotine & Tobacco Research, 15, 1737–1744. http://dx.doi. org/10.1093/ntr/ntt057. Emery, L. F., Romer, D., Sheerin, K. M., Jamieson, K. H., & Peters, E. (2014). Affective and cognitive mediators of the impact of cigarette warning labels. Nicotine & Tobacco Research, 16(3), 263–269. http://dx.doi.org/10.1093/ntr/ntt124. Etter, J. F. (2013). Should electronic cigarettes be as freely available as tobacco? Yes. BMJ, 346, f3845. http://dx.doi.org/10.1136/bmj.f3845. Etter, J. F., & Bullen, C. (2014). A longitudinal study of electronic cigarette users. Addictive Behaviors, 39(2), 491–494. http://dx.doi.org/10.1016/j.addbeh.2013.10. 028. Fairchild, A. L., Bayer, R., & Colgrove, J. (2014). The renormalization of smoking? Ecigarettes and the tobacco “endgame”. The New England Journal of Medicine, 293–295. http://dx.doi.org/10.1056/NEJMp1002530. Hajek, P., Foulds, J., JLe, H., Sweanor, D., & Yach, D. (2013). Should e-cigarettes be regulated as a medical device? The Lancet Respiratory Medicine, 1(6), 429–431. Hajek, P., Etter, J. F., Benowitz, N., Eissenberg, T., & McRobbie, H. (2014). Electronic cigarettes: Review of use, content, safety, effects on smokers and potential for harm and benefit. Addiction, 109, 1801–1810. http://dx.doi.org/10.1111/add.12659. Hammond, D., Fong, G. T., Borland, R., Cummings, K. M., Mcneill, A., & Driezen, P. (2007). Text and graphic warnings on cigarette packages: Findings from the international tobacco control four country study. American Journal of Preventive Medicine, 32(3), 202–209. http://dx.doi.org/10.1016/j.amepre.2006.11.011. Hine, D. W., Honan, C. A., Marks, A. D. G., & Brettschneider, K. (2007). Development and validation of the smoking expectancy scale for adolescents. Psychological Assessment, 19, 347–355. http://dx.doi.org/10.1037/1040-3590.19.3.347. Johnston, L. D., O'Malley, P. M., Bachman, J. G., Schulenberg, J. E., & Miech, R. A.. Monitoring the future: National survey results on drug use 1975–2014. (2014). Retrieved from http://www.monitoringthefuture.org/pubs/monographs/mtf-vol2_2014.pdf (Assessed February 12, 2016) . Kim, A. E., Arnold, K. Y., & Makarenko, O. (2014). E-cigarette advertising expenditures in the U.S., 2011–2012. American Journal of Preventive Medicine, 46(4), 409–412. http:// dx.doi.org/10.1016/j.amepre.2013.11.003. King, B. A., Patel, R., Nguyen, K. H., & Dube, S. R. (2015). Trends in awareness and use of electronic cigarettes among US adults, 2010–2013. Nicotine & Tobacco Research, 17, 219–227. http://dx.doi.org/10.1093/ntr/ntu191. Lee, H. Y., Lin, H. C., Seo, D. C., & Lohrmann, D. K. (2017). Determinants associated with E-cigarette adoption and use intention among college students. Addictive Behaviors, 65, 102–110. Moodie, C., MacKintosh, A. M., & Hammond, D. (2010). Adolescents' response to textonly tobacco health warnings: Results from the 2008 UK Youth Tobacco Policy Survey. European Journal of Public Health, 20, 463–469. http://dx.doi.org/10.1093/ eurpub/ckp199. Nu Mark LLC. MarkTen.com. (2007). Retrieved from https://www.markten.com/ gconnect/login_input.action (Accessed March 12, 2015) . O'Hegarty, M., Pederson, L. L., Yenokyan, G., Nelson, D., & Wortley, P. (2007). Young adults' perceptions of cigarette warning labels in the United States and Canada. Preventing Chronic Disease, 4, 1–9 (doi:A27 [pii]). Pierce, J. P., Choi, W. S., Gilpin, E. A., Farkas, A. J., & Merritt, R. K. (1996). Validation of susceptibility as a predictor of which adolescents take up smoking in the United States. Health Psychology, 15(5), 355–361. Pokhrel, P., Little, M. A., Fagan, P., Muranaka, N., & Herzog, T. A. (2014). Electronic cigarette use outcome expectancies among college students. Addictive Behaviors, 39(6), 1062–1065. http://dx.doi.org/10.1016/j.addbeh.2014.02.014. Richtel, M.. E-cigarette warning labels confound health officials. (2014). Retrieved from http://www.bostonglobe.com/business/2014/09/28/cigarette-makers-warningsconfound/Zfnbtsem2QFQZc1Zg8dTvN/story.html (Accessed March 12, 2015) . Romer, D., & Jamieson, P. (2001). Do adolescents appreciate the risks of smoking? Evidence from a national survey. The Journal of Adolescent Health, 29, 12–21. http:// dx.doi.org/10.1016/S1054-139X(01)00209-9. Singh, A., Owusu-Dabo, E., Britton, J., Munafo, M. R., & Jones, L. L. (2014). “Pictures

to subjects' recall and respondent bias. Second, convenience sampling was used, which could limit the generalizability of the results. Third, the sample is limited to only college students; the results may not be applied to other age groups. A larger-scale and randomized study is need in the future. Fourth, several important factors such as social approval were not included in this study. A future study examining the effect of other factors is needed to completely capture the psycho-social factors that influence e-cigarette use. Despite these limitations, this study provided information about the mechanism of e-cigarette warning labels influencing college students' intention to use e-cigarettes. The resultant findings could serve as direct evidence for future mandate establishment. 5. Conclusions To our knowledge, this is the first study examining the effect of ecigarette warning labels on college students' intention to use e-cigarettes, which illustrated the mechanism of the warning label swaying college students' e-cigarette use intention. Findings of this study provide direct empirical evidence showing the different effect of the two ecigarette warning labels. Overall, the warning label proposed by the FDA is more effective than the warning label created by e-cigarette companies. This proposed governmental label was found to reduce college students' intention to use e-cigarettes via increasing their perceived risk of e-cigarette use. In contrast, the label created by e-cigarette companies was found to be ineffective. Regarding determinants influencing label effect, the warning label created by e-cigarette companies had lower readability and comprehensibility. As previously found for combustible cigarettes, findings demonstrate that warning label information should be carefully selected and clearly presented on the packages where e-cigarette companies have overlooked, perhaps purposefully, these determining factors. As a result, rather than simply rely on warning labels created by e-cigarette companies, advocators should urge lawmakers to mandate the use of more effective-proven warning labels. Role of funding sources This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Contributors Lee and Lin designed the study. Lee conducted literature searches, provided summaries of previous research studies, and conducted the statistical analysis. Lin and Seo helped with conceptualization of the study, data analysis, interpretation of findings, and critical reviews of the manuscript. Lohrmann helped with the study design and critically revised the manuscript. All authors contributed to and have approved the final manuscript. Conflict of interests The authors have no competing interests or conflicts to declare. References Adkison, S. E., O'Connor, R. J., Bansal-Travers, M., et al. (2013). Electronic nicotine delivery systems: international tobacco control four-country survey. American Journal of Preventive Medicine, 44(3), 207–215. http://dx.doi.org/10.1016/j.amepre.2012.10. 018. Agaku, I. T., King, B. A., Husten, C. G., et al. (2014). Tobacco product use among adultsUnited States, 2012–2013. MMWR. Morbidity and Mortality Weekly Report, 63, 542–547. Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. Heidelberg: Springer. Baeza-Loya, S., Viswanath, H., Carter, A., et al. (2014). Perceptions about e-cigarette safety may lead to e-smoking during pregnancy. Bulletin of the Menninger Clinic, 78(3),

111


H.-Y. Lee et al.

29(6), 1143–1147. http://dx.doi.org/10.1016/j.addbeh.2004.03.007. US Food and Drug Administration. Family smoking prevention and tobacco control and federal retirement reform. (2009). Retrieved from https://www.gpo.gov/fdsys/pkg/PLAW111publ31/pdf/PLAW-111publ31.pdf (Accessed February 12, 2016) . US Food and Drug 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. (2014). Retrieved from https://www.fda.gov/downloads/AboutFD.../ UCM394933.pdf (Accessed July 28, 2017) . Wakefield, M., Kloska, D. D., O'Malley, P. M., et al. (2004). The role of smoking intentions in predicting future smoking among youth: Findings from Monitoring the Future data. Addiction, 99, 914–922. http://dx.doi.org/10.1111/j.1360-0443.2004.00742.x. Wogalter, M. S., Conzola, V. C., & Smith-Jackson, T. L. (2002). Research-based guidelines for warning design and evaluation. Applied Ergonomics, 33, 219–230. http://dx.doi. org/10.1016/S0003-6870(02)00009-1.

don't lie, seeing is believing”: Exploring attitudes to the introduction of pictorial warnings on cigarette packs in Ghana. Nicotine & Tobacco Research, 16, 1613–1619. http://dx.doi.org/10.1093/ntr/ntu127. Sutfin, E. L., McCoy, T. P., Morrell, H. E. R., Hoeppner, B. B., & Wolfson, M. (2013). Electronic cigarette use by college students. Drug and Alcohol Dependence, 131(3), 214–221. http://dx.doi.org/10.1016/j.drugalcdep.2013.05.001. Tan, A. S. L., & Bigman, C. A. (2014). E-cigarette awareness and perceived harmfulness: Prevalence and associations with smoking-cessation outcomes. American Journal of Preventive Medicine, 47(2), 141–149. http://dx.doi.org/10.1016/j.amepre.2014.02. 011. Trumbo, C. W., & Harper, R. (2013). Use and perception of electronic cigarettes among college students. Journal of American College Health, 61, 149–155. Tyc, V. L., Hadley, W., Allen, D., et al. (2004). Predictors of smoking intentions and smoking status among nonsmoking and smoking adolescents. Addictive Behaviors,

112