Next Article in Journal
Hepatitis C Virus Treatment Status and Barriers among Patients in Methadone Maintenance Treatment Clinics in Guangdong Province, China: A Cross-Sectional, Observational Study
Previous Article in Journal
Outdoor Endurance Training with Air Pollutant Exposure Versus Sedentary Lifestyle: A Comparison of Airway Immune Responses
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJERPH
Volume 16
Issue 22
10.3390/ijerph16224435
ijerph-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

An Eye Tracking Study of Anti-Smoking Messages on Toxic Chemicals in Cigarettes

by
Leah M. Ranney
1,*,
Sarah D. Kowitt
1,
Tara L. Queen
2,
Kristen L. Jarman
2 and
Adam O. Goldstein
1,2
1
Department of Family Medicine, UNC School of Medicine, University of North Carolina at Chapel Hill, 590 Manning Drive, CB# 7595, Chapel Hill, NC 27599, USA
2
Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, 101 Manning Drive, CB# 7295, Chapel Hill, NC 27599, USA
*
Author to whom correspondence should be addressed.
Int. J. Environ. Res. Public Health 2019, 16(22), 4435; https://doi.org/10.3390/ijerph16224435
Submission received: 22 August 2019 / Revised: 28 October 2019 / Accepted: 7 November 2019 / Published: 12 November 2019
(This article belongs to the Section Health Communication and Informatics)
Browse Figures
Versions Notes

Abstract

:
The US Food and Drug Administration is tasked with communicating information to the public about the harmful chemicals in cigarette smoke. Our study used eye tracking method to test the effectiveness of messages about the harmful chemicals in cigarettes smoke among adult smokers. A sample size of 211 current cigarette smokers viewed four communication messages that included: Health effects of a chemical in cigarette smoke and an image depicting the health effect. The messages focused on arsenic, formaldehyde, uranium, and general health. Eye tracking recorded the length of time participants viewed the text and the image. After each message, the participants were asked about the messages’ effectiveness in changing attitudes towards smoking. We analyzed the data using multilevel modeling, and of the 211 smokers, 59.7% were female, 36.5% were Black, and 21.3% had a high school degree or less. Compared to the general message, the messages about formaldehyde and uranium were more discouraging to smoking (p < 0.05). Messages about formaldehyde were more believable and made participants want to quit more than the general messages. Increasing message dose was significantly associated with discouraging participants from smoking and made participants want to quit (p < 0.05). Our findings suggest that anti-smoking messages, containing chemical information, can successfully increase negative attitudes toward smoking cigarettes and potentially encourage quitting.

1. Introduction

Cigarette smoking is the leading cause of preventable death in the USA [1,2]. Smoke from cigarettes and other tobacco products, like cigars and pipes, contain at least 70 cancer-causing chemicals known as carcinogens [3]. Chemicals created from the process of curing tobacco leaves, that are added to enhance flavor and reduce harshness (i.e., additives), and that are created when the cigarette is burned cause many serious health-related illnesses, such as cancer, heart disease, stroke, lung diseases, diabetes, and chronic obstructive pulmonary disease [1,4,5]. Since the 2009 Family Smoking Prevention and Tobacco Control Act, the US Food and Drug Administration (FDA) has been charged with regulating the manufacture, distribution, and marketing of tobacco products [6]. Under their purview, the FDA is required to publicly disclose the ingredients (i.e., constituents) found in tobacco products. As part of this requirement, the FDA has established a list of 93 harmful and potentially harmful constituents (HPHC) in tobacco products and tobacco smoke, and the tobacco industry must report HPHC in their existing and newly marketed tobacco products [7].
An important question raised by the FDA’s new requirement is how best to communicate tobacco constituents and their risks to the public. The FDA’s initial effort to communicate the risk associated with chemicals in cigarettes smoke was with The Real Cost campaign, which was targeted towards youth [8]. Several of these anti-smoking advertisements highlight the harms associated with chemicals in cigarettes, like the “7000 chemicals” ad. While, the evaluation of The Real Cost reported the campaign was highly successful, no outcomes were reported specific to the constituent messages [9,10,11].
Recent research has shed some light on how to communicate the dangers of the chemicals and chemical constituents in tobacco products [12,13,14,15]. For instance, research found that many people are unaware that burning the cigarette is the primary source of harmful chemicals, and few believe the harmful chemicals come from tobacco before cigarettes are manufactured [13,16]. An early study examining the optimal methods to display tobacco constituent levels on tobacco packaging found adult perceived descriptive information as more comprehensible than numerically displayed information [17]. Recent studies testing messages with cigarette constituents information among adults showed that the messages increased the constituents’ level of knowledge, reinforced smoking related-health outcomes, and discouraged people from wanting to smoke [18,19].
While, research is building around cigarette smoke constituents, few studies have examined the effectiveness of communication campaigns where the constituents are part of the messaging. Even fewer have examined this using eye tracking technology [20,21], which provides measurable data on visual attention, a precursor to information processing that can supplement evidence from self-report methods [22]. Measuring visual attention to specific components of communication campaigns can improve our understanding of how people attend to, and are influenced by, the messages [23]. Our study uses eye tracking technology to experimentally test the effectiveness of anti-smoking messages, containing constituent information on print ads, among adult cigarette smokers. We hypothesized that, (1) constituent messages increase negative attitudes toward smoking cigarettes more than a non-constituent general health message about the consequences of smoking, and (2) looking at the message text and images longer would be associated with higher self-reported measures of the messages’ impact.

2. Methods

2.1. Participants

Recruitment of participants occurred between July and December 2016 from the research triangle area of North Carolina. Participants were between the ages of 18–65 years old and current smokers (i.e., had smoked more than 100 lifetime cigarettes and smoked every day or some days in the past 30 days). Participants were included in the study if they were current smokers. While, participants were asked about their use of other tobacco products, other tobacco use was not an eligibility criterion for this study. Participants who were not able to be eye tracked (e.g., did not have vision in both eyes) or who had participated in a tobacco study in the past 3 months were excluded. The study initially enrolled 230 participants. Of these participants, 5 were removed because the eye tracker was not able to calibrate to their eye movements, 4 did not complete the study protocol, and 10 were removed due to incomplete eye tracking data. The final sample was 211 participants. The University of North Carolina Institutional Review Board approved study procedures, and participants provided their written informed consent.

2.2. Procedures

This was a one-time eye tracking study. Before the study session began, the participants were randomly allocated to 1 of 4 conditions and remained in the same condition group throughout the study session. The 4 conditions were created as part of a 2 × 2 experimental design. The two experimental factors were (1) whether participants saw messages with the FDA logo and (2) whether participants saw messages that included text about the benefits of quitting smoking: Additional details about the study methodology can be found in Jarman L, Kowitt S, Queen, L. et al. [20].
Within each of the 4 conditions, the participants saw 4 messages (Figure 1). The messages were formatted as hypothetical print campaigns about different constituents in cigarette smoke. All messages (regardless of condition), included (1) text about the health effects of a chemical constituent in cigarette smoking (e.g., cigarette smoke contains arsenic. This causes heart damage.) and (2) an image depicting the health effects of the chemical constituent. The 4 messages within each condition focused on the health effects of arsenic, formaldehyde, uranium, as well as a non-constituent general message about the health harms of smoking. Based on prior research assessing the understanding of 24 specific cigarette smoke constituents, the three constituents selected for the study messages were identified as most discouraging of smoking in a national probability sample of adults and adolescents [16]. Uranium, one of the top constituents from Brewer et al. contained two health effects of chemicals and was tested without altering this format. To develop the images and other relevant anti-smoking information, we used a series of focus groups and online experiments [24,25]. The images were intentionally selected to represent the message, and therefore, were not randomized with the message. At the beginning of the study session, participants’ eye movements were calibrated using a Tobii X2-60 eye tracker. Following calibration, participants then answered questions about their demographics and smoking behaviors. Participants were then shown the 4 constituent messages (i.e., in a counterbalanced order) and were allowed to view the messages for as long as they liked. After viewing each message, participants responded to questions about the persuasiveness and impact of these messages (see Measures). At the end of the eye tracking session, participants were asked additional questions about elements manipulated in the study (e.g., recall of source).

2.3. Measures

Our primary outcomes of interest were self-reported measures of the messages’ impact [26]. Our primary independent variables of interest were the constituent, attention to the image, attention to the constituent message text, and the effect of message dose.

2.3.1. Outcome: Measures about Message Impact

After each message was presented, the participants were asked to report on how believable the message was (hereafter believability), how much the message made them want to quit smoking (want to quit), and the extent to which the message discouraged them from smoking (discouragement). Responses were scored on a scale of 1–9 where 1 = Strongly Disagree, 5 = Neutral, 9 = Strongly Agree.

2.3.2. Message-Related Variables

In order to determine which constituent messages were more impactful, we created a categorical variable for the message (i.e., arsenic, formaldehyde, uranium, or general, with general being the referent group). Eye tracking variables of interest included the total length of time participants attended to the image (image dwell time) and text (text dwell time). Dwell time is a measure of visual attention that, by our definition, included both fixations within an area of interest, as well as saccades between those fixations [27]. Since dwell time for the entire print ad was highly correlated with both text and image dwell time, we analyzed the image dwell time and text dwell time separately. To determine whether there was a cumulative impact of messages over the course of the study, we created a continuous variable that captured message dose. For example, the first message that a participant viewed was coded as 0 previous messages, the second was coded as 1 previous messages, etc.

2.3.3. Control Variables: Demographics and Condition

Demographic measures, collected at the beginning of the study session, included: Age, sex at birth, sexual identity, ethnicity, race, education, income, and nicotine dependence. Nicotine dependence was assessed using the Fagerström Test for Nicotine Dependence (FTND), and responses were scored on a scale of 0–10 [28]. We controlled for these demographic measures in all models. In addition, we also controlled for the condition to which participants were assigned.

2.4. Data Analysis

2.4.1. Modeling Strategy

After examining the data descriptively, we analyzed data using 3 multilevel models that corresponded with each of the 3 outcomes. We used a multilevel modeling approach to account for multiple observations (4 messages) within participants over the course of the study. In order to parse between- versus within-person variance in dwell times, we grand and group mean-centered these variables. In grand mean-centering dwell time, we calculated the deviation of each participant’s score from the overall mean of each variable. In group mean centering, we calculated the deviation of each observation from the mean for the participant’s cluster (participants in this case). For clarity, we labeled all observation-level, stimuli specific variables (dwell time, believability, etc.) as “Level 1” variables and labeled all participant-level variables (demographics, nicotine dependence, etc.) as “Level 2” variables.

2.4.2. Multilevel Models

After centering, we examined the associations among the variables and outcomes, adjusting for control variables. We set critical α = 0.05 and used 2-tailed statistical tests. For all analyses, we used SAS version 9.4 procedures (SAS Inc., Cary, NC, USA). For the multilevel models, we used the PROC MIXED procedure. Since each participant saw 4 different messages and there were 211 participants, there were 844 unique observations (211 × 4).

3. Results

3.1. Participant Characteristics

Of the 211 participants, the majority were female (59.7%) and White (58.3%), and a substantial proportion of participants identified as African American (36.5%) (Table 1). The average age of participants was 36.2 years (standard deviation, SD: 12.4), and the average nicotine dependence score was 4.13 (SD: 2.59).

3.2. Descriptive Statistics

On average, the formaldehyde message scored the highest on all 3 outcomes. Specifically, the formaldehyde message was rated as more believable, made participants want to quit more, and discouraged them the most from smoking (Table 2). Interestingly, the arsenic message scored the lowest on all 3 outcomes.
Participants spent the most time looking at the image for the general message (4.39 s, SD: 4.61) and the least amount of time looking at the image for the uranium message (2.67 s, SD: 2.89). In contrast, we found that participants spent the most amount of time looking at the uranium message text (3.33 s, SD: 3.54) and the least amount of time looking at the general message text (0.41 s, SD: 0.65).

3.3. Multilevel Models

When we modeled believability as the outcome, we found that the arsenic message (B = −0.53, p < 0.001) and the uranium message (B = −0.42, p < 0.001) were less believable than the general message (Table 3). The formaldehyde message (B = 0.30, p < 0.02), was more believable than the general message. There were no effects of image or text dwell time on believability, nor was there an effect of message dose.
When we modeled wanting to quit as the outcome, we found that the formaldehyde message (B = 0.97, p < 0.001) made them want to quit more than the general message. There were no effects of image or text dwell time on the messages’ ability to help participants to quit. However, as message dose increased, they reported that the messages made them want to quit smoking more (B = 0.13, p < 0.001).
Finally, when we modeled discouragement as the outcome, we found that the formaldehyde message (B = 1.05, p < 0.001) and the uranium message (B = 0.43, p = 0.01) discouraged participants from smoking more than the general message. There were no effects of image or text dwell time on the messages’ ability to help participants to quit. However, as participants viewed more messages, they reported that the messages discouraged them more from smoking (B = 0.13, p = 0.004).

4. Discussion

This is one of the first studies using eye tracking technology to test the impact of cigarette smoke constituent messages among adult smokers. Using the results from previous research, we created four anti-smoking print ads by pairing text messages about constituents with images (three ads featured constituents and one featured a non-constituent, general health message). In addition to survey items, eye tracking technology was used to measure participants’ dwell time on the constituent text message and images contained within the ad. Our study indicates that constituent focused anti-smoking print ads performed differently than general anti-smoking print ads about the health harms of smoking among adult smokers, and that some constituent ads were more believable and influential than others. Specifically, the formaldehyde constituent ad was more believable, made smokers want to quit more, and discouraged smoking more, when compared to the general health ad (i.e., non-constituent ad). While the uranium constituent ad was also reported to be more discouraging and made participants want to quit more than general health ads (non-constituent), the arsenic and the uranium constituent ads were less believable than the general health ad. Research on constituent messages provides some insight into the nuances that may differentiate constituent messages from each other [12,13,15,16,29].
Research on cigarette smoke constituents suggest that, while knowledge of cigarette smoke constituents is low, most people have heard of nicotine, carbon monoxide, ammonia, arsenic, benzene, cadmium, and formaldehyde [13,15,16]. Brewer et al. found that the more recognizable constituents—ammonia, arsenic, formaldehyde, hydrogen cyanide, lead, and uranium—were most likely to discourage respondents from wanting to smoke [16]. Our study applied the findings from Brewer et al. and used the most familiar and effective constituents to produce the constituent print ads. An additional finding from Brewer et al. was that longer constituent names were more discouraging than shorter names. This may explain why the arsenic constituent ad scored the lowest on all of the outcomes (believability, wanting, discouragement) and the formaldehyde constituent ad was the most impactful constituent ad on all three outcomes. The image for the formaldehyde constituent ad, a man with a laryngectomy, was also more aversive or attention-grabbing than the other images, and may have contributed to higher message believability and greater desire to quit. Importantly, message believability has been shown to be associated with the intention to engage in smoking cessation behaviors [30].
Tobacco control mass media campaigns can promote quitting and reduce adult smoking prevalence when factors, such as audience reach, intensity, duration, and message type are successful [31]. Media campaigns have also linked cigarette smoking to a range of diseases with research suggesting that cancer-related anti-smoking messages are 1.5 times more believable than chronic disease messages [14]. Additionally, constituent messages, focused on cancer and health effects related to the lungs, mouth, and heart generally led to higher levels of smoking discouragement [15]. Our research supports these findings. Participants rated the formaldehyde and uranium constituent ads (which focused on cancer, and health effects related to the lungs, respectively) as more discouraging than the general message. Interestingly, the formaldehyde constituent ad had the only message containing the word cancer in the related-health effect, and it was the only constituent ad reported to be more believable than the non-constituent ad with the general message. Research suggests that pairing known information (i.e., smoking causes cancer) with new information (i.e., formaldehyde cause cancer) influences message processing and beliefs [14,32]. This may be an important principle for future constituent communication campaigns.
Notably, as participants in our study viewed more ads, they reported that the ad messages were more effective. This finding is consistent with evidence supporting the effect of increased message exposure on promoting smoking cessation [32,33]. Health message campaigns are often evaluated on the number of impressions, with the expectation that more campaign exposure may lead to increased effectiveness of the campaign [34,35]. Our findings provide initial support that constituent ads can successfully discourage smoking and potentially made smokers want to quit more with repeated message exposure.
Because previous research suggests that pictorial messages are more effective than text-only messages [36,37], our study paired each of the constituent messages with an image depicting the health effect to increasing ad effectiveness. As an indicator of visual attention, our study used eye tracking technology to measure participant dwell time on the constituent text and image for each constituent print ad. Eye tracking allows researchers to move beyond self-reported outcomes and observe participants’ actual visual attention [23]. Not surprisingly, participants spent more time on the longer text message on the uranium constituent ad. This ad contained two health effects opposed to the other constituent ad, with only one health effect and the non-constituent ad message. The least amount of time was spent on the text for the non-constituent ad, potentially due to the brevity of the message. Interestingly, the reverse was true for dwell time on the image; the image for the non-constituent ad had the longest dwell time, while the uranium constituent ad image had the shortest dwell time.
We found no effects from increasing image or text dwell time on any of the outcomes. The majority of participants did indeed look at the image and message text, and the research indicates that information acquisition from a visual stimuli can occur in a fraction of a second (i.e., 0.1 s) [38,39,40,41]. The participants in our study spent, on average, approximately 3–4 s on the image, and 1–3 s on the text for each message, which indicates that they spent a significant amount of time attending to the information and potentially processing it. Participants may have more carefully viewed each ad because research staff was present during the experiment, and because of the incentive offered for the study. However, participants were allowed to proceed with the experiment at their own pace. Hence, the absence of text and image dwell time effects might be due to study protocol and conscious participant behavior, resulting in the majority of participants adequately processing the message.
Overall our findings suggest that an anti-smoking campaign that uses constituent information that is familiar to the public and features the negative health effect of the constituent can successfully discourage smoking and persuade adult smokers to want to quit. Importantly, repeated ad exposure increased the impact of the constituent ads, indicating that multiple exposure to ads via a mass media campaign can increase persuasion over time.

Limitations

Our study is not without limitations. First, this was a one-time study in a laboratory setting where participants had to view the constituent messages. The results, therefore, may not be generalized to other settings. Relatedly, participants were recruited from one geographic area, which also may have limited the generalizability to other areas. While, there is a strong link between eye tracking fixations and attention, it is possible that, while, a person’s visual gaze is within an AOI, they are not actually processing that information. It is possible that when looking within our defined AOIs, that participants were processing peripheral information or thinking about something else entirely [42,43,44]. Therefore, we cannot determine whether participants were processing information associated with those areas of interest. The intention of the study was to assess whether messages as a whole increased negative attitudes and intentions to quit smoking rather than to choose the best image or message. Therefore, we intentionally did not randomize images with the messages and cannot disentangle the effects of the message text and image on outcomes. Finally, we assessed self-reported outcomes after each ad exposure. However, there were a number of factors that could have influenced these outcomes, including the message text and the message image.

5. Conclusions

Our study findings have significant implications for the FDA and their requirement to disclose cigarette smoke constituent information to the public in an understandable and not misleading manner. Further, our study anti-smoking constituent campaign ads are consistent with established standards for effective tobacco communication [45]. Formaldehyde, a familiar constituent, and throat cancer, a familiar health effect, produced the most believable and influential constituent ad message for adult smokers in this study. Educational campaigns that communicate well-designed constituent ads, with corresponding images, may help reduce smoking among adults.

Author Contributions

Conceptualization, L.M.R. and S.D.K.; data curation, S.D.K., T.L.Q., and K.L.J.; formal analysis, S.D.K. and T.L.Q.; funding acquisition, A.O.G.; investigation, S.D.K. and K.L.J.; methodology, L.M.R., S.D.K., T.L.Q., and K.L.J.; project administration, K.L.J. and A.O.G.; software, S.D.K., T.L.Q., and K.L.J.; supervision, L.M.R. and A.O.G.; validation, S.D.K.; visualization, L.M.R. and S.D.K.; writing—original draft, L.M.R. and S.D.K.; writing—review and editing, L.M.R., S.D.K., T.L.Q., K.L.J., and A.O.G.

Funding

Research reported in this publication was supported by grant number P50 CA180907 from the National Cancer Institute and the FDA Center for Tobacco Products (CTP). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the Food and Drug Administration.

Conflicts of Interest

The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication. The authors have no conflicts of interest to declare.

References

  1. US Department of Health and Human Services. The Health Consequences of Smoking—50 Years of Progress: A Report of the Surgeon General. Available online: www.cdc.gov/tobacco (accessed on 4 January 2019).
  2. Jamal, A.; Phillips, E.; Gentzke, A.S.; Homa, D.M.; Babb, S.D.; King, B.A.; Neff, L.J. Current Cigarette Smoking Among Adults—United States, 2016. MMWR 2018, 67, 53–59. [Google Scholar] [CrossRef] [PubMed]
  3. American Cancer Society Medical and Editorial Content Team. Harmful Chemicals in Tobacco Products. American Cancer Society. Available online: https://www.cancer.org/cancer/cancer-causes/tobacco-and-cancer/carcinogens-found-in-tobacco-products.html (accessed on 30 January 2019).
  4. US Food and Drug Administration. Products, Ingredients & Components—Chemicals in Every Cigarette. Available online: https://www.fda.gov/tobaccoproducts/labeling/productsingredientscomponents/ucm535267.htm (accessed on 30 January 2019).
  5. US Food and Drug Administration. Products, Ingredients & Components—Chemicals in Every Puff of Cigarette Smoke. Available online: https://www.fda.gov/TobaccoProducts/Labeling/ProductsIngredientsComponents/ucm535269.htm (accessed on 30 January 2019).
  6. US Food and Drug Administration. Compliance, Enforcement & Training—Family Smoking Prevention and Tobacco Control Act—An Overview. Available online: https://www.fda.gov/tobaccoproducts/guidancecomplianceregulatoryinformation/ucm246129.htm (accessed on 30 January 2019).
  7. US Food and Drug Administration. Rules, Regulations & Guidance—Harmful and Potentially Harmful Constituents in Tobacco Products and Tobacco Smoke: Established List. Available online: https://www.fda.gov/TobaccoProducts/Labeling/RulesRegulationsGuidance/ucm297786.htm (accessed on 30 January 2019).
  8. US Food and Drug Administration. The Real Cost Campaign. Available online: https://www.fda.gov/TobaccoProducts/PublicHealthEducation/PublicEducationCampaigns/TheRealCostCampaign/default.htm (accessed on 30 January 2019).
  9. Zhao, X.; Alexander, T.N.; Hoffman, L.; Jones, C.; Delahanty, J.; Walker, M.; Berger, A.T.; Talbert, E. Youth Receptivity to FDA’s The Real Cost Tobacco Prevention Campaign: Evidence from Message Pretesting. J. Health Commun. 2016, 21, 1153–1160. [Google Scholar] [CrossRef] [PubMed]
  10. Duke, J.C.; Alexander, T.N.; Zhao, X.; Delahanty, J.C.; Allen, J.A.; MacMonegle, A.J.; Farrelly, M.C. Youth’s Awareness of and Reactions to The Real Cost National Tobacco Public Education Campaign. PLoS ONE 2015, 10, e0144827. [Google Scholar] [CrossRef] [PubMed]
  11. Huang, L.-L.; Lazard, A.J.; Pepper, J.K.; Noar, S.M.; Ranney, L.M.; Goldstein, A.O. Impact of The real cost campaign on adolescents’ recall, attitudes, and risk perceptions about tobacco use: A national study. Int. J. Environ Res. Public Health 2017, 14, 42. [Google Scholar] [CrossRef] [PubMed]
  12. Moracco, K.E.; Morgan, J.C.; Mendel, J.; Teal, R.; Noar, S.M.; Ribisl, K.M.; Hall, M.G.; Brewer, N.T. “My First Thought was Croutons”: Perceptions of Cigarettes and Cigarette Smoke Constituents Among Adult Smokers and Nonsmokers. Nicotine Tob. Res. 2016, 18, 1566–1574. [Google Scholar] [CrossRef] [PubMed]
  13. Hall, M.G.; Ribisl, K.M.; Brewer, N.T. Smokers’ and nonsmokers’ beliefs about harmful tobacco constituents: Implications for FDA communication efforts. Nicotine Tob. Res. 2014, 16, 343–350. [Google Scholar] [CrossRef] [PubMed]
  14. Francis, D.B.; Noar, S.M.; Kowitt, S.D.; Jarman, K.L.; Goldstein, A.O. Believability of new diseases reported in the 2014 Surgeon General’s Report on smoking: Experimental results from a national survey of US adults. Prev. Med. 2017, 99, 94–98. [Google Scholar] [CrossRef] [PubMed]
  15. Kelley, D.E.; Boynton, M.H.; Noar, S.M.; Morgan, J.C.; Mendel, J.R.; Ribisl, K.M.; Stepanov, I.; Nylander-French, L.A.; Brewer, N.T. Effective Message Elements for Disclosures About Chemicals in Cigarette Smoke. Nicotine Tob. Res. 2018, 20, 1047–1054. [Google Scholar] [CrossRef] [PubMed]
  16. Brewer, N.T.; Morgan, J.C.; Baig, S.A.; Mendel, J.R.; Boynton, M.H.; Pepper, J.K.; Byron, M.J.; Noar, S.M.; Agans, R.P.; Ribisl, K.M. Public understanding of cigarette smoke constituents: Three US surveys. Tob. Control 2017, 26, 592–599. [Google Scholar] [CrossRef] [PubMed]
  17. Gallopel-Morvan, K.; Moodie, C.; Hammond, D.; Eker, F.; Beguinot, E.; Martinet, Y. Consumer understanding of cigarette emission labelling. Eur. J. Public Health 2011, 21, 373–375. [Google Scholar] [CrossRef] [PubMed]
  18. Cho, Y.J.; Thrasher, J.F.; Swayampakala, K.; Lipkus, I.; Hammond, D.; Cummings, K.M.; Borland, R.; Yong, H.H.; Hardin, J.W. Does Adding Information on Toxic Constituents to Cigarette Pack Warnings Increase Smokers’ Perceptions about the Health Risks of Smoking? A Longitudinal Study in Australia, Canada, Mexico, and the United States. Health Educ. Behav. 2018, 45, 32–42. [Google Scholar] [CrossRef] [PubMed]
  19. Baig, S.A.; Byron, M.J.; Boynton, M.H.; Brewer, N.T.; Ribisl, K.M. Communicating about cigarette smoke constituents: An experimental comparison of two messaging strategies. J. Behav. Med. 2017, 40, 352–359. [Google Scholar] [CrossRef] [PubMed]
  20. Jarman, K.L.; Kowitt, S.D.; Queen, T.L.; Ranney, L.M.; Kim, K.; Jones, E.E.; Donovan, E.; Goldstein, A.O. Do Smokers Recall Source or Quitline on Cigarette Constituent Messages? Tob. Regul. Sci. 2018, 4, 66–82. [Google Scholar] [CrossRef] [PubMed]
  21. Klein, E.G.; Quisenberry, A.J.; Shoben, A.B.; Cooper, S.; Ferketich, A.K.; Berman, M.; Peters, E.; Wewers, M.E. Health Warning Labels for Smokeless Tobacco: The Impact of Graphic Images on Attention, Recall, and Craving. Nicotine Tob. Res. 2017, 19, 1172–1177. [Google Scholar] [CrossRef] [PubMed]
  22. Higgins, E.; Leinenger, M.; Rayner, K. Eye movements when viewing advertisements. Front. Psychol. 2014, 5, 210. [Google Scholar] [CrossRef] [PubMed]
  23. Meernik, C.; Jarman, K.; Wright, S.T.; Klein, E.G.; Goldstein, A.O.; Ranney, L. Eye Tracking Outcomes in Tobacco Control Regulation and Communication: A Systematic Review. Tob. Regul. Sci. 2016, 2, 377–403. [Google Scholar] [CrossRef] [PubMed]
  24. Kowitt, S.; Sheeran, P.; Jarman, K.; Ranney, L.M.; Schmidt, A.M.; Huang, L.-L.; Goldstein, A.O. Effects of framing proximal benefits of quitting and motivation to quit as a query on communications about tobacco constituents. Nicotine Tob. Res. 2017, 19, 1178–1184. [Google Scholar] [CrossRef] [PubMed]
  25. Kowitt, S.D.; Sheeran, P.; Jarman, K.L.; Ranney, L.M.; Schmidt, A.M.; Noar, S.M.; Huang, L.L.; Goldstein, A.O. Cigarette Constituent Health Communications for Smokers: Impact of Chemical, Imagery, and Source. Nicotine Tob. Res. 2019, 21, 841–845. [Google Scholar] [CrossRef] [PubMed]
  26. Baig, S.A.; Noar, S.M.; Gottfredson, N.C.; Boynton, M.H.; Ribisl, K.M.; Brewer, N.T. UNC Perceived Message Effectiveness: Validation of a Brief Scale. Ann. Behav. Med. 2019, 53, 732–742. [Google Scholar] [CrossRef] [PubMed]
  27. Tobii Support User Manual. Available online: https://www.tobiipro.com/ (accessed on 30 January 2019).
  28. Heatherton, T.F.; Kozlowski, L.T.; Frecker, R.C.; Fagerström, K.O. The Fagerström Test for Nicotine Dependence: A revision of the Fagerström Tolerance Questionnaire. Br. J. Addict. 1991, 86, 1119–1127. [Google Scholar] [CrossRef] [PubMed]
  29. Morgan, J.C.; Byron, M.J.; Baig, S.A.; Stepanov, I.; Brewer, N.T. How people think about the chemicals in cigarette smoke: A systematic review. J. Behav. Med. 2017, 40, 553–564. [Google Scholar] [CrossRef] [PubMed]
  30. Cornacchione, J.; Smith, S.W. The Effects of Message Framing Within the Stages of Change on Smoking Cessation Intentions and Behaviors. Health Commun. 2012, 27, 612–622. [Google Scholar] [CrossRef] [PubMed]
  31. Durkin, S.; Brennan, E.; Wakefield, M. Mass media campaigns to promote smoking cessation among adults: An integrative review. Tob. Control 2012, 21, 127–138. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  32. Noar, S.M. A 10-Year Retrospective of Research in Health Mass Media Campaigns: Where Do We Go From Here? J. Health Commun. 2006, 11, 21–42. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  33. Duke, J.C.; Nonnemaker, J.M.; Davis, K.C.; Watson, K.A.; Farrelly, M.C. The Impact of Cessation Media Messages on Cessation-Related Outcomes: Results from a National Experiment of Smokers. Am. J. Health Promot. 2014, 28, 242–250. [Google Scholar] [CrossRef] [PubMed]
  34. McAfee, T.; Davis, K.C.; Alexander, R.L.; Pechacek, T.F.; Bunnell, R. Effect of the first federally funded US antismoking national media campaign. Lancet 2013, 382, 2003–2011. [Google Scholar] [CrossRef]
  35. McAfee, T.; Davis, K.C.; Shafer, P.; Patel, D.; Alexander, R.; Bunnell, R. Increasing the dose of television advertising in a national antismoking media campaign: Results from a randomised field trial. Tob. Control 2017, 26, 19–28. [Google Scholar] [CrossRef] [PubMed]
  36. Noar, S.M.; Hall, M.G.; Francis, D.B.; Ribisl, K.M.; Pepper, J.K.; Brewer, N.T. Pictorial cigarette pack warnings: A meta-analysis of experimental studies. Tob. Control 2016, 25, 341–354. [Google Scholar] [CrossRef] [PubMed]
  37. Noar, S.M.; Francis, D.B.; Bridges, C.; Sontag, J.M.; Ribisl, K.M.; Brewer, N.T. The impact of strengthening cigarette pack warnings: Systematic review of longitudinal observational studies. Soc. Sci. Med. 2016, 164, 118–129. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  38. Crespo, A.; Cabestrero, R.; Grzib, G.; Quirós, P. Visual attention to health warnings in tobacco advertisements: An eye-tracking research between smokers and non-smokers. Stud. Psychol. 2007, 49, 39–51. [Google Scholar]
  39. Bansal-Travers, M.; Hammond, D.; Smith, P.; Cummings, K.M. The impact of cigarette pack design, descriptors, and warning labels on risk perception in the US. Am. J. Prev. Med. 2011, 40, 674–682. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  40. Manor, B.R.; Gordon, E. Defining the temporal threshold for ocular fixation in free-viewing visuocognitive tasks. J. Neurosci. Methods 2003, 128, 85–93. [Google Scholar] [CrossRef]
  41. Pieters, R.; Wedel, M.; Zhang, J. Optimal Feature Advertising Design under Competitive Clutter. Manag. Sci. 2007, 53, 1815–1828. [Google Scholar] [CrossRef]
  42. Rosenholtz, R. Capabilities and Limitations of Peripheral Vision. Annu. Rev. Vis. Sci. 2016, 2, 437–457. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  43. Strasburger, H.; Rentschler, I.; Jüttner, M. Peripheral vision and pattern recognition: A review. J. Vis. 2011, 11, 13. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  44. Carrasco, M. Visual attention: The past 25 years. Vis. Res. 2011, 51, 1484–1525. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  45. Hammond, D.; Wakefield, M.; Durkin, S.; Brennan, E. Tobacco packaging and mass media campaigns: Research needs for Articles 11 and 12 of the WHO Framework Convention on Tobacco Control. Nicotine Tob. Res. 2013, 15, 817–831. [Google Scholar] [CrossRef] [PubMed] [Green Version]
Ijerph 16 04435 g001a 550Ijerph 16 04435 g001b 550
Figure 1. Stimuli by Condition for eye tracking experiment.
Figure 1. Stimuli by Condition for eye tracking experiment.
Ijerph 16 04435 g001aIjerph 16 04435 g001b
Table
Table 1. Participant characteristics, n = 211.
Table 1. Participant characteristics, n = 211.
VariableN (%) or Mean (SD)
 Age, years (mean, SD)36.23 (12.43)
Sex at Birth
 Female126 (59.72%)
 Male85 (40.28%)
Sexual Identity
 Straight or Heterosexual174 (82.46%)
 Gay, Lesbian, Bisexual, Other37 (17.54%)
Ethnicity
 Not Latino/Hispanic200 (94.79%)
 Latino/Hispanic11 (5.21%)
Race
 White123 (58.29%)
 Black or African American77 (36.49%)
 Other11 (5.21%)
Education
 Greater than high school166 (78.67%)
 High school graduate or less45 (21.33%)
Income
 Below $25,000 87 (41.23%)
 $25,000–$49,99984 (39.81%)
 Greater than $50,00040 (18.96%)
 Nicotine dependence score (mean, SD) (range = 0–10)4.13 (2.59)
Table
Table 2. Descriptive statistics for self-reported outcomes by message constituent, n = 211.
Table 2. Descriptive statistics for self-reported outcomes by message constituent, n = 211.
Message
Constituent		Dwell Time (Seconds), Mean (SD)Outcome, Mean (SD)
ImageTextFull AdBelievabilityWant to QuitDiscouragement
Arsenic2.74 (3.17)1.80 (1.80)9.88 (6.96)6.07 (2.07)4.75 (2.38)4.72 (2.36)
Formaldehyde3.32 (3.89)1.97 (1.92)9.00 (6.43)6.90 (1.39)5.91 (2.01)5.73 (2.24)
Uranium2.67 (2.89)3.33 (3.54)9.50 (6.77)6.16 (1.89)5.23 (2.26)5.08 (2.39)
General4.39 (4.61)0.41 (0.65)9.66 (8.38)6.61 (1.64)4.87 (2.17)4.73 (2.29)
Note: mean dwell time for image was 3.28 (SD: 3.76), mean dwell time for text was 1.88 (SD: 2.45), and mean dwell time for the full ad was 9.51 (SD: 7.17).
Table
Table 3. Multivariable repeated measures outcomes, n = 211 participants and n = 844 observations.
Table 3. Multivariable repeated measures outcomes, n = 211 participants and n = 844 observations.
Message
Constituent		Outcome
Believability
B (p-Value)		Want to Quit
B (p-Value)		Discouragement
B (p-Value)
Arsenic (ref. general)−0.53 (p < 0.001)−0.19 (p = 0.15)0.02 (p = 0.91)
Formaldehyde (ref. general)0.30 (p = 0.02)0.97 (p < 0.001)1.05 (p < 0.001)
Uranium (ref. general)−0.42 (p < 0.001)0.21 (p = 0.17)0.43 (p = 0.01)
Level 1 image dwell time 0.00 (p = 0.81)0.00 (p = 0.93)−0.02 (p = 0.30)
Level 2 image dwell time0.04 (p = 0.36)0.05 (p = 0.37)0.08 (p = 0.16)
Level 1 text dwell time −0.01 (p = 0.82)0.05 (p = 0.07)−0.04 (p = 0.22)
Level 2 text dwell time0.01 (p = 0.89)0.07 (p = 0.45)0.00 (p = 0.99)
Message dose−0.05 (p = 0.18)0.13 (p < 0.001)0.13 (p = 0.004)
Analyses control for sex at birth, age, sexual identity, race, ethnicity, income, education, nicotine dependence score, and condition. Boldface indicates significance at p < 0.05.

Share and Cite

MDPI and ACS Style

Ranney, L.M.; Kowitt, S.D.; Queen, T.L.; Jarman, K.L.; Goldstein, A.O. An Eye Tracking Study of Anti-Smoking Messages on Toxic Chemicals in Cigarettes. Int. J. Environ. Res. Public Health 2019, 16, 4435. https://doi.org/10.3390/ijerph16224435

AMA Style

Ranney LM, Kowitt SD, Queen TL, Jarman KL, Goldstein AO. An Eye Tracking Study of Anti-Smoking Messages on Toxic Chemicals in Cigarettes. International Journal of Environmental Research and Public Health. 2019; 16(22):4435. https://doi.org/10.3390/ijerph16224435

Chicago/Turabian Style

Ranney, Leah M., Sarah D. Kowitt, Tara L. Queen, Kristen L. Jarman, and Adam O. Goldstein. 2019. "An Eye Tracking Study of Anti-Smoking Messages on Toxic Chemicals in Cigarettes" International Journal of Environmental Research and Public Health 16, no. 22: 4435. https://doi.org/10.3390/ijerph16224435

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop