Abstract
Introduction.
The FDA has proposed new product standards regarding the availability of menthol cigarettes and flavored cigars in the US. However, it is unclear whether limiting characterizing flavors in cigarettes and cigars, as proposed, or across all tobacco products, produces differential effects on the tobacco use behaviors of African American/Black (AA/B) individuals who use menthol cigarettes. This study assessed whether limiting characterizing flavors in combusted products only, or across all tobacco products, produces differential effects on the tobacco use behaviors of AA/B individuals who use menthol cigarettes.
Methods.
Adult AA/B individuals who use menthol cigarettes in the US were recruited via Qualtrics (n=373) and in Richmond, Virginia (n=206) for an online experiment from September 2021-August 2022. Participants reported how their tobacco use behaviors would change under three scenarios: maintenance of status quo, limited flavor ban (ban characterizing flavors in cigarettes and cigars), and comprehensive flavor ban (ban characterizing flavors in all tobacco products). Seemingly unrelated regressions compared differences in expected responses to policy scenarios (p<0.05).
Results.
Both flavor ban scenarios resulted in higher quitting intentions for cigarettes and “all tobacco products” relative to status quo (ps<0.05). The comprehensive ban resulted in greater intentions to “quit all tobacco products” and lower intentions to switch to certain alternative products (e.g., electronic cigarettes, smokeless tobacco, heated tobacco products) than the limited ban (ps<0.05).
Conclusions.
AA/B individuals who use menthol cigarettes appear more likely to quit smoking if characterizing flavors in combusted products (e.g., menthol cigarettes) are banned, regardless of if characterizing flavors are available in non-combusted alternative tobacco products.
Introduction
In 2021, the US Food and Drug Administration (FDA) announced its intent to ban menthol as a characterizing flavor in cigarettes and all characterizing flavors in cigars.1 Due in part to menthol’s sensory and pharmacologic effects,2,3 those that use menthol (relative to non-menthol) cigarettes exhibit higher levels of nicotine dependence and greater difficulty quitting.4 Banning menthol in cigarettes could improve public health through primary and secondary disease prevention and redress tobacco-related health inequities. For example, nearly 90% of African American/Black (AA/B) individuals who smoke use menthol cigarettes compared to ~35% of similar White individuals.1,5,6
To inform rulemaking, the FDA needs evidence on the expected responses of people who use menthol cigarettes7 to the proposed ban and to potential modifications of the bans’ scope. Prior surveys suggest that many people who use menthol cigarettes (20–50%) would attempt to quit following a ban, while others would switch to using alternative (i.e., non-cigarette) tobacco products (10–20%) such as electronic (e-) cigarettes.8–10 It is unknown whether restricting access to characterizing flavors in alternative products would encourage more people who use menthol cigarettes to quit, or encourage substitution with combustible products. Such information is critical because the success of the FDA’s proposed ban hinges on optimizing quitting and switching to potentially-lower harm product among individuals affected by the ban.
Here, an online experiment assessed how AA/B individuals who use menthol cigarettes would respond to three policy scenarios (status quo, limited flavor ban [ban characterizing flavors in cigarettes and cigars], and comprehensive flavor ban [ban characterizing flavors in all tobacco/nicotine products]). It was hypothesized that AA/B individuals who use menthol cigarettes would be most likely to quit and least likely to switch to alternative products in the comprehensive flavor ban scenario.
Methods
Study Sample
This study was approved by the Virginia Commonwealth University IRB (HM299222619) and participants provided informed consent prior to enrollment. Data were collected September 30, 2021-August 11, 2022 from two samples: a “community-sample” and a “national-sample.” The community-sample was recruited through local and online advertising in Richmond, VA. The national-sample was recruited via a Qualtrics’ curated blended-panel. Eligible participants had to live in the US, be at least 18 years old, identify as AA/B, report every day or someday cigarette smoking in the past 30 days, and use a menthol or mint flavored cigarette brand regularly. Of the 1508 individuals screened, the final analytic sample size was N=579 (community-sample: N=206; national-sample: N=373) (see Appendix Figure 1). Participants were compensated with electronic giftcards ($5/screen; $35/survey).
Measures
Eligible participants were randomized to view one-of-three “information frames”: no frame (blank page; n=182), a “general” frame (information related to menthol cigarette harms and the consequences of a ban among all US adults; n=192), and an “enhanced” frame (general frame adapted to center on AA/B individuals who smoke; n=205). These frames were developed to reflect the information available on government websites (see Appendix Table 1). Participants then reported, using visual analog scales (0 [Not at all likely] – 100 [Extremely likely]), how 13 of their tobacco use behaviors would change under three scenarios: status quo, limited flavor ban, and comprehensive flavor ban (see Appendix Table 2). The status quo scenario was always presented first, but the order for the two ban scenarios was randomized. Neither the informational frames nor the scenario descriptions/outcomes were pretested prior to the study. Importantly, the outcome regarding willingness to “try to buy or import menthol cigarettes from unlicensed retailers” did not describe the legality of such behavior under these hypothetical scenarios.
Statistical Analysis
Because outcomes were believed to be interrelated, seemingly unrelated regressions jointly estimated a series of 13 linear regressions (one for each outcome) with a generalized least squares estimator.11 Regressions controlled for participants’ scenario order, age, income, gender, 4-item PROMIS nicotine dependence score,12 and sample (community vs. national). Standard errors were clustered at the individual-level. All analyses were conducted in Stata SE 17 (College Station, TX). Alpha was set to 0.05.
Results
The majority of the analytic sample were from the national-sample (64.42%) and identified as male (52.16%), with a median age of 35 and annual income of $38,000. The average PROMIS cigarette dependence raw score was 12.66 (SD: 3.92) and 82.5% used cigarettes daily. Regression results for selected behavioral outcomes are presented in Table 1 (see also Appendix Table 3 [regressions] and Appendix Table 4 [mean outcomes]). Relative to the status quo, AA/B individuals who use menthol cigarettes were more likely to quit using cigarettes and “all tobacco products” in both the limited and comprehensive ban scenarios (ps<0.05). Notably though, participants were more likely to quit all tobacco products and less likely to switch to or increase use of e-cigarettes, heated tobacco products, and smokeless tobacco, in the comprehensive ban than the limited ban (ps<0.05). No significant differences were detected in participant’s hypothetical likelihood of accessing menthol cigarettes through unlicensed retailers between the two ban scenarios (p>0.05). Conclusions were insensitive to stratification by sample, imputing missing values, or to using all available data for each regression.
Table 1.
Anticipated Responses to Hypothetical Bans on Flavored Tobacco Products
| Variable | Quit smoking cigarettes | Quit all tobacco products | Switch to or increase use of electronic cigarettes | Switch to or increase use of heated tobacco products | Switch to or increase use of smokeless tobacco | Buy or import menthol cigarettes from unlicensed retailers |
|---|---|---|---|---|---|---|
| Policy scenario (REF: Status quo) | ||||||
| Limited ban |
10.55
***
(1.23) |
8.59
***
(1.18) |
11.42
***
(1.30) |
2.99
**
(1.02) |
3.75
**
(1.11) |
8.50
***
(1.22) |
| Comprehensive ban |
11.86
***
(1.29) |
11.00
***
(1.25) |
5.30
***
(1.30) |
0.67 (0.91) |
0.02 (1.00) |
9.14
***
(1.30) |
| aLimited vs. Comprehensive | 0.1774 | 0.0235 | <0.0001 | 0.0122 | 0.0004 | 0.5044 |
| Information frame (REF: No Frame) | ||||||
| General frame | 5.13 (2.99) |
7.17
*
(3.07) |
0.29 (2.67) |
2.11 (2.42) |
1.36 (2.44) |
−2.39 (3.02) |
| Enhanced frame | 4.95 (3.02) |
5.11 (3.09) |
2.06 (2.76) |
1.09 (2.51) |
−0.26 (2.50) |
−3.06 (3.08) |
| Scenario order (REF: Limited, Comprehensive) | ||||||
| Comprehensive, Limited | 1.15 (2.46) |
0.24 (2.50) |
−1.73 (2.21) |
−1.54 (2.03) |
−0.12 (2.03) |
−0.48 (2.46) |
| Sample (REF: National) | ||||||
| Community | −16.51*** (2.74) |
−16.17*** (2.77) |
−2.82 (2.37) |
−2.39 (2.38) |
−3.40 (2.27) |
5.78
*
(2.70) |
| Gender (REF: Male) | ||||||
| Female | 4.37 (2.50) |
4.02 (2.56) |
1.62 (2.23) |
−5.26* (2.07) |
−4.91* (2.08) |
−0.08 (2.49) |
| PROMIS total score | −0.58 (0.33) |
−0.60 (0.34) |
1.47
***
(0.30) |
1.13
***
(0.27) |
0.96
***
(0.27) |
1.72
***
(0.33) |
| Age (years) | 0.11 (0.10) |
0.16 (0.11) |
−0.74*** (0.09) |
−0.66*** (0.08) |
−0.70*** (0.08) |
−0.39*** (0.11) |
| Income ($10,000 units) | 0.28 (0.1.’ |
0.20 (0.18) |
0.24 (0.13) |
0.23 (0.17) |
0.32
*
(0.14) |
0.35
*
(0.17) |
| Constant |
44.97
***
(6.57) |
41.91
***
(6.91) |
45.81
***
(6.13) |
38.06
***
(5.38) |
43.23
***
(5.33) |
24.73
***
(6.90) |
NOTE: Boldface indicates statistical significance:
p< 0.001
p<0.01,
p<0.05.
Coefficients and standard errors presented here are from seemingly unrelated linear regression with participant-level clustered standard errors.11 Sample size for each regression was based on a common support sample of 579 participants (206 from community-sample, 373 from national-sample). Each outcome variable was based on responses to visual analog scales, ranging from 0 [Not at all likely] to 100 [Extremely likely]. See Appendix Table 3 for results on remaining outcomes.
Limited vs. Comprehensive: displays the p-value from a post-estimation comparison of the coefficients associated with the limited and comprehensive ban scenarios to determine if a statistically significant difference existed between the two estimates.
Discussion
Findings from this blended national and community sample of AA/B individuals who use menthol cigarettes contribute several strengths and findings to the literature. First, focusing recruitment on AA/B individuals who use menthol cigarettes is an important contribution as responses to a menthol ban among impacted AA/B individuals may differ from the general population.13–15 Additionally, these analyses document anticipated responses across three hypothetical policy scenarios and capture a wider range of alternative products and behaviors than previous studies.8–10 Results suggest that by restricting access to characterizing flavors in alternative tobacco products concurrent to their removal in combusted products, policymakers may impair switching to non-combusted alternatives (e.g., e-cigarettes) but encourage more AA/B individuals who currently use menthol cigarettes to quit using tobacco altogether. Such an outcome could improve public health and narrow tobacco-related health disparities for AA/B individuals who smoke cigarettes.
These findings are largely consistent with previous research: a menthol ban would prompt a plurality of AA/B individuals who use menthol cigarettes to quit, but some would switch to alternative products (e.g., e-cigarettes) or continue smoking.14,15 Importantly, this work adds to the literature by demonstrating that quit intentions may not differ by whether flavored alternatives are available, but intentions to switch do. Thus, the public health benefits among AA/B individuals who use menthol cigarettes of permitting flavored alternatives to remain on the market following a ban on menthol cigarettes may be limited.
Limitations
Several limitations of these analyses should be noted. Foremost, data were generated from a cross-sectional, online convenience sample of self-reported AA/B individuals who use menthol cigarettes reporting potential changes in behavior under hypothetical policy scenarios. Thus, results should not be interpreted as “representative” and the behavioral “intentions” reported here may not equate to actual behaviors. However, hypothetical intentions to quit may actually underestimate true quit attempts following a menthol ban.16 Additionally, the text-based information frames and policy scenarios tested here were not pretested prior to implementation. Moreover, participant’s understanding of the scenarios was not assessed, no standard definitions of “all flavors” and “unflavored” were provided, and the intensity with which participants might use tobacco products in the hypothetical scenarios was not measured or defined. Finally, interest in switching to non-menthol cigarettes was not explicitly evaluated and can only be imprecisely extrapolated from the “smoke fewer cigarettes” outcome.
Conclusions
FDA’s proposed menthol cigarette and flavored cigar bans are likely to be effective policy levers for disease prevention but could be complemented by regulatory policies that limit access to flavors in alternative tobacco product.
Supplementary Material
Appendix Figure 1. Determination of Sample Size and Data Cleaning
Appendix Table 1. Information Frame Text
Appendix Table 2. Tobacco Regulatory Policy Scenarios Text and Behavioral Outcomes
Appendix Table 3. Anticipated Responses to Hypothetical Bans on Flavored Tobacco Products: Outcomes Not Presented in Table 1.
Appendix Table 4. Mean Outcomes for Each of The 13 Potential Behavioral Outcomes, Stratified by Sample
Acknowledgements
Acknowledgements & Thanks:
We would like to acknowledge and thank the participants of this study for their time, effort, and openness in sharing their experience and perspective with our team. We would also like to thank prior and current members of the Virginia Commonwealth University’s “HealthHub at 25th” (https://community.vcu.edu/health-hub/) for their contributions to the study and assistance in recruiting for this project.
Study Funding:
This work was supported by grant number U54DA036105 from the National Institute on Drug Abuse (all authors). Additional support was provided by UL1TR002649 from the National Center for Research Resources. The content is solely the responsibility of the authors and does not necessarily represent the views of the NIH or the FDA. The funding source had no other role than financial support.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Author statement File – CRediT
Augustus M White: Conceptualization, Methodology, Software, Verification, Formal Analysis, Data Curation, Writing – Original Draft, Visualization.
Akira Goden: Conceptualization, Methodology, Data Curation, Writing – Review & Editing.
Alyssa Rudy: Conceptualization, Methodology, Data Curation, Writing – Review & Editing.
Harlean Bajwa: Conceptualization, Methodology, Data Curation, Writing – Review & Editing.
Mignonne Guy: Conceptualization, Methodology, Data Curation, Writing – Review & Editing, Supervision, Project Administration.
Kristina Hood: Conceptualization, Methodology, Data Curation, Writing – Review & Editing, Supervision, Project Administration.
Caroline O Cobb: Conceptualization, Methodology, Data Curation, Writing – Review & Editing, Supervision, Project Administration, Funding acquisition.
Andrew J Barnes: Conceptualization, Methodology, Data Curation, Writing – Review & Editing, Supervision, Project Administration, Funding acquisition.
Conflicts of Interest: The authors have no affiliations with the tobacco industry, nor other conflicts of interest to disclose (all authors).
Financial Disclosure: No financial disclosures have been reported by the authors of this paper.
References
- 1.US Food and Drug Administration. FDA commits to evidence-based actions aimed at saving lives and preventing future generations of smokers. Silver Springs, MD: FDA. https://www.fda.gov/news-events/press-announcements/fda-commits-evidence-based-actions-aimed-saving-lives-and-preventing-future-generations-smokers. Published 2021. Accessed August 18, 2022. [Google Scholar]
- 2.Wickham RJ. The biological impact of menthol on tobacco dependence. Nicotine Tob Res. 2020;22(10):1676–1684. doi: 10.1093/ntr/ntz239. [DOI] [PubMed] [Google Scholar]
- 3.Ahijevych K, Garrett BE. The role of menthol in cigarettes as a reinforcer of smoking behavior. Nicotine Tob Res. 2010;12(Suppl 2):S110–S116. doi: 10.1093/ntr/ntq203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Fagan P, Moolchan ET, Hart A Jr., et al. Nicotine dependence and quitting behaviors among menthol and non-menthol smokers with similar consumptive patterns. Addiction. 2010;105(Suppl 1):55–74. doi: 10.1111/j.1360-0443.2010.03190.x. [DOI] [PubMed] [Google Scholar]
- 5.Weinberger AH, Giovenco DP, Zhu J, Lee J, Kashan RS, Goodwin RD. Racial/ethnic differences in daily, nondaily, and menthol cigarette use and smoking quit ratios in the United States: 2002 to 2016. Prev Med. 2019;125:32–39. doi: 10.106/j.ypmed.2019.04.009. [DOI] [PubMed] [Google Scholar]
- 6.Levy DT, Meza R, Yuan Z, et al. Public health impact of a US ban on menthol in cigarettes and cigars: a simulation study. Tob Control. 2021;0:1–8 (Epub ahead of print [3 October 2022]). doi: 10.1136/tobaccocontrol-2021-056604 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Williamson TJ, Riley KE, Carter-Harris L, Ostroff JS. Changing the language of how we measure and report smoking status: implications for reducing stigma, restoring dignity, and improving the precision of scientific communication. Nicotine Tob Res. 2020;22(12):2280–2282. doi: 10.1093/ntr/ntaa141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Yang Y, Lindblom EN, Ward KD, Salloum RG. How smokers of menthol cigarettes and flavored cigars might respond to FDA’s proposed bans. Nicotine Tob Res. 2022; 24(10):1645–1653. doi: 10.1093/ntr/ntac078. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Pearson JL, Abrams DB, Niaura RS, Richardson A, Vallone DM. A ban on menthol cigarettes: impact on public opinion and smokers’ intentions to quit. Am J Public Health. 2012;102(11):e107–e114. doi: 10.2105/AJPH.2012.300804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wackowski OA, Delnevo CD, Pearson JL. Switching to e-cigarettes in the event of a menthol cigarette ban. Nicotine Tob Res. 2015;17(10):1286–1287. doi: 10.1093/ntr/ntv021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zellner A. An efficient method of estimating seemingly unrelated regressions and tests for aggregation bias. J Am Stat Assoc. 1962;57(298):348–368. doi: 10.1080/01621459.1962.10480664. [DOI] [Google Scholar]
- 12.Shadel WG, Edelen MO, Tucker JS, Stucky BD, Hansen M, Cai L. Development of the PROMIS® nicotine dependence item banks. Nicotine Tob Res. 2014;16(Suppl 3):S190–S201. doi: 10.1093/ntr/ntu032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Gardiner P. The African Americanization of menthol cigarette use in the United States. Nicotine Tob Res. 2004;6(1):55–65. doi: 10.1080/14622200310001649478 [DOI] [PubMed] [Google Scholar]
- 14.Levy DT, Cadham CJ, Sanchez-Romero LM, Knoll M, Travis N, Yuan Z, Li Y, Mistry R, Douglas CE, Tam J, Sertkaya A, Warner KE, Meza R. An expert elicitation on the effects of a ban on menthol cigarettes and cigars in the United States. Nicotine Tob Res. 2021;23(11):1911–1920. doi: 10.1093/ntr/ntab121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Issabakhsh M, Meza R, Li Y, Yuan Z, Sanchez-Romero LM, Levy DT. The public health impact of a US menthol cigarette ban on the non-hispanic black population: A simulation study. Tob Control. 2022; Epub ahead of print [4 October 2022]. doi: 10.1136/tobaccocontrol-2022-057298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chaiton MO, Nicolau I, Schwartz R, et al. Ban on menthol-flavoured tobacco products predicts cigarette cessation at 1 year: A population cohort study. Tob Control. 2020;29(3):341–347. doi: 10.1136/tobaccocontrol-2018-054841. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Appendix Figure 1. Determination of Sample Size and Data Cleaning
Appendix Table 1. Information Frame Text
Appendix Table 2. Tobacco Regulatory Policy Scenarios Text and Behavioral Outcomes
Appendix Table 3. Anticipated Responses to Hypothetical Bans on Flavored Tobacco Products: Outcomes Not Presented in Table 1.
Appendix Table 4. Mean Outcomes for Each of The 13 Potential Behavioral Outcomes, Stratified by Sample