Abstract
Introduction
Tobacco companies have a history of making health claims about their new products. Such claims are now regulated by the US Food and Drug Administration. We examined consumer interest in hypothetical modified risk tobacco products (MRTPs) among current, former, and never established smokers and examined whether interest was associated with beliefs about tobacco and cancer.
Methods
Data were analyzed from the US nationally representative 2015 Health Information National Trends Survey (HINTS-FDA 2015; N = 3738). Interest in hypothetical MRTPs was assessed by asking participants their likelihood of using tobacco products claiming to be less addictive and less harmful than other products.
Results
About half of current smokers and a tenth of both former and never smokers reported they were “somewhat” or “very” likely to try hypothetical MRTPs claiming to be less harmful or less addictive. Female smokers, former smokers with lower smoking harm perceptions, and never smokers who are young adults or without college education expressed more interest in these products. Interest in using these products was positively associated with believing that smoking status is a changeable individual characteristic and that it is possible for tobacco products to be made without some harmful chemicals.
Conclusion
We identified several subgroups of current, former, and never smokers who may be particularly affected by the marketing of MRTPs and therefore important to study to inform models of the potential population health impact of authorizing the marketing of MRTPs.
Implications
Findings about interest in hypothetical MRTPs can inform models of how the marketing of MRTPs could affect population health. Understanding which subgroups are particularly interested in MRTPs can help determine who might be important to study to inform these models. We identified several groups who may warrant specific attention: smokers who are female, former smokers who hold low harm perceptions of smoking, never smokers who are young adults or have a high school education or less, people who believe that smoking is a changeable individual characteristic, and people who believe that it is possible to make low chemical tobacco products.
Introduction
The tobacco industry has had a history of marketing new products as less harmful or less addictive compared to conventional cigarettes in response to public concerns about smoking’s health effects.1,2 Filtered, “low tar,” and “light” cigarettes were marketed as less harmful than other cigarettes, despite not being less harmful.3–5 Products developed by tobacco companies ostensibly to be less harmful emerged on the market in the late 1990s and early 2000s6 and were termed potentially reduced exposure products (PREPs) by researchers.6,7 These products included cigarettes made with modified tobacco that reportedly contained lower levels of toxic substances (eg, Omni cigarettes) and cigarette-like products that delivered nicotine with less combustion than conventional cigarettes (eg, Eclipse). More recently, e-cigarettes had been advertised as a reduced harm alternative to cigarettes,8,9 before they were regulated by Food and Drug Administration (FDA).
Since 2009, the FDA has had regulatory oversight over tobacco products, and the marketing of regulated products with reduced harm claims has been prohibited without an FDA-issued modified risk tobacco product (MRTP) marketing authorization order.10 This order can be issued only if several requirements are met. One requirement is that the marketing of the product must benefit the health of the population as a whole, considering the effects on tobacco users and nonusers.10 As of the date this paper was published, FDA has not issued any MRTP marketing authorization orders, though several MRTP applications have been submitted to FDA for review.
Understanding who is interested in MRTPs can help inform how the marketing of authorized MRTPs could affect population health. Little research has been conducted on what subgroups are interested in MRTPs; however, one recent study examined demographic correlates of interest in hypothetical MRTPs (products claiming to be less addictive or less harmful than other tobacco products).11 The study found that among established cigarette smokers, interest in hypothetical MRTPs was higher among people who were female and older. The study also found that among never cigarette smokers, young adults, Asian Americans, African Americans, and people with low educational attainment were more likely to be interested in hypothetical MRTPs.
Prior research related to interest in an earlier wave of products, PREPs, may also help inform how the marketing of authorized MRTPs could affect population health. Several national surveys in the early 2000s found that few smokers had tried a PREP (2%–13%)12 though about half reported interest in trying one.13,14 National surveys found that daily smokers, smokers of “light” cigarettes, and smokers of at least a pack a day were more interested in trying PREPs compared with other smokers.12,13 National surveys also found that certain demographic groups expressed more interest in trying PREPs, particularly females, Whites,13 and people with low income.14 PREP use was higher in several demographic groups, including younger smokers, Whites, and those with either a GED or some college education (compared to those with a high school diploma or less than high school education).12
Smokers who were concerned about the health effects of their smoking may have been especially interested in PREPs. National surveys found that compared to smokers not interested in PREPs, smokers interested in PREPs thought about cancer more often, thought they had a higher risk of lung cancer, endorsed fatalistic beliefs about cancer,13 and were more concerned about smoking’s health effects.14,15 A focus group study16 and a national survey13 found that many smokers who had tried a PREP reported doing so to reduce the health risks of smoking or to quit smoking. Also, in national surveys, smokers who had contemplated quitting15 and made a quit attempt14 were more interested in trying PREPs.
The Current Study
This study examined US adults’ interest in products marketed as modified risk. Specifically, we assessed adults’ interest in two types of hypothetical MRTPs. We describe these products as “hypothetical MRTPs” to emphasize that this study did not assess interest in a specific product with a claim or FDA MRTP authorization order—rather, this study assessed interest in general “tobacco products” claiming to be “less harmful” or “less addictive.” The study addresses two gaps in the literature. First, with some exceptions,13–15 little research has assessed the extent to which beliefs about smoking, tobacco, or cancer relate to interest in PREPs or MRTPs. Second, most research conducted on PREPs or MRTPs has excluded never smokers.12,14,16–18 Identifying which people (including former and never smokers) may be particularly affected by the marketing of MRTPs could help inform FDA’s understanding of the potential impact of marketing such products on population health.
The purpose of this study was to identify demographic and belief correlates of interest in hypothetical MRTPs. We examined whether interest in hypothetical MRTPs was associated with beliefs about whether smoking is a permanent or changeable individual characteristic, beliefs about tobacco products’ harms and chemicals, and beliefs about the preventability of cancer (ie, cancer fatalism).19 To inform how the marketing of MRTPs would affect population health, we assessed these associations separately for current, former, and never smokers.
Methods
Participants and Design
We analyzed data from the 2015 Health Information National Trends Survey conducted by the National Cancer Institute (NCI) in partnership with FDA (HINTS-FDA 2015). HINTS-FDA 2015 (N = 3738) is a cross-sectional, probability-based nationally representative survey of US non-institutionalized civilians aged 18 or older. HINTS-FDA 2015 collected data through self-administered mail surveys sent to a random sample of non-vacant residential addresses. The overall weighted response rate was 33%. Additional methodological information is available elsewhere.20
Measures
Interest in Hypothetical MRTPs
Interest in hypothetical MRTPs was measured by two items, “If a tobacco product made a claim that it was less addictive than other tobacco products, how likely would you be to use that product?” and “If a tobacco product made a claim that it was less harmful to health than other tobacco products, how likely would you be to use that product?” Response options for both were Very likely, Somewhat likely, Somewhat unlikely, and Very unlikely and were reverse coded such that higher scores reflected greater interest.
Smoking Status
Respondents who had not smoked at least 100 cigarettes in their entire life were classified as never established smokers. Respondents who had smoked at least 100 cigarettes in their life but currently smoked “Not at all” were classified as former established smokers. Respondents who had smoked at least 100 cigarettes in their life and currently smoked “Some days” or “Every day” were classified as current established smokers.
Demographics
Demographic variables were recoded from original response options20 into discrete categories such that unweighted sample size was sufficient for analysis (n ≥ 50 per cell). These included sex (male, female), sexual identity (heterosexual, lesbian, gay, bisexual), race/ethnicity (Hispanic, White, Black, all others), and educational attainment (high school diploma, GED, or less; some college, vocational, or technical training; college graduate; postgraduate). Age was dichotomized because we were interested in comparing a high risk group, young adults aged 18–24, with older adults (aged 25+).
Beliefs
The belief that smoking is not a changeable characteristic was measured with a single item, “Smoking behavior is something basic about a person that they can’t change very much” (Strongly agree, Somewhat agree, Somewhat disagree, Strongly disagree).
Two items assessed harm beliefs: (1) “In your opinion, do you think that some types of cigarettes are less harmful to a person’s health than other types?” (Yes, No) and (2) “How much do you think people harm themselves when they smoke 10 or more cigarettes every day?” (No harm, Little harm, Some harm, A lot of harm).
Two items measured beliefs about chemicals in tobacco products: (1) “In the past year, how often have you thought about the chemicals contained in tobacco products?” (Never, Rarely, Sometimes, Often); and (2) “How likely do you think it is that tobacco products could be made without some of the chemicals that are harmful to health? (Very likely, Somewhat likely, Somewhat unlikely, Very unlikely).
Four items measured cancer fatalism: (1) “Cancer is most often caused by a person’s behavior or lifestyle” (reverse coded); (2) “It seems like everything causes cancer”; (3) “There’s not much you can do to lower your chances of getting cancer”; and (4) “There are so many different recommendations about preventing cancer, it’s hard to know which ones to follow.” Response options for all questions were Strongly agree, Somewhat agree, Somewhat disagree, and Strongly disagree. All likelihood and agreement responses were coded such that higher scores indicated more affirmative responses (ie, more agreement or higher likelihood).
Data Analysis
Analyses were conducted using SAS 9.3 and SAS-callable SUDAAN 11.0. Jackknife replicate weights were used in all analyses as recommended20 to generate nationally representative estimates and account for the complex sampling design. Missing data were handled using pairwise deletion to retain as many cases as possible.
We conducted weighted linear regressions to examine whether smoking status (never, former, current) predicted interest in less addictive and less harmful hypothetical MRTPs. Finding substantial differences in interest based on smoking status, we then conducted a series of regression analyses stratified by smoking status to assess whether demographic factors and beliefs predicted interest in each hypothetical MRTP. First, single-predictor models assessed the bivariate association between each demographic and belief variable and each dependent variable (interest in less addictive and less harmful hypothetical MRTPs). Second, all demographic and belief variables were entered into one model for each hypothetical MRTP to assess the unique contribution of each demographic and belief variable to explaining variance in product interest.
All of the above analyses used the full range of responses for interest in hypothetical MRTPs (ie, each variable included four levels: Very unlikely, Somewhat unlikely, Somewhat likely, or Very likely to use the product). For purely descriptive purposes, we also conducted analyses in which we dichotomized interest in each hypothetical MRTP. Specifically, we report the percentage of people who responded in the top two categories (ie, Somewhat likely or Very likely to use). We describe these individuals as having expressed an interest in using the hypothetical MRTP, and we present their percentages as a way to help readers understand the distribution of responses across smoking status groups.
Results
Table 1 provides a description of the demographic characteristics of the overall sample and by smoking status.
Table 1.
Description of Sample by Smoking Status
| Weighted % (unweighted n) | ||||
|---|---|---|---|---|
| Total | Never smokers | Former smokers | Current smokers | |
| Total | 60.3 (2041) | 24.9 (1132) | 14.8 (495) | |
| Age | ||||
| Younger than 25 | 10.6 (108) | 14.2 (81) | 2.8 (10) | 9.2 (16) |
| 25 or older | 89.4 (3520) | 85.8 (1902) | 97.2 (1089) | 90.8 (470) |
| Sex | ||||
| Male | 49.1 (1497) | 45.3 (717) | 56.2 (545) | 52.1 (208) |
| Female | 50.9 (2018) | 54.7 (1200) | 43.8 (520) | 48.0 (265) |
| Education | ||||
| HS education or less | 31.9 (964) | 27.5 (441) | 31.6 (289) | 49.0 (212) |
| Some college | 32.8 (1132) | 29.9 (547) | 38.5 (389) | 37.2 (178) |
| College degree | 20.3 (906) | 24.1 (564) | 18.3 (259) | 7.9 (70) |
| Graduate degree | 15.0 (672) | 18.5 (459) | 11.6 (174) | 5.9 (28) |
| Race/ethnicity | ||||
| White | 65.2 (2849) | 56.9 (1502) | 81.5 (936) | 72.3 (366) |
| Latino | 15.2 (241) | 19.6 (157) | 8.2 (47) | 8.9 (31) |
| Black | 12.7 (287) | 14.3 (169) | 6.4 (58) | 15.9 (52) |
| Other | 6.9 (255) | 9.2 (156) | 3.9 (60) | 2.9 (36) |
| Sexual identity | ||||
| Straight/heterosexual | 95.7 (3405) | 96.2 (1849) | 95.6 (1049) | 94.6 (448) |
| Lesbian, gay, or bisexual | 4.3 (105) | 3.8 (51) | 4.4 (32) | 5.4 (18) |
Interest in Hypothetical MRTPs by Smoking Status
Overall, levels of interest in using less addictive and less harmful MRTPs were similar. Tables 2 and 3 display the percentages of respondents expressing interest in using the hypothetical MRTP by smoking and demographic group. For both types of hypothetical MRTPs, about half of smokers reported being interested (answering either Somewhat or Very likely to use the product), while one-tenth or less of former and never smokers reported being interested.
Table 2.
Demographic Predictors of Interest in Using a Hypothetical Less Addictive MRTP (Weighted)
| Variable | Interesteda, % | Single predictor | All predictorsb |
|---|---|---|---|
| B (95% CI) | B (95% CI) | ||
| Never smokers | 7 | R 2 = .15 | |
| Age | |||
| Younger than 25 | 13 | 0.34 (0.03, 0.69)* | 0.22 (−0.12, 0.57) |
| 25 or older | 6 | Referent | Referent |
| Sex | |||
| Male | 5 | Referent | Referent |
| Female | 8 | 0.04 (−0.08, 0.16) | 0.01 (−0.11, 0.14) |
| Education | |||
| HS education or less | 12 | Referent | Referent |
| Some college | 4 | −0.23 (−0.43, −0.03)* | −0.17 (−0.35, 0.01)# |
| College degree | 4 | −0.25 (−0.45, −0.04)* | −0.14 (−0.28, 0.01)# |
| Graduate degree | 7 | −0.19 (−0.41, 0.03)# | −0.06 (−0.19, 0.08) |
| Race/ethnicity | |||
| White | 5 | Referent | Referent |
| Latino (vs. White) | 9 | 0.17 (−0.04, 0.39) | 0.03 (−0.15, 0.22) |
| Black (vs. White) | 9 | 0.18 (−0.11, 0.46) | −0.04 (−0.11, 0.03) |
| Other (vs. White) | 9 | 0.14 (−0.03, 0.31) | 0.14 (−0.08, 0.35) |
| Sexual identity | |||
| Heterosexual | 6 | Referent | Referent |
| Lesbian, gay, or bisexual | 4 | −0.09 (−0.24, 0.07) | 0.02 (−0.12, 0.17) |
| Former smokers | 8 | R 2 = .09 | |
| Age | |||
| Younger than 25c | 6 | 0.21 (−0.38, 0.80) | −0.13 (−0.89, 0.62) |
| 25 or older | 8 | Referent | Referent |
| Sex | |||
| Male | 7 | Referent | Referent |
| Female | 8 | −0.02 (−0.13, 0.09) | 0.04 (−0.12, 0.19) |
| Education | |||
| HS education or less | 10 | Referent | Referent |
| Some college | 7 | −0.10 (−0.32, 0.11) | −0.01 (−0.22, 0.19) |
| College degree | 10 | −0.05 (−0.29, 0.20) | 0.13 (−0.10, 0.37) |
| Graduate degree | 4 | −0.19 (−0.41, 0.04)# | −0.04 (−0.22, 0.14) |
| Race/ethnicity | |||
| White | 7 | Referent | Referent |
| Latinoc (vs. White) | 14 | 0.09 (−0.22, 0.41) | 0.16 (−0.26, 0.59) |
| Black (vs. White) | 5 | 0.02 (−0.18, 0.23) | 0.16 (−0.34, 0.66) |
| Other (vs. White) | 6 | 0.12 (−0.12, 0.37) | 0.12 (−0.15, 0.39) |
| Sexual identity | |||
| Heterosexual | 7 | Referent | Referent |
| Lesbian, gay, or bisexual c | 9 | −0.04 (−0.35, 0.27) | 0.04 (−0.36, 0.44) |
| Smokers | 46 | R 2 = .06 | |
| Age | |||
| Younger than 25c | 37 | −0.11 (−0.87, 0.64) | −0.08 (−0.77, 0.61) |
| 25 or older | 46 | Referent | Referent |
| Sex | |||
| Male | 39 | Referent | Referent |
| Female | 54 | 0.40 (0.03, 0.78)* | 0.47 (0.13, 0.81)* |
| Education | |||
| HS education or less | 46 | Referent | Referent |
| Some college | 50 | −0.04 (−0.46, 0.38) | −0.02 (−0.49, 0.44) |
| College degree | 32 | −0.29 (−0.72, 0.14) | −0.43 (−0.89, −0.03)# |
| Graduate degreec | 44 | 0.26 (−0.26, 0.78) | 0.36 (−0.43, 1.16) |
| Race/ethnicity | |||
| White | 47 | Referent | Referent |
| Latinoc (vs. White) | 38 | −0.12 (−0.71, 0.48) | −0.21 (−1.17, 0.76) |
| Black (vs. White) | 40 | 0 (−0.51, 0.51) | 0.31 (−0.31, 0.93) |
| Otherc (vs. White) | 62 | 0.08 (−0.41, 0.57) | −0.01 (−0.65, 0.63) |
| Sexual identity | |||
| Heterosexual | 47 | Referent | Referent |
| Lesbian, gay, or bisexualc | 36 | −0.49 (−1.48, 0.50) | −0.45 (−1.43, 0.53) |
CI, confidence interval; MRTP, modified risk tobacco product.
aResponded they were Somewhat likely or Very likely to use the product.
cSample size for this subgroup was less than 50; interpret results with caution.
# p < .10. *p < .05.
Table 3.
Demographic Predictors of Interest in Using a Hypothetical Less Harmful MRTP (Weighted)
| Variable | Interesteda, % | Single predictor | All predictorsb |
|---|---|---|---|
| B (95% CI) | B (95% CI) | ||
| Never smokers | 8 | R 2 = .17 | |
| Age | |||
| Younger than 25 | 18 | 0.38 (−0.01, 0.76)# | 0.29 (−0.13, 0.70) |
| 25 or older | 7 | Referent | Referent |
| Sex | |||
| Male | 8 | Referent | Referent |
| Female | 8 | 0.01 (−0.17, 0.20) | −0.03 (−0.19, 0.12) |
| Education | |||
| HS education or less | 14 | Referent | Referent |
| Some college | 4 | −0.25 (−0.50, 0.01)# | −0.13 (−0.32, 0.07) |
| College degree | 7 | −0.21 (−0.47, 0.06) | −0.03 (−0.20, 0.14) |
| Graduate degree | 7 | −0.22 (−0.49, 0.05) | −0.01 (−0.15, 0.13) |
| Race/ethnicity | |||
| White | 5 | Referent | Referent |
| Latino (vs. White) | 12 | 0.18 (−0.08, 0.44) | 0.03 (−0.18, 0.24) |
| Black (vs. White) | 12 | 0.23 (−0.06, 0.53) | 0.10 (−0.19, 0.39) |
| Other (vs. White) | 11 | 0.15 (−0.04, 0.34) | 0.10 (−0.14, 0.34) |
| Sexual identity | |||
| Heterosexual | 7 | Referent | Referent |
| Lesbian, gay, or bisexual | 3 | −0.16 (−0.31, −0.01)# | −0.06 (−0.23, 0.10) |
| Former smokers | 10 | R 2 = .08 | |
| Age | |||
| Younger than 25c | 17 | 0.06 (−0.51, 0.63) | 0.08 (−1.13, 1.29) |
| 25 or older | 10 | Referent | Referent |
| Sex | |||
| Male | 10 | Referent | Referent |
| Female | 9 | −0.03 (−0.16, 0.10) | 0.01 (−0.16, 0.18) |
| Education | |||
| HS education or less | 12 | Referent | Referent |
| Some college | 11 | −0.07 (−0.28, 0.13) | −0.01 (−0.21, 0.21) |
| College degree | 10 | −0.10 (−0.33, 0.13) | 0.03 (−0.19, 0.24) |
| Graduate degree | 5 | −0.20 (−0.41, 0.01)# | −0.10 (−0.28, 0.07) |
| Race/ethnicity | |||
| White | 9 | Referent | Referent |
| Latinoc (vs. White) | 16 | 0.11 (−0.23, 0.45) | 0.20 (−0.28, 0.68) |
| Black (vs. White) | 5 | 0.01 (−0.25, 0.28) | 0.20 (−0.40, 0.81) |
| Other (vs. White) | 10 | 0.20 (−0.12, 0.53) | 0.31 (−0.20, 0.82) |
| Sexual identity | |||
| Heterosexual | 10 | Referent | Referent |
| Lesbian, gay, or bisexualc | 10 | −0.05 (−0.36, 0.26) | −0.08 (−0.48, 0.31) |
| Smokers | 53 | R 2 = .07 | |
| Age | |||
| Younger than 25c | 40 | −0.05 (−0.81, 0.72) | −0.14 (−0.85, 0.56) |
| 25 or older | 54 | Referent | Referent |
| Sex | |||
| Male | 45 | Referent | Referent |
| Female | 61 | 0.46 (0.08, 0.84)* | 0.62 (0.28, 0.96)** |
| Education | |||
| HS education or less | 53 | Referent | Referent |
| Some college | 54 | −0.12 (−0.55, 0.31) | −0.03 (−0.47, 0.41) |
| College degree | 48 | −0.29 (−0.77, 0.19) | −0.23 (−0.75, 0.30) |
| Graduate degreec | 54 | −0.02 (−0.56, 0.51) | 0.10 (−0.54, 0.74) |
| Race/ethnicity | |||
| White | 55 | Referent | Referent |
| Latinoc (vs. White) | 50 | −0.18 (−0.82, 0.47) | −0.21 (−1.06, 0.64) |
| Black (vs. White) | 43 | −0.03 (−0.57, 0.50) | 0.12 (−0.47, 0.71) |
| Otherc (vs. White) | 63 | 0.07 (−0.38, 0.53) | 0.12 (−0.48, 0.72) |
| Sexual identity | |||
| Heterosexual | 53 | Referent | Referent |
| Lesbian, gay, or bisexualc | 39 | −0.59 (−1.57, 0.39) | −0.54 (−1.59, 0.50) |
CI, confidence interval; MRTP, modified risk tobacco product.
aResponded they were Somewhat likely or Very likely to use the product.
cSample size for this subgroup was less than 50; interpret results with caution.
# p < .10. *p < .05. **p < .01.
Mean interest in each MRTP differed by smoking status. Current smokers were more interested in using a less addictive hypothetical MRTP (M = 2.27, SE = 0.07) than were never smokers (M = 1.22, SE = 0.03; B [SE] = 1.05 [0.09], t = 12.35, p < .001, R2 = .24) and former smokers (M = 1.26, SE = 0.03; B [SE] = 1.01 [0.08], t = 12.72, p < .001, R2 = .24). Never and former smokers did not differ in their interest in using a less addictive hypothetical MRTP (B [SE] = 0.05 [0.04], n.s.).
Current smokers were also more interested in using a less harmful hypothetical MRTP (M = 2.41, SE = 0.08) than were never smokers (M = 1.25, SE = 0.04; B [SE] = 1.16 [0.09], t = 12.66, p < .001, R2 = .25) and former smokers (M = 1.30, SE = 0.04; B [SE] = 1.12 (0.09), t = 12.88, p < .001, R2 = .27). Never and former smokers did not differ in their interest in using a less harmful hypothetical MRTP (B [SE] = 0.05 [0.05], n.s.).
Demographic Correlates of Interest in Hypothetical MRTPs
Tables 2 and 3 show demographic correlates of interest in hypothetical less addictive and less harmful MRTPs by smoking status.
Never Smokers
Among never smokers, young adults were more interested than older adults in less addictive hypothetical MRTPs in bivariate and full models and were marginally more interested in less harmful hypothetical MRTPs in the bivariate model. Interest in less addictive hypothetical MRTPs decreased with education level in the bivariate model, though this was only marginally significant in the full model.
Former and Current Smokers
Among former smokers, no demographics were significantly associated with interest in hypothetical MRTPs. Among current smokers, females were more likely to be interested in both less addictive and less harmful hypothetical MRTPs compared to males in the bivariate and full models. No other demographics were significantly associated with interest in hypothetical MRTPs.
Beliefs Associated With Interest in Hypothetical MRTPs
Tables 4 and 5 show belief correlates of interest in hypothetical less addictive and less harmful MRTPs by smoking status.
Table 4.
Beliefs Associated With Interest in Using a Hypothetical Less Addictive MRTP (Weighted)
| Item | Not interested | Interesteda | Single predictor | All predictorsc |
|---|---|---|---|---|
| All scale response options range from 1 to 4, with 4 being most affirmativea | M (SE)b | M (SE)b | B (95% CI) | B (95% CI) |
| Never smokers R2 = .15 | ||||
| Smoking is not changeable | 3.29 (0.04) | 2.73 (0.22) | −0.11 (−0.21, −0.02)* | −0.06 (−0.13, 0.01)# |
| Harm beliefs | ||||
| Some cigs less harmful?—Yes | 94% | 6% | Referent | Referent |
| Some cigs less harmful?—No | 96% | 4% | −0.03 (−0.17, 0.12) | 0 (−0.11, 0.12) |
| 10 cigarettes/d harm | 3.92 (0.02) | 3.69 (0.21) | −0.29 (−0.58, −0.01)* | −0.21 (−0.46, 0.03)# |
| Tobacco chemical beliefs | ||||
| Think about chemicals | 1.75 (0.04) | 2.10 (0.25) | 0.07 (−0.02, 0.16) | 0.01 (−0.03, 0.06) |
| Low chemical products possible | 2.27 (0.05) | 2.79 (0.24) | 0.08 (0.02, 0.13)** | −0.01 (−0.05, 0.05) |
| Cancer beliefs | ||||
| Behavior does not cause cancer | 2.70 (0.04) | 2.59 (0.18) | 0 (−0.07, 0.07) | −0.01 (−0.12, 0.10) |
| Everything causes cancer | 2.42 (0.04) | 2.80 (0.18) | 0.06 (−0.02, 0.14) | 0.05 (−0.02, 0.11) |
| Prevention is impossible | 1.96 (0.04) | 2.58 (0.20) | 0.15 (0.03, 0.26)* | 0.08 (−0.03, 0.20) |
| Recommendations confusing | 2.95 (0.04) | 3.11 (0.15) | 0.01 (−0.07, 0.08) | 0.01 (−0.04, 0.07) |
| Former smokers R2 = .09 | ||||
| Smoking is not changeable | 3.39 (0.05) | 2.90 (0.21) | −0.15 (−0.26, −0.03)* | −0.05 (−0.19, 0.08) |
| Harm beliefs | ||||
| Some cigs less harmful?—Yes | 88% | 12% | Referent | Referent |
| Some cigs less harmful?—No | 94% | 6% | −0.22 (−0.40, −0.05)* | −0.11 (−0.33, 0.11) |
| 10 cigarettes/d harm | 3.82 (0.02) | 3.71 (0.07) | −0.20 (−0.38, −0.01)* | −0.20 (−0.45, 0.05) |
| Tobacco chemical beliefs | ||||
| Think about chemicals | 1.94 (0.06) | 2.26 (0.25) | 0.06 (−0.01, 0.13)# | 0.06 (−0.04, 0.15) |
| Low chemical products possible | 2.49 (0.05) | 2.90 (0.20) | 0.07 0 (.01, 0.13)* | 0.02 (−0.03, 0.07) |
| Cancer beliefs | ||||
| Behavior does not cause cancer | 2.59 (0.05) | 2.45 (0.22) | −0.03 (−0.12, 0.07) | −0.07 (−0.18, 0.04) |
| Everything causes cancer | 2.58 (0.05) | 2.71 (0.21) | 0.02 (−0.07, 0.11) | 0.05 (−0.02, 0.12) |
| Prevention is impossible | 1.97 (0.04) | 2.06 (0.20) | 0.03 (−0.06, 0.11) | −0.01 (−0.16, 0.13) |
| Recommendations confusing | 2.97 (0.05) | 2.94 (0.22) | −0.01 (−0.11, 0.10) | 0.03 (−0.04, 0.11) |
| Smokers R2 = .30 | ||||
| Smoking is not changeable | 3.01 (0.10) | 2.71 (0.09) | −0.24 (−0.39, −0.09)** | −0.33 (−0.53, −0.12)*** |
| Harm beliefs | ||||
| Some cigs less harmful?—Yes | 49% | 51% | Referent | Referent |
| Some cigs less harmful?—No | 57% | 43% | −0.08 (−0.63, 0.47) | 0.29 (−0.29, 0.87) |
| 10 cigarettes/d harm | 3.67 (0.05) | 3.68 (0.05) | 0.06 (−0.18, 0.31) | −0.17 (−0.54, 0.21) |
| Tobacco chemical beliefs | ||||
| Think about chemicals | 2.55 (0.11) | 2.70 (0.11) | 0.07 (−0.08, 0.23) | 0.03 (−0.16, 0.23) |
| Low chemical products possible | 2.56 (0.16) | 3.41 (0.09) | 0.34 (0.20, 0.49)*** | 0.38 (0.21, 0.55)*** |
| Cancer beliefs | ||||
| Behavior does not cause cancer | 2.59 (0.10) | 2.64 (0.12) | 0.02 (−0.15, 0.18) | 0 (−0.19, 0.18) |
| Everything causes cancer | 2.85 (0.14) | 2.72 (0.08) | −0.04 (−0.21, 0.14) | −0.04 (−0.22, 0.15) |
| Prevention is impossible | 2.28 (0.14) | 2.03 (0.10) | −0.08 (−0.24, 0.09) | −0.14 (−0.33, 0.04) |
| Recommendations confusing | 3.02 (0.13) | 3.12 (0.09) | 0.05 (−0.18, 0.27) | 0.03 (−0.23, 0.28) |
CI, confidence interval; MRTP, modified risk tobacco product.
aResponded they were Somewhat likely or Very likely to use the product.
bExcept for whether or not they believed some cigarettes were less harmful than others.
# p < .10. *p < .05. **p < .01. ***p < .001.
Table 5.
Beliefs Associated With Interest in Using a Hypothetical Less Harmful MRTP (Weighted)
| Item | Not interested | Interestedb | Single predictor | All predictorsc |
|---|---|---|---|---|
| All scale response options range from 1 to 4, with 4 being most affirmativea | M (SE)a | M (SE)a | B (95% CI) | B (95% CI) |
| Never smokers R2 = .18 | ||||
| Smoking is not changeable | 3.32 (0.04) | 2.53 (0.21) | −0.16 (−0.27, −0.05)** | −0.06 (−0.13, 0.01)# |
| Harm beliefs | ||||
| Some cigs less harmful?—Yes | 93% | 7% | Referent | Referent |
| Some cigs less harmful?—No | 92% | 8% | 0.05 (−0.12, 0.21) | 0.06 (−0.10, 0.23) |
| 10 cigarettes/d harm | 3.93 (0.02) | 3.62 (0.23) | 0.16 (−0.68, 0.12) | −0.46 (−1.04, 0.13) |
| Tobacco chemical beliefs | ||||
| Think about chemicals | 1.75 (0.04) | 2.01 (0.25) | 0.06 (−0.04, 0.16) | 0.01 (−0.05, 0.08) |
| Low chemical products possible | 2.25 (0.05) | 2.97 (0.21) | 0.10 (0.03, 0.18)** | 0.01 (−0.05, 0.07) |
| Cancer beliefs | ||||
| Behavior does not cause cancer | 2.70 (0.04) | 2.62 (0.19) | 0.01 (−0.08, 0.10) | −0.03 (−0.15, 0.08) |
| Everything causes cancer | 2.43 (0.04) | 2.60 (0.22) | 0.05 (−0.05, 0.15) | 0.07 (−0.01, 0.14)# |
| Prevention is impossible | 1.98 (0.04) | 2.29 (0.23) | 0.12 (−0.01, 0.24)# | 0.05 (−0.08, 0.18) |
| Recommendations confusing | 2.96 (0.04) | 2.98 (0.20) | 0.02 (−0.09, 0.12) | 0.01 (−0.05, 0.08) |
| Former smokers R2 = .08 | ||||
| Smoking is not changeable | 3.41 (0.05) | 2.86 (0.18) | −0.17 (−0.29, −0.05)** | −0.10 (−0.24, 0.04) |
| Harm beliefs | ||||
| Some cigs less harmful?—Yes | 84% | 16% | Referent | Referent |
| Some cigs less harmful?—No | 93% | 7% | −0.22 (−0.43, −0.02)* | −0.17 (−0.44, 0.10) |
| 10 cigarettes/d harm | 3.82 (0.02) | 3.72 (0.07) | −0.20 (−0.40, 0.01)# | −0.17 (−0.43, 0.10) |
| Tobacco chemical beliefs | ||||
| Think about chemicals | 1.92 (0.06) | 2.39 (0.21) | 0.07 (−0.01, 0.14)# | 0.04 (−0.06, 0.15) |
| Low chemical products possible | 2.47 (0.05) | 2.93 (0.16) | 0.08 (0.02, 0.14)** | 0.03 (−0.03, 0.09) |
| Cancer beliefs | ||||
| Behavior does not cause cancer | 2.58 (0.05) | 2.44 (0.19) | −0.01 (−0.11, 0.10) | −0.03 (−0.16, 0.10) |
| Everything causes cancer | 2.57 (0.06) | 2.66 (0.16) | 0.02 (−0.06, 0.11) | 0.04 (−0.04, 0.12) |
| Prevention is impossible | 1.96 (0.04) | 2.08 (0.15) | 0.05 (−0.05, 0.15) | 0.02 (−0.15, 0.18) |
| Recommendations confusing | 2.96 (0.06) | 2.95 (0.18) | −0.01 (−0.11, 0.09) | 0.01 (−0.08, 0.10) |
| Smokers R2 = .29 | ||||
| Smoking is not changeable | 3.01 (0.11) | 2.76 (0.09) | −0.23 (−0.39, −0.08)** | −0.30 (−0.51, −0.09)** |
| Harm beliefs | ||||
| Some cigs less harmful?—Yes | 47% | 53% | Referent | Referent |
| Some cigs less harmful?—No | 48% | 52% | 0.05 (−0.50, 0.61) | 0.37 (−0.21, 0.94) |
| 10 cigarettes/d harm | 3.64 (0.06) | 3.70 (0.05) | 0.09 (−0.20, 0.37) | −0.04 (−0.38, 0.30) |
| Tobacco chemical beliefs | ||||
| Think about chemicals | 2.55 (0.12) | 2.68 (0.10) | 0.05 (−0.10, 0.21) | 0 (−0.20, 0.19) |
| Low chemical products possible | 2.55 (0.17) | 3.31 (0.09) | 0.33 (0.18, 0.48)*** | 0.36 (0.18, 0.53)*** |
| Cancer beliefs | ||||
| Behavior does not cause cancer | 2.63 (0.12) | 2.60 (0.10) | −0.03 (−0.20, 0.14) | −0.03 (−0.24, 0.18) |
| Everything causes cancer | 2.81 (0.17) | 2.77 (0.07) | −0.02 (−0.21, 0.16) | −0.07 (−0.25, 0.11) |
| Prevention is impossible | 2.27 (0.15) | 2.07 (0.09) | −0.08 (−0.27, 0.10) | −0.18 (−0.37, 0.01)# |
| Recommendations confusing | 2.99 (0.15) | 3.13 (0.08) | 0.08 (−0.16, 0.31) | 0.03 (−0.22, 0.28) |
CI, confidence interval; MRTP, modified risk tobacco product.
aExcept for whether or not they believed some cigarettes were less harmful than others.
bResponded they were Somewhat likely or Very likely to use the product.
# p < .10. *p < .05. **p < .01. ***p < .001.
Never Smokers
Among never smokers, several beliefs were associated with interest in using both hypothetical MRTPs in the bivariate models, though not in the full models (Tables 4 and 5). Interest in less addictive hypothetical MRTPs was higher among those believing that (1) smoking status is a changeable characteristic, (2) people harm themselves little when they smoke 10 or more cigarettes every day, (3) it is likely that tobacco products can be made without some of their harmful chemicals, and (4) there is not much you can do to lower your chance of getting cancer (Table 4). In the full model, these associations did not remain significant, though the first two beliefs were marginally significant.
In the bivariate models, interest in less harmful hypothetical MRTPs was higher among those believing that (1) smoking status is a changeable characteristic and (2) it is likely that tobacco products can be made without some of their harmful chemicals. However, in the full model, the former was only marginally significant and the latter was not significant.
Former Smokers
Among former smokers, several beliefs were associated with interest in both hypothetical MRTPs in the bivariate models, though no associations were significant in the full models (Tables 4 and 5). Interest in less addictive hypothetical MRTPs was higher among those believing that (1) smoking status is a changeable characteristic, (2) some cigarettes are less harmful than others, (3) people harm themselves little when they smoke 10 or more cigarettes every day, and (4) it is likely that tobacco products can be made without some of their harmful chemicals (Table 4).
Interest in less harmful hypothetical MRTPs was higher among those believing that (1) smoking status is a changeable characteristic, (2) some cigarettes are less harmful than others, and (3) it is likely that tobacco products can be made without some of their harmful chemicals (Table 5).
Smokers
Among smokers, interest in both hypothetical MRTPs was higher among respondents believing that smoking status is a changeable characteristic and that it is possible for low chemical tobacco products to be made in both bivariate and full models. No other beliefs were significantly associated with interest in hypothetical MRTPs.
Discussion
This research used nationally representative data to assess the extent to which smoking status, demographic characteristics, and beliefs about tobacco and cancer were associated with interest in less addictive and less harmful hypothetical MRTPs. We found that smoking status was the single biggest predictor of self-reported interest in both types of hypothetical MRTPs: current smokers were far more interested in both types of hypothetical MRTPs compared to never established smokers and former smokers, with about half of current smokers and about one-tenth of former and never smokers reporting interest. While it is encouraging that current smokers are more interested than never or former smokers, this difference alone does not indicate that marketing MRTPs would benefit population health. Models of population health impact should also consider factors such as the proportions of the population that are current, never, and former smokers; the percentage of current smokers who would completely switch to the MRTP; and the proportion of nonsmokers who would begin regularly using it. Although interest in both types of hypothetical MRTPs was much lower among never smokers compared to current smokers, certain subsets of never smokers were especially likely to express interest in hypothetical MRTPs. For example, based on the dichotomized measure of interest, about twice as many young adults and people without college education were interested in hypothetical MRTPs compared to older adults or people with at least some college education. Regardless of smoking status, interest in hypothetical MRTPs was higher among people who believed smoking status is a changeable characteristic and that it is possible for tobacco products to be made without some of their harmful chemicals.
Demographic Characteristics Associated With Interest in Hypothetical MRTPs
Many of our findings were consistent with previous research on interest in PREPs and hypothetical MRTPs. This includes the finding that about half of smokers were interested in trying PREPs/hypothetical MRTPs,11–14 that smokers were more interested than never smokers,11,13 and that female smokers were more interested than male smokers.11,13 Consistent with a recent publication on interest in hypothetical MRTPs,11 we also found that that never-smoking young adults and people with low educational attainment were more interested than older adults and people with high educational attainment. These groups could be considered vulnerable subpopulations in future research on potential MRTPs. Overall, our results indicate that the marketing of MRTPs has the potential to benefit the individual health of smokers if interested smokers could completely switch.
Tobacco, Harm, and Cancer Beliefs Associated With Interest in Hypothetical MRTPs
Previous research has not assessed how interest in hypothetical MRTPs relates to believing that smoking status is a changeable individual characteristic, and we found that this belief was positively associated with interest in hypothetical MRTPs regardless of smoking status. Smokers who think of smoking as a changeable behavior could be interested in hypothetical MRTPs as a way to quit, consistent with the prior finding that about one-fifth of smokers who had tried a PREP did so to assist quitting.13 Consistent with research on quitting beliefs and smoking initiation, never smokers who believe that smoking is a changeable behavior could also be interested if they see less danger in experimenting with a product that they view as easier to quit.21,22
Several findings on harm and cancer beliefs differed from those of previous studies. We found that beliefs about the harmfulness of cigarettes and cancer fatalism were not consistently associated with interest in hypothetical MRTPs across smoking statuses. Specifically, harm beliefs were negatively associated with interest in hypothetical MRTPs consistently in former smokers, less consistently for nonsmokers, and not at all for current smokers. Our null finding among smokers is surprising, given the past finding that smokers who were concerned about the harms of smoking were particularly interested in PREPS14 and MRTPs11 and frequently mentioned harm reduction as their reason for trying PREPs.13,16 Also, in this study, cancer beliefs did not appear to be related to interest in hypothetical MRTPs. This differs from an analysis of previous HINTS data reporting that PREPs interest was higher among those with fatalistic attitudes towards preventing cancer.13 These conflicting findings underscore the need for additional research on interest in MRTPs and harm and cancer beliefs.
Limitations
Although our results were largely consistent with previous research, one set of results was inconsistent: we found no associations between smokers’ cancer and harm beliefs and interest in hypothetical MRTPs. One potential explanation is that previous studies have measured interest in specific types of tobacco products (eg, cigarettes)13 or specific products that were on the market (eg, Eclipse),14 while our study used a more distal and abstract measure of interest in hypothetical general tobacco products. This could have resulted in our study finding weaker effects, as it may have been challenging for respondents to imagine a product when answering the question. Future studies could consider more concrete measures of interest, such as asking participants about their interest in a specific marketed product, or providing more information about a specific hypothetical MRTP (eg, a hypothetical new cigarette brand).
Implications
If FDA determines that a product submitted as the subject of an MRTP application would benefit the health of the population as a whole, then FDA can authorize that product for marketing. Our findings about interest in hypothetical MRTPs among current, former, and never smokers can inform models of how the marketing of MRTPs could affect population health. Understanding which subgroups are particularly interested in MRTPs can help determine who might be important to study to inform these models. We identified several subgroups of current, former, and never smokers who may warrant specific attention: smokers who are female, former smokers who hold low harm perceptions of smoking, and never smokers who are young adults or have a high school education or less. Population health could benefit if MRTPs can be marketed in a way that minimizes appeal to all former and never smokers, particularly to never smokers who are young adults and have a high school education or less. Further, studying people who believe that smoking is a changeable individual characteristic and those who believe that it is possible to make low chemical tobacco products may be informative, as these beliefs were associated with interest in using both hypothetical MRTPs.
Conclusions
This analysis of nationally representative US data found that about half of current smokers and a tenth of former and never smokers reported interest in hypothetical MRTPs. Female smokers, former smokers with lower harm perceptions of smoking, and never smokers who were young adults or had a high school education or less expressed more interest in hypothetical MRTPs. Across different categories of smoking status, believing that smoking status is a changeable individual characteristic and that it is possible for tobacco products to be made with fewer chemicals were positively associated with interest in using hypothetical MRTPs. Results have implications for understanding populations that may be particularly affected by the marketing of MRTPs, and can help to inform models of the potential population health impact of authorizing the marketing of MRTPs.
Funding
Publication of this article was supported by the US Food and Drug Administration, Center for Tobacco Products, and the US National Cancer Institute, National Institutes of Health. No funding was provided specifically for conducting the analysis, drafting the manuscript, or submitting this paper for publication.
Declaration of Interests
None declared.
Acknowledgment
This publication represents the views of the authors and does not represent the position or policy of the Food and Drug Administration Center for Tobacco Products or the National Cancer Institute.
References
- 1. Brandt A. The Cigarette Century: The Rise, Fall, and Deadly Persistence of the Product That Defined America. New York: Basic books; 2009. [Google Scholar]
- 2. Kluger R. Ashes to Ashes: America’s Hundred-Year Cigarette War, the Public Health, and the Unabashed Triumph of Philip Morris. New York: Vintage; 2010. [Google Scholar]
- 3. Parascandola M. Tobacco harm reduction and the evolution of nicotine dependence. Am J Public Health. 2011;101(4):632–641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Pollay RW, Dewhirst T. The dark side of marketing seemingly “Light” cigarettes: successful images and failed fact. Tob Control. 2002;11(Suppl. 1):I18–I31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. National Cancer Institute. Risks associated with smoking cigarettes with low machine-measured yields of tar and nicotine. Smoking and Tobacco Control Monograph; 2001. [Google Scholar]
- 6. Pederson LL, Nelson DE. Literature review and summary of perceptions, attitudes, beliefs, and marketing of potentially reduced exposure products: communication implications. Nicotine Tob Res. 2007;9(5): 525–534. [DOI] [PubMed] [Google Scholar]
- 7. Stratton K, Shetty P, Wallace R, Bondurant S. Clearing the smoke: the science base for tobacco harm reduction–executive summary. Tob Control. 2001;10(2):189–195. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Richardson A, Ganz O, Vallone D. Tobacco on the web: surveillance and characterisation of online tobacco and e-cigarette advertising. Tob Control. 2015;24(4): 341–347. doi: 10.1136/tobaccocontrol-2013-051246 [DOI] [PubMed] [Google Scholar]
- 9. Grana RA, Ling PM. “Smoking revolution”: a content analysis of electronic cigarette retail websites. Am J Prev Med. 2014;46(4):395–403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Family Smoking Prevention and Tobacco Control Act of 2009. Pub. L 111-31, 123 Stat 1776. [Google Scholar]
- 11. Pearson JL, Johnson AL, Johnson SE, et al. Adult interest in using a hypothetical modified risk tobacco product: findings from Wave 1 of the Population Assessment of Tobacco and Health Study (2013–2014). Addiction. 2017. doi: 10.1111/add.13952. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Parascandola M, Augustson E, Rose A. Characteristics of current and recent former smokers associated with the use of new potential reduced-exposure tobacco products. Nicotine Tob Res. 2009;11(12): 1431–1438. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Parascandola M, Augustson E, O’Connell ME, Marcus S. Consumer awareness and attitudes related to new potential reduced-exposure tobacco product brands. Nicotine Tob Res. 2009;11(7):886–895. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Hund LM, Farrelly MC, Allen JA, et al. Findings and implications from a national study on potential reduced exposure products (PREPs). Nicotine Tob Res. 2006;8(6):791–797. [DOI] [PubMed] [Google Scholar]
- 15. Shiffman S, Pillitteri JL, Burton SL, Di Marino ME. Smoker and ex-smoker reactions to cigarettes claiming reduced risk. Tob Control. 2004;13(1):78–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Caraballo RS, Pederson LL, Gupta N. New tobacco products: do smokers like them?Tob Control. 2006;15(1):39–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Hamilton WL, Norton GD, Ouellette TK, Rhodes WM, Kling R, Connolly GN. Smokers’ responses to advertisements for regular and light cigarettes and potential reduced-exposure tobacco products. Nicotine Tob Res. 2004;6(Suppl. 3):S353–S362. [DOI] [PubMed] [Google Scholar]
- 18. O’Connor RJ, Hyland A, Giovino GA, Fong GT, Cummings KM. Smoker awareness of and beliefs about supposedly less-harmful tobacco products. Am J Prev Med. 2005;29(2):85–90. [DOI] [PubMed] [Google Scholar]
- 19. Niederdeppe J, Levy AG. Fatalistic beliefs about cancer prevention and three prevention behaviors. Cancer Epidemiol Biomarkers Prev. 2007;16(5):998–1003. [DOI] [PubMed] [Google Scholar]
- 20. Blake KD, Portnoy DB, Kaufman AR, et al. Rationale, Procedures, and Response Rates for the 2015 Administration of NCI’s Health Information National Trends Survey: HINTS-FDA 2015. Journal Health Commun. 2016;21(12):1269–1275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Arnett JJ. Optimistic bias in adolescent and adult smokers and nonsmokers. Addict Behav. 2000;25(4):625–632. [DOI] [PubMed] [Google Scholar]
- 22. Gerking S, Khaddaria R. Perceptions of health risk and smoking decisions of young people. Health Econ. 2012;21(7):865–877. [DOI] [PubMed] [Google Scholar]