Article Text


Misperceptions of harm among Natural American Spirit smokers: results from wave 1 of the Population Assessment of Tobacco and Health (PATH) study (2013–2014)
  1. Jennifer L Pearson1,2,
  2. Amanda Johnson1,
  3. Andrea Villanti1,2,
  4. Allison M Glasser1,
  5. Lauren Collins1,
  6. Amy Cohn1,3,
  7. Shyanika W Rose1,
  8. Raymond Niaura1,2,
  9. Cassandra A Stanton1,3,4
  1. 1Schroeder Institute for Tobacco Research and Policy Studies at Truth Initiative, Washington, DC, USA
  2. 2Department of Health, Behavior, and Society, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
  3. 3Department of Oncology, Georgetown University Medical Center/Cancer Prevention and Control Program, Lombardi Comprehensive Cancer Center, Washington, DC, USA
  4. 4Westat, Rockville, Maryland, USA
  1. Correspondence to Dr Jennifer Pearson, Schroeder Institute for Tobacco Research and Policy Studies at Truth Initiative, 900 G St NW, Fourth floor, Washington, DC 20001, USA; jpearson{at}


Introduction This study estimated differences in cigarette harm perceptions among smokers of the Natural American Spirit (NAS) brand—marketed as ‘natural’, ‘organic’ and ‘additive-free’—compared to other smokers, and examined correlates of NAS use.

Methods Data were drawn from wave 1 of the Population Assessment of Tobacco and Health (PATH) study, a nationally representative study of US adults (2013–2014). Weighted analyses using a subset of current adult smokers (n=10 565) estimated the prevalence of NAS use (vs all other brands) and examined associations between NAS use and sociodemographics, tobacco/substance use, tobacco harm perceptions, quit intentions, quit attempts and mental/behavioural health.

Results Overall, 2.3% of adult smokers (920 000 people in the USA) reported NAS as their usual brand. Nearly 64% of NAS smokers inaccurately believed that their brand is less harmful than other brands compared to 8.3% of smokers of other brands, after controlling for potential confounders (aOR 22.82). Younger age (18–34 vs 35+; aOR 1.54), frequent thinking about tobacco harms (aOR 1.84), past 30-day alcohol use (aOR 1.57), past 30-day marijuana use (aOR 1.87) and sexual orientation (lesbian, gay, bisexual, ‘other’ or ‘questioning’ vs heterosexual; aOR 2.07) were also associated with increased odds of smoking NAS.

Conclusions The majority of NAS smokers inaccurately believes that their cigarettes are less harmful than other brands. Given the brand's rapid growth and its more common use in vulnerable groups (eg, young adults, lesbian, gay, bisexual, ‘other’ or ‘questioning’ adults), corrective messaging and enforcement action are necessary to correct harm misperceptions of NAS cigarettes.

  • Advertising and Promotion
  • Packaging and Labelling
  • Tobacco industry

Statistics from

Video abstract


Natural American Spirit (NAS), produced by Santa Fe Natural Tobacco Company (SFNTC), is owned by Reynolds American and sells cigarettes and roll your own tobacco with labelling and advertising that features the descriptors ‘natural’, ‘organic’ and ‘additive-free’. Consumers believe that cigarettes marketed with these and similar descriptors are significantly more appealing, healthier or less harmful than packages without these descriptors.1–4 Such misperceptions may encourage smokers to switch cigarette brands rather than quit smoking entirely, or may increase intent to try a product.5 No research has determined that NAS cigarettes are less harmful to human health than other cigarette brands. Carbon monoxide, tar, heavy metals and other particulate matter are present in conventional and in ‘organic’, ‘natural’ or ‘additive-free’ cigarette smoke.6

Existing federal and state law prohibits tobacco companies from misleading consumers through their products' labelling and advertising.7–9 The Family Smoking Prevention and Tobacco Control Act (Tobacco Control Act) also prohibits tobacco companies from making explicit or implicit reduced-risk claims through their product labelling or advertising without first obtaining a permissive order from the Food and Drug Administration (FDA).10 Based on these laws, US states and the federal government have taken some action towards correcting consumers' inaccurate beliefs concerning NAS cigarettes and towards preventing NAS cigarettes' labelling and advertising from further misleading consumers or otherwise violating the law. In June 2000, the Federal Trade Commission issued a complaint against SFNTC, asserting that their advertising improperly represented that NAS was less harmful than other cigarette brands.11 SFNTC subsequently agreed, among other actions, to include a disclaimer that “No additives in our tobacco does NOT mean a safer cigarette” on all cigarette advertisements.12 In March 2010, as a result of a settlement agreement with states' attorneys general, SFNTC was required to include an additional disclaimer on advertising stating that “Organic tobacco does NOT mean a safer cigarette.”13 In August 2015, the FDA issued a warning letter to SFNTC and two other tobacco manufacturers, stating that their labelling and advertising of products as ‘natural’ and ‘additive-free’ is in violation of section 911 of the Tobacco Control Act, because their labelling improperly implies that “the tobacco products present a lower risk of tobacco-related disease or is less harmful than one or more other commercially marketed tobacco products” and is thus considered ‘adulterated’ under section 902(8) of the Tobacco Control Act.14 ,15 FDA and SFNTC have not resolved the issues raised in the 2015 warning letter.

Despite these actions, the popularity of the NAS brand has increased in recent years. Market share for NAS grew 554% between 2002 and 2013 among all age groups and, according to the most recently available data, was the fifth most popular cigarette brand among US young adults aged 18–34 years between 2011 and 2014 (following Marlboro, Newport, Camel and Pall Mall).16 ,17 This recent increase in market share and popularity with young consumers may be due to the brand's packaging and advertising conveying unsubstantiated reduced harm messages, which have not been corrected by the settlement-required disclaimers.3 ,4 Indeed, misperceptions of harm associated with ‘natural’ and ‘additive-free’ cigarettes have also been shown to be associated with intention to purchase the brand.5 We know little about the health-risk beliefs or demographic characteristics of individuals who use NAS cigarettes. Such information could not only add to the existing evidence that NAS labelling and advertising are misleading consumers, despite corrective efforts, but could also be used to identify target audiences for improved corrective public health messaging about the risks associated with NAS cigarette smoking or other tobacco products making similar ‘organic,’ ‘natural’ or ‘additive-free’ claims. It is especially important to examine the prevalence of NAS brand use among groups with high smoking prevalence such as those with mental health or substance use disorders, or among the lesbian, gay, bisexual, transgender or queer (LGBTQ) community, as NAS marketing to these vulnerable groups could be a mechanism for further widening tobacco-related health disparities if use discourages quitting smoking, cutting back or switching to less harmful tobacco products.18–21 This study sought to fill these gaps by examining the prevalence of NAS brand preference and its correlates, including sociodemographic characteristics, tobacco and other substance use, tobacco harm perceptions and mental, behavioural and physical health using data from US adult smokers in the Population Assessment of Tobacco and Health (PATH) Study.


Data are from wave 1 of the PATH Study, a nationally representative, longitudinal cohort study of 32 320 adults in the USA, aged 18 years and older conducted from 12 September 2013 to 15 December 2014. Recruitment involved address-based, area probability sampling, using an in-person household screener to select participants. Adult tobacco users, young adults aged 18–24 years and African-Americans were oversampled relative to population proportions. The weighting procedures adjusted for oversampling and non-response; combined with the use of a probability sample, the weighted data yield representative estimates of the non-institutionalised, civilian US population in 2012–2013. The weighted response rate for the household screener was 54.0%. Among households that were screened, the overall weighted response rate was 74.0% for the adult interview. Further details regarding the PATH Study design and methods are available elsewhere.22 This study used de-identified PATH Study data available from the National Addiction & HIV Data Archive Program (NAHDAP) ( PATH Restricted Use File (RUF) and was determined to be exempt as non-human research by Chesapeake IRB (Pro00015910).


Tobacco use

The current study focuses on cigarette smokers with a usual brand of cigarettes. Current cigarette smoking was defined as now smoking cigarettes ‘some days’ or ‘every day’ and smoking at least 100 lifetime cigarettes. Participants were also asked about their use of e-cigarettes, traditional cigars, cigarillos, filtered cigars, hookah tobacco, pipe tobacco, smokeless tobacco, snus and dissolvable tobacco (‘Do you now use [product]…’). For each non-cigarette product, respondents who reported ‘some days’ or ‘every day’ use were classified as current users of that product.

Brand preference

Since PATH is administered via laptop, the in-person interview included a robust assessment of smokers' cigarette brand preference. Smokers who reported that they had a regular cigarette brand were shown images with cigarette brand logos; 658 brands and sub-brands were included in the PATH wave 1 assessment. Participants chose the logo of their preferred brand, which took them to a second screen where individuals could identify their sub-brand.

Current cigarette smokers were asked to identify their usual brand of cigarettes, with those reporting ‘don't know’ or refusing to answer excluded from the analytical sample (2.0%). The number of cigarettes smoked per day, age of first tobacco trial, whether they had attempted to quit tobacco in the past year (yes/no), intention to quit smoking/using tobacco products (collapsed into ‘in the next 6 months’, ‘not in the next 6 months, but within the year’, ‘more than a year’ and ‘no intention to quit’) and how soon they used each tobacco product on waking (dichotomised to ‘use in the first 5 min’ (yes/no)) were also assessed.

Harm perceptions

Current cigarette smokers were asked about the relative harm of their usual cigarette brand compared to other brands with the following item: ‘Do you think the brand of cigarettes you usually smoke might be less harmful, no different, or more harmful compared to other cigarette brands?’ with responses dichotomised to less harmful (yes/no) for regression analyses. ‘Don't know,’ missing and refused responses were collapsed into one response category in the PATH RUF; participants with this response were excluded from analyses as the percentage of missing data was low (0.6% overall; no missing data among American Spirit smokers). Current cigarette smokers were also asked about how often they thought about the harm of their tobacco use in the past 30 days (‘never’ to ‘very often’; collapsed into ‘never/rarely,’ ‘sometimes’ and ‘often/very often’ for analyses).

Mental health, substance use and physical health

Items from the Global Appraisal of Individual Needs—Short Screener (GAIN) were used to capture internalising disorders (eg, ‘When was the last time that you had significant problems with…Feeling trapped, lonely, sad, blue, depressed, or hopeless about the future?’ (4 items)), externalising disorders (eg, ‘When was the last time that you did the following things two or more times…Lied or conned to get things you wanted or to avoid having to do something?’ (5 items)) and substance use problems (eg, ‘When was the last time you…used substances at least weekly, spent a lot of time obtaining substances, reduced involvement in activities due to the use of substances, had problems with withdrawal’ (7 items)).23 For each item, participants indicated when they most recently experienced the problem: ‘Never’, ‘1+ years ago’, ‘2 to 12 months ago’ and ‘Past month’. The GAIN scale uses the following severity threshold cut-points, assessing symptoms in the past year: 0–1 symptoms (low), 2–3 symptoms (moderate) and 4/4+ (high, depending on the scale). Dichotomised measures of any past 30-day alcohol (0/1) and marijuana (0/1) use were also assessed. We also included perceived overall health status (‘poor’ to ‘excellent’), which is a valid assessment of actual health status.24

Sociodemographic characteristics

Sociodemographic variables used in these analyses included age (grouped 18–34, 35+), gender, race (White and non-White), ethnicity (Hispanic, Latino/Latina or of Spanish origin vs non-Hispanic), educational attainment (dichotomised as completed at least some college (yes/no)) and sexual orientation (lesbian, gay, bisexual, ‘other’ or ‘questioning’ vs heterosexual).

Statistical analyses

The analytical sample for correlates of NAS was limited to current cigarette smokers who reported 100 lifetime cigarettes, who ‘usually purchase’ manufactured cigarettes and who identified their usual cigarette brand; these were the only smokers asked about the perceived relative harm of their own brand of cigarettes. Analyses were conducted using SVY procedures in Stata/MP V.14.1 to account for weighting. Individuals were excluded from regression analyses if they were missing data on one or more variables included in the regression. As per the NAHDAP data use agreement, cell counts below minimum disclosure protection requirements were suppressed in all tables. Pearson's χ2 tests and Student's t-tests were used to assess differences in means or prevalence estimates with statistical significance of p<0.05. Bivariate analyses were used to estimate associations between NAS brand preference with sociodemographics, tobacco/substance use, harm perceptions, quit intentions, quit attempts and mental/behavioural health. Next, using a stepwise model building approach, multivariable logistic regression models were conducted to examine the relative (adjusted) odds of NAS brand preference controlling for correlates in the following domains: (1) sociodemographics, (2) tobacco/substance use, (3) tobacco harm perceptions and (4) mental, behavioural and physical health. In each step, correlates of NAS brand preference with statistical significance of p<0.05 were included in subsequent models. Sociodemographic variables were retained in all models; other non-significant correlates were dropped from the final adjusted model for parsimony. We describe our findings as ‘likelihoods’ rather than ‘odds’ because our outcome (NAS brand preference) is rare; in these cases, discussing ORs as risk ratios does not misrepresent the degree of association between the outcome and the independent variable.25


Correlates of NAS brand preference

Table 1 presents the characteristics of current cigarette smokers overall and compares NAS smokers with smokers of other brands. Overall, 32.7% (unweighted) of the PATH wave 1 adult sample were current established cigarette smokers with a reported brand preference (n=10 565). Approximately 2.3% (weighted) of current adult smokers reported NAS as their usual brand of cigarettes, representing ∼920 000 adults in the USA. Figure 1 presents the proportions of current NAS and other brand smokers who believe that their preferred brand is less harmful, ‘no difference’ or more harmful than other cigarette brands. Among smokers of other brands, 8.3% (95% CI 7.66% to 8.94%) believed their brand was less harmful than other brands; 82.8% (95% CI 81.91% to 83.69%) believed their brand to be as harmful as other brands and 8.4% believed their brand was more harmful than other brands (95% CI 9.95% to 9.01%). In comparison, 63.9% (95% CI 56.72% to 70.56%) of NAS smokers believed their brand was less harmful than other brands; 35.6% (95% CI 29.04% to 42.66%) believed their brand to be as harmful as other brands and 0.5% believed their brand was more harmful than other brands (95% CI 0.06% to 3.77%).

Table 1

Characteristics of current cigarette smokers by brand preference (other brands vs American Spirit) (n=10 565)

Figure 1

Percentage of smokers who believe that their own brand of cigarettes is less harmful, no difference or more harmful than other brands, by brand preference (NAS vs all other brands) (n=10 565). (NAS smokers did not have any don't know/refused/missing values). NAS, Natural American Spirit.

In addition to believing that their brand was less harmful than other brands, NAS smokers were also significantly more likely to report thinking about the harms of their tobacco use more frequently than cigarette smokers of other brands (46.1% vs 35.2%, p<0.001). With respect to demographics, NAS smokers were more likely to be 18–34 years (48.2% vs 39.8%, p=0.02), White (88.0% vs 77.2%, p=0.002), have attended at least some college (74.7% vs 45.0%, p<0.0001) and identify as lesbian, gay, bisexual, ‘other’ or ‘questioning’ (16.1% vs 7.3%, p<0.0001). NAS smokers were also more likely to report past 30-day use of alcohol (76.0% vs 58.0%, p<0.0001) or marijuana (34.2% vs 18.9%, p<0.0001). NAS brand preference was also correlated with increasing internalising and externalising symptom counts (p's=0.006).

Table 2 presents the final adjusted multivariable logistic regression model of NAS brand preference among current cigarette smokers, controlling for significant covariates in the stepwise models (not shown). Among current cigarette smokers, the strongest correlate of NAS brand preference before and after adjustment was the perception that their usual brand was less harmful than other cigarette brands (aOR 22.82, 95% CI 15.98 to 32.59). Smokers who ‘often’ or ‘very often’ thought about the harms of their tobacco use were significantly more likely to report NAS brand preference compared to those who ‘never’ or ‘rarely’ thought about the harms of their tobacco use (aOR 1.84, 95% CI 1.20 to 2.83). Younger age (aOR 1.54, 95% CI 1.08 to 2.20), male gender (aOR 1.57, 95% CI 1.15 to 2.14), having attended at least some college (aOR 3.06, 95% CI 2.15 to 4.36) and lesbian, gay, bisexual, ‘other’ or ‘questioning’ identity (aOR 2.07, 95% CI 1.22 to 3.51) were also associated with NAS brand preference. Additionally, past 30-day marijuana users (aOR 1.87, 95% CI 1.31 to 2.65) and alcohol users (aOR 1.57, 95% CI 1.12 to 2.21) were more likely to report NAS as their usual brand compared to other cigarette brands.

Table 2

Adjusted logistic regression of NAS brand preference on sociodemographics, substance/tobacco use, harm perceptions and mental, physical and behavioural health among current cigarette smokers; Population Assessment of Tobacco and Health, 2013–2014 (n=10 289*)


Nearly 1 million US adult smokers prefer NAS. NAS smokers are 22 times more likely than other smokers to believe that their brand is less harmful than other cigarette brands. Taken in context with prior research,3 ,4 these results suggest that NAS smokers are concerned about the harms associated with their smoking. Adult NAS smokers may choose the NAS brand because they perceive it as a less harmful cigarette product as a result of NAS branding and/or the descriptors ‘organic’, ‘natural’ and ‘additive-free’ on product packaging and advertising. These findings directly support FDA's assertion in their 2015 letter to SFNTC that American Spirit cigarettes “sell or distribute cigarette products, the label, labeling, or advertising of which represents explicitly and/or implicitly that the products or their smoke do not contain or are free of a substance and/or that the products present a lower risk of tobacco related disease or are less harmful than one or more other commercially marketed tobacco products.”14 The prevalence of reduced harm perceptions among NAS smokers also clearly demonstrates that the disclaimer statements on NAS packs and advertisements, which have been in their most recent form since 2010 and predate data collection for the current study by several years, are not an effective means to correct consumers' inappropriate harm perceptions.

As the data are cross-sectional, we cannot determine a causal sequence between lowered harm perceptions and NAS use. However, prior research has shown that NAS packs convey decreased harm messages to non-smokers, current smokers and former smokers alike;4 thus, it is reasonable that at least some current smokers chose their brand due to a pre-existing and inaccurate belief that NAS cigarettes are less harmful than other brands. Another possible mechanism driving inaccurate harm perceptions among NAS smokers could be their repeated exposure to NAS cigarette packs, which prominently feature ‘organic’, ‘natural’ and ‘additive-free’ descriptors. It is also possible that the sensory experience of using NAS affects harm perceptions, as has been found among smokers who prefer ‘light’ cigarettes.26 Among youth and young adults, lowered cigarette harm perceptions are correlated with susceptibility to smoking27–29 and tobacco use.30 Furthermore, among youth and adults, lower tobacco product harm perceptions more generally are associated with ever or current use of these products.31–38 Lower perceived harm of one's own tobacco use is associated with decreased intention to quit, fewer quit attempts and lower odds of cessation.39–43 Given the well-understood relationship of lower tobacco harm perceptions to continued tobacco use and delayed cessation, it is possible that many of these NAS smokers would have tried to quit, cut back their smoking or switched to less harmful forms of tobacco if they did not mistakenly believe that NAS cigarettes were less harmful than other cigarette brands.39–44 As the long-term health effects of continued smoking are well documented,45 these findings highlight the need for immediate action to correct the inaccurate harm perceptions held by the majority of NAS smokers and to prevent NAS (and similar tobacco product) labelling and advertising from misleading consumers in the future.

We found that the adjusted odds of NAS cigarette use were twice as high among lesbian, gay, bisexual, ‘other’ or ‘questioning’ smokers than heterosexual smokers. NAS is one of several brands that has appropriated the language of the LGBTQ rights movement to sell tobacco products.46 In a 2005 NAS advertisement in The Advocate, an LGBTQ-interest magazine, phrases including freedom ‘to marry’, ‘to speak’ and ‘to inhale’ were prominently displayed next to a pack of NAS cigarettes as a message from ‘the people of Santa Fe Natural Tobacco Company.’47 Given the already high burden of smoking-related and other poor health outcomes in this community,20 ,48 ,49 preference for a cigarette brand that misleads the majority of its consumers into believing that it is less harmful than other brands should be of particular concern to public health researchers, tobacco control advocates, policymakers and relevant regulatory agencies, such as the FDA.

In addition to sexual identity and harm perceptions, NAS smokers are distinct from smokers of other brands in several ways. Natural American Spirit smokers are more likely than smokers of other brands to be young adults aged 18–34 years, male, White, to have completed at least some college and to report past 30-day use of marijuana or alcohol. They are also more likely than smokers of other brands to think about the harms for their tobacco use ‘often’ or ‘very often’. Together, these findings describe a group of smokers who are relatively younger, more affluent, more educated and more concerned about their health than smokers of other cigarette brands. One of the SFNTC's marketing efforts has been to portray itself as environmentally friendly,50 ,51 which resonates with the 40% of adults who consider their generations to be environmentally conscious and Millennials who consider themselves idealistic.52 Future research on SFNTC's marketing strategy could illuminate how the brand has marketed to these groups and inform future corrective messaging campaigns.

Our findings have US and global implications, with NAS growing in popularity in the USA and expanding its distribution internationally. In April 2016, Reynolds American's subsidiary SFNTC reported strong year-over-year growth in the USA. NAS sales and profit margins, despite an overall decrease in domestic smoking prevalence.53 ,54 Globally, Reynolds American sold non-US rights to the NAS brand (including the trademarks and all outstanding shares of Reynolds American subsidiaries outside the USA, which sell the brand)55 to Japan Tobacco for $5 billion in September 2015.56 Press releases on the sale note that the largest international markets for the NAS brand are in Germany, Switzerland56 ,57 and Japan, where sales increased 2.5 times to 1.5 billion sticks between 2011 and 2014.58 Continued branding and messaging portraying NAS as a lower harm cigarette could facilitate the introduction or perpetuation of smoking in global markets like the USA where smoking prevalence has fallen in recent years.


The majority of NAS smokers inaccurately believes that their cigarettes are less harmful than other brands. These findings directly support FDA's assertion in their 2015 letter to SFNTC that American Spirit cigarettes ‘explicitly and/or implicitly’ represent that “products present a lower risk of tobacco-related disease or are less harmful than one or more commercially marketed tobacco products.”14 NAS use is higher in vulnerable groups like young adults, lesbian, gay, bisexual, ‘other’ or ‘questioning’ smokers, and those who use marijuana and alcohol. Given the brand's rapid domestic and global expansion, regulatory efforts to prevent NAS labelling and advertising from misleading consumers, as well as other corrective messaging and policy action on the part of FDA and other regulatory bodies, are urgently needed to ensure that consumers are not misled about the risks of smoking NAS cigarettes.

What this paper adds

  • Prior research has revealed that Natural American Spirit (NAS) cigarette pack descriptors and advertising convey reduced harm messages to study participants. Using recent (2013–2014) nationally representative data from the USA, this study shows that, compared to smokers of other brands, NAS smokers are over 22 times more likely to inaccurately believe that their brand is less harmful than other brands. NAS smokers are more likely to be young adults, non-Hispanic White, college educated, past 30-day marijuana and alcohol users, and identify as lesbian, gay, bisexual, ‘other’ or ‘questioning’ than smokers of other brands.


View Abstract


  • Twitter Follow Shyanika Rose at @ShyanikaRose

  • Contributors JLP conceived the manuscript topic, lead the analysis team and wrote the first draft of the paper. AJ conducted the analyses. AJ, AV, AMG, LC, AC, SWR, RN and CAS contributed to the analysis plan, interpretation of results and manuscript writing.

  • Funding This research was supported by internal funds from Truth Initiative.

  • Competing interests JLP is a pending expert witness in litigation against Santa Fe Natural Tobacco Company.

  • Ethics approval The study was determined to be exempt as non-human research by Chesapeake IRB (Pro00015910).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement PATH data are available from the National Addiction & HIV Data Program at

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.