Youths’ Perceptions of Nicotine Harm and Associations With Product Use

Although nicotine is the main addictive substance in tobacco, tobacco combustion is responsible for most tobacco-related diseases. U.S. adults hold misperceptions about nicotine harm. However, little is known about youth nicotine perceptions.

To address this gap, we assessed U.S. youths’ nicotine perceptions and how these perceptions relate to tobacco use. Participants were youth (ages 12–17) in waves 4 (w4; December 2016–January 2018; N = 14 798) and 4.5 (w4.5; December 2017–December 2018; N = 12 918) of the Population Assessment of Tobacco and Health Study, a nationally representative longitudinal cohort study. We describe beliefs about nicotine; perceptions of the nicotine harm in cigarettes, e-cigarettes, and nicotine replacement therapy (NRT); and demographic differences. Regressions assess whether w4 nicotine perceptions predicted w4.5 tobacco use.

Most youth correctly responded that nicotine is the main cause of addiction (77.1%) but incorrectly responded that nicotine is the main substance that causes smoking-related cancer (74.7%). Youth distinguished between the harm of nicotine in different products, and on average rated the nicotine in cigarettes as most harmful, followed by e-cigarettes and NRT. Among youth who did not use at w4, greater harm perceptions of nicotine in cigarettes, e-cigarettes, and NRT were associated with lower likelihood of reporting current tobacco use at w4.5. Among youth who currently used cigarettes or e-cigarettes at wave 4, nicotine perceptions did not predict switching to e-cigarettes or cigarettes, respectively, at wave 4.5.

These findings underscore the challenge of developing effective and comprehensive communication strategies that accurately convey the effects of nicotine without encouraging tobacco use.

Many U.S. adults have misperceptions about nicotine, incorrectly believing it is the substance that causes most smoking-related cancers; studies have not assessed youth’s perceptions of nicotine and how these perceptions relate to tobacco use. This study found that similar to adults, most youth incorrectly believed nicotine is the main substance that causes smoking-related cancer; youth also distinguish between the harmfulness of nicotine in different products, and rated the nicotine in cigarettes as most harmful, followed by e-cigarettes and NRT. Perceptions of the harm in different nicotine and tobacco products negatively predicted becoming a person who used tobacco a year later, but did not predict switching between e-cigarettes and cigarettes. Findings highlight the challenges of accurately communicating about the harms of nicotine without encouraging tobacco use; findings can be considered in the context of FDA’s potential nicotine product standard that would lower nicotine levels in combustible tobacco products to a minimally or nonaddictive level.

Nicotine is the main addictive substance in tobacco products. It can cause negative cardiovascular effects,1 and exposure to nicotine during youth may prime the brain for the rewarding effects of other drugs, lead to difficulty with maintaining focus, and cause mood changes.2 However, it is tobacco combustion and the accompanying release of thousands of harmful chemicals, rather than nicotine itself, that is responsible for most of the negative tobacco-related health outcomes (eg, the myriad of cancers related to tobacco use).3,4

In 2018, the U.S. Food and Drug Administration announced a potential nicotine product standard that would lower nicotine levels in combustible tobacco products to a minimally or nonaddictive level.5 Given nicotine’s role in addiction, this proposed policy could dramatically increase the number of adults who are able to successfully quit or switch to less harmful tobacco products; it could also dramatically reduce the number of youths who use combustible tobacco products.6 However, given that reducing nicotine levels would not decrease the toxicity or harmfulness of combusted tobacco products, there would be little health benefit to individuals who continue using the reduced nicotine products with the same frequency and duration as current combustible products. Thus, the impact of this product standard could depend, in part, on public understanding of the harms of nicotine and reduced nicotine products.

Many studies have found that tobacco risk perceptions (eg, tobacco harm or addictiveness) are associated with and predict tobacco use behavior.7,8 However, less research has been done looking at nicotine perceptions. Research has found that many U.S. adults have misperceptions about the harms of nicotine and reduced nicotine tobacco products. Among U.S. adults in 2015, about half (49%) incorrectly believed that nicotine is the main substance that causes smoking-related cancers9 (as found in other studies10–12). Moreover, almost one-third (30%) incorrectly believed that a “low nicotine” cigarette would be less harmful than other cigarettes.9 Studies of U.S. adult smokers suggest that labeling or advertising cigarettes as low in nicotine may lead smokers to incorrectly perceive these cigarettes as less carcinogenic or lower in disease risk even when smoked for many years13,14 and at the same rate as other cigarettes.15

Few studies have examined the relationship between adult perceptions of nicotine and use of nicotine and tobacco products. One study16 found that compared to young adults who had not used tobacco in the past 30 days, people who had used tobacco had lower odds of endorsing that nicotine was responsible for smoking-related diseases and cancer, that is, they were more likely to correctly understand that nicotine was not the main substance responsible. Similar studies find high rates of the misperception that nicotine replacement therapy (NRT) is just as dangerous as cigarettes, and this belief is associated with reduced intentions to use NRT in a quit attempt.17–19

Although these studies have documented young adult and adult perceptions of nicotine, and youths’ perceptions of nicotine in the United Kingdom,20 studies have not assessed youth’s perceptions of nicotine in the United States. The current project is the first to address this gap by describing nicotine perceptions in a nationally representative sample of youth (ages 12–17). This study describes demographic differences in youth nicotine perceptions, assesses how perceptions relate to one another, and tests whether these perceptions predict future tobacco use. We hypothesized that perceptions of the harm of nicotine in different products would be positively correlated with one another, consistent with research finding that smokers who think nicotine in cigarettes is harmful also tend to think nicotine in NRT is harmful.,15,21,22 Finally, consistent with research finding that risk perceptions predict behavior,7,8 we expected that youth who have lower (vs. higher) nicotine harm perceptions would be more likely to initiate tobacco product use above and beyond risk perceptions.

The Population Assessment of Tobacco and Health (PATH) Study is an ongoing, nationally representative, longitudinal cohort study of U.S. adults (ages 18 and older) and youth (ages 12–17). This paper reports on restricted use file youth data from waves 4 and 4.5, collected approximately 1 year apart from December 2016–January 2018 and December 2017–November 2018, respectively. Individuals who turn 18 while in the study transition from the youth version of the study to the adult version. For the purposes of this study, we only analyzed youth data sets; thus, longitudinal analyses excluded participants who were age 17 at wave 4 because they aged up to young adults (age 18 years) at the time of the wave 4.5 interviews and were not eligible for the youth interview. The PATH Study employed a stratified, address-based, area-probability sampling design that oversampled adults who use tobacco, African Americans, and young adults (ages 18–24). Audio computer-assisted self-interviews were conducted in person with 14 798 youth at wave 4 and 12 918 youth at wave 4.5. The wave 4 cohort included a replenishment sample (N = 3739) that was combined with the remaining cohort recruited at wave 1 (N = 11 059) to form a new cohort that is nationally representative. The youth-weighted wave 4 response rate for the wave 1 cohort was 79.5%. Among youth selected during screening for the replenishment sample, the weighted response rate was 70.6%. At wave 4.5, the weighted response rate for the wave 4 cohort was 89.1%, corresponding to 12 918 youth interview respondents. These were the most recent waves that were available at the time of analysis. The PATH Study was conducted by Westat and approved by Westat’s institutional review board. Further details and study instruments are available elsewhere23,24 (https://doi.org/10.3886/Series606).

The PATH Study’s design allowed us to (1) describe nicotine beliefs and perceptions, including assessing key demographic differences in these beliefs and perceptions and describing how these perceptions relate to one another, and (2) determine how these beliefs and perceptions predict future tobacco use behaviors in a nationally representative sample of youth.

Supplementary Table 1 lists the variable names and questions participants were asked assessing nicotine beliefs, nicotine harm perceptions, absolute product harm perceptions, and relative harm perceptions.

Participants were classified as currently using a tobacco product if they affirmed they used the product in the past 30 days. E-products were described as e-cigarettes, vape pens, personal vaporizers and mods, e-cigars, e-pipes, e-hookahs, and hookah pens; cigars included traditional cigars, cigarillos, or little filtered cigars; and any tobacco use included cigarettes, e-cigarettes or other electronic nicotine products, cigars, pipe, hookah, snus, or other types of smokeless tobacco (such as dip, spit, or chew), and dissolvable tobacco. Analyses used derived variables in the restricted use file are as follows: Wave 4/wave 4.5 current cigarettes, e-cigarettes, cigars, and any tobacco use. To summarize results, we use the term “switching” to mean the participant went from being a wave 4 past 30-day user of one product to a nonuser of that product at wave 4.5, AND the participant went from being a wave 4 nonuser of a different product to a wave 4.5 past 30-day user of that different product.

At wave 4, participants responded to four items asking about intentions to use cigarettes and e-cigarettes: “Do you think you will (smoke cigarettes/use e-products) in the next year?”; “Do you think you will try a (cigarette/e-product) soon?”; “If your best friend offered you an (cigarette/e-product), would you smoke it?”; and “Have you ever been curious about (smoking a cigarette/using an e-product)?. Consistent with previous research,25 participants were categorized as non-susceptible to the product if they responded on lowest scale point (scored as 1) ‘Definitely not’ and ‘Not at all curious’ on all four items, and as susceptible to the product if they selected any other response on any item (scored as 2–4; ie, ‘Probably not,’ ‘Probably yes,’ or ‘Definitely yes’,” or “A little,” “Somewhat,” or “Very curious”).

Participants reported their sex (male or female); age; race/ethnicity (recoded as Hispanic, or Non-Hispanic: White, Black, Asian, or Other); and sexual orientation identity (straight, gay/lesbian, or bisexual).

Analyses were conducted in SAS 9.4 and SUDAAN 11.0.3 and were weighted using balanced repeated replication weights with Fay’s adjustment.

We reported weighted frequencies for nicotine belief and perception items and reported weighted descriptive statistics and 95% confidence intervals by the following wave 4 participant characteristics: sex, age (12–14, 15–17), race/ethnicity, sexual orientation identity, cigarette, and e-cigarette susceptibility, and current (past 30 days) cigarette and e-cigarette use status. We reported weighted Pearson correlations between wave 4 nicotine belief and perception items and other risk perception items. In these analyses, we used wave 4 cohort cross-sectional weights (R04_Y_C04WGT).

Using weighted binary logistic regression, we used wave 4 nicotine beliefs to predict (1) wave 4.5 current use of any tobacco, cigarettes, cigars, and e-cigarettes among people who did not use those products at wave 4, (2) whether a person switched from smoking cigarettes at wave 4 to using e-cigarettes at wave 4.5, and (3) whether a person switched from using e-cigarettes at wave 4 to smoking cigarettes at wave 4. In these analyses, the outcome variable was current product use at wave 4.5, with current use coded as 1 and current nonuse coded as 0. In these longitudinal analyses, we used wave 4.5 single-wave weights for the wave 4 cohort (X04_Y_S04WGT), consistent with recommendations.26 For analyses examining harm perceptions of nicotine in e-cigarettes and cigarettes, we controlled for absolute harm perceptions of e-cigarettes and cigarettes, respectively, so the analysis examines the extent to which nicotine beliefs predict future use above and beyond absolute harm perceptions (as they have already established predictors of behavior,7,8). Analyses also controlled for demographic characteristics (age, sex, and race/ethnicity).

The wave 4 cohort was weighted to be nationally representative. The weighted sample characteristics were: 50.8% female and 49.2% male; 53.1% non-Hispanic white, 13.2% non-Hispanic black, 10.0% non-Hispanic other or mixed race, and 23.6% Hispanic. Moreover, 39.9% of the sample was in middle school, and 60.1% of the sample was in high school.

Overall, in wave 4 most youth correctly responded that nicotine causes addiction (77.1% responded “definitely” or “probably yes”; Table 1). However, most youth also incorrectly responded that nicotine is the main substance that causes smoking-related cancer (74.70% responded “definitely yes” or “probably yes”). Overall, most youth rated the nicotine in cigarettes, e-cigarettes, and NRT as “very harmful” or “extremely harmful.” Based on comparing means and 95% confidence intervals (Table 1), participants rated the harm of nicotine in cigarettes higher than the harm of nicotine in e-cigarettes; the harm of NRT was rated lower than both products. Across products, few participants responded “don’t know” in response to these questions (about 1%–3%; Table 1).

Distribution of Responses to Youth Nicotine Beliefs and Perceptions, Weighted Percent, 95% Confidence Intervals, and Unweighted Sample Size, PATH Study Wave 4

PATH = Population Assessment of Tobacco and Health Study.

Distribution of Responses to Youth Nicotine Beliefs and Perceptions, Weighted Percent, 95% Confidence Intervals, and Unweighted Sample Size, PATH Study Wave 4

PATH = Population Assessment of Tobacco and Health Study.

Based on wave 4 descriptive statistics, perceptions of the harms of nicotine and beliefs about nicotine being the main substance that causes smoking-related cancer appeared to differ by demographic variables; however, beliefs about nicotine causing addiction tended to be more similar across demographic groups (Supplementary Table 2). On average, older youth (ages 15–17) appeared to rate all products as less harmful than their younger peers (ages 12–14). Consistent with this, fewer older youth thought that nicotine was the main substance that causes smoking-related cancer compared to younger youth. Females consistently rated the harms of nicotine higher than males did, and more females thought nicotine was the main substance that causes smoking-related cancer and addiction compared to males. White youth perceived nicotine in different products as less harmful compared to black youth; additionally, more Hispanic youth believed nicotine was the main substance that causes smoking-related cancer compared to whites, but fewer Hispanic youth thought nicotine was the substance that caused addiction. Youth identifying as bisexual appeared to have lower nicotine harm perceptions than their heterosexual counterparts, and fewer youth identifying as bisexual believed nicotine was the main substance that causes smoking-related cancer compared to their heterosexual counterparts. Youth categorized as susceptible to cigarettes or e-cigarettes had lower harm perceptions compared to those categorized as non-susceptible; furthermore, fewer susceptible youth believed nicotine was the main substance that causes smoking-related cancer. Finally, youth who currently smoke cigarettes and youth who currently use e-cigarettes had lower nicotine harm ratings compared to youth who did not use, and fewer youth who did not use believed nicotine causes cancer.

Nicotine beliefs and harm perceptions were generally correlated with one another (Table 2). The belief that nicotine was the main substance that causes smoking-related cancer was weakly, but significantly, positively correlated with the belief that nicotine is addictive. Harm perceptions of nicotine in different tobacco products were moderately correlated with one another. Absolute harm perceptions of cigarettes and e-cigarettes were moderate to strongly positively correlated with the perceptions of harms of nicotine in the three products.

Weighted Correlation Between Youth Nicotine Belief and Perception Items, r (95% Confidence Intervals), PATH Study, Wave 4

Bolded numbers are statistically significant at p < .05. Analyses exclude participants who selected the “don’t know” response option for each nicotine belief and perception item. Waves 4 was collected from December 2016 to January 2018.

1 Do you believe nicotine is the chemical that causes most of cancer caused by smoking cigarettes? 1= Definitely yes to 4= Definitely no.

2 Do you believe nicotine is the main substance in tobacco that makes people want to use tobacco products?

1 = Definitely yes to 4 = Definitely no.

PATH = Population Assessment of Tobacco and Health Study.

Weighted Correlation Between Youth Nicotine Belief and Perception Items, r (95% Confidence Intervals), PATH Study, Wave 4

Bolded numbers are statistically significant at p < .05. Analyses exclude participants who selected the “don’t know” response option for each nicotine belief and perception item. Waves 4 was collected from December 2016 to January 2018.

1 Do you believe nicotine is the chemical that causes most of cancer caused by smoking cigarettes? 1= Definitely yes to 4= Definitely no.

2 Do you believe nicotine is the main substance in tobacco that makes people want to use tobacco products?

1 = Definitely yes to 4 = Definitely no.

PATH = Population Assessment of Tobacco and Health Study.

Harm perceptions of nicotine in cigarettes, e-cigarettes, and NRT at wave 4 were negatively associated with currently using any tobacco product, e-cigarettes, cigarettes, and cigars about 1 year later at wave 4.5, controlling for wave 4 absolute harm perceptions (Table 3). Among youth who currently used e-cigarettes but not cigarettes at wave 4 (unweighted n = 251), harm perceptions of nicotine in cigarettes and e-cigarettes did not predict switching to cigarette smoking at wave 4.5 (Table 4). Similarly, among youth who currently smoked cigarettes but did not use e-cigarettes at wave 4 (unweighted n = 150), nicotine harm perceptions did not predict switching to e-cigarette use at wave 4.5 (Table 4). These analyses controlled for absolute harm perceptions (eg, analyses with harm perceptions of nicotine in cigarettes as the predictor controlled for absolute harm perceptions of cigarettes).

PATH Study Wave 4 Youth Current Non-tobacco Users’ Perceived Nicotine Harm Predicting Wave 4.5 Current Product Use (Current Nonuser = 0, Current User = 1), Odds Ratios and 95% Confidence Intervals

Bolded numbers were statistically significant at p < .05. Both adjusted and unadjusted analyses exclude w4 participants aged 17 as these participants are moved to the adult data set at age 18. Adjusted analyses controlled for age sex, and ethnicity. Analyses exclude participants who selected the “don’t know” response option for each nicotine belief and perception item (shown in Table 1). Wave 4 was collected from December 2016 to January 2018; wave 4.5 was collected from December 2017 to December 2018.

1 Current use is defined as any use in the past 30 days.

2 Analyses control for absolute harm perceptions of cigarettes.

3 Analyses control for absolute harm perceptions of e-cigarettes.

4 PATH defines use of any tobacco product as use of cigarettes, e-cigarettes or other electronic nicotine products, cigars, pipe, hookah, snus, or other types of smokeless tobacco (such as dip, spit, or chew), and dissolvable tobacco. This is a derived variable.

5E-cigarettes in PATH are described with the following preamble: “The next questions are about electronic nicotine products, such as e-cigarettes, vape pens, personal vaporizers and mods, e-cigars, e-pipes, e-hookahs, and hookah pens. These products are battery-powered and produce vapor or aerosol instead of smoke. They contain nicotine liquid, sometimes called “e-liquid” or “e-juice,” although the amount of nicotine can vary and some may not contain any nicotine at all. Some can be bought as one-time, disposable products, while others can be bought as reusable kits that are rechargeable. Some common brands include Vuse, Blu, Logic, MarkTen, NJOY, eGo, and iTaste. such as e-cigarettes, vape pens, personal vaporizers and mods, e-cigars, e-pipes, e-hookahs, and hookah pens.

6Any cigar use includes traditional cigars, cigarillos, or little filtered cigars. Cigars in PATH are described with the following preamble: The next questions are about traditional cigars, cigarillos, and filtered cigars. These products go by lots of different names, so please use these descriptions and photos to understand what they are. Traditional cigars contain tightly rolled tobacco that is wrapped in a tobacco leaf. Some common brands of cigars include Macanudo, Romeo y Julieta, Arturo Fuente, and Cohiba, but there are many others… Cigarillos and filtered cigars are smaller than traditional cigars. They are usually brown. Some are the same size as cigarettes, and some come with filters or with plastic or wooden tips. Some common brands are black & Mild, Swisher Sweets, Dutch Masters, Phillies Blunts, Zig Zag, and Cheyenne.

PATH = Population Assessment of Tobacco and Health Study.

PATH Study Wave 4 Youth Current Non-tobacco Users’ Perceived Nicotine Harm Predicting Wave 4.5 Current Product Use (Current Nonuser = 0, Current User = 1), Odds Ratios and 95% Confidence Intervals

Bolded numbers were statistically significant at p < .05. Both adjusted and unadjusted analyses exclude w4 participants aged 17 as these participants are moved to the adult data set at age 18. Adjusted analyses controlled for age sex, and ethnicity. Analyses exclude participants who selected the “don’t know” response option for each nicotine belief and perception item (shown in Table 1). Wave 4 was collected from December 2016 to January 2018; wave 4.5 was collected from December 2017 to December 2018.

1 Current use is defined as any use in the past 30 days.

2 Analyses control for absolute harm perceptions of cigarettes.

3 Analyses control for absolute harm perceptions of e-cigarettes.

4 PATH defines use of any tobacco product as use of cigarettes, e-cigarettes or other electronic nicotine products, cigars, pipe, hookah, snus, or other types of smokeless tobacco (such as dip, spit, or chew), and dissolvable tobacco. This is a derived variable.

5E-cigarettes in PATH are described with the following preamble: “The next questions are about electronic nicotine products, such as e-cigarettes, vape pens, personal vaporizers and mods, e-cigars, e-pipes, e-hookahs, and hookah pens. These products are battery-powered and produce vapor or aerosol instead of smoke. They contain nicotine liquid, sometimes called “e-liquid” or “e-juice,” although the amount of nicotine can vary and some may not contain any nicotine at all. Some can be bought as one-time, disposable products, while others can be bought as reusable kits that are rechargeable. Some common brands include Vuse, Blu, Logic, MarkTen, NJOY, eGo, and iTaste. such as e-cigarettes, vape pens, personal vaporizers and mods, e-cigars, e-pipes, e-hookahs, and hookah pens.

6Any cigar use includes traditional cigars, cigarillos, or little filtered cigars. Cigars in PATH are described with the following preamble: The next questions are about traditional cigars, cigarillos, and filtered cigars. These products go by lots of different names, so please use these descriptions and photos to understand what they are. Traditional cigars contain tightly rolled tobacco that is wrapped in a tobacco leaf. Some common brands of cigars include Macanudo, Romeo y Julieta, Arturo Fuente, and Cohiba, but there are many others… Cigarillos and filtered cigars are smaller than traditional cigars. They are usually brown. Some are the same size as cigarettes, and some come with filters or with plastic or wooden tips. Some common brands are black & Mild, Swisher Sweets, Dutch Masters, Phillies Blunts, Zig Zag, and Cheyenne.

PATH = Population Assessment of Tobacco and Health Study.

PATH Study Wave 4 Nicotine Harm Perceptions Predicting E-cigarette Users and Smokers Becoming Smokers and E-cigarette Users, Respectively, at Wave 4.5; Odds Ratios and 95% Confidence Intervals, (Current Nonuser = 0, Current User =1)

Adjusted analyses controlled for age group (ages 12–14 or 15–17), sex, and ethnicity. Analyses exclude participants who selected the “don’t know” response option for each nicotine belief and perception item. Wave 4 was collected from December 2016 to January 2018; Wave 4.5 was collected from December 2017 to December 2018.

1 Analyses control for absolute harm perceptions of cigarettes.

2 Analyses control for absolute harm perceptions of e-cigarettes.

3 E-cigarettes in PATH are described with the following preamble: “The next questions are about electronic nicotine products, such as e-cigarettes, vape pens, personal vaporizers and mods, e-cigars, e-pipes, e-hookahs, and hookah pens. These products are battery-powered and produce vapor or aerosol instead of smoke. They contain nicotine liquid, sometimes called “e-liquid” or “e-juice,” although the amount of nicotine can vary and some may not contain any nicotine at all. Some can be bought as one-time, disposable products, while others can be bought as reusable kits that are rechargeable. Some common brands include Vuse, Blu, Logic, MarkTen, NJOY, eGo, and iTaste. such as e-cigarettes, vape pens, personal vaporizers and mods, e-cigars, e-pipes, e-hookahs, and hookah pens.

PATH = Population Assessment of Tobacco and Health Study.

PATH Study Wave 4 Nicotine Harm Perceptions Predicting E-cigarette Users and Smokers Becoming Smokers and E-cigarette Users, Respectively, at Wave 4.5; Odds Ratios and 95% Confidence Intervals, (Current Nonuser = 0, Current User =1)

Adjusted analyses controlled for age group (ages 12–14 or 15–17), sex, and ethnicity. Analyses exclude participants who selected the “don’t know” response option for each nicotine belief and perception item. Wave 4 was collected from December 2016 to January 2018; Wave 4.5 was collected from December 2017 to December 2018.

1 Analyses control for absolute harm perceptions of cigarettes.

2 Analyses control for absolute harm perceptions of e-cigarettes.

3 E-cigarettes in PATH are described with the following preamble: “The next questions are about electronic nicotine products, such as e-cigarettes, vape pens, personal vaporizers and mods, e-cigars, e-pipes, e-hookahs, and hookah pens. These products are battery-powered and produce vapor or aerosol instead of smoke. They contain nicotine liquid, sometimes called “e-liquid” or “e-juice,” although the amount of nicotine can vary and some may not contain any nicotine at all. Some can be bought as one-time, disposable products, while others can be bought as reusable kits that are rechargeable. Some common brands include Vuse, Blu, Logic, MarkTen, NJOY, eGo, and iTaste. such as e-cigarettes, vape pens, personal vaporizers and mods, e-cigars, e-pipes, e-hookahs, and hookah pens.

PATH = Population Assessment of Tobacco and Health Study.

This study of youth complements prior research on nicotine beliefs among U.S. adults.,9,16 Four-fifths of U.S. youth correctly believed that nicotine is the main substance in tobacco products that makes people want to use tobacco (80.1%), which is similar to the percentage of U.S. adults who endorse this belief (83%).9 Conversely, a large proportion of youth incorrectly agreed that nicotine is the substance responsible for most cancer caused by smoking (74.7%), compared to adults in a prior study (49%9). This incorrect belief was especially common among the youngest youth (ages 12–14) and those who were female, among whom approximately four out of five agreed with the statement. Youth who were Hispanic or non-Hispanic black were also particularly likely to hold this belief; this pattern was similar to prior findings among U.S. adults.9 Finally, this belief was more common among youth who currently did not smoke compared to youth who did, as previously observed among adults,9 and more common among youth who did not use e-cigarettes compared to youth who did.

In this study, there were clear differences in how participants viewed nicotine, with the nicotine in cigarettes being viewed as most harmful, followed by the nicotine in e-cigarettes and then NRT. Similarly, perceptions of the harm of nicotine in one tobacco product consistently correlated with other measures of harm. For example, beliefs about the harms of nicotine in e-cigarettes positively correlated with perceptions of the overall harm of e-cigarettes as well as perceptions of the overall harm of cigarettes. Other studies show that both youth and adults perceive e-cigarettes as less dangerous than cigarettes.20,27,28 It is possible that participants read the question about the harms of nicotine and still considered the harm of the product in its entirety, rather than nicotine specifically.

The current study also found that higher perceptions of the risks of nicotine in tobacco products were associated with lower odds of subsequently moving from being a youth who did not currently use it to a person who did. This finding is similar to findings from other studies that evaluated the relationship between risk perceptions of tobacco use and subsequent use among youth20,29,30 For example, Strong and colleagues20 found that youth at wave 1 of PATH with lower perceptions of harm or addictiveness of tobacco products were more likely to report trying the product at wave 2. Findings such as these are generally framed as indicating that high perceptions of risks are protective against subsequent use. However, controlling for absolute harm perceptions, nicotine harm perceptions did not predict youth product switching, either from cigarettes to e-cigarettes or from e-cigarettes to cigarettes. Although findings may be limited by reduced statistical power because of smaller unweighted samples of both youth who currently used e-cigarettes and youth who currently smoked cigarettes, it is also possible that nicotine harm perceptions do not play a strong role in product switching above and beyond absolute harm perceptions.

Findings from the current study, however, still raise the possibility of future negative unintended consequences to youth having high perceptions of risks of nicotine Over half the participants in this study believed that the nicotine in NRT was very or extremely harmful. Although higher perceptions of the harms related to nicotine may be associated with reduced odds of becoming a current user of different tobacco products, these same perceptions may reduce the likelihood that individuals wanting to quit tobacco will try to do so using NRT. Consistent with this possibility, Shiffman et al.17 found that, in a sample of current and former smokers, two-thirds believed that NRT was just as dangerous as cigarettes, and this belief was associated with reduced odds of using NRT in a quit attempt. This is a critical consideration as NRT increases the chance of successful cessation, and combinations of different types of NRT can further enhance a person’s likelihood of quitting.18 Thus, from a public health standpoint, it is important for individuals to understand that nicotine is addictive, and this addiction is the main reason why people who smoke continue to do so31; however, it is also important for individuals to understand that nicotine is not the main reason why using tobacco is harmful to health, and that NRT is a safe and effective treatment for tobacco dependence.

Findings from this study highlight one potential challenge of effectively communicating about the harms of nicotine, as such communication may reduce a misbelief that may be protective against future tobacco use. Studies of adults have considered how to accurately describe the harm of nicotine. For example, Byron et al.13 studied how to effectively communicate about the nicotine and harm in low-nicotine cigarettes. The authors highlighted the need for a comprehensive approach that describes the role nicotine plays in addiction as well as the harm of the tobacco product itself. For example, low-nicotine cigarettes can be less addictive but no less deadly than conventional cigarettes if smoked in the same way. Yang et al.32 found that participants who viewed a nicotine fact sheet were more likely than control participants to disagree that nicotine was the main cause of smoking-related disease (26.2% vs. 12.7%); however, the majority of participants who saw the fact sheet still held this incorrect belief. Future research could explore how youth understand the role that nicotine plays in both addiction and harm of tobacco use as well as how to communicate this information effectively to youth.

This study has several limitations. The longitudinal analyses excluded participants who were aged 17 at wave 4 because they aged up to young adults (age 18) at the time of the wave 4.5 interviews and were not eligible for the youth interview. Finally, analyses predicting product initiation were each conducted independently, so analyses of one product did not address a participant’s use of another product.

Findings from this nationally representative study further our understanding of youths’ perceptions of nicotine. Lower perceptions of the harm of nicotine in cigarettes, e-cigarettes, and NRT were all associated with higher odds of currently using any tobacco product, e-cigarettes, cigarettes, or cigars about one year later. These findings are consistent with the possibility that greater harm perceptions of nicotine are potentially both protective (preventing initiation among youth who do not use tobacco) and harmful (preventing youth who use combustible tobacco from switching to NRT). This underscores the challenge of developing effective and comprehensive communication strategies that accurately communicate the effects of nicotine without encouraging tobacco use.

When this work was initiated, this author was employed at the Center for Tobacco Products. She completed it while at her current position at the National Cancer Institute, Tobacco Control Research Branch, Rockville, MD, USA.

This manuscript is supported with Federal funds from the Center for Tobacco Products (CTP), Food and Drug Administration (FDA), Department of Health and Human Services (HHS), through an interagency agreement between the FDA Center for Tobacco Products and the National Institute on Drug Abuse, National Institutes of Health, and through an interagency agreement between the FDA Center for Tobacco Products and the Centers for Disease Control and Prevention. The PATH Study is supported with federal funds from NIH NIDA and FDA CTP, under contract to Westat (contract nos. HHSN27101100027C and HHSN27120160001C) (NIDA Federal Author name) was substantially involved in the scientific management of and providing scientific expertise for contract nos. HHSN27101100027C and HHSN27120160001C.

We acknowledge and thank Tesfa Alexander, who brought the author team together to study this important topic.

The authors have no conflicts of interest.

EO led project initiation, conducted analyses, and drafting the first draft of the manuscript. MR led project planning, drafted portions of the introduction and discussion, and coordinated revisions among the authorship team. AP provided analytic and methodological direction and support and wrote sections of the introduction and discussion. KM contributed to the planning, framing, and writing portions of the introduction and discussion.

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the U.S. Department of Health and Human Services or any of its affiliated institutions or agencies.

The data used in these analyses are from the Population Study of Tobacco and Health, available to the public at https://www.icpsr.umich.edu/web/NAHDAP/studies/36231. The Methods describes the waves of data used, and Supplementary Table 1 describes the exact variable names used in analyses.

Benowitz   NL.  

Google Scholar

WorldCat

Leslie   FM.  

Google Scholar

WorldCat

Apelberg   BJ, Feirman   SP, Salazar   E, et al. .

Google Scholar

WorldCat

O’Brien   EK, Persoskie   A, Tam   J.  

Google Scholar

WorldCat

Strong   DR, Leas   E, Elton-Marshall   T, et al. .

Google Scholar

WorldCat

O’Brien   EK, Nguyen   AB, Persoskie   A, Hoffman   AU.  

Google Scholar

WorldCat

Bansal-Travers   M, Hammond   D, Smith   P, Cummings   KM.  

Google Scholar

WorldCat

Borrelli   B, Novak   SP.  

Google Scholar

WorldCat

Mutti   S, Hammond   D, Borland   R, Cummings   MK, O’Connor   RJ, Fong   GT.  

Google Scholar

WorldCat

Byron   MJ, Hall   MG, King   JL, Ribisl   KM, Brewer   NT.  

Google Scholar

WorldCat

Byron   MJ, Jeong   M, Abrams   DB, Brewer   NT.  

Google Scholar

WorldCat

Denlinger-Apte   RL, Joel   DL, Strasser   AA, Donny   EC.  

Google Scholar

WorldCat

Villanti   AC, Naud   S, West   JC, et al. .

Google Scholar

WorldCat

Shiffman   S, Ferguson   SG, Rohay   J, Gitchell   JG.  

Google Scholar

WorldCat

Ferguson   SG, Gitchell   JG, Shiffman   S, Sembower   MA, Rohay   JM, Allen   J.  

Google Scholar

WorldCat

Vogt   F, Hall   S, Marteau   TM.  

Google Scholar

WorldCat

East   K, Brose   LS, McNeill   A, Cheeseman   H, Arnott   D, Hitchman   SC.  

Google Scholar

WorldCat

Carpenter   MJ, Ford   ME, Cartmell   K, Alberg   AJ.  

Google Scholar

WorldCat

Hyland   A, Ambrose   BK, Conway   KP, et al. .

Google Scholar

WorldCat

Pierce   JP, Choi   WS, Gilpin   EA, Farkas   AJ, Merritt   RK.  

Google Scholar

WorldCat

Fong   GT, Elton-Marshall   T, Driezen   P, et al. .

Google Scholar

WorldCat

Persoskie   A, O’Brien   EK, Nguyen   AB, Tworek   C.  

Google Scholar

WorldCat

Song   AV, Morrell   HER, Cornell   JL, et al. .

Google Scholar

WorldCat

Chaffee   BW, Couch   ET, Urata   J, Gansky   SA, Essex   G, Cheng   J.  

Google Scholar

WorldCat

Creamer   MR, Wang   TW, Babb   S, et al. .

Google Scholar

WorldCat

Yang   B, Owusu   D, Popova   L.  

Google Scholar

WorldCat