According to the Family Smoking Prevention and Tobacco Control Act of 2009 (FSPTCA),1 consumer perceptions of labels or marketing statements for modified risk tobacco products (MRTPs) should be tested to show that they will not mislead the consumer to believe that the product is less harmful or demonstrates less risk than is actually true. As such, on an annual basis, pre- and postmarket studies should be conducted to demonstrate that current and potential consumers of each MRTP understand the actual and relative risks of the product. As discussed in Chapter 1, the FSPTCA articulates a public health standard whereby product sponsors must conduct studies on the effect of the product on the population as a whole. As outlined in the law, this evaluation of the health of the population must include studies demonstrating that (1) perceptions of less risk from the MRTP do not result in nontobacco users initiating tobacco use, (2) existing tobacco users who would otherwise consider quitting all tobacco products do not switch to this new MRTP, and (3) usage of tobacco products does not increase as a result of this new product.
This chapter begins with a brief review of how users and nonusers perceive tobacco-related outcomes, including perceptions of epidemiologic data, short- and long-term risks to the individual, addiction, and potential benefits. Careful attention is given regarding perceptions of different types of tobacco products, as well as how perceptions of tobacco
1 Family Smoking Prevention and Tobacco Control Act of 2009, Public Law 111-31, 123 Stat. 1776 (June 22, 2009)
use outcomes vary by age and demographics. Next, the chapter outlines the standards for studies on risk perceptions, including the questions that should be addressed through the studies, standards for the research designs, participant recruitment, measurement, and analysis.
Judgments about risk, otherwise known as risk perceptions, are viewed as a fundamental element of most theoretical models of health behavior and behavioral decision making, including social cognitive theory (Bandura, 2001), the health belief model (Rosenstock, 1974), the theory of reasoned action (Fishbein and Ajzen, 1975), the theory of planned behavior (Ajzen, 1985), self-regulation theory (Kanfer, 1970), and subjective culture and interpersonal relations theory (Triandis, 1977). In general, these models argue that individuals’ perceptions about the value and likelihood of behavior-related positive and negative consequences and their vulnerability to those consequences play a key role in behavioral choices. As such, understanding individuals’ perceptions of tobacco- related products, including MRTPs, whether such perceptions change over time with the introduction of MRTPs, and whether such perceptions play a role in tobacco use behavior, is critical. The committee also acknowledges, as the 2007 Institute of Medicine (IOM) report articulated, that perceptions of risk (and benefit) may have differing implications for product use among different consumers. It is important to understand both the risk (and benefit) perceptions of the consumer and the value that is placed upon these perceptions.
In the next few sections, the committee provides an overview of the literature on tobacco-related perceptions, followed by methodological considerations to design studies to determine perceptions and behavioral implications of MRTPs.
It is critical to first understand the extent to which both tobacco users and nonusers understand the actual risks of tobacco use, compared to epidemiologic data. Much of the literature comparing perceptions to actual data suggests that, on average, smokers overestimate the risks of smoking (Borland, 1997; Johnson et al., 2002; Kristiansen et al., 1983; Viscusi, 1990, 1991, 1992), while other studies show that smokers underestimate them (Arnett, 2000; Hansen and Malotte, 1986; Leventhal et al., 1987; Schoenbaum, 1997; Sutton, 1998; Virgili et al., 1991). Among adolescents and young adults (ages 18-22), Jamieson and Romer (2001) found that
70 percent of smokers and 79 percent of nonsmokers overestimated the risk of contracting lung cancer from smoking. Just over a third of the smokers and more than 40 percent of nonsmokers overestimated the risk of death from smoking, and 41 percent of smokers and 27 percent of nonsmokers either underestimated or did not know this rate (Jamieson and Romer, 2001). About a quarter of the nonsmoking participants and 21 percent of the smokers also underestimated the number of years of life that would be lost due to smoking, and they inaccurately perceived more deaths caused by gunshots, car accidents, alcohol, and other drug use than by smoking cigarettes (Jamieson and Romer, 2001). Given people’s limited understanding of tobacco-related risk, MRTP labels and advertisements should be careful to convey information on tobacco-related risks accurately and in a manner that can be fully comprehended by the general population.
A great number of studies have examined both smokers’ and non-smokers’ perceptions of tobacco-related outcomes, including perceived short- and long-term health risks, social risks, risks of becoming addicted, risks from secondhand smoke, and cumulative risks. Findings on these tobacco-related perceptions as well as the important relationship between perceptions and tobacco use are reviewed next.
Historically, studies of tobacco-related perceptions were largely focused on perceptions of long-term health risks associated with smoking, such as heart attack and lung cancer. More recently, there has been an emphasis on short-term health and social risks that are more pertinent to adolescents and even adults, such as the smell of cigarettes, the yellowing of teeth, and the possibility of getting into trouble (Gritz et al., 2003; Halpern-Felsher et al., 2004; IOM, 2007; Prokhorov et al., 2002).
Studies have also examined whether such tobacco-related perceptions are related to actual tobacco use. There have been a number of studies that have relied on cross-sectional data to test the relationship between adolescents’ perceived tobacco risk and actual tobacco use. The bulk of these findings indicate that adolescents who have smoked hold lower perceptions of risk than adolescents who have not smoked (Jamieson and Romer, 2001; Romer and Jamieson, 2001).
Using prospective, longitudinal data to examine whether perceptions actually predict the initiation of tobacco use, Song and colleagues (2009b) showed that, compared to adolescents with the highest perceptions of tobacco-related risks, adolescents with the lowest perceptions of tobacco- related long-term risks were 3.64 times more likely to initiate tobacco use. The same relationship was observed with perceptions of short-term risks,
whereby the adolescent participants who believed that tobacco-related short-term risks were unlikely were 2.68 times more likely to initiate smoking compared to adolescents with higher perceptions of short-term risks (Song et al., 2009b).
In addition to understanding the extent to which adolescent and adult smokers and nonsmokers perceive tobacco-related risks and whether these risk perceptions deter tobacco use, it is critical to learn the extent to which perceived tobacco-related benefits motivate individuals to use a tobacco product. Indeed, studies provide support that perceived benefits are an equally, if not more important, component of the decision equation. For example, Prokhorov and colleagues (2002) found that scores on a smoking-related pros or benefits scale increased and con scores decreased as adolescents became more susceptible to smoking. Pallonen et al. (1998) showed that nonsmokers were more likely to initiate tobacco use if they perceived more smoking benefits, whereas perceived smoking risks were less related to smoking onset.
Halpern-Felsher et al. (2004), as well as Goldberg et al. (2002), found that participants who have smoked perceive benefits more likely to occur, and risks less likely to occur, compared to adolescents who have not smoked. Results from more recent longitudinal studies have demonstrated that adolescents who report the highest perceptions of smoking-related benefits are as much as 3.3 times more likely to initiate smoking (Song et al., 2009b), and that adolescents who have experimented with as little as one puff of cigarette have greater perceptions of benefits compared to those who have never smoked (Morrell et al., 2010).
In summary, adolescents’ perceptions of the risks and benefits of cigarette smoking play an important role in adolescents’ decisions to smoke, and adolescents with lower perceptions of tobacco risks are more likely to initiate tobacco use. It is therefore essential that studies of consumer perceptions examine whether the information about MRTPs that is provided to consumers affects the perceived risks and benefits of the products, and what implications these perceptions have for subsequent use of the MRTP in relation to pre-existing tobacco products. Given that adolescence is a period of heightened vulnerability for the initiation of tobacco use, it is particularly important to evaluate whether adolescents accurately understand the purported benefits of an MRTP. The ethical considerations for studies involving populations at high risk for tobacco initiation, such as adolescents, are discussed in Chapter 2 and Chapter 6.
Other aspects of tobacco-associated risks that are not fully understood by many adolescents and young adults include misunderstandings about nicotine addiction and the ability to quit using tobacco products. Studies suggest that smokers and nonsmokers are not fully aware of the addictive nature of smoking (Arnett, 2000; DiFranza et al., 2011; Halpern-Felsher et
al., 2004; Leventhal et al., 1987; Slovic, 1998, 2001). It is argued that adolescent smokers may be less concerned about the long-term risks of smoking partly because they believe that they can stop smoking easily and at any time (Arnett, 2000; Halpern-Felsher et al., 2004; IOM, 2007; Slovic, 1998).
Perceptions of addiction go beyond the physical need to smoke, and include fulfilling an emotional or social need, such as avoiding unpleasant mood states or wanting to socially relate to others (Johnson et al., 2003). Rugaska et al. (2001) concluded that youth perceive dependence risks to be associated solely with adult smoking; the authors found that adolescents believe their underage smoking for social settings was safe, in contrast to adults who smoke to cope with everyday life stress.
Weinstein et al. (2004) examined smokers’ beliefs concerning the ease of quitting and the nature of addiction. They found that more than 96 percent of the adolescents and adults in their study agreed with the statement, “the longer you smoke, the harder it is to quit,” and most believed that addiction develops quickly. Other analyses have found that smokers are relatively optimistic about the idea of addiction, believing that smoking cessation is not that difficult (Jamieson and Romer, 2001) and overestimating the ease with which a smoker can quit (Weinstein et al., 2004).
When inquired about the ease of quitting smoking, adolescents with smoking experience believed they will find it easier to quit and will be more likely to quit smoking compared to adolescents without smoking experience (Halpern-Felsher et al., 2004). Arnett (2000) found that 60 percent of the adolescents and almost half of the adults in their study endorsed the idea that they could smoke for a few years and then quit if and when they wanted. Weinstein et al. (2005) found differences in perceptions of risks between smokers who did and did not plan to quit smoking, with those planning to quit recognizing higher risks of lung cancer.
In addition to examining perceptions of personal risk from smoking, a few studies have examined perceptions of risk from secondhand smoke, including risk to others if you smoke, and personal risk from others’ smoke. Glantz and Jamieson (2000) found that youth who smoked were less likely than nonsmoking youth to believe that secondhand smoke leads to thousands of deaths each year. They also showed that adolescents who planned to quit smoking were more aware of the effects of secondhand smoke than were smokers without quit intentions. Romer and Jamieson (2001) found that knowledge of secondhand smoke harm was indirectly related to intentions to quit because of its relationship with perceived risk of smoking overall. Kurtz and colleagues (2001) showed that elementary, middle, and high school students with smoking experience were less knowledgeable about and had less negative views of secondhand smoke compared to students without smoking experience. Similarly, Halpern-Felsher and Rubinstein (2005) found that adolescents with
smoking experience perceived less risk from secondhand smoke than did adolescents without smoking experience. In a follow-up study, Song et al. (2009a) showed that perceptions of risk from secondhand smoke predicted smoking initiation, with adolescents with the lowest perceived risk of secondhand smoke being the most likely to subsequently try smoking.
Taken together, this set of literature demonstrates the need to understand and describe perceptions of tobacco-related outcomes, including perceptions of short- and long-term risks, addiction, and potential benefits. It is also important to understand perceptions concerning secondhand smoke as well as other tobacco products. These studies aid in identifying critical perceptions held by smokers and nonsmokers; perceptions are also instrumental in predicting subsequent tobacco use and changes in patterns of use that are important to capture. Data from these studies should be included in the portfolio of evidence submitted to the Food and Drug Administration (FDA) when applying for a modified risk claim on a tobacco product.
Differences in Perceptions of Risks and Benefits by Type of Tobacco Product
A small set of literature has examined whether perceptions of risks and benefits vary by the type of tobacco product. Most of this research has examined perceptions of so called “light,” “ultra light,” and “low- tar” cigarettes. The studies show that adults have misperceptions about the health risks associated with smoking light and ultra light cigarettes; most adult smokers believe these cigarettes deliver less tar and nicotine, produce milder sensations, and result in less health consequences (Etter et al., 2003; Shiffman et al., 2001; Slovic, 2001). Studies have also shown that smokers have switched to these so-called lighter cigarettes to reduce the health risks of smoking (Slovic, 2001). Shiffman et al. (2001) examined the perceptions of light, ultra light, and regular cigarettes among adult daily smokers; participants believed that lights and ultra lights were less risky compared to regular cigarettes and that the ultra light cigarettes were the least harmful. Similarly, Etter et al. (2003) quantified the perceptions of smoking different cigarettes, showing that participants believed they needed to smoke two light cigarettes or four ultra light cigarettes to inhale the same amount of nicotine as one would inhale from a single regular cigarette. Etter and colleagues (2003) also found that current adult light cigarette smokers believed they were at less risk of developing lung cancer than did smokers of regular cigarettes.
Kropp and Halpern-Felsher (2004) extended these studies to examine adolescents’ perceptions of light cigarettes. In their study, adolescents believed they were significantly less likely to have a heart attack, get lung
cancer, have trouble breathing, get a bad cough, and die from a smoking-related disease if smoking light cigarettes compared with smoking regular cigarettes. The participants also believed that light cigarettes have less tar and nicotine than regular cigarettes and that it would be easier to quit smoking light compared to regular cigarettes.
A study of Norwegian older adolescents and young adults (aged 16-20 years) examined perceptions of different tobacco products, including roll-your-own tobacco, factory-made cigarettes, low-tar factory-made cigarettes, pipe tobacco, cigars or cigarillos, loose snus, prepackaged snus, and nicotine replacement therapies (NRTs). Participants rated roll-your-own tobacco as most harmful and NRTs less harmful (0verland et al., 2008). In a direct comparison, snus was considered less harmful than cigarettes on average, and participants who used snus rated it less harmful than did nonusers of snus (0verland et al., 2008). Callery and colleagues (2011) examined the relative health risk beliefs among a group of adult Canadian smokers (aged 18-30 years). They found that between 30 percent and 47 percent of the participants wrongly believed that smokeless tobacco and cigarettes are equally harmful, and some wrongly noted that smokeless tobacco is more harmful than cigarettes (Callery et al., 2011).
Other studies have examined whether smokers believe there are differences in harm based on type, brand, or color packaging of tobacco products. Mutti and colleagues (2011) showed that adult smokers attributed differential risks based on cigarette brands and packaging color (e.g., gold or silver compared to red or black). Smokers of light and mild cigarettes perceived their cigarette brand to be less harmful compared to others, as did smokers of cigarettes found in gold, silver, purple, or blue packages. Similarly, Bansal-Travers et al. (2011) perceived differences in risk based on color of the cigarette package, with white coloring denoting less risk.
These studies confirm that adults and adolescents, as well as smokers and nonsmokers, harbor misconceptions about tobacco products based on the packaging coloring or descriptors. As noted by a previous IOM committee (2007), “such perceptions are likely the result, in part, of the tobacco industry’s marketing of light cigarettes as the healthier smoking choice, a safer alternative to cessation, and a first step toward quitting smoking altogether.” More favorable perceptions of light, ultra light, and low-tar cigarettes are important to note, since many smokers have made the choice to smoke light cigarettes because they believe such cigarettes are less addictive or safer than regular cigarettes (Etter et al., 2003). Further, adults who smoke light or ultra light cigarettes might be less likely to attempt to quit smoking, believing that their cigarette choices provide a safer alternative to regular cigarette smoking (Etter et al., 2003; Shiffman et al., 2001).
Demographic Differences in Tobacco-Related Perceptions
With the exception of identifying age differences, there are surprisingly few studies that have examined differences in tobacco-related perceptions by other demographic variables, such as gender, race/ethnicity, or socioeconomic status. The small literature on these topics is reviewed next.
Previous studies have found limited gender-specific differences among smokers with regards to benefit perceptions of smoking. Among adults, women are more likely than men to be concerned about post-cessation weight gain, women are more likely to identify weight gain as the cause for relapse to smoking, and women are less likely to be motivated to quit smoking if they fear subsequent weight gain (Swan et al., 1993; Weekley, 1992). McKee et al. (2005) showed that adult females perceived both greater risk and greater benefits from smoking than did adult males. Others have found that women are less likely to acknowledge the health benefits of smoking cessation (Sorensen and Pechacek, 1987) and that men are more likely to quit smoking in order to have better health (Curry et al., 1997). Adolescent males report fewer health concerns than females, and they perceive fewer risks and greater benefits associated with a variety of health-related risky behaviors (Millstein and Halpern-Felsher, 2002). Taken together, these studies provide evidence to support the existence of gender-based differences in perceptions of the risks and benefits of smoking. These differences may also relate to why females have poorer smoking cessation outcomes as compared to males (Perkins, 2001). Thus, consumer perceptions of tobacco products applying for the modified risk claim should be explored separately for males and females in adolescent and adult samples.
Surprisingly few studies have examined cultural variation (including race, ethnicity, country of origin, acculturation, language usage, and social class) in perceptions, especially related to tobacco use. As described in a previous IOM report (2007),
it is possible that the level of perceived risk (and benefit) may differ across groups of individuals, possibly as a factor of culture, socioeconomic status, or differences in exposure to behavior-related outcomes, for example. Alternatively, groups of adolescents or young adults might perceive the same level of risk, but these perceptions might have different implications for their smoking, in part due to differences in perceived control, risk-reducing strategies used, or value placed on the negative outcome (e.g., bad breath or trouble breathing) compared to the value placed on the benefit (e.g., looking cool) of smoking.
Future studies are needed.
Adolescents’ Reasons for Smoking
Qualitative studies have used methods such as one-on-one interviewing or focus groups to understand the motivations for smoking (IOM, 2007). Based on these studies, the most commonly identified reasons for smoking include: to satisfy curiosity, to fit in with peers, to relieve stress and boredom, to decrease appetite, to increase the high from alcohol and drugs, and because parents smoke (Clark et al., 2002; Dunn and Johnson, 2001; Gittelsohn et al., 2001; Kegler and Cleaver, 2000; Nichter et al., 1997; Vuckovic et al., 2003). A previous IOM committee (2007) noted that “adolescents form perceptions of smoking images, such as nonsmokers being more mature (Lloyd et al., 1997), and adolescents recognize that different types of smoking identities (beyond the usual categories of nonsmokers, experimenters, and smokers) exist for adolescents (Johnson et al., 2003).” A number of studies indicate that such images have an impact on adolescents’ smoking. Gerrard and colleagues’ (2008) Prototype Willingness Model of adolescent risk behavior postulates that an adolescent’s image of a typical smoker or nonsmoker will influence his or her willingness to smoke, and ultimately his or her actual smoking behavior. Research confirms that adolescents who hold more favorable images of a typical smoker are more willing to smoke and accept the consequences of smoking (Gerrard et al., 2008).
Advertisements for tobacco products have targeted reasons for smoking across a variety of groups defined by demographic characteristics such as age (adolescents, young adults, and adults), gender, race, socioeconomic status, and psychosocial needs; they have also been directed at creating favorable images of smokers in order to increase sales (Anderson et al., 2005; Balbach et al., 2003; Carpenter et al., 2005; Cook et al., 2003; Cummings et al., 2002; Landrine et al., 2005; Ling and Glantz, 2004; Wakefield et al., 2002; Wayne and Connolly, 2002). Pre- and postmarket studies should show that perceptions of MRTPs do not cause consumers to increase use of harmful tobacco products or lead to dual use of MRTPs and conventional tobacco products.
Study Questions to Address the Risk Perceptions of Modified Risk Tobacco Products
With reference to each MRTP, it will be important to identify consumers’ perceptions of disease risk, likelihood of addiction, likelihood of reducing or increasing others’ exposure to potentially hazardous compounds (e.g., secondhand smoke), and perceptions of risk compared to
other products that are already on the market. Perceptions of general harm, such as overall risk of harm or addiction, as well as perceptions of specific harm, such as risk of lung cancer or heart disease, should be studied. It is also important to establish consumers’ intentions of using the product, both for consumers who do and do not currently use any other tobacco product. Of particular importance are adolescents’ perceptions of the risks and benefits of using the product, and whether they intend to initiate tobacco use with the MRTP rather than a traditional tobacco product because they believe the latter is a “safe” alternative. These issues should be addressed in both pre- and postmarket studies.
Studies of risk perception should also include comprehensive questions that address the many aspects of risk perceptions, including areas which researchers may ordinarily regard as self-evident. For example, it is important to include questions about perceived risks of secondhand smoke to nonusers for all MRTPs, regardless if the product is inhaled or non-inhaled. Such a comprehensive approach will allow researchers and regulators to better understand all components of perceived risk reduction. In addition, longitudinal postmarket studies should address whether differences in perceptions and/or intentions among different age, racial, socioeconomic status, or education groups predict later product use, change in product use, or progression to dual use of MRTPs and traditional tobacco products.
This section outlines the committee’s review of research designs for use in pre- and postmarket studies of consumer perceptions of MRTPs. The focus of the discussion is on specific issues related to ethical procedures, target population selection and recruitment, construct measurement, and analysis.
To determine perceptions of MRTPs, as well as whether such perceptions influence tobacco use behavior, studies will need to occur both pre- and postmarket for each MRTP. Premarket research will play an essential role in developing the messages that the tobacco industry can use to communicate information about the MRTPs to consumers. This research will determine consumers’ ability to accurately understand messages that communicate information about the risks, benefits, and conditions of use pertaining to the MRTP itself and compared to existing tobacco products. Studies should also test how these messages influence consumers’ perceptions of the risks, benefits, and likelihood of addiction related to the MRTP. Clearly, no message developed can result in any significant misunderstanding, misinterpretation, or generalization of what exactly the MRTP is supposedly modifying. For example, if the
tobacco company claims that the product contains less nicotine, then the consumer or potential consumer cannot believe that the product also reduces the risk of lung cancer. Thus, the perceived influence of the new product on health and other outcomes should match the actual difference in health effects.
The first stage of premarket research will involve formative work using focus groups. Focus groups are useful for offering depth and insight from similar groups of people, especially when the intent is to gather general themes and ideas on topics not yet well studied. Focus groups are particularly useful when no existing research can provide the information, and they are an ideal way to generate new ideas that will be relevant for subsequent larger-scale studies, surveys, and future research (Krueger, 2000). These focus groups should consist of the target populations described below. The first phase of focus group research should include discussions with various groups of individuals regarding the best, most effective, and most comprehensible messaging that should be used to market and to label the product if the product is later approved as an MRTP. That is, what is the most accurate and easily comprehended message? The second phase should include discussions with groups of similar individuals to assess how the messages that were developed in phase 1 are received by consumers. Specifically, do potential consumers understand the messages correctly? Do the messages change intentions to use this MRTP or any other tobacco product?
Once messages that communicate potential risks and benefits of use are developed using the focus groups, the effects of these messages on consumer perceptions should be tested. Statements to be tested should include not only product labels or inserts intended to convey health information about the product, but also marketing statements that will appear on any form of advertisement of the MRTP. Nonverbal messages should be tested as well. For example, when banned from using labels such as “light” or “mild” on cigarette packages in countries other than the United States, the industry switched to using lighter colors to indicate “lighter” cigarettes. As a result, smokers perceived cigarettes in the lighter colored packs to be less harmful and easier to quit (Hammond and Parkinson, 2009; Hammond et al., 2009); this phenomenon has been replicated in a recent U.S. study as well (Bansal-Travers et al., 2011). Therefore, if the industry decides to use imagery, color-coding, or any other visual (but nonverbal) means of conveying information about the MRTP, then they should also test the influence of this type of messaging on consumer perceptions in pre- and postmarket studies, as well as its influence on use of the MRTP in postmarket studies.
The minimum standards to test consumer perceptions and understanding of messages about MRTPs include showing these messages to
participants in randomized order and then evaluating participants’ understanding of the messages and health outcomes affected by the message (see later section on measuring risk communication) and their subsequent perceptions of the product in terms of its potential risks, benefits, and likelihood of being addictive (see later section on measuring perceptions). Techniques such as eye-tracking could also be employed by researchers to study how research participants react to and understand warning labels, texts, or advertisements. It will be important to compare consumers’ perceptions of the MRTP to selected comparison products that are currently on the market, using experimental designs. Additionally, it will be informative to investigate how perceptions are linked to product use by the consumer. The relevance of behavioral economic self-administration studies in evaluating the reinforcement potency of a product is discussed in Chapter 4.
It will also be important to test consumers’ intentions to use the MRTP in general and compared to current products on the market (see later section on measuring intentions). That is, given information about a specific MRTP, questions to be investigated include (1) do participants plan to start using tobacco for the first time by using the MRTP, (2) do they intend to use it to help them quit smoking regular cigarettes or other traditionally available tobacco products, (3) do they intend to use both products concurrently, or (4) do they not intend to use the MRTP at all?
The studies required by FDA for products applying to switch from a prescription to over-the-counter (OTC) product may be useful in setting standards for studies on risk perceptions and risk communication. Under this requirement, prescription drug sponsors must conduct labeling comprehension studies to provide data on how the candidate OTC product label can inform the consumer about the product, including how the consumer can understand and apply the information presented on the drug label. The product itself does not need to be administered to the research participants. Although label comprehension studies may not fully predict consumer behavior once a prescription drug reaches the market as an OTC product, they can assist in creating a label that communicates effectively. The committee believes that the standards for the label comprehension studies required for a prescription-to-OTC switch can be useful in the regulation of MRTPs.
After the product has been approved by the FDA as an MRTP and released for general sale, it is vital to continue monitoring consumer perceptions and behavior related to that product via ongoing postmarket research. Conducting nationally representative cohort-sequential longitudinal surveys a minimum of three times per year (every 4 months) will be useful, with longitudinal studies ongoing until sufficient time has passed to be able to observe changes in tobacco use patterns. The longitudinal
aspect of the design will allow researchers to track changes in consumer perceptions and intentions over time; it will also determine how these perceptions influence subsequent usage of the new MRTP, initiation of other tobacco use, and changes in overall patterns of tobacco usage. The cohort-sequential aspect of the design will allow researchers to control for historical or age effects that may affect real and perceived outcomes (e.g., effects on perceived health risks, addiction risks, and actual usage). How long a particular cohort should be followed depends on the age group of the cohort. Ideally, children and adolescents should be followed at least through young adulthood (e.g., age 25) because this is the period in which most people begin to use tobacco products. Adults who begin the survey after age 25 may be followed for a shorter period of time, perhaps 3 to 5 years. The next section will provide more information on participants and sampling.
Populations to Be Studied
In a preceding section, a number of questions about consumer perceptions that should be addressed were outlined. Each of these questions should be asked and answered across a variety of important study populations.
Based on the scientific literature discussed earlier in this chapter, perceptions of MRTPs, including interpretation of marketing and health messages regarding particular MRTPs, and whether such perceptions influence changes in tobacco use, are likely to differ depending on whether or not consumers are current tobacco users, and whether or not current users desire to quit. Therefore, perceptions should be studied among people who have never used a tobacco product; people who have used any tobacco product in the past, but not currently; people who currently use a tobacco product and do not intend to quit; and people who currently use a tobacco product and do intend to quit, either with or without the use of NRT or other approved smoking cessation aids. Assessment of tobacco use with items from previously validated measures and surveys is standard. A list of sample items and their sources are listed in Box 5-1. Tobacco use should be assessed for each category of tobacco product separately: cigarettes, cigars, chewing tobacco, snuff, and pipe tobacco.
Among tobacco users, level of nicotine dependence should also be assessed and included as a potential predictor of differential perceptions toward the MRTP. Levels of nicotine dependence can be investigated by employing widely used and well-validated measures of nicotine dependence, such as the Fagerstrom Test for Nicotine Dependence and the Hooked on Nicotine Checklist (HONC) (DiFranza et al., 2002; Heatherton
• Have you smoked at least 100 cigarettes in your entire life? (CDC, 2010)
• Do you now smoke cigarettes every day, some days, or not at all? (CDC, 2010)
• How long has it been since you last smoked part or all of a cigarette? (during the past 30 days, more than 30 days ago but within the past 12 months, more than 12 months ago but within the past 3 years, more than 3 years ago) (SAMHSA, 2009)
• What is your best estimate of the number of days you smoked part or all of a cigarette during the past 30 days? (1 or 2 days, 3 to 5 days, 6 to 9 days, 10 to 19 days, 20 to 29 days, all 30 days) (SAMHSA, 2009)
• How many cigarettes per day do you smoke? (Heatherton et al., 1991)
• During your entire life, about how many times have you smoked a few puffs of a cigarette? (Lee and Halpern-Felsher, 2011; Song et al., 2009a)
• During your entire life, about how many times have you smoked a whole cigarette? (Lee and Halpern-Felsher, 2011; Song et al., 2009a)
• Have you ever smoked part or all of any type of cigar (including big cigars, cigarillos, or even little cigars that look like cigarettes)? (SAMHSA, 2009)
• Have you ever smoked tobacco in a pipe, even once? (SAMHSA, 2009)
• Have you ever used chewing tobacco, even once? (SAMHSA, 2009)
• Have you ever used snuff, even once? (SAMHSA, 2009)
et al., 1991). Additional measures of nicotine dependence include the nicotine dependence criteria of the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV), as well as the Nicotine Dependence Syndrome Scale, Minnesota Withdrawal Scale, and Shiffman-Jarvik Withdrawal Scale (American Psychiatric Association, 2000; Hughes and Hatsukami, 1986; Shiffman and Jarvik, 1976; Shiffman et al., 2004). Many of these measures, such as the HONC, DSM-IV measures, and the modified Fagerstrom Tolerance Questionnaire, were developed for or have been adapted for use among adolescents (Prokhorov et al., 1996). At present, the only known reliable measures of nicotine dependence for smokeless tobacco use are the HONC and the Autonomy Over Smoking Scale, and they have only been tested for reliability in adolescents (DiFranza et al., 2011).
Smoking behavior can be characterized through an assessment of the frequency, timing, and duration of prior quit attempts; this should be incorporated into the minimum standards. Having experienced an unsuccessful quit attempt versus never having tried to stop smoking may differentially influence smokers’ perceptions of an MRTP, and thus have an impact on their responses to marketing messages and subsequent
product use. For example, tobacco users who were unsuccessful in their quit attempts may perceive an MRTP as a potential cessation aid, even if the product is not marketed as such, which would have important implications for use. Alternately, this type of smoker may believe he or she will never be able to quit using tobacco, and therefore view the MRTP as an option to continue using tobacco with less risk. Having experienced more than one failed attempt at smoking cessation may serve to solidify any beliefs smokers may have about their likelihood of success in the future, and affect their perceptions of the MRTP and their behavior accordingly. Finally, quit attempts made more recently may have a stronger effect on perceptions and behavioral outcomes than those made in the more distant past because of the salience of the event (see Box 5-2 for sample questions to assess prior cessation attempts).
Perceptions of and intentions to use a given MRTP are also likely to differ by age group. Thus, it is critical that studies include participants in the following age groups: children (≤ 12 years old), adolescents (13-17 years old), young or emerging adults (18-25 years old), and adults (≥ 25 years old). Studies should compare perceptions, intentions, and actual tobacco use patterns within and across the age groups.
Research has shown that tobacco use and perceptions of tobacco- related risks/benefits may also differ by race and ethnicity, thus placing certain ethnic groups at increased risk for tobacco use and subsequent disease. Evaluation of differences in perceptions by racial or ethnic categories is standard in all studies of consumer perception. The basic racial or ethnic categories recommended by the IOM Subcommittee on Standardized Collection of Race/Ethnicity Data for Healthcare Quality Improvement is appropriate for these studies: Asian, Black or African American, Native Hawaiian or Other Pacific Islander, White, American Indian or Alaska Native, and Hispanic or Latino, plus additional categories for Other and Two or More Races (IOM, 2009).
• How many times in the past have you made a serious attempt to quit smoking?
• What was the longest period of time that you were able to quit smoking?
• When was your most recent serious attempt to quit smoking?
• How long were you able to quit smoking during your most recent quit attempt?
SOURCE: Abrams et al. (2003).
Studies also show that individuals with low socioeconomic status are more likely to use tobacco and carry a disproportionate amount of the health burden associated with tobacco use. As a result, it is imperative that the potential influence of socioeconomic status on consumer perceptions and use of MRTPs is understood. The most recent reported estimates of family income and poverty thresholds published by the U.S. Census Bureau can help researchers to understand the influence of socioeconomic status.2
Finally, numerous studies show that individuals with less education are more likely to use tobacco; thus, they are more likely to suffer the health consequences of tobacco use. Researchers investigating these tobacco products should evaluate potential differences in consumer perceptions of MRTPs by level of education. For studies on consumer perceptions of MRTPs, it is standard to include the use of the Substance Abuse and Mental Health Services Administration education categories: less than high school, high school graduate, some college, and college graduate. Adding a category for individuals who have completed graduate school will strengthen these studies.
Study participants should be recruited such that there are satisfactory numbers of participants falling into each tobacco use, age, racial/ethnic, socioeconomic status, or educational category described above. A sample size will be considered satisfactory based on a priori statistical power analyses to ensure that the sample adequately reflects the demographic characteristics of the population of interest. Study participants should not have any affiliation with the tobacco industry, the FDA, or any tobacco control agency.
For focus groups and experiments, the samples should be drawn from multiple sites across the United States because of the regional differences in tobacco use and exposure to pro- and antitobacco marketing and campaigns. Each focus group should contain 8-12 participants, with participants within a given focus group having similar age, racial/ethnic, socioeconomic status, or educational category described above. Multiple focus groups should be conducted, each one representing different demographic characteristics, to ensure the results are generalizable across each group. For experimental designs, participants should be randomly assigned to each group, with numbers of each demographic group equally represented. For surveys, the samples should be nationally representative; however, certain groups may be oversampled because of low prevalence
2 These estimates can be found on the U.S. Census Bureau website: http://www.census.gov/hhes/www/poverty/methods/definitions.html
rates in the general population, such as minority racial or ethnic groups. Participants should be recruited for surveys using the random digit dialing method.
Specific information on measuring tobacco use and sample demographic characteristics was discussed above. Here details on measuring perceptions, risk communication, and tobacco use intentions are provided.
Inclusion of conditional risk assessments is standard for evaluations of participants’ perceptions of risks, benefits, and likelihood of addiction associated with a given MRTP. This type of risk assessment uses scenarios to explicitly place the outcomes under investigation in the context of a specific behavior. Previous research shows that conditional risk assessments more closely reflect health risk behavior models and are stronger predictors of behavior than unconditional risk assessments, which do not place outcomes in a precise behavioral context (Halpern-Felsher et al., 2001; Ronis, 1992; Van Der Velde et al., 1996). As an example, for evaluating short- and long-term risks and benefits, the committee suggests using a conditional risk scenario such as the following: “Imagine that you just began smoking. You smoke about 2 or 3 cigarettes each day. Sometimes you smoke alone, and sometimes you smoke with friends. If you smoke about 2 or 3 cigarettes each day, what is the chance that…?” (Halpern-Felsher et al., 2004). The second scenario for evaluating long-term risks can include “Imagine that you continue to smoke about 2 or 3 cigarettes each day for the rest of your life. What is the chance that…?” (Halpern-Felsher et al., 2004).
After reading the scenarios, research participants should then be evaluated on their perceptions of relevant outcomes occurring to them, others, and to smokers and nonsmokers. Research participants can be asked about perceptions of general harm or specific harm (such as lung cancer, heart attack, etc.). The committee believes that researchers should ask about specific tobacco-related outcomes (rather than more general perceptions of harm), given that generalized outcomes can be vague, can lead to misperceptions, or can produce results that are difficult to interpret. Inquiring about specific outcomes reduces misinterpretation of the questions, and allows the investigator to determine the domains in which the misperceptions of the MRTP are most likely to occur.
In addition to considering the types of risks to assess, it is important to utilize the best response set for each question. Perceptions can be
assessed using a number of scales, including likelihood scales (such as: “what is the likelihood or chance that [a specific outcome] will occur?”), log linear scales, lexical scales (such as: “very likely to not very likely” or “small chance to large chance”), comparative scales (such as: “compared to [another product], is this MRTP more or less likely to cause [a specific outcome]”). The committee supports the use of likelihood estimates assessed through numerical scales (such as: “please estimate the chance that [a specific outcome] will occur using any number between 0 percent and 100 percent”) or comparative risk assessments (Biehl and Halpern-Felsher, 2001; Halpern-Felsher et al., 2004).
Numeracy, or the ability to understand and use numbers, is very low in the general population (Gigerenzer et al., 2007; Reyna et al., 2009). As a result, a large proportion of the population, including health professionals and educated laypeople, has difficulty comprehending numerical information about risks and benefits to health (Gigerenzer et al., 2007; Reyna et al., 2009). It is important that the tobacco product sponsor communicates the risks and benefits associated with a given MRTP accurately and in a way that the general population can clearly understand. Thus, it is essential that the product sponsor carefully crafts messages about the risks and benefits of any MRTP and then demonstrates through rigorous testing that people correctly understand and interpret such messages.
There is a significant public health interest in ensuring that consumers accurately understand the risks if they use the MRTP. This includes understanding their increased level of risk if they are not current tobacco users, or their presumed decreased level of risk if they are already tobacco users (this requires comparisons of risk between the MRTP and specific tobacco products that are currently on the market). It also includes understanding changes in specific types of risk, such as risk of carbon monoxide exposure, risk of heart attack, or risk of specific types of cancer. In addition, consumers should comprehend what conditions of use are associated with the stated risks and benefits of the product (e.g., daily use, hourly use, or as indicated on package). Finally, consumers should be able to understand how these risks and benefits relate to groups of people similar to themselves. For example, what are the risks for female, African-American adolescents? Conveying such information of course assumes the tobacco product sponsor has already completed the appropriate and scientifically sound research that will allow it to make claims about the risks and benefits associated with using the MRTP under specific conditions of use and across a variety of relevant populations.
Research indicates that the best way to promote an accurate understanding of risk is to report absolute rather than relative risks (Gigerenzer et al., 2007). For instance, it is better to state that 5 in 100 people will develop shortness of breath when using the MRTP as compared to 10 in 100 people who smoke a traditional cigarette, than to state that there is a 50 percent reduction in risk for shortness of breath when using the MRTP compared to smoking a traditional cigarette.
Based on studies of exposure and risk, the industry should first generate statements that communicate the risks and/or benefits of using the MRTP and include the following elements:
• Use statements of absolute rather than relative risk.
• Clearly state what type of risk and outcome is being addressed.
• Clearly state under what conditions of use the risks/benefits are incurred.
• Clearly state what comparison is being made (i.e., among which alternative products).
• Clearly state what populations incur the risks/benefits (e.g., people who do versus do not use tobacco, males versus females, certain age or racial/ethnic groups).
Once these statements of risk are generated, they should be presented to research participants, and the participants’ understanding of the statements should be assessed using the research designs discussed earlier.
Several types of intentions to engage in MRTP use should be assessed, including intent to try the new product, intent to use the MRTP to aid in tobacco use cessation (related to intent to quit tobacco use), and intent to use the MRTP while continuing to use current tobacco product(s). The general format of the questions may include, but is not limited to the sample questions in Box 5-3.
In addition to the methods for assessing risk perceptions outlined above, other factors that may relate to the likelihood of trying or adopting use of an MRTP should be considered. A number of research groups have examined outcome expectancies as predictors of both smoking uptake and relapse after cessation. Expectancies are a class of attitude, formed from previous knowledge, beliefs, and experiences, that serve to guide behavior (Del Boca et al., 2002). The most widely used measure for ciga
• What is the chance that you will try [the MRTP] sometime in the next 6 months?
• What is the chance that you will try [the MRTP] in your life?
• What is the chance that you will ever use [the MRTP]?
• What is the chance that you will use [the MRTP] to help you quit smoking cigarettes/chewing tobacco/etc.?
• What is the chance that you will use [the MRTP] in addition to other tobacco products that you already use?
• If one of your best friends were to offer you [the MRTP] in the next 6 months, would you use it? (for adolescents)
rette expectancies has been the Smoking Consequences Questionnaire (Brandon and Baker, 1991) and its various derivatives for adults, adolescents, and children (Copeland et al., 1995; Lewis-Esquerre et al., 2005; Rash and Copeland, 2008). Broadly, expectancies can be divided into positive outcomes (i.e., anticipated benefits) and negative outcomes (i.e., anticipated harms). Wetter and colleagues (1994) established that positive expectancies (positive reinforcement, negative reinforcement, and appetite-weight control) predicted withdrawal severity while negative expectancies predicted cessation success. Studies in adolescent populations have shown outcome expectancies (those relating to negative affect relief in particular) are related to smoking uptake, behavior, and nicotine dependence (Heinz et al., 2010; Wahl et al., 2005). One study that examined expectancies in relation to modified tobacco products showed that positive expectancies predicted interest in trying both Quest and Eclipse, regardless of level of smoking experience (O’Connor et al., 2007). The committee suggests that studies of MRTP perceptions include a measure of outcome expectancies.
Evidence that has emerged over the past decade points to the importance of affect in shaping decisions about a wide array of health behaviors, including tobacco use (Keer et al., 2010; Kiviniemi et al., 2007; Lawton et al., 2009). While judgment and decision making have most widely been regarded as rational processes, accumulating evidence suggests an important role for affective processes and emotions in guiding decisions, primarily through heuristics (Greifeneder et al., 2011; Slovic et al., 2005). According to this model, affective reactions to stimuli, which are often
automatic, can become salient in guiding decisions based on individual and situational conditions, particularly those requiring complex analysis or under time pressure. Broadly speaking, activities viewed positively are seen as low risk/high benefit, whereas those viewed negatively are seen as high risk/low benefit. Other evidence suggests that messages that evoke emotional responses are better remembered (Lang and Dhillon, 1995) and promote higher order cognitive processing (Donohew et al., 1998; Keller and Block, 1996). Thus, affective heuristics, and emotional factors more broadly, can be important to consider in user and nonuser reactions to MRTPs. A number of measures have been developed to assess affective reactions. Some measures are scales that ask participants to rate their feelings by responding to descriptive statements or words along unipolar or bipolar (semantic differential) axes, either numeric or visual-analog. Validated clinical measures such as the Profile of Mood States or Positive and Negative Affect Schedule can also be employed to measure current feelings among participants (McNair et al., 1971; Watson et al., 1988). Affect can also be measured using pictograms to assess affect valence, arousal, and dominance brought about by a particular stimulus (Bradley and Lang, 1994). This measure has been validated against the International Affective Picture System (IAPS) (Lang et al., 1997). The IAPS images cover five domains: pleasant-aroused, pleasant-calm, neutral, unpleasant-calm, and unpleasant-aroused. The committee suggests that studies examining affective reactions to MRTP-related stimuli (e.g., advertising, packaging, marketing) include a set of IAPS images from each domain for comparative purposes.
Consistent with the importance of affect and outcome expectancies, Wakefield and colleagues have been working to evaluate youth reactions to smoking messaging (Wakefield et al., 2003, 2005). They have noted five key considerations to understand ad impact and facilitate comparison of different ads: (1) previous exposure and reactions to general antismoking information and to test ads, (2) comprehension, (3) specific ad appraisals, (4) relative utility of target ad compared to generic antismoking information, and (5) recall of the test ad within 1 week. The proposed metrics are broadly applicable across media types (e.g., video, print, Internet) and include both cognitive and emotional responses. A sample questionnaire used by the Wakefield et al. research team to assess youth responses to anti-smoking ads is provided in Figure 5-1.
The industry should hire independent, professional biostatisticians to aid in initial measurement design and all analyses following completion of data collection. The biostatisticians should conduct a priori power
What is the MAIN point that this ad is trying to make?
What ELSE is it trying to say?
How well do the following phrases describe this ad? (Circle one number for each phrase)
|This ad….||Strongly Disagree||Disagree||Neither
|…had a message that is
important to me
|…said things that were hard to believe||1||2||3||4||5|
|…made me stop and think||1||2||3||4||5|
|… made me curious to know if what the ad says is true||1||2||3||4||5|
|…is one that I would talk to other people about||1||2||3||4||5|
|…told me something new||1||2||3||4||5|
|…talked down to me||1||2||3||4||5|
|This ad made me feel…|
|This ad was…|
|Overall, I thought this ad was a very good anti-smoking advertisement…|
|What makes it that way?|
|Have you seen this ad on TV before today?|
analyses for all studies in order to determine appropriate sample size and level of acceptable power.
In general, the analysis of the focus group data should involve a continuum from the raw data to descriptive statements to interpretation. Analysis of the data should occur in three steps: (1) identification of participants’ concepts, (2) organization of participants’ concepts into a hierarchy (a model), and (3) quantitative analysis of frequency of participants’ concepts. Analyses should first identify concepts (e.g., health) used by participants in the course of the interview using a variety of techniques drawn from grounded theory (Strauss and Corbin, 1997). Next, the identified concepts should be organized into a hierarchy, making use of diagrams and other comparative analytic techniques. Once a hierarchy of participants’ concepts is completed, the entire dataset should be coded for participants’ concepts using an appropriate software package, such as NVivo (NUD*IST). Participant concept data should then be exported into a database, allowing for analyses of whether frequency of participant concepts varies by individual-level characteristics, and so on.
The issues discussed in this chapter are relevant for the interpretation of data generated from scientific studies and for the evaluation of product applications at the FDA. The next chapter discusses the cross-cutting issues presented in this chapter as well as earlier chapters. The next chapter will also focus on methods to integrate information, and present the committee’s findings and recommendations.
Abrams, D. B., R. Niaura, R. A. Brown, K. M. Emmons, M. G. Goldstein, P. M. Monti, and L. A. Linnan. 2003. The tobacco dependence treatment handbook: A guide to best practices. New York, NY: The Guilford Press.
Ajzen, I. 1985. From intentions to actions: A theory of planned behavior. In Action control, from cognition to behavior, edited by J. Kuhl, and J. Beckmann. Berlin: Springer-Verlag.
American Psychiatric Association. 2000. Diagnostic and statistical manual of mental disorders: DSM-IV-TR. 4th ed. Washington, DC: American Psychiatric Publishing, Inc.
Anderson, S. J., S. A. Glantz, and P. Ling. 2005. Emotions for sale: Cigarette advertising and women’s psychosocial needs. Tobacco Control 14(2):127.
Arnett, J. J. 2000. Optimistic bias in adolescent and adult smokers and nonsmokers. Addictive Behaviors 25(4):625-632.
Balbach, E. D., R. J. Gasior, and E. M. Barbeau. 2003. R. J. Reynolds’ targeting of African Americans: 1988-2000. American Journal of Public Health 93(5):822-827.
Bandura, A. 2001. Social cognitive theory of mass communication. Media Psychology 3(3):265-299.
Bansal-Travers, M., R. O’Connor, B. V. Fix, and K. M. Cummings. 2011. What do cigarette pack colors communicate to smokers in the U.S.? American Journal of Preventive Medicine 40(6):683-689.
Biehl, M., and B. L. Halpern-Felsher. 2001. Adolescents’ and adults’ understanding of probability expressions. Journal of Adolescent Health 28(1):30-35.
Borland, R. O. N. 1997. Tobacco health warnings and smoking-related cognitions and behaviours. Addiction 92(11):1427-1435.
Bradley, M. M., and P. J. Lang. 1994. Measuring emotion: The self-assessment manikin and the semantic differential. Journal of Behavior Therapy and Experimental Psychiatry 25(1):49-59.
Brandon, T. H., and T. B. Baker. 1991. The smoking consequences questionnaire: The subjective expected utility of smoking in college students. Psychological Assessment: A Journal of Consulting and Clinical Psychology 3(3):484.
Callery, W. E., D. Hammond, R. J. O’Connor, and G. T. Fong. 2011. The appeal of smokeless tobacco products among young Canadian smokers: The impact of pictorial health warnings and relative risk messages. Nicotine & Tobacco Research 13(5):373-383.
Carpenter, C. M., G. F. Wayne, and G. N. Connolly. 2005. Designing cigarettes for women: New findings from the tobacco industry documents. Addiction 100(6):837-851.
CDC (Centers for Disease Control and Prevention). 2010. Healthy people 2010 operational definition. ftp://ftp.cdc.gov/pub/health_statistics/nchs/datasets/data2010/focusarea27/o2701a.pdf (accessed October 5, 2011).
Clark, V. L. P., D. L. Miller, J. W. Creswell, K. McVea, R. McEntarffer, L. M. Harter, and W. T. Mickelson. 2002. In conversation: High school students talk to students about tobacco use and prevention strategies. Qualitative Health Research 12(9):1264.
Cook, B. L., G. F. Wayne, L. Keithly, and G. Connolly. 2003. One size does not fit all: How the tobacco industry has altered cigarette design to target consumer groups with specific psychological and psychosocial needs. Addiction 98(11):1547-1561.
Copeland, A. L., T. H. Brandon, and E. P. Quinn. 1995. The smoking consequences questionnaire—adult: Measurement of smoking outcome expectancies of experienced smokers. Psychological Assessment 7(4):484.
Cummings, K. M., C. Morley, J. Horan, C. Steger, and N. R. Leavell. 2002. Marketing to America’s youth: Evidence from corporate documents. Tobacco Control 11(Suppl. 1):i5.
Curry, S. J., L. Grothaus, and C. McBride. 1997. Reasons for quitting: Intrinsic and extrinsic motivation for smoking cessation in a population-based sample of smokers. Addictive Behaviors 22(6):727-739.
Del Boca, F. K., J. Darkes, M. S. Goldman, and G. T. Smith. 2002. Advancing the expectancy concept via the interplay between theory and research. Alcoholism: Clinical and Experimental Research 26(6):926-935.
DiFranza, J. R., J. A. Savageau, K. Fletcher, J. K. Ockene, N. A. Rigotti, A. D. McNeill, M. Coleman, and C. Wood. 2002. Measuring the loss of autonomy over nicotine use in adolescents: The DANDY (Development and Assessment of Nicotine Dependence in Youths) study. Archives of Pediatrics and Adolescent Medicine 156(4):397.
DiFranza, J. R., M. Sweet, J. A. Savageau, and W. Ursprung. 2011. The assessment of tobacco dependence in young users of smokeless tobacco. Tobacco Control. Published Online June 28, 2011.
Donohew, L., E. P. Lorch, and P. Palmgreen. 1998. Applications of a theoretic model of information exposure to health interventions. Human Communication Research 24(3):454-468.
Dunn, D. A., and J. L. Johnson. 2001. Choosing to remain smoke-free: The experiences of adolescent girls. Journal of Adolescent Health 29(4):289-297.
Etter, J.-F., L. T. Kozlowski, and T. V. Perneger. 2003. What smokers believe about light and ultralight cigarettes. Preventive Medicine 36(1):92-98.
Fishbein, M., and I. Ajzen. 1975. Belief, attitude, intention, and behavior: An introduction to theory and research. Reading, MA: Addision-Wesley.
Gerrard, M., F. X. Gibbons, A. E. Houlihan, M. L. Stock, and E. A. Pomery. 2008. A dual-process approach to health risk decision making: The prototype willingness model. Developmental Review 28(1):29-61.
Gigerenzer, G., W. Gaissmaier, E. Kurz-Milcke, L. M. Schwartz, and S. Woloshin. 2007. Helping doctors and patients make sense of health statistics. Psychological Science in the Public Interest 8(2):53-96.
Gittelsohn, J., K. M. Roche, C. S. Alexander, and P. Tassler. 2001. The social context of smoking among African-American and white adolescents in Baltimore City. Ethnicity and Health 6(3-4):211-225.
Glantz, S. A., and P. Jamieson. 2000. Attitudes toward secondhand smoke, smoking, and quitting among young people. Pediatrics 106(6):e82.
Goldberg, J. H., B. L. Halpern-Felsher, and S. G. Millstein. 2002. Beyond invulnerability: The importance of benefits in adolescents’ decision to drink alcohol. Health Psychology 21(5):477.
Greifeneder, R., H. Bless, and M. T. Pham. 2011. When do people rely on affective and cognitive feelings in judgment? A review. Personality and Social Psychology Review 15(2):107.
Gritz, E. R., A. V. Prokhorov, K. S. Hudmon, M. M. Jones, C. Rosenblum, C.-C. Chang, R. M. Chamberlain, W. C. Taylor, D. Johnston, and C. de Moor. 2003. Predictors of susceptibility to smoking and ever smoking: A longitudinal study in a triethnic sample of adolescents. Nicotine & Tobacco Research 5(4):493-506.
Halpern-Felsher, B. L., and M. L. Rubinstein. 2005. Clear the air: Adolescents’ perceptions of the risks associated with secondhand smoke. Preventive Medicine 41(1):16-22.
Halpern-Felsher, B. L., S. G. Millstein, J. M. Ellen, N. E. Adler, J. M. Tschann, and M. Biehl. 2001. The role of behavioral experience in judging risks. Health Psychology 20(2):120-126.
Halpern-Felsher, B. L., M. Biehl, R. Y. Kropp, and M. L. Rubinstein. 2004. Perceived risks and benefits of smoking: Differences among adolescents with different smoking experiences and intentions. Preventive Medicine 39(3):559-567.
Hammond, D., and C. Parkinson. 2009. The impact of cigarette package design on perceptions of risk. Journal of Public Health 31(3):345-353.
Hammond, D., M. Dockrell, D. Arnott, A. Lee, and A. McNeill. 2009. Cigarette pack design and perceptions of risk among UK adults and youth. European Journal of Public Health 19(6):631-637.
Hansen, W. B., and C. K. Malotte. 1986. Perceived personal immunity: The development of beliefs about susceptibility to the consequences of smoking. Preventive Medicine 15(4):363-372.
Heatherton, T. F., L. T. Kozlowski, and R. C. Frecker. 1991. The Fagerstrom test for nicotine dependence: A revision of the Fagerstrom tolerance questionnaire. British Journal of Addiction 86(9):1119-1127.
Heinz, A. J., J. D. Kassel, M. Berbaum, and R. Mermelstein. 2010. Adolescents’ expectancies for smoking to regulate affect predict smoking behavior and nicotine dependence over time. Drug and Alcohol Dependence 111(1-2):128-135.
Hughes, J. R., and D. Hatsukami. 1986. Signs and symptoms of tobacco withdrawal. Archives of General Psychiatry 43(3):289-294.
IOM (Institute of Medicine). 2007. Ending the tobacco problem: A blueprint for the nation. Washington, DC: The National Academies Press.
IOM. 2009. Race, ethnicity, and language data: Standardization for health care quality improvement. Washington, DC: The National Academies Press.
Jamieson, P., and D. Romer. 2001. What do young people think they know about the risks of smoking. In Smoking: Risk, perception & policy, edited by P. Slovic. Thousand Oaks, CA: Sage Publications.
Johnson, J. L., J. L. Bottorff, B. Moffat, P. A. Ratner, J. A. Shoveller, and C. Y. Lovato. 2003. Tobacco dependence: Adolescents’ perspectives on the need to smoke. Social Science and Medicine 56(7):1481-1492.
Johnson, R. J., K. D. McCaul, and W. M. Klein. 2002. Risk involvement and risk perception among adolescents and young adults. Journal of Behavioral Medicine 25(1):67-82.
Kanfer, F. H. 1970. Self-regulation: Research, issues, and speculations. Edited by Institute for Research in Clinical Psychology, C. Neuringer, and J. L. Michael, Behavior modification in clinical psychology. New York: Appleton-Century-Crofts.
Keer, M., B. Putte, and P. Neijens. 2010. The role of affect and cognition in health decision making. British Journal of Social Psychology 49(1):143-153.
Kegler, M., and V. Cleaver. 2000. The social context of experimenting with cigarettes: American Indian “start stories.” American Journal of Health Promotion 15(2):89-92.
Keller, P. A., and L. G. Block. 1996. Increasing the persuasiveness of fear appeals: The effect of arousal and elaboration. Journal of Consumer Research 22(4):448-459.
Kiviniemi, M. T., A. M. Voss-Humke, and A. L. Seifert. 2007. How do I feel about the behavior? The interplay of affective associations with behaviors and cognitive beliefs as influences on physical activity behavior. Health Psychology 26(2):152.
Kristiansen, C. M., C. M. Harding, and J. R. Eiser. 1983. Beliefs about the relationship between smoking and causes of death. Basic and Applied Social Psychology 4(3):253-261.
Krueger, R. A. 2000. Focus groups: A practical guide for applied research. Thousand Oaks, CA: Sage Publications.
Kurtz, M. E., J. Kurtz, S. M. Johnson, and W. Cooper. 2001. Sources of information on the health effects of environmental tobacco smoke among African-American children and adolescents. Journal of Adolescent Health 28(6):458-464.
Landrine, H., E. A. Klonoff, S. Fernandez, N. Hickman, K. Kashima, B. Parekh, K. Thomas, C. R. Brouillard, M. Zolezzi, and J. A. Jensen. 2005. Cigarette advertising in black, latino, and white magazines, 1998-2002: An exploratory investigation. Ethnicity and Disease 15(1):63-67.
Lang, A., and K. Dhillon. 1995. The effects of emotional arousal and valence on television viewers’ cognitive capacity and memory. Journal of Broadcasting & Electronic Media 39(3):313.
Lang, P., M. Bradley, and B. Cuthbert. 1997. International affective picture system (IAPS): Technical manual and affective ratings. Gainesville, FL: National Institute of Mental Health Center for the Study of Emotion and Attention.
Lawton, R., M. Conner, and R. McEachan. 2009. Desire or reason: Predicting health behaviors from affective and cognitive attitudes. Health Psychology 28(1):56.
Lee, J., and B. L. Halpern-Felsher. 2011. What does it take to be a smoker? Adolescents’ characterization of different smoker types. Nicotine & Tobacco Research 13(11):1106-1113.
Leventhal, H., K. Glynn, and R. Fleming. 1987. Is the smoking decision an “informed choice”? JAMA 257(24):3373.
Lewis-Esquerre, J. M., J. R. Rodrigue, and C. W. Kahler. 2005. Development and validation of an adolescent smoking consequences questionnaire. Nicotine & Tobacco Research 7(1):81.
Ling, P. M., and S. A. Glantz. 2004. Tobacco industry research on smoking cessation. Journal of General Internal Medicine 19(5p1):419-426.
Lloyd, B., K. Lucas, and M. Fernbach. 1997. Adolescent girls’ constructions of smoking identities: Implications for health promotion. Journal of Adolescence 20(1):43-56.
McKee, S. A., S. S. O’Malley, P. Salovey, S. Krishnan-Sarin, and C. M. Mazure. 2005. Perceived risks and benefits of smoking cessation: Gender-specific predictors of motivation and treatment outcome. Addictive Behaviors 30(3):423-435.
McNair, D. M., M. Lorr, and L. F. Droppleman. 1971. Manual for the profile of mood states. San Diego, CA: Educational and Industrial Testing Service.
Millstein, S. G., and B. L. Halpern-Felsher. 2002. Judgments about risk and perceived invulnerability in adolescents and young adults. Journal of Research on Adolescence 12(4):399-422.
Morrell, H. E. R., A. V. Song, and B. L. Halpern-Felsher. 2010. Predicting adolescent perceptions of the risks and benefits of cigarette smoking: A longitudinal investigation. Health Psychology 29(6):610-617.
Mutti, S., D. Hammond, R. Borland, M. K. Cummings, R. J. O’Connor, and G. T. Fong. 2011. Beyond light and mild: Cigarette brand descriptors and perceptions of risk in the international tobacco control (ITC) four country survey. Addiction 106:1166-1175.
Nichter, M., N. Vuckovic, G. Quintero, and C. Ritenbaugh. 1997. Smoking experimentation and initiation among adolescent girls: Qualitative and quantitative findings. Tobacco Control 6(4):285.
O’Connor, R. J., R. L. Ashare, B. V. Fix, L. W. Hawk, K. M. Cummings, and W. C. Schmidt. 2007. College students’ expectancies for light cigarettes and potential reduced exposure products. American Journal of Health Behavior 31(4):402-410.
0verland, S., J. Hetland, and L. E. Aar0. 2008. Relative harm of snus and cigarettes: What do Norwegian adolescents say? Tobacco Control 17(6):422-425.
Pallonen, U. E., J. O. Prochaska, W. F. Velicer, A. V. Prokhorov, and N. F. Smith. 1998. Stages of acquisition and cessation for adolescent smoking: An empirical integration. Addictive Behaviors 23(3):303-324
Perkins, K. A. 2001. Smoking cessation in women: Special considerations. CNS Drugs 15(5):391-411.
Prokhorov, A. V., U. E. Pallonen, J. L. Fava, L. Ding, and R. Niaura. 1996. Measuring nicotine dependence among high-risk adolescent smokers. Addictive Behaviors 21(1):117-127.
Prokhorov, A. V., C. A. de Moor, K. S. Hudmon, S. Hu, S. H. Kelder, and E. R. Gritz. 2002. Predicting initiation of smoking in adolescents: Evidence for integrating the stages of change and susceptibility to smoking constructs. Addictive Behaviors 27(5):697-712.
Rash, C. J., and A. L. Copeland. 2008. The Brief Smoking Consequences Questionnaire-Adult (BSCQ-A): Development of a short form of the SCQ-A. Nicotine & Tobacco Research 10(11):1633.
Reyna, V. F., W. L. Nelson, P. K. Han, and N. F. Dieckmann. 2009. How numeracy influences risk comprehension and medical decision making. Psychological Bulletin 135(6):943-973.
Romer, D., and P. Jamieson. 2001. Do adolescents appreciate the risks of smoking? Evidence from a national survey. Journal of Adolescent Health 29(1):12-21.
Ronis, D. L. 1992. Conditional health threats—health beliefs, decisions, and behaviors among adults. Health Psychology 11(2):127-134.
Rosenstock, I. M. 1974. The health belief model and preventive health behavior. In The health belief model and personal health behavior, edited by M. H. Becker. Thorofare, NJ: Slack.
Rugkasa, J., B. Knox, J. Sittlington, O. Kennedy, M. P. Treacy, and P. S. Abaunza. 2001. Anxious adults vs. cool children: Children’s views on smoking and addiction. Social Science and Medicine 53(5):593-602.
SAMHSA (Substance Abuse and Mental Health Services Administration). 2009. 2010 national survey on drug use and health: CAI specifications for programming. Rockville, MD: Substance Abuse and Mental Health Services Administration.
Schoenbaum, M. 1997. Do smokers understand the mortality effects of smoking? Evidence from the Health and Retirement Survey. American Journal of Public Health 87(5):755-759.
Shiffman, S. M., and M. E. Jarvik. 1976. Smoking withdrawal symptoms in two weeks of abstinence. Psychopharmacology 50(1):35-39.
Shiffman, S., J. L. Pillitteri, S. L. Burton, J. M. Rohay, and J. G. Gitchell. 2001. Smokers’ beliefs about “light” and “ultra light” cigarettes. Tobacco Control 10:I17-I23.
Shiffman, S., A. Waters, and M. Hickcox. 2004. The nicotine dependence syndrome scale: A multidimensional measure of nicotine dependence. Nicotine & Tobacco Research 6(2):327-348.
Slovic, P. 1998. Do adolescent smokers know the risks? Duke Law Journal 47:1133-1141.
Slovic, P. 2001. Smoking: Risk, perception, & policy. Thousand Oaks, CA: Sage Publications.
Slovic, P., E. Peters, M. L. Finucane, and D. G. MacGregor. 2005. Affect, risk, and decision making. Health Psychology 24(4, Suppl. 1):S35-S40.
Song, A. V., S. A. Glantz, and B. L. Halpern-Felsher. 2009a. Perceptions of second-hand smoke risks predict future adolescent smoking initiation. Journal of Adolescent Health 45(6):618-625.
Song, A. V., H. E. R. Morrell, J. L. Cornell, M. E. Ramos, M. Biehl, R. Y. Kropp, and B. L. Halpern-Felsher. 2009b. Perceptions of smoking-related risks and benefits as predictors of adolescent smoking initiation. American Journal of Public Health 99(3):487-492.
Sorensen, G., and T. F. Pechacek. 1987. Attitudes toward smoking cessation among men and women. Journal of Behavioral Medicine 10(2):129-137.
Strauss, A. L., and J. M. Corbin. 1997. Grounded theory in practice. Thousand Oaks, CA: Sage Publications, Inc.
Sutton, S. 1998. How ordinary people in Great Britain perceive the health risks of smoking. Journal of Epidemiology and Community Health 52(5):338-339.
Swan, G. E., M. M. Ward, D. Carm Elli, and L. M. Jack. 1993. Differential rates of relapse in subgroups of male and female smokers. Journal of Clinical Epidemiology 46(9):1041-1053.
Triandis, H. C. 1977. Subjective culture and interpersonal relations across cultures. Annals of the New York Academy of Sciences 285(1):418-434.
Van Der Velde, F. W., C. Hooykaas, and J. Van Der Pligt. 1996. Conditional versus unconditional risk estimates in models of AIDS-related risk behaviour. Psychology & Health 12(1):87-100.
Virgili, M., N. Owen, and H. H. Sverson. 1991. Adolescents’ smoking behavior and risk perceptions. Journal of Substance Abuse 3(3):315-324.
Viscusi, W. K. 1990. Do smokers underestimate risks? Journal of Political Economy 98(6):1253- 1269.
Viscusi, W. K. 1991. Age variations in risk perceptions and smoking decisions. The Review of Economics and Statistics 73(4):577-588.
Viscusi, W. K. 1992. Smoking: Making the risky decision. New York: Oxford University Press.
Vuckovic, N., M. R. Polen, and J. F. Hollis. 2003. The problem is getting us to stop: What teens say about smoking cessation. Preventive Medicine 37(3):209-218.
Wahl, S. K., L. R. Turner, R. J. Mermelstein, and B. R. Flay. 2005. Adolescents’ smoking expectancies: Psychometric properties and prediction of behavior change. Nicotine & Tobacco Research 7(4):613.
Wakefield, M., C. Morley, J. K. Horan, and K. M. Cummings. 2002. The cigarette pack as image: New evidence from tobacco industry documents. Tobacco Control 11(Suppl. 1):i73-i80.
Wakefield, M., R. Durrant, Y. Terry-McElrath, E. Ruel, G. Balch, S. Anderson, G. Szczypka, S. Emery, and B. Flay. 2003. Appraisal of anti-smoking advertising by youth at risk for regular smoking: A comparative study in the United States, Australia, and Britain. Tobacco Control 12(Suppl. 2):ii82-ii86.
Wakefield, M., G. I. Balch, E. Ruel, Y. Terry McElrath, G. Szczypka, B. Flay, S. Emery, and K. Clegg Smith. 2005. Youth responses to anti smoking advertisements from tobacco control agencies, tobacco companies, and pharmaceutical companies. Journal of Applied Social Psychology 35(9):1894-1910.
Watson, D., L. A. Clark, and A. Tellegen. 1988. Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology 54(6):1063-1070.
Wayne, G. F., and G. N. Connolly. 2002. How cigarette design can affect youth initiation into smoking: Camel cigarettes 1983-93. Tobacco Control 11(Suppl. 1):i32-i39.
Weekley, C. K. 1992. Smoking as a weight-control strategy and its relationship to smoking status. Addictive Behaviors 17(3):259-271.
Weinstein, N. D., P. Slovic, and G. Gibson. 2004. Accuracy and optimism in smokers’ beliefs about quitting. Nicotine & Tobacco Research 6(Suppl. 3):S375-S380.
Weinstein, N. D., S. E. Marcus, and R. P. Moser. 2005. Smokers’ unrealistic optimism about their risk. Tobacco Control 14(1):55-59.
Wetter, D. W., S. S. Smith, S. L. Kenford, D. E. Jorenby, M. C. Fiore, R. D. Hurt, K. P. Offord, and T. B. Baker. 1994. Smoking outcome expectancies: Factor structure, predictive validity, and discriminant validity. Journal of Abnormal Psychology 103(4):801.