Article Text

Quantifying how smokers value attributes of electronic cigarettes
1. James Nonnemaker,
2. Annice E Kim,
3. Youn Ok Lee,
4. Anna MacMonegle
1. RTI International, RTP, North Carolina, USA
1. Correspondence to Dr James Nonnemaker, RTI International, 3040 E. Cornwallis Road, Research Triangle Park, RTP, NC 27709, USA; jnonnemaker{at}rti.org

## Abstract

Introduction Rates of electronic cigarette (e-cigarette) use have increased quickly among US adults (3.3% in 2010 to 8.5% in 2013) and youth (4.5% in 2013 to 13.4% in 2014). As state and local governments consider regulatory policies, understanding what smokers believe about e-cigarettes and how they value e-cigarettes is important.

Methods Using data from a convenience sample of Florida adult smokers (N=765), we investigated the value smokers place on specific attributes of e-cigarettes (availability of flavours, effectiveness of e-cigarettes as a cessation aid, healthier alternative to regular cigarettes, ability to use e-cigarettes in public places) by asking smokers how much they would be willing to pay for e-cigarettes with and without each of these attributes.

Results For cigarette-only and dual users, losing the ability to use an e-cigarette as a quit aid and losing the harm reduction of an e-cigarette significantly reduced the price respondents were willing to pay for an e-cigarette. For cigarette-only users, not being able to use an e-cigarette indoors and losing flavours also significantly reduced the price respondents were willing to pay for an e-cigarette.

Conclusion Our results suggest that smokers value multiple attributes of e-cigarettes. Our valuation measures also appear to align with smokers' beliefs about e-cigarettes.

• Cessation
• Electronic nicotine delivery devices
• Tobacco industry

## Request Permissions

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

## Introduction

Electronic cigarettes (e-cigarettes) deliver nicotine-containing aerosol and are rapidly gaining in popularity.1 Rates of e-cigarette use have increased quickly among US adults (3.3% in 2010 to 8.5% in 2013)2–4 and youth (4.5% in 2013 to 13.4% in 2014 among high school students),5 predominantly among current smokers, raising concerns that e-cigarettes will lead to dual use (cigarette and e-cigarette use) rather than cessation.6 Despite limited evidence of their safety, e-cigarettes are marketed by an unregulated industry that is growing quickly with little oversight.7 The Food and Drug Administration (FDA) does not currently regulate e-cigarettes, although under a new proposed rule (deeming rule) e-cigarettes would be deemed a tobacco product and come under FDA's authority to regulate as such.8 ,9 In the absence of regulation, advertising expenditures for these products have increased dramatically,10 and manufacturers, including tobacco companies, are marketing these products online, in print and on television.10

The increase in e-cigarette use raises several public health concerns. Currently, very little is known about the potential health effects of e-cigarettes.11 E-cigarette use may serve as a gateway to nicotine addiction and tobacco use, particularly for youth.12 E-cigarettes are most popular among current smokers; however, there are concerns that they may undermine social norms and existing smoking policies13 and delay cigarette smoking cessation. Many consumers believe e-cigarettes are a healthier alternative to traditional cigarettes, and reports show that some smokers use them as cessation aids, despite the limited evidence regarding their safety and efficacy.14–16 Beliefs and use behaviours related to e-cigarettes are forming without conclusive evidence about potential health impacts. E-cigarettes are aggressively marketed as a healthier alternative to combustible cigarettes that can help smokers quit and can be used anywhere smoking is not allowed.17

In response to these growing concerns, some local and state governments are taking action to regulate e-cigarettes. In 2010, California successfully sued e-cigarette retailer Smoking Everywhere for deceptive marketing, and e-cigarette companies are now required to place the following warning on their products to comply with California Proposition 65: “Warning: This product contains nicotine, a chemical known to the state of California to cause birth defects or other reproductive harm.”18 As of 1 December 2014, three state laws and 260 local laws were in effect that prohibit e-cigarette use in public places,19 ranging from workplaces, restaurants and bars (eg, Boston, New Jersey, North Dakota) to school property (Arkansas and Colorado), state workplace facilities (eg, Delaware) and correctional facilities (eg, Oklahoma, South Dakota). Other localities have recently passed more restrictive bills, including New York City's ban on sales of e-cigarettes to anyone younger than age 21 (City of New York, 2014) and Chicago's mandate requiring retailers to place e-cigarette displays behind the counter.20

To understand which policies work best or how smokers might respond to such policies, understanding what smokers believe about e-cigarettes and how they value e-cigarettes is important. Many smokers regret their decision to start smoking and express a desire to quit.21 However, quitting smoking is difficult for many smokers. Surveys of US smokers suggest that, although 70% of smokers say they want to quit and 34% of smokers try to quit each year, only 10% remain tobacco free for at least 1 year.22 Thus, a significant population will remain at risk of the negative health effects of smoking, suggesting that e-cigarettes could result in harm reduction (ie, offering smokers who cannot quit a ‘safer’ alternative to cigarette smoking).22 Currently, little is known about why people use e-cigarettes. In a study using a convenience internet sample of European e-cigarette users, Etter and Bullen23 examined the reasons participants gave for using e-cigarettes. The authors found that most participants used e-cigarettes because they were perceived to be less harmful than regular cigarettes (84%), helped with quitting smoking or avoiding relapse (77%), helped to deal with cravings (79%) and withdrawal symptoms (67%), and because they were cheaper than regular cigarettes (57%). Fewer participants reported using e-cigarettes to avoid smoking restrictions (39%).

To date, no studies have investigated the value people place on attributes of e-cigarettes. In this study, we used data from a convenience sample of smokers to investigate the value smokers place on different attributes of e-cigarettes. We measured how much smokers value specific attributes of e-cigarettes (availability of flavours, effectiveness of e-cigarettes for quitting smoking regular cigarettes, healthier alternative to regular cigarettes, ability to use e-cigarettes in public places) by asking them how much they would be willing to pay for e-cigarettes with and without each of these attributes. In so doing, we gain a better understanding of the relative value smokers place on various attributes of e-cigarettes and what may motivate e-cigarette use.

## Methods

### Design overview and sample

The study was conducted as part of an annual cross-sectional survey of Florida adult smokers conducted for the Florida Department of Health Bureau of Tobacco Free Florida. The survey was conducted with individuals who (1) are aged 18 or older, (2) indicate that they are current or former smokers and (3) reside in Florida. Participants were recruited from an online convenience panel, comScore Inc, a worldwide online market research firm that maintains a panel of approximately one million US internet users. Study invitations were emailed to comScore panellists. Interested panellists answered a series of screening questions to determine their eligibility. Confirmed eligible participants who provided informed consent were directed to complete a survey online that took about 21 min to complete, on average. A total of 765 Florida adult smokers completed the survey in December 2013–January 2014. The study protocol was approved by RTI International's Institutional Review Board.

### Measures

We measured respondents’ (1) valuation of e-cigarette attributes; (2) beliefs related to e-cigarettes; and (3) demographics, e-cigarette use, and smoking and cessation behaviours.

#### Valuation of e-cigarette attributes

To assess how important various e-cigarette attributes are to smokers, survey participants were presented with information about e-cigarettes and a list of four potential attributes (ie, flavours, helps to quit smoking, reduces harm relative to cigarettes and can be used indoors). E-cigarettes are currently available in a wide variety of device types. Disposable e-cigarettes typically resemble conventional cigarettes and are widely available at convenience stores, grocery stores and online. Many national marketing campaigns for e-cigarette brands feature disposable devices (eg, blu eCigs, NJOY). Owing to its wide availability and marketing, we chose a hypothetical disposable e-cigarette device for the valuation exercise. Respondents first reported the price they would pay for a single disposable e-cigarette unregulated and with the full set of attributes. Respondents then answered a series of questions asking them how much they would be willing to pay for a single disposable e-cigarette when a specific characteristic changed (figure 1). These questions assessed how much the respondent values each e-cigarette characteristic.

Figure 1

Sample willingness to pay questions (e-cigarette, electronic cigarette).

We measured the value of each attribute (1) as an absolute difference (ie, value=WTPbase−WTPrestricted) and (2) as a relative difference (value=((WTPbase−WTPrestricted)/WTPbase)) where WTPbase stands for the base price respondents are willing to pay for an e-cigarette with all attributes, and WTPrestricted stands for the price respondents are willing to pay for an e-cigarette with an attribute removed.

#### Demographics, e-cigarette use, and smoking and cessation behaviours

Demographics: Demographic characteristics included gender, race/ethnicity, age, education and income. Race/ethnicity was categorised as non-Hispanic Caucasian, non-Hispanic African-American, Hispanic and non-Hispanic other. Income was categorised as less than $25 000,$25 000–$34 999,$35 000–$49 999,$50 000–$74 999 and$75 000 or more. Education was categorised as never attended, elementary, some high school, high school graduate, some college and college graduate.

E-cigarette use: We assessed whether survey respondents had ever used e-cigarettes (yes/no); whether they were currently using e-cigarettes (every day, some days or not at all); whether they had ever used e-cigarettes to quit smoking cigarettes (yes/no); and for non-e-cigarette users, whether they planned on trying e-cigarettes in the next 12 months (yes/no). For current e-cigarette users, we assessed on how many of the past 30 days they used e-cigarettes, brand of e-cigarettes they used, flavour of e-cigarette liquids they used (mint/menthol, candy, tobacco, alcohol, other), and time between waking and using first e-cigarette (within 5 min, between 6 and 30 min, between 31 and 59 min, after 60 min).

Type of e-cigarette/cigarette user and tobacco use behaviours: Based on standard questions to assess tobacco use, we constructed a product use measure that separates current smokers into three categories of product use: dual use of e-cigarettes and cigarettes, cigarette only, and e-cigarette only. We defined current smoking as having smoked at least 100 cigarettes lifetime and current use (some days or every day). We also asked whether the respondent made a quit attempt in the past 12 months (yes/no).

### Analytic approach

We report descriptive statistics for all key measures collected in the survey and stratify results by (1) dual users (N=390), (2) cigarette-only users (N=362) and (3) e-cigarette-only users (N=13). To assess how e-cigarette attributes influence willingness to pay (WTP) for e-cigarettes, we calculated coefficients and corresponding 95% CIs for a series of multivariate linear regression models. We regressed indicators for each characteristic (flavour, quit aid, harm reduction, ability to use indoors) controlling for demographic characteristics (gender, race/ethnicity, age, income) on respondents’ reported WTP for an e-cigarette with a specific set of attributes. Using this approach, the coefficient on each attribute indicator represents how each respondent values an attribute relative to the e-cigarette with all attributes.

Each participant answered the WTP questions for each e-cigarette characteristic, a within-person design, so regressions were clustered on the individual to account for within-person correlation among responses. To compare the relationship between these factors and WTP for an e-cigarette for dual users and cigarette-only users, we ran the same model overall and among each of the three subgroups (dual users, cigarette-only users and e-cigarette-only users). However, due to a small sample size (N=13), results for e-cigarette-only users may not be reliable and therefore were not reported. For the models that included cigarette-only users (overall and exclusive cigarette users), we included an indicator variable for those who had ever tried an e-cigarette.

Out of the 765 respondents in our sample, 105 people reported that they would pay $0.00 for the base e-cigarette. This indicated that they were not interested in buying a single disposable e-cigarette. These respondents were set to missing for any analyses regarding the valuation of e-cigarettes. Because this is a convenience sample, we used unweighted proportions in our analysis. Seventeen respondents had missing values for demographic variables. These cases were excluded in the multivariate analyses. ## Results Table 1 summarises the demographic and e-cigarette and tobacco use characteristics of the sample. Of the 765 Florida tobacco-using adults surveyed, 52.4% were male and 84.7% were Caucasians. More than three-quarters (85.1%) of respondents were between the ages of 25 and 64, almost three-quarters (73.2%) had some college or a college degree, and 82.4% reported a household income of$25 000 or more. Among the full sample, 55.9% had made a quit attempt in the past year.

Table 1

Respondents’ demographics and e-cigarette tobacco use characteristics

Of those who made a quit attempt in the past year, 21.5% tried e-cigarettes as a cessation tool. The proportion of respondents having made a quit attempt was highest for dual users (73.9%), followed by e-cigarette-only users (54.5%) and cigarette-only users (36.5%). One-quarter of dual users reported using e-cigarettes as a cessation resource (25.4%), whereas less than 10% of cigarette-only users reported the same.

Two-thirds (65.0%) of respondents reported having ever tried an e-cigarette. Among respondents who reported never trying an e-cigarette, 53.4% planned on trying an e-cigarette in the next 12 months. Among respondents who reported ever trying an e-cigarette, almost three-quarters (73.4%) reported currently using e-cigarettes every day or some days. Among current e-cigarette users, 11.9% used e-cigarettes on 20 or more of the past 30 days. The top brands used by current e-cigarette users were blu (59.9%), Logic (41.4%) and Nicotek (35.6%). The majority of current e-cigarette users (60.4%) reported using the mint-flavoured or menthol-flavoured e-cigarettes, followed by candy-flavoured e-cigarettes (56.6%), traditional or tobacco-flavoured e-cigarettes (31.8%), and alcohol-flavoured e-cigarettes (8.5%). To measure dependence on e-cigarettes, respondents were asked how soon they have their first e-cigarette on waking. Approximately 65% of current e-cigarette users reported having their first e-cigarette within 6–59 min of waking up, 21.3% had their first cigarette 60 min or more after waking, and 14.4% reported having their first e-cigarette within 5 min of waking.

### Valuation of e-cigarette attributes

Table 2 summarises the value respondents (dual users, cigarette-only users, e-cigarette-only users) placed on a single disposable e-cigarette with various attributes. Among all respondents who said they would be interested in buying a single disposable e-cigarette, the mean price they would be willing to pay for the hypothetical ‘base’ disposable e-cigarette—which came in flavours, helped you quit, was less harmful than cigarettes and could be used indoors—was $6.64. The mean price for the base e-cigarette was$7.37 for dual users, $5.66 for cigarette-only users and$5.52 for e-cigarette-only users. For each characteristic removed, the WTP was lower, and, on average, dual users were willing to pay a higher price for a single disposable e-cigarette than cigarette-only or e-cigarette-only users. The attributes ‘being less harmful’ and ‘being able to use indoors’ had a relatively large effect on WTP. The attribute ‘helping you quit’ had a relatively large effect on WTP for cigarette smokers, and ‘flavours’ had a relatively large effect on WTP for e-cigarette-only users. When an e-cigarette did not have the attributes of being less harmful or of being able to be used indoors, WTP was lower for the full sample and cigarette-only users.

Table 2

Mean price willing to pay for a single disposable e-cigarette for overall sample, dual users, cigarette-only users and electronic cigarette (e-cigarette)-only users

Table 3 summarises the results from the multivariate regressions using WTP for an e-cigarette as the outcome. Recall that WTP is measured as the self-reported price each respondent is willing to pay for an e-cigarette with a specified set of attributes. The model includes indicators for each e-cigarette attribute and controls for gender, race/ethnicity, age and income. Among the full sample, the attributes ‘coming in flavours’, ‘helping you quit’, ‘being less harmful’ and ‘being able to use indoors’ all significantly reduced the price respondents were willing to pay for an e-cigarette (p<0.05).

Table 3

Multivariate models for willingness to pay

These results were similar among cigarette-only users. While the base price they were willing to pay was lower compared with dual users and the overall sample, removing attributes had a larger effect on their WTP. Removing the attribute ‘being less harmful’ had the largest significant effect on WTP. The three additional attributes—‘coming in flavours’, ‘helping you quit’ and ‘being able to use indoors’—also had a significant effect on WTP when removed. Among dual users, removing the attribute ‘being less harmful’ had the largest significant effect on WTP, although the effect was smaller than among cigarette-only users. Removing the attribute ‘being able to use indoors’ also had a significant effect on WTP among dual users.

## Discussion

In this paper, we attempt to measure the relative value that smokers place on certain attributes of disposable e-cigarettes. Our results suggest that the most valued attributes among dual users and cigarette-only users are the potentially less harmful effects of e-cigarettes compared with cigarettes and the ability to use e-cigarettes in indoor public places. Cigarette-only users also valued potential use of e-cigarettes as a cessation aid and flavours as attributes.

E-cigarettes are marketed as being safer than regular cigarettes, and our study suggests that smokers value this attribute. However, uncertainty remains as to the actual harm posed by e-cigarettes. If new information becomes available confirming the relative safety of e-cigarettes, more smokers may start using and perhaps switch to e-cigarettes given that they value this attribute. If, however, studies suggest that e-cigarettes are not as safe as currently thought, this could change smokers’ beliefs and values regarding the relative safety of e-cigarettes. Such a change in beliefs and values might affect smokers’ decisions to use or switch to e-cigarettes. Smokers also appear to value the perceived ability to use e-cigarettes in indoor public places where they could not use cigarettes. As localities and states enact regulations that make it harder to use e-cigarettes in public places, the relative value smokers place on e-cigarettes might decrease, and smokers might be less likely to use e-cigarettes for this reason.

Cigarette-only users, but not dual users, appear to value e-cigarettes’ potential as a cessation aid. Interestingly, dual users were more likely to try to quit than cigarette-only users (73.9% vs 36.5%), yet they did not value this e-cigarette attribute. A possible explanation for this result is that most of the cigarette-only users had no experience with e-cigarettes and these e-cigarette naïve smokers may be more optimistic about the efficacy of e-cigarettes as a cessation aid. Dual users, having experience with e-cigarettes, may feel that they will not actually help them quit. As with information about the relative safety of e-cigarettes, not enough information is currently available to know the extent to which e-cigarettes can help smokers quit. Regardless of the evidence base, e-cigarettes are to some extent being marketed as cessation aids, and smokers perceive this to be true. If information becomes available that supports this contention, more smokers may use e-cigarettes as a cessation aid. If, however, new information casts doubt on this potential benefit of e-cigarettes, then fewer smokers may consider using e-cigarettes as a cessation aid.

Cigarette-only users, but not dual users, significantly valued flavour as an attribute. This result might be explained by the fact that cigarettes are not sold in flavours (except for menthol) and so for cigarette users flavours might seem appealing as something different to try. For dual users, flavours might not be valued as much because they are a common characteristic of e-cigarettes.

As with any study, our results and conclusions must be viewed within the context of a number of limitations. First, our measure of value may be biased. We used a direct method of measuring WTP for an e-cigarette with specified attributes (ie, we asked them what amount they would pay) that has known limitations.24 ,25 However, the results for our measure of value agree in general with the importance smokers placed on these attributes are also consistent with smokers’ beliefs about e-cigarettes (see online supplementary appendix). Assuming smokers place some value on the attributes of e-cigarettes we have measured, we would expect the value to be negative (ie, smokers would offer to pay less if an attribute is removed from e-cigarettes). We find all values to be negative and thus consistent with expectations. Also, the rank order in which each group rates the importance of attributes in online supplementary table A1 corresponds with how they price each respective characteristic; for example, the highest proportion of dual smokers responded that using e-cigarettes for harm reduction is ‘very important’, and these same people are willing to pay less for an e-cigarette without harm reduction than for an e-cigarette without the other attributes. Our estimates of the value placed on these attributes are also consistent with other studies examining the reasons people might use e-cigarettes.2 ,26 Also, the estimated WTP for a disposable e-cigarette with the full set of attributes was close to the actual price estimated using scanner data for Florida convenience stores during the study time frame, giving us additional confidence in our method for assessing WTP (results available on request). Second, our study is limited by the small number of e-cigarette-only users. It would be interesting to compare e-cigarette-only users to dual users to see whether the value they place on e-cigarettes differs. Former smokers who now use e-cigarettes might place similar values on e-cigarette attributes as smokers; however, e-cigarette users who have never smoked cigarettes might value different attributes. Third, we only asked participants to value a disposable e-cigarette, and our results might not generalise to other types of e-cigarette devices. Users of disposable devices might differ from users of other (eg, tank) devices. For example, smokers trying to quit might be more likely to use a disposable device because it is smoked like a cigarette. Alternatively, tank devices might deliver nicotine more efficiently, making them a more appealing cessation aid. Also, disposable devices come in fewer flavours than tank devices. Fourth, we always started with the e-cigarette with all attributes and then removed flavours, help quitting, less harmful and ability to use to avoid restrictions in that order when asking respondents to provide a price they would be willing to pay. This could potentially bias the stated WTP. Finally, our sample is a convenience sample from Florida, and thus may not be representative of smokers in the USA.

• There is an emerging literature examining reasons that people use electronic cigarettes (e-cigarettes).

• Despite some literature on the reasons people give for using e-cigarettes, much remains to be learned about the relative value people place on different attributes of e-cigarettes.

• This study directly asks smokers what they would be willing to pay for disposable e-cigarettes with different attributes and from this information estimates the relative value that smokers place on each attribute.

## Acknowledgments

The authors wish to acknowledge Susan Murchie for editorial assistance with the manuscript.

• ## Supplementary Data

This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

## Footnotes

• Contributors JN, YOL and AEK designed the study. AM conducted the analyses. All co-authors consulted in the writing of the manuscript.

• Funding This study was funded by the Florida Department of Health.

• Competing interests None declared.

• Ethics approval RTI IRB.

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