Article Text

Download PDFPDF

Unplanned cigarette purchases and tobacco point of sale advertising: a potential barrier to smoking cessation
  1. Eben J Clattenburg1,
  2. Jessica L Elf2,
  3. Benjamin J Apelberg3
  1. 1Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
  2. 2Institute for Global Tobacco Control, Johns Hopkins University, Baltimore, Maryland, USA
  3. 3Johns Hopkins School of Public Health, Baltimore, Maryland, USA
  1. Correspondence to Eben J Clattenburg, Johns Hopkins University School of Medicine, Doctors’ Lounge, 110 Harvey/Nelson Building, Johns Hopkins Hospital, 600 N. Wolfe Street, Baltimore, MD 21287, USA; eclatte1{at}jhmi.edu

Abstract

Background In the USA, tobacco marketing expenditure is increasingly concentrated at the point of sale (POS). Previous studies have demonstrated an association between exposure to tobacco POS advertising and increased smoking initiation, but limited evidence is available on adult smokers’ decisions and behaviours.

Methods An immediate post-cigarette purchase survey was administered to 301 cigarette purchasers outside of two grocery stores in Vermont to assess the prevalence of unplanned purchases and opinions about POS tobacco advertising and displays.

Results In total, 11.3% of purchases were reported as unplanned. Certain groups were more likely to make unplanned purchases including: 18–24-year-olds (OR: 2.1, 95% CI 1.0 to 4.4), less than daily smokers (OR: 5.6, 95% CI 1.9 to 16.9), smokers who made 3+ quit attempts in the previous year (OR: 2.4, 95% CI 0.9 to 6.0), those who plan to quit in the next month (OR: 3.7, 95% CI 1.6 to 9.0), and those who agreed that tobacco POS advertising makes quitting smoking harder (OR: 2.3, 95% CI 1.1 to 4.8). Overall, 31.2% of participants agreed that tobacco POS advertising makes quitting smoking harder. Individuals who intended to quit within the next month, made 3+ quit attempts in the last year, or made an unplanned cigarette purchase were the most likely to agree.

Conclusions Young adults and individuals making multiple quit attempts or planning to quit in the next month are more likely to make unplanned cigarette purchases. Reducing unplanned purchases prompted by tobacco POS advertising could improve the likelihood of successful cessation among smokers.

  • Advertising and Promotion
  • Prevention
  • Public opinion
  • Cessation

Statistics from Altmetric.com

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

In the USA, major restrictions on tobacco advertising were implemented following the Master Settlement Agreement of 1998. The Master Settlement Agreement banned outdoor advertising, branded merchandise, product placement in movies and television shows, the use of cartoon characters in advertising, and prohibited brand-name sponsorship of events, among other stipulations.1 As these advertising restrictions were implemented, tobacco industry marketing shifted to unregulated point of sale (POS) advertising, cigarette displays and promotions.2 In 2008, tobacco companies spent $163 million on advertising at the POS and $7.1 billion on price discounts.3

Research among teens has demonstrated an association between exposure to tobacco POS advertising and increased smoking initiation.4 ,5 However, there is limited research on the association between tobacco advertising at the POS and adult smokers’ decisions and behaviours in the USA.6 One mechanism by which POS advertising and displays could negatively impact smokers is by prompting unplanned purchases of cigarettes among those attempting to quit. Among adult smokers in Houston, Texas, cigarette-related pictures caused cravings, and these cravings were more intense among nicotine deprived smokers.7 Outside of the USA, more extensive research on the POS environment has been conducted. Wakefield et al found that 38% of smokers in Victoria, Australia who made a quit attempt in the past 12 months reported an increased urge to buy cigarettes when they saw tobacco POS displays.8 In western Australia, post-purchase interviews were conducted to assess the rate of unplanned tobacco purchases and whether or not the POS display was reported to prompt the purchase.9 The study found that 22% of cigarette purchases were unplanned and that 47% of these purchases were reported to be prompted by the POS display. Unlike the USA, however, western Australia has strict POS restrictions limiting the display to one example of each tobacco product, with a total area not exceeding one square metre and prohibitions on any other types of advertising materials.9

In the USA, the extent to which unplanned cigarette purchases are made and the effect of POS advertising and displays is unknown. This study seeks to begin to fill this knowledge gap by administering immediate post-purchase surveys in rural Vermont in order to assess: (1) the proportion of cigarette purchases at grocery stores reported as unplanned and (2) factors associated with unplanned cigarette purchases, including beliefs about POS tobacco advertising and displays.

Methods

Data collection consisted of immediate post-cigarette purchase surveys of adults at two grocery stores in Vermont during July of 2010. Ethics approval for this study was obtained from the Johns Hopkins Bloomberg School of Public Health Institutional Review Board.

Study location

In each of two communities in Windsor County, Vermont, one grocery store that sells cigarettes was selected for recruitment. These two rural communities of approximately 4000 and 10 000 inhabitants, respectively, were selected for ease of access and perceived generalisability to rural Vermont and low socioeconomic status areas. Grocery stores from two separate communities were chosen in order to prevent duplicate recruitment of participants into the study, since individual identifying information was not collected. The two towns have similar economic status (inflation adjusted per capita incomes $20 899 and $24 313, respectively, compared the national average of $27 041.) and ethnic compositions (97.7% and 97.6% American).10

Participants

As individuals exited the grocery store, they were asked to participate in an anonymous survey pertaining to ‘consumer choices’. No mention of POS tobacco advertising was made. Eligible participants were at least 18 years of age, had just purchased cigarettes from the store and had not already taken the survey. If interested in participating, oral informed consent was obtained, and participants received a healthy energy bar as compensation.

Survey instrument and data collection

A 5-minute interview was designed to assess the prevalence of unplanned cigarette purchases and participant opinions about POS tobacco advertising and displays. The survey instrument included the following demographic characteristics: participant age, sex and race/ethnicity. Questions regarding smoking behaviour included: smoking frequency (daily vs occasional), intention to quit, number of quit attempts in the past year, usual and purchased brands, and plan to purchase cigarettes prior to store arrival (table 1). Unplanned cigarette purchases were identified using the following question: ‘Before you arrived at this store just now, did you plan on purchasing cigarettes?’ Participants who made unplanned purchases were asked: ‘What if anything in this store prompted you to buy cigarettes?’ and asked to select one of the following response options: ‘lower price’, ‘recommendation from a friend’, ‘special promotion/coupon’, ‘tobacco advertising or display at the store’ or ‘other’, which was then specified by the participant. The response: ‘tobacco advertising or display’ qualified the participant as being influenced by advertising.

Table 1

Demographic characteristics of recent cigarette purchasers, Vermont (n=301)

Opinions about POS advertising were assessed by asking respondents whether they agreed or disagreed with the following statements: ‘Tobacco advertising or displays in stores influences the brand or type of product I purchase’, ‘Tobacco advertising or displays in stores makes it more difficult for me to quit smoking’, and ‘There should be a law prohibiting the advertisement and display of tobacco products in stores’. Responses to the statements were assessed using a 5-point scale, with 1 as ‘disagree’ and 5 as ‘agree’. For data analysis, responses of 1 through 3 were grouped and categorised as disagree/neutral and responses of 4 or 5 were categorised as ‘agree’.

After the first 100 surveys of the study were administered, three additional questions were added to capture the following characteristics: educational attainment, cigarettes per day and time to first smoke after waking. Data was collected between 8:00 and 17:00 over a period of 31 days, primarily during the workweek. Data was also collected on 3 weekend days to broaden the participant demographics. A total of 380 individuals were approached and 301 participants were successfully interviewed, yielding a response rate of 79.2%.

Data analysis

Unplanned cigarette purchases and agreement with the statement that POS advertising and displays makes quitting smoking more difficult were the two outcomes of interest in this study. Proportions of these outcomes were calculated and compared across subgroups. χ2 test was used for variables with cell sizes greater than five and Fischer's Exact Test was used for variables with cell sizes less than five. Logistic regression was conducted to assess the association between each of the demographic characteristics, smoking behaviours and opinions on the two outcomes of interest, after adjustment for age and store location. Age and store location were included in regression models because they were hypothesised to be associated with the outcomes of interest. Education was not included in regression models because it was only collected on a subsample of the study population. Sensitivity analyses conducted on this subsample found that the adjustment for education did not materially affect the associations observed. Study data were collected and managed using REDCap electronic data capture tools hosted at Johns Hopkins Bloomberg School of Public Health.11 All data analysis was conducted using Stata 11.12

Results

Participants

Characteristics of the 301 participants are summarised in table 1. Nearly all were daily smokers, over half smoked more than 15 cigarettes per day and 69% made at least one quit attempt in the previous 12 months.

Unplanned purchases

Eleven percent (11.3%, n=34) of participants reported making unplanned purchases. Of those, 76.5% (n=26) identified POS tobacco advertising or display as the precipitating factor, 8.8% (n=3) ‘saw someone else purchase tobacco’, and 14.7% (n=5) responded ‘other’. ‘Other’ reasons for unplanned purchases of tobacco include ‘remembered I needed them’, ‘bought beer and remembered I needed smokes’, ‘had enough money left after groceries’, ‘had extra money after lunch’ and ‘promotion’.

Young adults (18–24-year-olds) were twice as likely to have made an unplanned purchase compared with those aged 25 and older (OR: 2.1, 95% CI 1.0 to 4.4). After adjustment for age and store location, several factors were associated with an increased likelihood of having made an unplanned purchase of cigarettes. Less than daily smokers (OR: 5.6, 95% CI 1.9 to 16.9), those who plan to quit within the next month (OR: 3.7, 95% CI 1.6 to 9.0), those who made 3+ quit attempts in the past year (OR: 2.4, 95% CI 0.9 to 6.0), those who believe that tobacco POS advertising and displays makes quitting harder (OR: 2.3, 95% CI 1.1 to 4.8), and those who believe that tobacco POS advertising and displays influences brand choice (OR: 2.6, 95% CI 1.0 to 6.4) were more likely to have made an unplanned purchase (see table 2). No significant differences in unplanned purchasing were found by sex, cigarettes per day, time to first smoke, support for POS advertising ban, education, race/ethnicity or usual versus different brand of purchase.

Table 2

Percentages and ORs for unplanned cigarette purchase, by demographic and smoking characteristics, Vermont (n=301)

Opinions related to POS advertising and displays

Three 5-point scale questions were included to assess if the participants support a ban on tobacco POS advertising, believe that tobacco POS advertising makes quitting smoking more difficult and believe that tobacco POS advertising influences brand choice (table 1). Approximately one out of five smokers (18.9%, n=57) supported a POS tobacco advertising and display ban and another 22% (n=66) were neutral. There was no statistically significant difference in percentage of unplanned cigarette purchases between those who supported the ban and those who did not.

Over 30% of participants agreed that tobacco POS advertising or display makes quitting smoking more difficult (n=31.2%, n=94). Table 3 describes the likelihood of agreeing to this statement among different subgroups. Participants who made three or more quit attempts in the past year (OR: 18.7, 95% CI 8.4 to 42.0) and participants who stated that they are planning to quit within the next month (OR: 13.0, 95% CI 5.7 to 29.6) were also more likely to agree that tobacco POS advertising or display makes quitting smoking more difficult (table 3). Those who are not interested in quitting were less likely to agree (OR: 0.1, 95% CI 0.03 to 0.5).

Table 3

Percentages and ORs for agreeing that POS tobacco advertising makes quitting smoking more difficult, by demographic and smoking characteristics, Vermont (n=301)

Discussion

This study assessed the proportion of cigarette smokers making unplanned purchases at two grocery stores in Vermont and identified factors associated with those unplanned cigarette purchases. A total of 11.3% of the purchases were reported as unplanned, and several associations with unplanned purchases were identified. Young adults, less than daily smokers, smokers who made three or more quit attempts in the previous year, smokers who intend to quit in the next month and those who agreed that tobacco POS advertising or display makes quitting smoking harder were more likely to make unplanned purchases. Additionally, nearly a third of the smokers interviewed agreed that tobacco POS advertising and displays makes quitting smoking harder. Individuals who intended to quit within the next month, made three or more quit attempts in the last year, or made an unplanned cigarette purchase were the most likely to agree.

In a study in western Australia, Carter et al found through post-purchase surveys that 20% of participant purchases were unplanned and that 50% of those purchases were reported as prompted by tobacco POS advertising.9 Carter et al also identified a trend towards 18–24-year-olds making more unplanned tobacco purchases, similar to the association presented here.9 In this study, 11.3% of purchases were reported unplanned, and 76.5% of unplanned cigarette purchases were reported to be prompted by tobacco POS advertising. Therefore, 10% of all cigarette purchases in the western Australian study and 8.6% of all cigarette purchases in this study were reported to have been prompted by tobacco POS advertising or displays. Our study was relatively small and limited in geography and therefore not intended to be representative of the USA. There is likely to be a large degree of variation in unplanned cigarette purchases across the US, as a function of both the store environment and types of consumers frequenting the stores. In addition to small sample sizes, possible differences in unplanned purchase rates between our study and Carter et al could stem from differences in disposable income or level of nicotine addiction among purchasers. The percentage of unplanned cigarette purchases could also be lower because studies have shown that grocery stores have a lower concentration of tobacco POS advertising than other store types, such as convenience stores.13 In addition, not all checkout lanes at the grocery stores sold cigarettes, therefore not all individuals visiting the store would be exposed to tobacco displays and advertising.

In addition to young adult smokers, the results of this study also show that smokers who have made three or more quit attempts in the past, those who are planning to quit in the next month, and less than daily smokers are more likely to make unplanned cigarette purchases. Nearly one third of respondents (31.2%) agreed with the statement that POS tobacco advertising or displays makes quitting harder, which is similar to what was found in the Wakefield's (31%) and Carter's (28%) study.8 ,9 In this study, the same subgroups who made unplanned purchases at a higher rate (participants who made at least three quit attempts in the past year and those who are planning to quit in the next month) were also the most likely to agree with this statement. In fact, those who agreed were twice as likely to make an unplanned purchase. This association suggests that among participants who are actively trying to quit smoking, POS advertising may be a significant perceived barrier to their success. Among those who believe that POS advertising makes it more difficult to quit smoking, 43% supported a ban on POS advertising.

An association has been demonstrated between exposure to tobacco advertising and increased use of tobacco products.14 Several recent studies have examined the potential role of POS advertising. McNeill et al examined the effect of legislation requiring the removal of tobacco POS displays from retailers in Ireland. The removal of the displays led to drastically decreased recall of tobacco and cigarette pack displays by adult smokers from 59% to 20%.15 A second study by Hoek et al explored how tobacco POS advertising affects smokers trying to quit using indepth interviews of 20 participants. The interviews found that tobacco POS advertising evoked emotional and physical reactions in the participants that create an ‘ongoing temptation’ to buy cigarettes or smoke.16 A third study assessed smokers who have a medium or high sensitivity to tobacco POS advertisement (they notice the advertisements more), and it found that they were less likely to quit smoking than smokers with low tobacco POS sensitivity.17 These three studies suggest that smokers who are more aware of tobacco POS advertising and displays may have a harder time quitting, that removing tobacco displays reduces awareness of the advertising and that both qualitative and quantitative studies found that advertising makes quitting more difficult. The data from our study are consistent with the notion that smokers trying to quit find tobacco POS advertisements a barrier to successful smoking cessation. As a result, restrictions limiting the display and marketing of tobacco products at the POS could reduce unplanned cigarette purchases and improve cessation outcomes.

The findings from this study are limited in their generalisability since the study was conducted in a small, mostly white population from two nearby towns in Vermont. Vermont's smoking prevalence among adults is 16.8% (lower than the US median, 18.4%) and Vermont has comprehensive smoking bans in the workplace, restaurants and bars.18 The state has a $2.24 tax per pack of cigarettes, which is the ninth highest in the USA.19 This is more stringent than many states, and the high tax could limit the number of unplanned purchases.

On average, grocery stores accounted for only 12.4% of total cigarette sale volume between 1994 and 2002, whereas convenient stores and gas stations accounted approximately 75% of sales.20 In addition, the characteristics and purchasing patterns of smokers may be different at grocery stores compared with other types of tobacco retail outlets. For example, the types and extent of tobacco advertising at these locations that are visible from outside of the store or from the street may contribute to the amount of unplanned purchases that are made. We did not measure the quantity of advertising in the stores in this study. Studies from California, Ontario, Massachusetts, Guatemala, Argentina and Ireland indicate that the quantity of tobacco POS advertising and displays vary greatly by store type, proximity to schools and neighbourhood socioeconomic status.13 ,19 ,21 In addition, sampling primarily occurred during weekdays, and only limited survey collection was performed on weekends, which could result in an oversampling of unemployed individuals. Future studies should be repeated in additional locations among a more diverse study population and include locations such as gas stations and convenience stores to better characterise the extent to which unplanned purchases occur and the potential influence of the POS environment.

This study was also limited by its small sample size. Only 34 of participant purchases were unplanned (11% of total), which reduced the ability to conduct extensive multivariate analyses. Furthermore, many of the smoking-related indicators, such as intention to quit and previous quit attempts, are highly correlated with one another. As a result, these variables were assessed in separate models, rather than attempting to identify the independent associations with each of the outcomes under study. Lack of information on education among all participants limited our ability to include it as a potential confounder in regression models. Sensitivity analyses conducted on this subsample found that the adjustment for education did not materially affect the associations observed. Finally, all measures, including unplanned purchases and the extent to which POS advertising and displays influenced these purchases, were based on self-report. It is possible that after spending time purchasing groceries, one could buy cigarettes and forget that they had not planned to purchase the cigarettes before entering the store. In addition, smokers may over or underestimate the role that product advertising plays on their behaviour.

Despite these limitations, there was a trend towards smokers with greater interest in quitting and previous quit attempts being more likely to report making unplanned cigarette purchases. In particular, we found that smokers who intend to quit within the next month or had made three or more quit attempts in the past year were more likely to report unplanned cigarette purchases and attribute POS advertising and displays to prompting those purchases. While this study is not able to identify POS advertising and displays as a cause of unplanned purchases, the findings provide insights into perceived barriers to quitting. In the USA, almost 70% of adults who smoke report wanting to quit, and 45% have made at least one attempt to quit within the past year.22 ,23 Despite this desire and effort, only 4% of smokers successfully quit from 2009 to 2010.24 Our findings suggest that the display and advertising of tobacco products may act as a barrier to cessation among smokers trying to quit and that reducing unplanned purchases prompted by tobacco POS advertising and displays could improve the likelihood of successful cessation.

What this paper adds

  • The retail outlet has become an increasingly important avenue for tobacco industry advertising. To our knowledge, this is the first study to conduct immediate post-purchase surveys to assess influence of point of sale (POS) advertising on unplanned purchases by smokers in the USA.

  • Most surveyed participants who made unplanned purchases reported POS advertising as the precipitating factor. Additionally, young adults, smokers who are trying to quit, and smokers who agreed that cigarette advertising makes quitting harder were more likely to make unplanned cigarette purchases.

  • Reducing exposure to cigarette POS advertising and eliminating visible product displays may help reduce unplanned cigarette purchases and increase success among smokers trying to quit.

Acknowledgments

We would like to thank Dr Eric Bass and Dr David Friedman for their support of the project.

References

Footnotes

  • Contributors EJC: Development of study design, data collection, analysis, interpretation of data and drafting the paper. JLE: Contributed to development of study design, data analysis, interpretation and revision of the paper. BJA: Development of concept and study design, overview of data analysis and interpretation and revision of paper. All authors reviewed the final version of the paper and approved it for publication.

  • Competing interests None.

  • Ethics approval The Johns Hopkins University School of Public Health IRB.

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