Youth and Young Adult Convenience Store Behaviors: Findings From Intercept Surveys

Abstract

Introduction

To inform tobacco policy and prevention efforts, we examined youth and young adult behaviors at gas-station-convenience stores and whether these behaviors varied by demographics or tobacco use.

Aims and Methods

Between June and July 2017, we conducted 990 intercept surveys at convenience stores in North Carolina among 16- to 25-year olds who were susceptible to or used tobacco. We used logistic regression to examine whether demographics or tobacco use predicted pumping gas, entering the store, or purchasing tobacco, gas, food, lottery tickets, or other items inside the store.

Results

Most participants (85.3%) reported ever using tobacco. Most visited that store at least once per week (40.3%) or once within the past month (41.2%). Just over half (55.0%) reported pumping gas at the store. Of those who went or planned to go inside (68.8%), 43.2% purchased food, 39.2% purchased gas, 33.1% purchased tobacco, and 6.8% purchased lottery tickets. Those who were aged 21–25, visited the store two to three times a week, and went inside already were more likely to purchase tobacco (all p < .05). Past 30-day cigarette, cigar, smokeless, and other tobacco uses were positively associated with purchasing tobacco (all p < .05).

Conclusions

Our sample of young people who were susceptible to or ever used tobacco regularly visited gas-station-convenience stores, and one-third purchased or planned to purchase tobacco during their visit. Convenience stores appear to be an important access point for young people. Practitioners and policy makers should consider a comprehensive set of strategies to reduce access among youth and young adults.

Implications

We conducted 990 intercept surveys among youth and young adults at gas-station-convenience stores. Among our sample of those who had ever used tobacco or were susceptible to use, most visited the store frequently and one-third purchased tobacco, particularly those who used cigarettes and cigars. Tobacco companies heavily market in convenience stores, and our results show tobacco is a product often purchased by susceptible young adults. Practitioners and policy makers should consider prioritizing efforts at convenience stores to reduce product purchase and use.

Introduction

Convenience stores are one of the few remaining locations for tobacco advertisements within the United States. Tobacco companies spend $50 million per year marketing at the point-of-sale, including to attract new customers.¹ Youth and young adults frequent these stores, particularly those who currently or have ever smoked.² Thus, exposure to retail marketing is considered a causal factor for tobacco use initiation.^3–12

Understanding young peoples’ behaviors at these establishments could provide important information to direct policy and communication efforts. If these establishments are a tobacco retail source, particularly for underage youth, policy makers may be more likely to enact minimum age of sale laws or retailer-licensing restrictions to reduce youth access. Additionally, findings may direct communication efforts. The US Food and Drug Administration and other agencies (eg, New York City, Florida Department of Health) have conducted tobacco education campaigns at convenience stores.^13–15 These interventions have shown some success in increasing knowledge and reducing tobacco purchases.^13,15 Identifying where young people spend their time (eg, the gas pumps and food aisles) could inform message placement for future interventions. Few studies have reported on what customers do at these establishments. Two studies among adults at convenience stores (in Pennsylvania and New Zealand) found 13%–14% of store purchases included tobacco.^16,17 The objectives of this study were to determine (1) the content and prevalence of youth and young adult behaviors at gas-station-convenience stores and (2) whether behaviors varied by demographics or tobacco use.

Methods

Participants and Procedures

Data were collected between June and July 2017, as part of two cluster randomized trials examining the effect of tobacco education campaigns, one for cigarillos and one for waterpipe tobacco, at 20 gas-station-convenience stores. Stores were selected based on their location, agreement to post tobacco education messages, and number of touchpoints (ie, on premise locations where the store was willing to post messages, including gas pump toppers, door signs, exterior posters, and interior floor graphics). Stores were randomized to intervention with messages posted or no-message control. We conducted intercept surveys with approximately 50 participants per store at baseline and follow-up (separate samples). The present analyses include only data collected at baseline, prior to message exposure. Data collectors approached anyone appearing to be between the ages of 16 and 25 to invite for screening prior to completing the 5-minute survey on an iPad. Eligible participants were between 16 and 25, had ever used or were susceptible to using tobacco, and had not participated in the study. Participants received a $10 Starbucks gift card after completing the survey. Wake Forest School of Medicine Institutional Review Board approved study procedures.

Measures

Convenience Store Behavior

Participants were asked how often they visited that store in the past month, not including the present day (never, once, two or three times, once a week, two or three times a week, or almost every day). Participants were asked whether they put gas in their car and whether they had been or planned to go inside. Those who had been or planned to go inside were asked what they had already purchased or planned to purchase. Options included gas (pay inside), tobacco products, lottery tickets, food or drink, other product(s), or no purchase. Due to state law at the time that prohibited tobacco purchasing for those younger than 18 years, we did not ask 16- to 17-year olds about purchasing tobacco; they were coded as missing for that item.

Tobacco Use

To determine eligibility, participants were asked whether they ever used any of a list of tobacco products even one or two times. The list included pictures and sample brands for cigarettes, e-cigarettes, cigars, smokeless tobacco (SLT), and other tobacco (pipe, roll-your-own, loose tobacco). If participants reported never using tobacco, they were asked whether they would use each of the products if offered by one of their best friends (susceptibility).^18,19 We asked those who reported use how many of the past 30 days they used each product. We coded use as yes/no per product.

Demographics

We asked participants whether they were male or female, their race and ethnicity, and as a proxy for socioeconomic status, the highest degree or level of school their mother or female guardian completed.

Analyses

We conducted descriptive statistics for all demographic characteristics and outcomes. We conducted adjusted logistic regression models for each outcome, controlling for demographics, tobacco use, number of store visits, and store characteristics. Store characteristics controlled for included the individual store (stores varied in size and products sold), whether it was located in an urban or rural area, whether the store was in the cigarillo or waterpipe trial, and whether the store was randomized to the intervention or control conditions.

Results

Sample Characteristics

Of 6509 individuals approached, 3947 (60.6%) declined to be screened, citing reasons such as being in a hurry or not being interested in the incentive. A total of 2562 (39.4%) agreed to be screened. Of those screened, 1024 were eligible (40.0%). Individuals were ineligible due to age (56.2%), being a nonsusceptible nonuser (16.3%), having already completed the survey (5.7%), or some combination of these reasons (21.8%). Of those eligible, 990 (96.7%) completed the survey (range 47–52 completions per store).

Most participants were between ages 21 and 25 (54.7%), male (56.3%), white (51.3%), non-Hispanic (85.7%), and had a mother with at least a college degree (57.1%; Table 1). Most (85.3%) reported ever using tobacco, with 40.0% reporting past 30-day use of a single product and 31.7% reporting past 30-day use of multiple products. Past 30-day use was highest for cigarettes (39.7%), then cigars (38.5%), e-cigarettes (20.0%), waterpipe (11.3%), SLT (9.2%), and other tobacco products (6.8%). Participants visited that store at least once per week (40.3%) or within the past month (41.2%); 18.5% reported not visiting within the past month.

Table 1.

Sample Characteristics

	n (%)
Age, years (n = 990)
16–17	125 (12.6)
18–20	323 (32.6)
21–25	542 (54.7)
Gender (n = 980)
Male	552 (56.3)
Female	428 (43.7)
Race (n = 982)
White	504 (51.3)
Black	309 (31.5)
Other	169 (17.1)
Ethnicity (n = 927)
Non-Hispanic	794 (85.7)
Hispanic	133 (14.3)
Mother’s education (n = 939)
≥College degree	403 (42.9)
<College degree	536 (57.1)
Tobacco use status (n = 990)
Single product use	396 (40.0)
Polytobacco use	314 (31.7)
Lifetime, but no past 30-day use	134 (13.5)
Susceptible never use	146 (14.7)
Past 30-day cigarette use (n = 986)
No	595 (60.3)
Yes	391 (39.7)
Past 30-day e-cigarette use (n = 986)
No	789 (80.0)
Yes	197 (20.0)
Past 30-day cigar use (n = 986)
No	606 (61.5)
Yes	380 (38.5)
Past 30-day hookah use (n = 986)
No	875 (88.7)
Yes	111 (11.3)
Past 30-day SLT use (n = 986)
No	895 (90.8)
Yes	91 (9.2)
Past 30-day OTP use (n = 986)
No	919 (93.2)
Yes	67 (6.8)
Store type (n = 990)
Rural	48 (4.8)
Urban	942 (95.2)
Past month store visits (n = 986)
Never	182 (18.5)
Once	140 (14.2)
Two to three times	266 (27.0)
Once a week	101 (10.2)
Two to three times a week	146 (14.8)
Almost every day	151 (15.3)
Got gas (n = 984)
No	443 (45.0)
Yes	541 (55.0)
Went/going inside (n = 985)
Not going inside	307 (31.2)
Went inside	539 (54.7)
Going inside	139 (14.1)
Purchase tobacco (n = 602)
No	403 (66.9)
Yes	199 (33.1)
Purchase gas (n = 678)
No	412 (60.8)
Yes	266 (39.2)
Purchase food (n = 678)
No	385 (56.8)
Yes	293 (43.2)
Purchase lottery (n = 678)
No	632 (93.2)
Yes	46 (6.8)
Purchase other (n = 678)
No	601 (88.6)
Yes	77 (11.4)
No purchase (n = 678)
No	630 (92.9)
Yes	48 (7.1)

SLT = smokeless tobacco; OTP = other tobacco products, include products not listed above (eg, pipe, roll-your-own, clove cigarettes).

Convenience Store Behavior

Over half (55.0%) reported getting gas. Most (68.8%) reported they went or planned to go inside. Of those, the most common purchase or planned purchase was food (43.2%), followed by gas (39.2%), tobacco (33.1%), something not listed (11.4%), and lottery tickets (6.8%); 7.1% purchased nothing. Among participants who purchased or planned to purchase tobacco (n = 199), 65.3% did or purchased something else: 42.2% got gas, 32.7% paid for gas inside, 34.2% purchased food, 8.0% purchased lottery tickets, and 8.0% purchased something not listed.

Factors Associated With Convenience Store Behaviors

In adjusted models, we identified factors associated with convenience store behaviors (Table 2). Past 30-day cigar use (adjusted odds ratio [AOR] = 2.02, 95% confidence interval [CI] = 1.21–3.37) and visiting the store two to three times per week (AOR=1.97, 95% CI=1.13–3.43), or almost every day (AOR=2.96, 95% CI=1.63–5.39), were positively associated with going inside the store. Past 30-day SLT use (AOR = 0.46, 95% CI = 0.22–0.94) was negatively associated with purchasing gas. Visiting the store once in the past month (AOR = 2.01, 95% CI = 1.05–3.85) was positively associated with purchasing gas. Female gender (AOR = 0.69, 95% CI = 0.48–0.996) was negatively associated with purchasing food. Having already been inside (AOR = 1.78, 95% CI = 1.16–2.74) was positively associated with purchasing food. We did not identify significant factors associated with purchasing lottery tickets or purchasing something other than the products listed. The only store-level characteristic associated with convenience store behavior was the individual store (eg, some stores had greater food availability and were therefore more likely to have participants report purchasing food; data available on request).

Table 2.

Adjusted Logistic Regression Models

	Put gas in car (n = 873) AOR (95% CI)	Went/going inside (n = 874) AOR (95% CI)	Purchase tobacco (n = 542) AOR (95% CI)	Purchase gas (n = 610) AOR (95% CI)	Purchase food (n = 610) AOR (95% CI)	Purchase lottery (n = 610) AOR (95% CI)	Purchase other (n = 610) AOR (95% CI)
Age, years
16–17	Ref	Ref	N/A	Ref	Ref	Ref	Ref
18–20	1.44 (0.90–2.29)	1.30 (0.79–2.15)	Ref	0.87 (0.48–1.59)	0.69 (0.38–1.26)	5.10 (0.64–40.50)	0.51 (0.22–1.20)
21–25	1.11 (0.71–1.74)	21.30 (0.80–0.11)	1.98 (1.24–3.16)	0.58 (0.33–1.42)	1.24 (0.70–2.20)	3.58 (0.45–28.46)	0.57 (0.25–1.28)
Gender
Female vs. male	1.08 (0.80–1.44)	0.85 (0.62–1.17)	1.57 (0.996–2.48)	0.98 (0.68–1.42)	0.69 (0.48–0.996)	0.84 (0.40–1.72)	1.04 (0.59–1.86)
Race
Black vs. white	0.81 (0.57–1.15)	1.40 (0.95–2.08)	1.17 (0.68–2.02)	1.25 (0.75–1.82)	0.85 (0.54–1.32)	0.61 (0.25–1.49)	0.69 (0.32–1.48)
Other vs. white	0.91 (0.58–1.43)	0.81 (0.50–1.33)	1.31 (0.64–2.70)	1.25 (0.70–2.22)	1.11 (0.63–1.96)	0.67 (0.20–2.21)	1.60 (0.74–3.47)
Ethnicity
Hispanic vs. non-Hispanic	1.09 (0.68–1.74)	1.20 (0.72–2.00)	0.51 (0.24–1.11)	0.75 (0.42–1.34)	1.03 (0.58–1.81)	0.36 (0.08–1.68)	1.15 (0.51–2.62)
Mother’s education
<College vs. ≥college	1.29 (0.97–1.72)	0.88 (0.64–1.22)	1.05 (0.68–1.62)	1.38 (0.96–1.96)	1.18 (0.83–1.67)	0.92 (0.46–1.88)	1.37 (0.79–2.37)
Tobacco use
Single product use	Ref	Ref	Ref	Ref	Ref	Ref	Ref
Polytobacco use	0.63 (0.36–1.11)	0.94 (0.49–1.81)	0.33 (0.15–0.72)	0.82 (0.42–1.61)	0.96 (0.49–1.86)	0.50 (0.13–1.90)	0.90 (0.34–2.38)
Lifetime, no current use	1.52 (0.88–2.63)	1.34 (0.73–2.44)	0.14 (0.03–0.63)	1.56 (0.78–3.10)	1.16 (0.59–2.30)	1.54 (0.38–6.28)	0.68 (0.23–2.03)
Susceptible never use	1.21 (0.70–2.08)	0.99 (0.55–1.79)	0.16 (0.03–0.74)	0.93 (0.46–1.88)	1.38 (0.69–2.74)	1.29 (0.31–5.30)	0.96 (0.35–2.66)
Past 30-day cigarette use
Yes vs. No	0.93 (0.61–1.43)	1.35 (0.82–2.24)	2.90 (1.57–5.36)	0.89 (0.52–1.53)	1.03 (0.61–1.75)	0.93 (0.33–2.61)	0.65 (0.31–1.40)
Past 30-day e-cigarette use
Yes vs. No	1.36 (0.88–2.12)	1.09 (0.65–1.85)	1.05 (0.57–1.94)	1.08 (0.62–1.87)	1.21 (0.71–2.07)	0.96 (0.32–2.80)	1.22 (0.56–2.66)
Past 30-day cigar use
Yes vs. No	1.21 (0.78–1.87)	2.02 (1.21–3.37)	3.21 (1.72–5.97)	1.09 (0.64–1.88)	0.82 (0.48–1.37)	2.07 (0.73–5.85)	0.88 (0.41–1.88)
Past 30-day waterpipe use
Yes vs. No	1.12 (0.69–1.84)	1.30 (0.72–2.35)	1.12 (0.56–2.24)	0.90 (0.50–1.63)	1.37 (0.73–2.45)	2.81 (0.97–8.12)	1.22 (0.52–2.86)
Past 30-day SLT use
Yes vs. No	0.85 (0.49–1.46)	1.69 (0.87–3.30)	3.53 (1.64–7.60)	0.46 (0.22–0.94)	1.37 (0.73–2.60)	0.75 (0.19–2.95)	1.98 (0.84–4.68)
Past 30-day OTP use
Yes vs. No	1.14 (0.63–2.07)	0.66 (0.33–1.28)	4.61 (1.95–10.94)	1.84 (0.87–3.89)	0.54 (0.25–1.18)	1.19 (0.31–4.64)	1.33 (0.47–3.74)
Past month store visits
Never	Ref	Ref	Ref	Ref	Ref	Ref	Ref
Once	1.55 (0.94–2.55)	1.09 (0.65–1.83)	1.20 (0.52–2.81)	2.01 (1.05–3.85)	0.77 (0.41–1.46)	1.34 (0.27–6.55)	1.63 (0.48–5.52)
Two to three times	1.08 (0.71–1.65)	1.24 (0.79–1.94)	1.04 (0.50–2.17)	0.98 (0.55–1.77)	1.01 (0.58–1.76)	2.60 (0.66–10.21)	3.60 (1.28–10.19)
Once a week	2.00 (1.15–3.50)	1.15 (0.65–2.04)	0.51 (0.19–1.35)	1.43 (0.70–2.92)	1.20 (0.60–2.40)	1.60 (0.29–8.68)	2.48 (0.71–8.16)
Two to three times a week	1.04 (0.64–1.70)	1.97 (1.13–3.43)	2.50 (1.13–5.53)	1.33 (0.70–2.52)	1.23 (0.66–2.28)	2.35 (0.54–10.24)	1.94 (0.60–6.30)
Almost every day	1.19 (0.73–1.95)	2.96 (1.63–5.39)	1.78 (0.82–3.83)	1.72 (0.92–3.21)	1.08 (0.60–1.98)	3.20 (0.78–13.08)	2.92 (0.97–8.81)
Going vs. went inside	Not Applicable	Not Applicable	0.49 (0.28–0.88)	0.71 (0.45–1.12)	1.78 (1.16–2.74)	0.79 (0.30–2.08)	1.51 (0.81–2.82)
Rural vs. urban	0.57 (0.27–1.19)	0.63 (0.29–1.38)	0.80 (0.22–2.92)	1.07 (0.43–2.69)	0.46 (0.18–1.13)	0.91 (0.09–8.89)	1.59 (0.44–5.76)

Each column is a separate outcome. SLT = smokeless tobacco; OTP = other tobacco products, include products not listed above (eg, pipe, roll-your-own, clove cigarettes). Models also control for an individual store, whether the store was a waterpipe or cigarillo store, and whether it was designated as an intervention or a control store. . All bold values are p < .05.

Factors Associated With Tobacco Purchases

Age 21–25 (AOR = 1.98, 95% CI = 1.24–3.16), past 30-day cigarette use (AOR = 2.90, 95% CI = 1.57–5.36), past 30-day cigar use (AOR = 3.21, 95% CI = 1.72–5.97), past 30-day SLT use (AOR = 3.53, 95% CI = 1.64–7.60), past 30-day other tobacco product use (AOR = 4.61, 95% CI = 1.95–10.94), and visiting the store two to three times per week in the past month (AOR = 2.50, 95% CI = 1.13–5.53) were positively associated with purchasing tobacco. Compared to those who reported past 30-day use of a single tobacco product, those who reported polytobacco use (AOR = 0.33, 95% CI = 0.15–0.72), lifetime but not current use (AOR = 0.14, 95% CI = 0.03–0.63), or never use (AOR = 0.16, 95% CI = 0.03–0.74) were less likely to purchase tobacco. Compared to those who had already been inside, those planning to go inside were less likely to report planning to or already purchasing tobacco (AOR = 0.49, 95% CI = 0.28–0.88).

Discussion

Young people who have ever used or are susceptible to using tobacco visit gas-station-convenience stores and report purchasing tobacco products. Almost 33% of 18- to 25-year olds in the present sample reported purchasing tobacco products. This is higher than the 13%–14% reported by two studies among adults in New Zealand¹⁶ and Pennsylvania¹⁷; however, neither of those studies restricted the sample to youth and young adults who had used or were susceptible to tobacco use.

Past month use of traditional products (ie, cigarettes, cigars, and SLT) was associated with planned or reported tobacco purchases at the store. We did not find an association between tobacco purchases and current e-cigarette use, the product used most often among this age group within the United States.^20,21 Our sample reported higher use of cigarettes and cigars than national samples, suggesting a preference for these products over e-cigarettes. Additionally, these surveys were conducted in summer 2017, which appears to have been before increases in youth e-cigarette use in North Carolina.²² Interestingly, those who reported single product use were more likely than those who reported never or polytobacco use to report planning to or purchasing tobacco that day. Given the high rates of polytobacco use among youth and young adults,^23,24 this is surprising. However, it is possible that youth and young adults are accessing some tobacco products, particularly e-cigarettes, from sources other than gas-station-convenience stores.²⁵

Most participants went inside the store, with cigar users more likely than nonusers. Thus, placing communication messages inside or on the door may reach this population. Posting messages inside near other products (eg, food) might be effective as most young adults who purchased tobacco were likely to also purchase something else while in the store. However, one-third of respondents purchased only tobacco; for these young adults, it may be best to place antitobacco messages near where tobacco products are sold (ie, the tobacco power wall).

Due to legal restrictions at the time of data collection prohibiting tobacco purchases younger than 18, we did not directly ask 16- and 17-year olds about their tobacco purchases. In our study, 21- to 25-year olds were more likely than 18- to 20-year olds to report intending to or purchasing tobacco, but nearly 25% of 18- to 20-year olds did report purchasing tobacco. Now that the minimum legal age of sale for tobacco has increased to 21 nationwide, monitoring how the law impacts those 18- to 20-year olds who use tobacco will be critical.

Findings should be considered alongside limitations. First, behaviors were self-reported. The larger study examined exposure to a tobacco education campaign, and some survey items asked participants about their knowledge and perceptions of different tobacco products; it is possible that some participants chose not to report their actual or intended tobacco purchases. For example, those who had already been in the store were more likely than those who had not yet been in the store to report purchasing or intending to purchase tobacco. Second, the sample consisted of a convenience sample of young people in North Carolina, a state rich in tobacco history, with preemption and no tobacco retailer licensing requirements; therefore, the results may not be generalizable to other states. Additionally, stores that opted to participate may differ from those who did not, which could also limit generalizability.

Conclusions

Young people who use tobacco or are susceptible to tobacco use are visiting gas-station-convenience stores and purchasing tobacco. Convenience stores represent key locations for tobacco product marketing.^2,11 Thus, practitioners and policy makers should consider a comprehensive set of strategies to reduce the tobacco availability among youth and young adults, including retailer licensing and tobacco use prevention campaigns at gas-station-convenience stores.

Funding

The research reported in this publication was supported by grant number P50CA180907, from the National Cancer Institute and the FDA Center for Tobacco Products. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the Food and Drug Administration.

Declaration of Interests

The authors listed have no conflicts of interest to declare.