Changes in the Point-of-Sale Among Vape Shops in Six U.S. Metropolitan Areas Over Time, 2018–2021

Abstract

Introduction

E-cigarette retail surveillance is needed during regulatory changes, like the U.S. increasing minimum legal sales age to 21 (T21) and flavor restrictions (2019 and 2020) and certain state/localities increasing related restrictions.

Aims and Methods

We examined regulatory compliance (eg, minimum-age signage), promotional strategies (eg, health claims), and products at 2 timepoints among vape shops across six U.S. metropolitan statistical areas (MSAs; Atlanta, Boston, Minneapolis, Oklahoma City, San Diego, Seattle). In summer 2018, pairs of trained auditors assessed randomly selected shops (n = ~30/MSA). In fall 2021, audits were conducted among 2018 shops (if open and allowed) and additional randomly selected shops (n = ~20/MSA). Data from 179 shops in 2018 and 119 in 2021 (43 from the 2018 sample) were compared.

Results

There were decreases (p < .01) in the proportion of shops with (1) minimum-age signs (90.5% vs. 73.9%), (2) their own e-liquid brand (68.2% vs. 44.5%), onsite vaping (73.2% vs. 46.2%), counter seating (65.2% vs. 34.5%), and e-liquid sampling (90.0% vs. 33.6%), and (3) signs with product/price promotions (89.9% vs. 65.5%), health/cessation claims (29.1% vs. 12.6%), and cartoon imagery (27.4% vs. 11.8%). The proportions selling wet/dry vaporizers (26.4% vs. 39.5%), CBD products (23.3% vs. 71.4%), and pipes/glassware/papers (18.4% vs. 52.9%) increased. In 2021, many sold THC (12.6% e-liquids, 62.2% other products) and kratom (40.3%).

Conclusions

With increasing restrictions (eg, on flavors, sampling, and T21), fewer shops sold their own e-liquid brands or accommodated onsite use/sampling, but fewer also posted minimum-age signage. Notably, more offered cannabis-related products. These changes underscore the need for comprehensive surveillance to assess regulatory impact.

Implications

The past 6 years marked increasing e-cigarette sales restrictions in the United States, yet limited research has examined the implications for tobacco specialty shops selling e-cigarettes. This study found that, from 2018 to 2021, there were significant decreases in the proportion of vape shops with their own e-liquid, onsite vaping, e-liquid sampling, lounge/counter seating, and price promotions, as well as minimum-age signs. There were increases in the proportion selling cannabis-derived products and related paraphernalia. Tobacco control research and regulatory agencies must consider how tobacco specialty stores have evolved alongside legislative changes that impact them and consumers.

Introduction

Marketing in the retail environment is a key driver of e-cigarette use.^1–3 Marketing (eg, products, price, and promotion) aims to influence how and why consumers vape (eg, perceptions of safety or use for cessation of combustibles), attract new users, and promote continued use.^1–4 Moreover, retail settings are prominent sources of tobacco and e-cigarette advertising exposure and influences on uptake among youth^2,3,5,6 and adults.^7–10 Thus, data regarding e-cigarette retail and marketing strategies are critical for understanding industry target markets and strategies for shaping product perceptions and use.

E-cigarette products are sold via various channels, including online, traditional retailers (eg, convenience, food, liquor, and drug stores), and tobacco specialty shops, particularly “vape shops” (sell e-cigarettes but not other tobacco products¹¹) and “smoke shops” (sell e-cigarettes and other tobacco products and accessories [eg, hookah, cigars, and pipes]). After cigarettes, e-cigarettes are the most widely advertised product in tobacco retailers.¹² In 2020, brick-and-mortar retailers (including vape and smoke shops) accounted for >80% of the e-cigarette market share globally.¹³ Notably, tobacco specialty shops represent roughly a third of all tobacco retailers in the United States,¹⁴ and the most common retail source for youth¹⁵ and young adults¹⁶ who purchased e-cigarettes in the past year. In particular, vape shops have played a critical role in e-cigarette uptake, enabling users to sample products before purchasing,¹³ which is especially relevant given the popularity of modular devices (facilitate customizing nicotine levels, flavors, etc.)¹³ and users making their own e-liquid by combining flavors, nicotine.¹³

The past 6 years have marked increased restrictions on e-cigarette sales and distribution, including the federal ban on flavored cartridge-based e-cigarettes (January 2020)¹⁷ and several states and local jurisdictions implementing flavored e-cigarette restrictions,^18,19 as well as various other tobacco-related restrictions (eg, federal increase in the minimum legal sales age to December 21, 2019²⁰). Perhaps relatedly, many vape shops transitioned to smoke shops that sell other tobacco, cannabis-derived products (ie, cannabidiol [CBD], tetrahydrocannabinol [THC]),^21,22 and kratom (marketed to treat pain, mental health, opioid withdrawal, etc.^23,24). Indeed, hemp-derived CBD can be legally purchased in most states,²⁵ and CBD popularity and product diversity have increased rapidly.²⁶ Furthermore, as of November 2022, 21 states and the District of Columbia legalized non-medical cannabis,²⁷ noteworthy given increases in vaping cannabis²⁸ and vaping-related lung injury particularly connected to THC.²⁹ These policy changes are anticipated to alter the tobacco retail landscape, particularly for e-cigarette retailers, as many vape and smoke shops sell CBD and/or THC products (eg, synthetic THC products like delta-8 THC),³⁰ show interest in CBD/THC product sales,²² and name stores with cannabis-related terms.²¹

Longitudinal surveillance of e-cigarette retailers, particularly specialty stores, is needed to understand changes in e-cigarette marketing and consumer exposure, particularly as the regulatory landscape is evolving for tobacco and cannabis. The current study compares data from retail audits conducted in 2018⁹ and 2021 across the same six U.S. metropolitan statistical areas (MSAs; a geographical regions with a relatively high core population density and close economic ties throughout the area). Specifically, we examined differences in compliance with federal regulations, product offerings, prices, and promotional strategies.

Methods

Study Settings

The current analyses report data from retail audits conducted in summer of 2018 and fall of 2021 for the Vape shop Advertising, Place characteristics, and Effects Surveillance (VAPES) study. VAPES examines the vape shop retail environment in six MSAs: Atlanta-Sandy Springs-Roswell (Georgia); Boston-Cambridge-Newton (Massachusetts); Minneapolis-St. Paul-Bloomington (Minnesota); Oklahoma City (Oklahoma); San Diego-Carlsbad (California); and Seattle-Tacoma-Bellevue (Washington). These MSAs were selected for representation across U.S. regions and their variability in state tobacco, e-cigarette, and cannabis legislation (see Supplementary Table 1).¹⁸ In summer of 2018, none of the six MSAs had local sales restrictions on vaping products or flavored tobacco, and California and Washington had non-medical cannabis retail. By 2021, Massachusetts and several jurisdictions in California and Minnesota had implemented sales restrictions on flavored e-cigarettes, and Massachusetts had non-medical cannabis retail.

Sample

The sampling frame for vape shops assessed in 2018 is detailed elsewhere.⁹ In brief, we searched “vaporizer store” on Google and “vape shops” on Yelp to identify stores tagged by retailers or customers as vape shops, restricted lists to stores with complete addresses, eliminated duplicate entries, geocoded records to latitude/longitude using ArcGIS v10.1, used a telephone protocol to confirm that stores fit our definition of “vape shop” (ie, sold vape products but not other tobacco products), and then restricted the sampling frame to vape shops within 25 miles of each MSA centroid. A similar process was used in 2021 prior to conducting audits.²¹

Of the randomly selected vape shops in 2018 (n = 180, 30/MSA), audits were completed in 163; in cases where the originally selected shops were not audited, randomly selected shops in each MSA were assessed (n = 16), for a total of 179 audits (additional details elsewhere⁹). In 2020, we re-contacted the vape shops to determine if they were still open and met eligibility criteria (ie, sold e-cigarette products but not other tobacco products). In 2021, 59.2% (n = 106 of 179 stores audited in 2018) were ineligible: The majority were closed (n = 75) and the remainder no longer sold e-cigarettes (n = 5) or could not be confirmed (n = 26; Supplementary Table 2). Of the 73 shops attempted, 19 asked data collectors to leave (excluded from analyses). Of the 54 completed stores, 43 fit our definition of a vape shop; 11 sold other tobacco products (excluded from analyses). Using the updated 2021 retailer list, we randomly selected replacement stores to reach a target sample of ~20/MSA (replacements ranged from eight in Oklahoma City to 17 in Boston). In 2021, data collectors completed 119 audits. The samples were roughly evenly divided across MSAs in both years (Table 1).

Table 1.

Contextual Factors Among Vape Shops Assessed in Summer 2018 (N = 179) and Fall 2021 (N = 119)

	2018		2021
	N = 179		N = 119
Variable	N (%)	Kappa	N (%)	Kappa	p
MSA*
Atlanta	29 (16.2)	—	19 (16.0)	—	—
Boston	29 (16.2)	—	20 (16.8)	—	—
Minneapolis	30 (16.8)	—	21 (17.6)	—	—
Oklahoma City	31 (17.3)	—	20 (16.8)	—	—
San Diego	30 (16.8)	—	20 (16.8)	—	—
Seattle	30 (16.8)	—	19 (16.0)	—	—
Any minimum-age signage	171 (95.5)	0.83	107 (89.9)	0.67	.058
Minimum age to enter signage	109 (60.9)	0.68	58 (48.7)	0.72	.038
Minimum age to purchase signage	162 (90.5)	0.83	88 (73.9)	0.63	<.001
Vaping/smoking signage and behaviors
No vaping signs	18 (10.1)	0.81	7 (5.9)	0.79	.218
No smoking signs	39 (21.8)	0.79	8 (6.7)	0.85	.001
Vaping in shop	131 (73.2)	0.77	55 (46.2)	0.77	<.001
Smoking in shop	3 (1.7)	1.00	3 (2.5)	0.79	.604
Consumer setting
Customer lounge seating	159 (88.8)	0.96	63 (52.9)	0.74	<.001
Counter/tasting bar seating	117 (65.2)	1.00	41 (34.5)	0.72	<.001
Access to e-liquids
Offer free e-liquid samples	29 (16.3)	0.98	24 (20.2)	0.80	.327
E-liquids sampling (in ways below)	135 (90.0)	0.96	40 (33.6)	0.93	<.001
Use shop’s device and e-liquid w/o nicotine	69 (46.0)	0.95	4 (3.4)	0.39	<.001
Use shop’s device and e-liquid w/ nicotine	24 (16.0)	1.00	9 (7.6)	0.78	.115
Use own device and e-liquid w/o nicotine	30 (20.0)	0.93	1 (0.8)	1.00	<.001
Use own device and e-liquid w/ nicotine	6 (4.0)	0.91	5 (4.2)	0.32	.703
Smell	82 (54.7)	0.99	37 (31.1)	0.77	.011
Taste	36 (24.0)	1.00	25 (21.0)	0.73	.851
Parameters (eg, membership, cost/trial)	5 (3.3)	0.89	1 (0.8)	1.00	.240

^*Between May–July 2018, data collection were conducted in Atlanta, Oklahoma City, San Diego, Seattle, Boston, and Minneapolis-St. Paul, respectively. Between October and December 2021, data collection were conducted in Minneapolis-St. Paul, Oklahoma City, Seattle, San Diego, Boston, and Atlanta, respectively. n/a = not assessed.

^§Significance level set at alpha = 0.01. intra-class correlations (ICCs) among metropolitan statistical areas were <0.20, except for: (0.20–0.50) no vaping or smoking signs and lounge/bar seating; and (>0.50) for sampling with nicotine liquids.

Data Collection

Training and Quality Control

Methods for baseline audits are reported elsewhere⁹ and were replicated for data collection in fall 2021. Pairs of trained data collectors recorded data via assessment forms either: (1) by hand on printed forms, or (2) electronically on iPads using forms programmed in Alchemer.com. All questions offered checkboxes for response options and open fields for notes.

Data collectors were pairs of MPH students in summer of 2018 and pairs of professional data collectors in each MSA recruited by Headway Workforce Solutions in fall of 2021. All data collectors participated in a 2-day training (in-person in 2018, virtual in 2021) that culminated with field practice at retail locations where data collectors lived (in Atlanta in 2018, in various MSAs in 2021). The data collector pairs practiced gathering data at the initial store to examine initial reliability, and then each pair met with the training team to debrief, compare answers, and discuss any discrepancies. Once data collection began in each MSA, the data collectors checked in at least weekly with the first and second authors (CB, KR) to address data collection issues, ensure data quality, and provide feedback on emergent themes potentially integrated into surveillance tools.

Assessment Form

The form⁹ was adapted from the standardized tobacco assessment for retail setting adaptation for vape shops.³¹ This form demonstrated high inter-rater reliability in 2018 and took 15–45 minutes (average of ~30 minutes) to complete.⁹

Age Verification and Onsite Use.

We assessed the presence of signage indicating minimum-age requirements for entering and purchasing, respectively. We also assessed no vaping signs and no smoking signs and actual onsite use, respectively, interior settings that facilitate onsite use (ie, lounge seating, tasting bars), and e-liquid sampling (free or otherwise) and conditions under which sampling could occur.

Product Availability.

We assessed availability of devices and types of e-liquids (shop’s own brand, other vendors’ brands, e-liquids containing nicotine salt, ranges of nicotine, and nicotine salt). Device classifications were revised between 2018 and 2021 to reflect new product introductions. In 2018, we coded devices as closed systems (eg, disposable) and open systems (eg, devices with refillable tanks); in 2021, we coded devices as disposables, cartridge-based, pod mods, and box mods. At both timepoints, we also coded the availability of herbal/dry chamber vaporizers and wet/dry vaporizers. In 2018, we assessed the availability of other CBD products (e-liquids and other products); in 2021, we also assessed the availability of THC products (e-liquids and other products) and kratom. We also recorded the availability of pipes, glassware, or wrapping papers.

Price.

In 2018, we assessed lowest and highest prices for closed and open-system devices, with and without starter kits. In 2021, we assessed lowest and highest prices for disposables, cartridge-based products, pod mods, and box mods. In both years, we assessed lowest and highest prices for e-liquids with and without nicotine salt.

Promotions.

We recorded the presence of price specials on devices or e-liquids (eg, buy one, get one; two for one); happy hour/early bird specials; daily/weekly/monthly specials; e-liquid bargain bins; drawings/raffles for discounts/coupons; discounts for military/veterans; and discounts for college students. We also recorded advertising signage, defined as 8.5 × 11 inches or larger, branded with the intent to sell product, and professionally produced and/or amateur/hand-written.⁹ We assessed presence of health warnings posted on signage (eg, contraindications, cautions) and messaging related to health claims (eg, safer than other tobacco, safe in general, health benefits, and cessation aid). We also recorded the presence of signage promoting products or price promotions and signage using cartoon imagery (eg, use of comically exaggerated features, attribution of human characteristics to animals, etc.⁹). In addition, we assessed shop/product-branded non-tobacco products available for purchase, such as apparel or paraphernalia, membership/loyalty/rewards programs, promotion via social media, and delivery options.

Data Analysis

To determine inter-rater reliability, we computed kappas for categorical variables estimated in PROC FREQ and intra-class correlations for continuous variables in PROC MIXED. Using SPSS v26.0, we conducted descriptive analyses to characterize vape shop practices at the two timepoints. Bivariate analyses identified differences in regulatory compliance (eg, age verification) and marketing characteristics between 2018 and 2021, using chi-square tests for categorical outcomes, and independent t-tests for continuous outcomes, even though 43 of 139 stores (30.9%) were observed in both years. When cell sizes were small (ie, ≤5), Fisher’s exact test was used. We also examined intra-class correlations (percent of variation attributable to MSA) for each 2021 outcome. Alpha = 0.01 was considered as the cutoff for statistical significance due to multiple comparisons.

Results

Inter-rater Reliability

Inter-rater reliability for the majority of assessments at both timepoints were good to excellent (kappas ranging 0.79–1.00; Tables 1–3).³² Items with kappas below 0.60 may have resulted from low-prevalence occurrences, discrepant records for categorical variables, and/or interpretation, for example, regarding signage with health warnings, cartoons, or product/price promotions (eg, disagreement as to whether prices posted on signage constituted special promotional prices).

Table 3.

Signage and Promotions at Vape Shops Assessed in Summer 2018 (N = 179) and Fall 2021 (N = 119)

	2018		2021
	N = 179		N = 119
Variables	N (%)	Kappa	N (%)	Kappa	p
Any e-cigarette price promotions	149 (83.2)	0.69	24 (20.2)	0.53	<.001
Any signage w/ e-cigarette product/price promotions	161 (89.9)	0.62	78 (65.5)	0.72	<.001
Devices	84 (46.9)	0.70	48 (40.3)	0.64	.262
Device price promos	36 (20.1)	0.39	16 (13.4)	0.49	.138
E-liquids	151 (84.4)	0.64	65 (54.6)	0.70	<.001
E-liquid price promos	82 (45.8)	0.52	16 (13.4)	0.60	<.001
Signage promoting CBD, THC, Kratom, etc.*
CBD products	31 (26.1)	0.86	36 (30.3)	0.64	.009
CBD price promos	n/a	—	5 (4.2)	0.74	—
THC products	n/a	—	5 (4.2)	0.74	—
THC price promos	n/a	—	0 (0)	n/a	—
Kratom products	n/a	—	6 (5.0)	0.31	—
Kratom price promos	n/a	—	0 (0)	n/a	—
Signage with health warnings*
E-cigarette health warnings	34 (19.0)	0.42	32 (26.9)	0.75	.107
CBD health warnings	n/a	—	2 (1.7)	1.00	—
THC health warnings	n/a	—	3 (2.5)	0.66	—
Kratom health warnings	n/a	—	0 (0)	n/a	—
Signage with health claims
Any signage with e-cigarette health or cessation claims	52 (29.1)	0.69	15 (12.6)	0.63	.001
Any signage with CBD health claims	n/a	—	10 (8.4)	0.64	—
Any signage with THC health claims	n/a	—	4 (3.4)	0.49	—
Any signage with kratom health claims	n/a	—	0 (0)	n/a	—
Signage with cartoon imagery	49 (27.4)	0.46	14 (11.8)	0.84	.001
Shop/product apparel or paraphernalia	104 (58.1)	0.61	77 (64.7)	0.77	.253
Membership, loyalty, or rewards programs	74 (41.3)		25 (21.0)	0.75	<.001
Advertising on social media	47 (26.3)	0.70	39 (32.8)	0.65	.224
Delivery options (eg, free delivery) *	5 (2.8)	0.74	n/a	—	—
Home delivery	n/a	—	0 (0)	n/a	—
Curbside pickup	n/a	—	14 (11.8)	0.91	—

^*Different assessments in 2018 versus 2021. n/a = not assessed.

^§Significance level set at alpha = 0.01. intra-class correlations (ICCs) among metropolitan statistical areas were < .20, except for: (0.20–0.50) signage for device and e-liquid price promotions, CBD and THC products; and (>0.50) Kratom products/promotions, CBD health warnings and claims, membership/loyalty/rewards programs, and curbside pickup.

Age Verification and Onsite Use-Related Characteristics

There were decreases in proportions of vape shops with minimum age to purchase signage (90.5% vs. 73.9%, p < .001) and “no smoking” signs (21.8% in 2018 to 6.7% in 2021, p = .001). The proportion with “no vaping” signs did not decline significantly (10.1% vs. 5.9%, p = .218). There were decreases in proportions with vaping inside the shop (73.2% vs. 46.2%, p < .001), lounge seating (88.8% vs. 52.9%, p < .001), and counter/bar seating (65.2% vs. 34.5%, p < .001). Some shops violated FDA regulations by offering free e-liquid samples, but the proportion did not increase significantly (16.3% vs. 20.2%, p = .327). However, the proportion that offered ways to sample e-liquids (eg, loyalty programs, low-cost sampling) dramatically decreased (90.0% vs. 33.6%, p < .001).

Product Availability

The majority of shops carried closed-system (64.8% in 2018), disposable (88.2% in 2021), and cartridge-based devices (52.9% in 2021), and almost all carried open-system devices (100% in 2018), specifically pod mods and box mods (93.3%, respectively, in 2021; Table 2). From 2018 to 2021, the proportion carrying combined wet/dry vaporizers increased (26.4% vs. 39.5%, p = .004).

Table 2.

Product Availability Among Vape Shops Assessed in Summer 2018 (N = 179) and Fall 2021 (N = 119)

	2018		2021
	N = 179		N = 119
Variables	N (%)	Kappa	N (%)	Kappa	p
Device products *
Closed systems (2018)	116 (64.8)	0.96	n/a	n/a	<.001
Disposables (2021)	n/a	n/a	105 (88.2)	0.93	—
Cartridge-based (2021)	n/a	n/a	63 (52.9)	0.96	—
Open systems (2018)	179 (100.0)	n/a	n/a	—	—
Pod mods (2021)	n/a	n/a	111 (93.3)	0.92	—
Box mods (2021)	n/a	n/a	111 (93.3)	0.92	—
Herbal/dry chamber vaporizers	56 (34.4)	0.97	51 (42.9)	0.90	.041
Combined wet/dry vaporizers	43 (26.4)	0.93	47 (39.5)	0.94	.004
E-liquid products
Sells vape shop’s own brand	122 (68.2)	0.99	53 (44.5)	0.86	<.001
Sells other vendors’ brands	160 (89.4)	1.00	105 (88.2)	0.77	.757
Sells e-liquids with nicotine salt	145 (81.0)	0.96	81 (68.1)	0.71	.011
CBD, THC, and Kratom
Sells e-liquids containing CBD	76 (42.5)	1.00	37 (31.1)	0.72	.048
Other CBD products	35 (23.3)	0.92	85 (71.4)	0.86	<.001
Sells e-liquids containing THC	n/a	n/a	15 (12.6)	0.63	—
Other THC products	n/a	n/a	74 (62.2)	0.88	—
Kratom	n/a	n/a	48 (40.3)	0.89	—
Pipes/glassware/papers	33 (18.4)	0.94	63 (52.9)	0.82	<.001

^*Different assessments of e-cigarette devices in 2018 versus 2021. n/a = not assessed.

^§Significance level set at alpha = 0.01. Intra-class correlations (ICCs) among metropolitan statistical areas were < .20, except for: (0.20–0.50) cartridge-based devices, own brand of e-liquids or e-liquids with nicotine salts, CBD products other than e-liquids, and THC products.

From 2018 to 2021, the proportion of shops carrying their own brand of e-liquids decreased (68.2% vs. 44.5%, p < .001). At both timepoints, the majority sold other vendors’ brands of e-liquids (≥88.2%, respectively). On average, lowest and highest levels of nicotine (without salt) in e-liquids were 0.02 mg (SD = 0.22) and 20.03 mg (SD = 8.92) in 2018, compared to 1.16 mg (SD = 2.99) and 18.16 mg (SD = 14.66) in 2021, representing a significant increase in lowest nicotine levels (without salt; p < .001). On average, the lowest and highest nicotine salt levels were 20.45 mg (SD = 13.55) and 39.28 mg (SD = 20.46) in 2018, compared to 16.51 mg (SD = 10.70) and 48.15 mg (SD = 12.86) in 2021, representing significant increases in the highest nicotine salt level (p < .001) and decreases in the lowest (p = .006) over time.

From 2018 to 2021, there were increases in the proportion selling CBD products (other than e-liquids; 23.3% vs. 71.4%, p < .001) and selling pipes/glassware/rolling papers (18.4% vs. 52.9%, p < .001). In 2021, 12.6% of shops sold e-liquids containing THC, 62.2% other THC products, and 40.3% kratom.

Pricing

In 2018, average least and most expensive prices for each product category were: closed-system devices, $37.12 (SD = 15.26) to $48.36 (SD = 13.75); open-system starter kits, $29.46 (SD = 12.09) to $101.98 (SD = 43.22); open-system devices (no starter kit), $54.38 (SD = 18.95) to $130.17 (SD = 39.48); and e-liquids, $0.32/mL to $0.44/mL. In 2021 (using the revised device categories to reflect the current e-cigarette market), price ranges were: disposables, $12.15 (SD = 5.61) to $26.62 (SD = 5.93); cartridge-based, $24.67 (SD = 12.93) to $45.32 (SD = 24.91); pod mods, $31.81 (SD = 10.24) to $65.22 (SD = 37.57); box mods, $58.82 (SD = 18.91) to $122.72 (SD = 53.42); and e-liquids, $0.38/mL to $0.45/mL.

Promotional Strategies

There were dramatic reductions from 2018 to 2021 in the presence of e-cigarette price promotions (83.2% vs. 20.2%, p < .001; Table 3). Significant (p < .01) reductions occurred in the proportion of shops with price specials on e-liquids (eg, buy one, get one; 55.9% vs. 9.2%) and daily, weekly, or monthly specials on e-liquids (24.0% vs. 9.2%), but no significant changes regarding price specials on devices (26.3% vs. 19.3%), e-liquid bargain bins (25.7% vs. 13.4%), happy hour or early bird specials on e-liquids (13.4% vs. 6.7%), and discounts for military/veterans (13.4% vs. 17.6%).

From 2018 to 2021, we observed reductions in the proportion of shops with signage promoting products or product price promotions (89.9% vs. 65.5%, p < .001), particularly for e-liquids (84.4% vs. 54.6%, p < .001) and e-liquid price promotions (45.8% vs. 13.4%, p < .001). Over time, the proportion with signage promoting CBD increased significantly (26.1% vs. 30.3%, p = .009), with ~5% promoting THC or kratom in 2021. From 2018 to 2021, there were decreases in the proportion of shops with signs including cartoon imagery (27.4% vs. 11.8%, p = .001) and with signs with e-cigarette health or cessation claims (29.1% vs. 12.6%, p = .001). However, nearly 10% of 2021 shops had signs promoting health claims for CBD (8.4%) and/or THC (3.4%).

At both timepoints, a majority of shops (58.1% vs. 64.7%, p = .253) carried apparel or paraphernalia promoting the shop or its products, and a smaller proportion advertised via social media (26.3% in 2018 vs. 32.8% in 2021, p = .224). Over time, the proportion with membership, loyalty, or rewards programs decreased significantly (41.3% vs. 21.0%, p < .001). In 2021, 11.8% of shops offered curbside pickup.

Discussion

The current study adds to a growing literature about the practices of tobacco specialty stores in general and vape shops in particular.^21,33,34 Key findings comparing vape shop practices in 2021 and 2018 include less (or less noticeable) age verification signage, fewer vape shops with onsite vaping, bar or lounge seating, e-liquid sampling, changes in e-cigarette devices, and e-liquids available (including more closed-system/disposable devices and less shop-branded e-liquids), decreases in several price promotional strategies and signs with product or price promotions, and decreases in signs with prohibited content (eg, harm/cessation claims, cartoon imagery). Alongside these changes, we also documented that vape shops offered a greater diversity of products, particularly cannabis-derived products.

Findings regarding decreases in the proportion of vape shops with minimum-age signs are difficult to interpret. Prior research indicates high rates of noncompliance with age verification policies, underestimates of noncompliance, gaps in enforcement,^35,36 and particularly high noncompliance rates for e-cigarette sales,³⁷ especially in tobacco specialty shops,^9,10,38 but this does not explain the decrease in signage (or noticeability of signage) over time. This may be due to the lack of retail surveillance/enforcement during the coronavirus disease 2019 (COVID-19) pandemic.

It is also notable that there were decreases in onsite vaping, lounge seating, counter seating, and e-liquid sampling. Historically, vape shops appealed to consumers by allowing sampling of products before purchasing and encouraging socialization through bar/lounge seating. Despite greater restrictions on e-liquid sampling (eg, FDA prohibiting free samples, COVID-19 transmission concerns), 16%–20% of vape shops offered free e-liquid samples in 2018 and 2021, as found previously,³³ underscoring the ongoing need to monitor enforcement.

The proportion of vape shops that sold their own e-liquid brand decreased. Historically, a substantial proportion of vape shop revenues came from e-liquids.³⁹ However, 2018 FDA regulations required greater oversight of e-liquid manufacturers and market approval to sell customized e-liquids, which has financial implications.¹¹ This change occurred alongside state and local sales restrictions on flavored e-cigarette products,^17–19 which have been shown to reduce flavored tobacco availability and tobacco advertising.⁴⁰ Collectively, these changes may have impacted vape shops’ decisions to sell their own e-liquid brands, particularly since many vape shops are small businesses or single-store owners.³⁹ Perhaps relatedly, decreases in the proportion of vape shops that offered price promotions and posted signs with product/price promotions may have been related to the loss of revenues overall, but particularly e-liquid revenues. On the other hand, the proportion selling closed-system/disposable devices increased, perhaps as a result of the loophole in FDA restrictions that allowed flavored disposable vaping products to continue being sold.

Another important finding is decreases observed in the proportion of vape shops posting claims of e-cigarettes’ health or cessation benefits or cartoon imagery. FDA regulation that went into effect with the Deeming Rule included prohibition of such health and cessation claims, and in 2018, FDA issued warnings to manufacturers, distributors, and retailers for selling e-liquids with cartoon-like imagery on advertising/labeling in 2018.⁴¹ However, such health/cessation claims may be communicated by shop personnel, which happened when mystery shoppers interacted with vape shop employees in 100% of the shops we studied in 2018,⁹ thus warranting continued surveillance. Additionally, youth and young adults can be targeted using various strategies other than cartoon imagery (eg, with flavors,¹⁷ social media marketing,⁹ retail proximity to youth-oriented places⁹). These critical regulatory issues should be monitored to protect consumers and prevent youth e-cigarette use.

Although this study focused on brick-and-mortar retailers, online e-cigarette retail is expected to be the fastest-growing segment in the future.^13,42 Half of U.S. vape shops adapted to the COVID-19 pandemic by doing business online,⁴³ which has been shown to be a high priority across vape shops.⁴⁴ Delivery options are more commonly offered, perhaps due to pandemic-related restrictions.⁴⁴ It is critical to examine how this evolves given the 2020 extension of the Prevent All Cigarette Trafficking Act (which prohibits USPS delivery of cigarettes and smokeless tobacco to consumers) to include e-cigarettes,⁴⁵ especially given high rates of noncompliance for cigarettes.⁴⁶

A key finding from this study includes potential vape shop diversification of product offerings beyond e-cigarettes. Specifically, increases were found in the proportion of vape shops selling wet/dry vaporizers, CBD products, and pipes/glassware/papers. There was a concomitant increase in posting signs promoting CBD. In 2021, a large proportion of shops sold THC products and kratom. Perhaps in response to e-cigarette sales restrictions, many vape shops have or are transitioning to smoke shops that sell other tobacco and cannabis-derived products.^21,22 Smoke shops are of particular concern to public health due to higher rates of noncompliance, underage access, and availability of a broader array of tobacco products and other un-/under-regulated products, such as delta-8-THC, delta-10-THC, delta-O THC, HHC, and kratom.³⁰ The hemp-derived products were not specifically referenced in the Agriculture Improvement Act of 2018,²⁵ leaving their legality at the federal level ambiguous, resulting in over 20 states now restricting or prohibiting these products altogether.⁴⁷ Top among the concerns regarding these products are their contents⁴⁸ and safety.⁴⁹ Thus, tobacco control researchers and practitioners must be aware of other substances of concern in the tobacco retail environment as well as the dynamic interplay between regulations, retail, and consumers in order to be relevant in addressing how retailers and consumers navigate new policy contexts and whether retailers comply with regulations. Moreover, researchers and practitioners need to be cognizant of other policy areas within and outside of a single jurisdiction, as tobacco-, marijuana-, and alcohol-related policies likely interplay, and policies from one state or jurisdiction have implications for others.

Note that only 43 vape shops in this study had data at both time points. To provide broader context, between 2019 and 2020, about 80% of vape shops reported a decline in revenue (avg. 18% decline),⁴³ largely attributed to COVID-19; however, an estimated 7% was because of state/local advances in tobacco control, particularly ENDS product restrictions.⁴³ Furthermore, our prior research found that, within 16 months (from 2018 to 2019), 11.5% of tobacco specialty stores in these six MSAs closed, and while no neighborhood factors (eg, racial/ethnic mix, age groups) predicted closings, stores selling more diverse products (outside of e-cigarettes) were more likely to remain open,²¹ which coincides with current findings in these six MSAs.

Areas for Future Research

The surveillance methods used in this study allow further examination of tobacco specialty retailers over time and across contexts, particularly critical given the range of regulatory environments across U.S. localities and states. Although this study did not examine retail websites, future research should extend surveillance of brick-and-mortar to online retailers to assess compliance with relevant restrictions on sales and marketing, including industry advertising with safety, and cessation claims or cartoon imagery, in order to inform regulatory efforts to minimize youth access and exposure more broadly. Additionally, strategies used to communicate health/cessation claims or to entice young people are critical to examine—both in paid or posted ads and marketing and via other channels (eg, social media, interpersonal communication with retail personnel). Further assessment is also needed regarding cannabis-derived products, kratom, and other un-/under-regulated products in tobacco specialty stores to inform federal, state, and local regulations and their enforcement.

Limitations

Findings have limited generalizability and should be considered within the relevant policy and social contexts (eg, COVID-19). Also, note that only 30% of stores had data at 2 timepoints. Furthermore, despite rigorous data collection training and quality control, some data (eg, measures with low inter-rater reliability) may have been impacted by differences in how research staff approached assessment or how to shop personnel interacted with research staff. Also, to keep pace with a rapidly changing retail landscape, some measures (eg, device types) were revised from 2018 to 2021, which precluded some comparisons over time. Nonetheless, it is critical that surveillance efforts adapt to the evolving market for e-cigarettes and other products to maintain relevance. Finally, high intra-class correlations for some variables underscore the need for comparisons of these factors across MSAs, which is the focus of additional analyses.

Conclusion

Alongside increasing e-cigarette restrictions, consumer vape shop retail experiences have changed. For example, vape shops display fewer promotions, particularly those related to price, and the social context has changed due to restrictions on e-liquid sampling and flavored e-cigarette products. Additionally, while the presence of age verification signs decreased, signs with health/cessation claims, and cartoon imagery also decreased. Notably, current findings contribute to the growing literature underscoring the need for surveillance of the broader range of tobacco specialty stores, particularly smoke shops, which sell e-cigarettes alongside other tobacco products, and un-/under-regulated products (eg, cannabis-derived, kratom), which may influence consumer behavior in ways that policymakers did not consider and need to anticipate.

Supplementary Material

A Contributorship Form detailing each author’s specific involvement with this content, as well as any supplementary data, are available online at https://academic.oup.com/ntr.

Funding

This work was supported by the US National Cancer Institute (R01CA215155-01A1, PI: Berg). Dr. Berg is also supported by other funding from the US National Institutes of Health, including the National Cancer Institute (R01CA239178-01A1, MPIs: Berg, Levine; R01CA179422-01, PI: Berg; R21CA261884-01A1, MPIs: Berg, Arem; R01CA278229-01, MPIs: Berg, Kegler), the Fogarty International Center (R01TW010664-01, MPIs: Berg, Kegler; D43TW012456-01, MPIs: Berg, Paichadze, Petrosyan), the National Institute of Environmental Health Sciences/Fogarty (D43ES030927-01; MPIs: Berg, Marsit, Sturua), and the National Institute on Drug Abuse (R01DA054751-01A; MPIs: Berg, Cavazos-Rehg). Dr. Romm is supported by the National Institute on Drug Abuse (F32DA055388-01; PI: Romm).

Declaration of Interests

The authors declare no conflicts of interests.

Data Availability

Data not publicly available (available upon request).