Age Verification and Online Sales of Little Cigars and Cigarillos to Minors

Abstract

Objectives:

In our cross-sectional study, we aimed to determine age verification and sales of little cigars and cigarillos (LCCs) online to underage teens.

Methods:

We selected 100 popular Internet Little Cigar and Cigarillo Vendors (ILVs) for order attempts. From August to December 2015, we supervised 14 teens 14–17 years old making order attempts for LCCs.

Results:

Of the 91 valid orders attempted, we received 89. For the valid orders attempted, 9.9% of ILVs used no age verification strategies at all, 84.6% used less effective forms of age verification, and 50.5% used more effective ones. Only one order was blocked during the order attempt and only one attempt was made to verify age at delivery. Most (79.8%) deliveries were left at the door and only 2 order attempts were rejected because of age verification strategies, resulting in a successful or valid buy rate of 97.8%.

Conclusions:

Our study demonstrated that ILVs selling LCCs were not making adequate efforts to verify the age of their customers, at the point-of-sale or point-of-delivery, facilitating easy access by minors. Few ILVs utilized age verification strategies that prevented online sales of LCCs to underage teens.

Estimated revenue from cigar sales in the United States (US) is expected to top $9.1 billion in 2019, up from $7.5 billion in 2010, with continuing growth anticipated.¹ Cigar sales increased by 29% during 2012–2016, largely a result of increasing sales of little cigars and cigarillos (LCCs)² in response to cigarette excise tax increases³ and the ban on the sale of flavored cigarettes via the 2009 Family Smoking Prevention Tobacco Control Act (FSPTCA).⁴ In the wake of these regulatory efforts, little cigars and cigarillos have been marketed as a cheaper alternative to combustible cigarettes, making flavored cigarette-like products more readily available to US consumers than banned, flavored cigarettes.⁵

In 2018, the National Youth Tobacco Survey (NYTS) found that 9.2% of all middle and high school students (equating to 1.3 million students) were cigar smokers.⁶ Cigar use has been shown to be higher among youth who also use other tobacco products or other drugs (eg, alcohol, marijuana, and inhalants) compared to youth who do not.⁷

Cigars, including LCCs, often have been falsely viewed as less harmful than cigarettes,⁸ despite their well-documented adverse health risks, including cancer.⁹ More specifically, adolescents perceive cigars as having lower health risks than cigarettes.^10–14 These erroneous views and the availability of regular and flavored LCCs at lower cost than cigarettes have induced many smokers to switch.¹⁵ Youth who smoke cigars report doing so for social and financial reasons; they think it makes them appear more mature,¹⁴ and they associate smoking LCCs with a relaxing time. They further note that LCCs can be purchased individually or in packs of 2, resulting in a lower, more accessible cost.¹⁶

Currently no federal legislation governs characterizing flavors in LCCs, although the FDA recently announced their intent to ban flavors in LCCs.¹⁷ Youth, in general, have been shown to be more inclined toward flavored tobacco products than adults.⁷ An ongoing concern is the availability of flavored cigar products which are very appealing to youth^18–20 especially in light of the FSPTCA ban on characterizing flavors in cigarettes.²¹ The first tobacco product used by 4 out of 5 12-to-17-year-olds is flavored.²⁰ Furthermore, youth who first tried a flavored tobacco product have been more likely to become current tobacco users than those whose first product was not flavored.¹⁸ In 2014, 73.8% of 12-to-17-year-olds who smoked cigars reported that flavoring is a primary reason for using a tobacco product,²⁰ and 64% of current youth tobacco users reported using a flavored cigar in the past 30 days.¹⁹ In addition, youth reported that a flavored tobacco product was easier to use than a non-flavored one and those who reported that preference were significantly more likely to be current users of LCCs.^18,22 Furthermore, adolescents and young adults who were more regular cigar smokers preferred flavored cigars.²³ Monitoring the Future investigators noted in 2018 their concern that LCCs “will have the effect of reversing the hard-won gains in reducing cigarette smoking among youth.”²⁴

Various states and the Federal government have recognized the dangers of youth exposure to nicotine and implemented point-of-sale age restrictions, thereby supporting hard-won gains in reducing smoking. The Internet, however, remains a viable option for underage teens to access LCC products because of vendors’ use of less effective age verification strategies.²⁵ Research has shown that teens are more likely to attempt online orders if they perceive restricted access to tobacco products in retail stores.²⁶

Given the public health concerns that arise from youth access to LCCs, we wanted to: (1) assess the extent to which underage teens can purchase LCCs online; and (2) examine age verification strategies used by Internet Little Cigar and Cigarillo Vendors (ILVs). This is the first study to specifically assess online sales of LCCs to minors and age verification strategies of ILVs. This study was conducted before the 2016 FDA Deeming Regulations²⁷ requiring age verification for online tobacco product sales (without specifying how age should be verified for online sales), thus providing a baseline against which to compare the effects of the Deeming Regulations on ILV age verification strategies in future studies.

METHODS

ILV Sample

We identified a sample of popular English-language vendors of LCCs via an exhaustive search of Internet Tobacco Vendors (ITVs) from our larger-scale study. As Figure 1 depicts, we used 3 major sources to identify a total of 31,239 potential ITVs in January 2014. The first source was our highly complex, automated search algorithms developed and updated annually since 2004 to search 180 million websites, message boards, spam emails, and news groups which yielded 15,285 potential ITVs. Our second source was a set of 15,515 Internet Cigarette Vendor websites identified in the previous wave of our ongoing ITV Study, which tracks the sales and marketing practices of English-language ITVs.²⁸ Our third source was 439 websites identified from promotional websites listing multiple ITVs found during screening of the other 2 sources. As Figure 1 shows, we then screened the 31,239 websites manually for eligibility, resulting in the exclusion of 26,798 websites because they were defunct, duplicates, or did not sell tobacco products. From the set of 4441 ITVs, we then excluded those that did not sell LCCs for home delivery or which received so little visitor traffic that their traffic rank could not be calculated, yielding 312 popular ILVs. We further screened for the most popular ILVs, arriving at 204 (based on Alexa.com visitor traffic rankings²⁹). Exclusion criteria led to the removal of 104 sites because they: did not ship to North Carolina (N = 36), no longer sold LCCs (N = 13), were not functioning at checkout (N = 12), did not accept credit cards or e-checks (N = 11), were cost-prohibitive for most teens (products cost more than $200; N = 7), or were auction sites (N = 3), or because of a minor issue with aggregating popularity data (N = 22), resulting in a final vendor sample of 100 popular ILVs for underage order attempts. (The issue with aggregating popularity data resulted in the omission of 22 ILVs with the following popularity ranks: 58^th, 85^th, 87^th-91^st, 95^th, 96^th, 104^th, 106^th, 108^th-116^th, 118^th, and 120^th. As a result of their omission, the least popular site included in our sample for purchases ranked 122^nd. With fewer than 10 sites with popularity ranks below 100 that were omitted and the lowest rank of 122 for sites included in lieu of the 22 excluded sites, we feel that our study’s sample remains representative of popular ILVs.)

Figure 1. Sampling Diagram for Study of Little Cigar and Cigarillo Internet Purchases by Youth.

a: An error in aggregating popularity data resulted in the omission of 22 sites with the following popularity ranks among ILVs: 58th, 85th, 87th-91st, 95th, 96th, 104th, 106th, 108th-116th, 118th, and 120th. As a result of their omission, the least popular site included in our sample for purchases ranked 122nd. With fewer than ten sites with popularity ranks below 100 that were omitted and the lowest rank of 122 for sites included in lieu of the 22 excluded sites, we feel that our study’s sample remains representative of popular ILVs.

Buyers

To carry out order attempts, we recruited 14 English-speaking teens who were 14–17 years old via email flyers posted to university listservs, local youth groups, and in local teen hangouts. These teens attempted to order LCCs from popular English-language ILVs to assess the vendors’ age verification practices.

In accordance with our study procedures approved by the University of North Carolina’s Institutional Review Board, teens and their parents provided written assent and consent, respectively, to participate in the study, including consent for their child to use a parent’s driver’s license and identity to attempt to bypass age verification during order attempts. Multiple steps were taken to protect study participants including:

• Creating monitored email addresses and phone numbers for teen buyers to prevent communication from ILVs with them.

• Preparing each order under direct supervision of an adult staff member, who clicked the final ‘submit’ button, so that, technically, the order was placed by an adult.

• Procuring letters from the local district attorney and police chiefs providing immunity from arrest or prosecution for all study staff and participants because of their involvement in the study.

Measures

Although differing definitions exist,^24,30 we grouped little cigars and cigarillos together for the purposes of our research because ILVs and manufacturers often conflate the terms, labeling cigarette-sized products as cigarillos, or smaller cigar-shaped products as little cigars.

For this study, we assessed the following: age warnings featured, age verification strategies in use by vendors, and details of order attempts (including products ordered, prices paid, and methods of delivery). We calculated the overall buy rate (percentage of order attempts successfully delivered) as well as valid order attempts and a valid buy rate, which excluded orders that failed for reasons unrelated to age verification, such as problems with processing payment.

Age Verification Strategies

We tracked age warnings and all strategies ILVs used for age verification, using discrete variables representing the types of age verification seen in our prior ITV studies since 1999 as well as an “other, please specify” variable, which is used to identify emerging age verification strategies. ILVs employed 2 types of age verification strategies, those that: (1) cannot effectively verify age, and (2) can potentially effectively verify age. These 2 types of strategies are described below.

Strategies that cannot effectively verify age.

The following age verification strategies are categorized as being unable to effectively verify age:

• Posting language indicating that by submitting their order (or even viewing their products) the buyer is certifying that he/she is of legal purchasing age

• Providing a checkbox or a button that a buyer selects to indicate he/she is 18 years old or older

• Requiring the buyer to type in his/her age (in years)

• Stating that the ILV ensures that buyers are adults by only accepting credit card payment (asserting incorrectly that minors cannot obtain credit or debit cards)

• Stating that they verify age without indicating the strategy.

Strategies that can potentially effectively verify age.

Age verification strategies that can potentially effectively verify age if used properly (although not proven to effectively or consistently block youth access either by the companies offering them or by research studies) include:

• Seeking Date Of Birth (DOB), which could be used in an ineffective way by checking if the DOB entered makes the user of legal age, which can be thwarted by lying, or which can be used in conjunction with name and address to verify a customer’s age with a Public Records Database (PRD)

• Requesting a driver’s license image (which might be taken at face value, and could be thwarted by digitally altering the image, or which might be verified with a PRD)

• Using an Online Age Verification (OAV) service, which typically collects customer information or license images to verify the information with a PRD

• Requiring the buyer to input the last 4 digits of his/her social security number (presumably used in conjunction with a PRD to verify age)

• Requiring the buyer to enter his/her driver’s license number (presumably used in conjunction with a PRD to verify age)

• Using challenge questions, which are deployed after an ID has been verified with a PRD as belonging to a legitimate adult, asking multiple-choice questions based on PRD information intended to verify that the submitter of the ID is the owner of the ID, such as “With which company do you have a mortgage?”

• Confirming age at delivery (conducted by the delivery driver checking the recipient’s ID before delivering the package)

It should be noted that some ILVs claimed to use age verification strategies that were not used when an order attempt was made.

Study Procedures

Between August and December 2015, teen buyers were supervised one-on-one by adult staff as they attempted to place orders for flavored LCCs from the selected 100 ILV websites (if flavored products were not available, unflavored products were ordered). During each order attempt, staff tracked details of the order encounter such as product cost, age verification strategies used by the vendor, and shipping information. ILV interactions via phone or email were handled by staff to minimize youth interaction with vendors and tobacco products.

We tested the vendors’ use of DOB for age verification by having the teens initially attempt to place an order using their DOB. If rejected, and to emulate likely real-world behavior of underage teens attempting to buy age-restricted products online, they then attempted the order misrepresenting their age with a fake DOB (or in other cases, selecting checkboxes indicating they were over 18, similar to teen-generated strategies to bypass age verification used in a previous study assessing age verification by tobacco company corporate websites³¹). If the fake DOB was rejected, the teen then used a parent’s identity and DOB.

In addition to misrepresenting age, teens were also permitted to misrepresent their identity. We learned from our previous study³¹ that teens had easy access to, and were willing to use, a parent’s driver’s license to bypass age verification strategies which may have led to underestimated, successful order rates. Consequently, all teen buyers in the current study had a parent who gave written permission for their child to attempt to use their identity and driver’s license to bypass age verification when needed to complete an order (when a driver’s license was requested and a teen’s ID/license rejected). Doing so allowed us to assess whether vendors would return challenge questions and, if so, whether they thwarted underage teens’ access.

For all of the orders, we used credit cards issued to each buyer in his or her own name; if, however, they attempted the order again using a parent’s ID because they had failed to pass age verification with their own information, they used a credit card issued in their parent’s name.

Delivery

Teen buyers were asked to answer the door for deliveries when they were home and to present their own ID, if requested. For all deliveries, teen buyers noted the date, whether the package was discovered/left at the door, with the teen buyer, or with an adult, and any details of age verification efforts. Parents took control of all deliveries upon receipt and staff regularly retrieved packages from their homes.

RESULTS

ILV Demographics

Most orders (75%) were from ILVs in the US; 19% in Florida, 10% in Pennsylvania, 9% in NY, 7% in multiple locations, and the remainder (1%-5%) in 16 other states. Of the remaining ILVs, 10% had indicators of multiple country locations (for example, a US phone number or address and details about products being shipped from overseas) and the others had 1–3 sites in each of 10 different countries. Overall, half of the ILVs reported having retail stores. Overall, 92% also sold large cigars, 30% sold e-cigarette products, and 7% sold cigarettes. Demographic data were missing for one ILV.

Products Ordered

From the study’s 100 selected sites, the teen buyers attempted 55 orders of flavored LCCs and 45 of non-flavored LCCs. The minimum product order was placed at each website; for cigarette-sized LCCs, this was typically a carton (usually 200 sticks), and for larger cigarillo/small cigar-sized LCCs, this was typically a tin of 10 cigarillos or multipacks of tins of up to 60 cigarillos per order. Overall, flavored LCCs could be obtained more cheaply (mean cost: $22.22) than non-flavored LCCs (mean cost: $28.55).

Order Attempts

Of the 100 ILVs in our study sample from which orders were attempted, 9 orders were considered invalid after failing for reasons unrelated to age verification, such as the vendor charging for the order without delivering any products. Of the 91 valid order attempts, 89 (97.8%) were successfully delivered. Fifty of the 51 valid orders for flavored LCC products (98.0%) and 39 of 40 valid orders of non-flavored LCCs, (97.5%) were delivered. As Figure 2 depicts, age verification strategies successfully blocked only 2 orders.

Figure 2.

Outcomes of Online Purchase Attempts of Little Cigars and Cigarillos by Youth

Online Tobacco Sales: Fraud Risk

Some of the 9 unsuccessful, invalid orders were from sites likely collecting credit card information for fraudulent purposes, as happened in a similar study of Internet Cigarette Vendors, where 70 order attempts resulted in 22 orders received and hundreds of instances of credit card fraud totaling over $7000.³² For our current study, we used each credit card for only a few order attempts. Five of the credit cards experienced a total of 26 fraud attempts (3 of which were major fraud attempts over $700). A 6th card was double charged by one ILV and charged for a product never received by another ILV.

Age Verification

Table 1 describes the types of age verification strategies used by ILVs relative to our order attempts, overall buy rate, valid order attempts, and valid buy rate.

Table 1.

Age Verification Strategies Used by Internet Vendors of Little Cigars and Cigarillos (N = 100)

Type of Strategyd	Order Attempts (N = 100) N (%)	Overall Buy Ratea N (%)	Valid Order Attemptsb (N = 91) N (%)	Valid Buy Ratec N (%)
No attempts to verify age at all	10 (10.0)	9 (90.0)	9 (9.9)	9 (100.0)
Age warning on home page	66 (66.0)	59 (89.4)	60 (65.9)	59 (98.3)
Age verification strategies that cannot effectively verify age	85 (85.0)	76 (89.4)	77 (84.6)	76 (98.7)
”Submitting order” certifies age	82 (82.0)	73 (89.0)	74 (81.3)	73 (98.6)
Checkbox/button certifies age	54 (54.0)	47 (87.0)	48 (52.7)	47 (97.9)
Input current age	3 (3.0)	3 (100.0)	3 (3.3)	3 (100.0)
Accepts only credit cards ensuring buyers are adults	1 (10)	1 (100.0)	1 (11)	1 (100.0)
Site states they verify age	1 (10)	1 (100.0)	1 (11)	1 (100.0)
Only age verification strategies that cannot effectively verify age	42 (42.0)	36 (85.7)	36 (40.0)	36 (100.0)
Age verification strategies that can potentially effectively verify agee	48 (48.0)	44 (91.7)	46 (50.5)	44 (95.7)
Date of Birth (DOB)	39 (39.0)	36 (92.3)	37 (40.7)	36 (97.3)
Send a copy of driver’s licensef	16 (16.0)	12 (75.0)	14 (15.4)	12 (85.7)
Online Age Verification service (OAV)	13 (13.0)	11 (84.6)	12 (13.2)	11 (91.7)
Input last 4 digits of SSN	2 (2.0)	2 (100.0)	2 (2.2)	2 (100.0)
Enter driver’s license number	1 (10)	1 (100.0)	1 (11)	1 (100.0)
Challenge questions	1 (10)	1 (100.0)	1 (11)	1 (100.0)
Age verified at delivery	10 (10.0)	9 (90.0)	10 (11.0)	9 (90.0)g

Note.

^a:Overall buy rate represents the proportion of orders received from among order attempts.

^b:Valid order attempts excludes the orders that failed for reasons unrelated to age verification.

^c:Valid buy rate represents the proportion of orders received from among valid order attempts. Some vendors used more than one age verification strategy; reported age verification strategy types were not mutually exclusive.

^d:Age verification strategies were not mutually exclusive, some vendors used more than one.

^e:These strategies have the potential to restrict youth access if they are employed appropriately.

^f:We considered the presence of a “We Card” logo on the site to indicate that the vendor would request a copy of the buyer’s driver’s license.

^g:One vendor/site did not make any claims about age verification strategies, although they sent a text message requesting a copy of a driver’s license after the order was placed. Though the vendor shipped the package without receiving a copy of the license, they shipped the package requiring age verification at delivery; the delivery driver performed the delivery correctly by refusing to give the package to the underage teen buyer.

In examining overall order attempts, 10.0% of vendors did not employ any age verification strategies. Passive age warnings were used by two-thirds (66.0%) of ILVs. Most vendors (85.0%) employed age verification strategies that cannot effectively verify age, such as a checkbox to indicate that the buyer is 18 years old or older. Overall, 42% of ILVs used only forms of age verification that cannot effectively verify age (eg, the ILVs did not also use a more effective strategy). We received all 36 orders (100% valid buy rate) from vendors using only ineffective age verification strategies.

Approximately half of vendors (48.0% of all order attempts, 50.5% of valid order attempts) used potentially effective age verification strategies. DOB was the most common strategy of the ILVs using age verification strategies that could potentially block underage teens’ access--featured by 39 vendors (39.0%). Of those, 12 sites (30.8%) accepted teen’s underage DOB and 17 (43.6%) accepted a fake DOB, all of which shipped the orders. One of the 12 that accepted an underage DOB contacted us via email to inform us that we were underage, but then shipped the order. One of the 17 that accepted a fake DOB followed up to confirm the fake DOB and released the order when the fake DOB was verbally confirmed over the phone (without any further checking via a PRD or of an ID). Beyond the 29 sites described above, one site accepted an underage DOB, but did not ship the product nor make any attempts to follow up, leaving the order status listed as “in process” without charging the buyer; it was unclear whether the order was not completed due to age verification or a problem with the website. Of the vendors that sought an adult DOB (which was encountered only after attempts failed using an underage DOB and then a fake DOB), 5 of them shipped the product when we provided a parent’s DOB, name, and shipping address. Another of these vendors purporting to require an adult DOB cancelled the order due to self-described fraud prevention measures rather than age verification. Of the remaining 2 vendors of the 39 purporting to use DOB, one did not require a DOB at all and the other requested an “optional” DOB and shipped products without submission of a DOB. Notably, we had only one valid order attempt blocked as a result of DOB used for age verification.

Among the other potentially effective age verification strategies, 13 vendors (13.0%) purported to use an OAV service. One site claiming to use an OAV cancelled the order as a fraud prevention measure, not age verification. Of the 12 valid orders placed with sites claiming to use an OAV, only one site (8.3%) successfully blocked the order.

Of the 16 vendors (16.0%) that purported to request a copy of the buyer’s driver’s license for age verification, we found that only 7 (43.8%) of them did. In 5 of these cases, the buyer submitted a parent’s license and the order was released. The sixth vendor cancelled the order after receiving a copy of the ID because the billing address for the order did not match the address on the ID (a common credit card fraud prevention measure). The seventh case is the only vendor that requested a copy of a license and then, without receiving a copy of the license, paid for the delivery company’s age verification at delivery service for the order. One vendor accepted our payment, but did not ship the order, and another vendor rejected the order because the billing and shipping addresses on the order did not match, without requesting ID. All of the other vendors (43.8%) shipped the orders without requesting or receiving ID. Overall, we received 12 (85.7%) of the 14 valid orders that purported to require a driver’s license for age verification, 7 (58.3%) of which shipped without the vendor having received a copy of the buyer’s license.

As a part of this study, we identified 2 rarely-used age verification strategies in our sample of ILVs: inputting the last 4 digits of the buyer’s social security number (SSN) and entering a driver’s license number. Although these strategies were used by only one (driver’s license number) or 2 (inputting SSN) vendors, neither strategy blocked the order.

We found one vendor that claimed to use challenge questions for age verification, but none were presented, and the buyer was able to complete their order unimpeded.

Ten ILVs purported to use age verification at delivery, although none of them did. Instead, one vendor who did not advertise using age verification at delivery did employ it (and would have blocked the delivery if a parent had not been there to step in and sign after the teen could not provide proof of age). Orders were delivered by a combination of USPS (85.4%), UPS (13.5%), and FedEx (1.1%). Eleven of the 89 (12.4%) packages were handed to an adult in the household when the teen buyer was unavailable. Two of the 89 (2.2%) were handed directly to a teen buyer at their home with no effort to verify age; another 2 orders were handed to another underage child in the household, also with no effort to verify age. Of the 3 orders retrieved from shipping centers (3.4%), youth buyers retrieved 2 of them for which they were required to present a photo ID to verify identity, not age; the third was retrieved by an adult (who also had to present ID).

DISCUSSION

The overall buy rate for this study was 89.0%, with a higher valid buy rate of 97.8%. Ten percent of ILVs made no attempts to verify age at all. Most ILVs (84.6%) from which valid order attempts were placed used ineffective age verification methods involving passive verification or self-report of age (selecting checkboxes to verify age, submitting order implies age), which are easy for teens to bypass. Teens successfully ordered LCCs from nearly all of these sites, with popular flavored products available more cheaply than non-flavored products. Our findings show that although about half (50.5%) of sites included in the valid order attempts used age verification strategies that could potentially effectively verify age, underage teens were able to successfully order LCCs from those sites at a high rate (95.7%), indicating that these potentially effective age verification strategies (including OAVs which advertise themselves as an effective way to prevent youth access) either do not work effectively or were not used consistently and rigorously by the ILVs. Not surprisingly, all orders from vendors who relied on ineffective age verification strategies were successfully received, underscoring the need for age verification to prevent sales to minors. Only 2 of the 91 valid order attempts were blocked: one via DOB (presumably checked with a public records database) and one that verified age at delivery. Some cards used for orders were subject to credit card fraud, which as has previously been demonstrated, is a significant risk in online tobacco orders,^23,32 and should be a concern to consumers.

Our study underscored ILVs’ lax age verification strategies at point-of-order and point-of-delivery. We received all orders (100% valid buy rate) from vendors that employed only ineffective age verification strategies; the fact that we received all our orders clearly demonstrated their lack of effectiveness. About half of vendors purported to use potentially effective age verification strategies, but still had a high valid buy rate of 95.7%. Overall, we noted that the age verification strategies that could potentially block youth access such as DOB, OAV, and sending a copy of a driver’s license were not used in a way that effectively prevented youth access. It is unknown whether the vendors had the intention to implement these strategies poorly or if they made good faith efforts to implement strategies that turned out to be less effective in practice. For example, none of the 10% of ILVs that claimed that they would verify age at delivery did so (the only vendor who did so did not state that they would). Although this study attempted to determine the proportion of ILVs using challenge questions to verify age and how effective they were, as with other studies using a similar protocol,^23,32 we were unable to assess the effectiveness of challenge questions due to their lack of use. Our previous study published in 2015 about e-cigarette sales to minors did not find any sites using challenge questions.²⁵ Only one ILV in the study claimed to use challenge questions, and in practice, the vendor did not. Overall, our findings about age verification suggest that youth access to LCCs online can be prevented by more consistent use of effective age verification strategies (that have been proven to work).

Regarding the order that was blocked at point-of-sale, we want to highlight the unusual details of that order. This ILV provided exceptional attention to age verification by identifying a submitted order that shared the same address as a previous order that they had flagged as potentially being an underage order attempt. The ILV rejected the second order based on that information alone, indicating a willingness to lose business to prevent youth access. In addition, the same ILV rejected another order attempt by a different buyer with their parent’s ID because of a mismatch between billing and shipping addresses. This example suggests that rigorous age verification, including potentially low-tech strategies, is possible, and that at least one ILV took the steps needed to put it into practice.

Despite regulators’ efforts to reduce youth access to cigarettes, including flavored ones, via 2009’s FSPTCA⁴ and the Prevent All Cigarette Trafficking (PACT) Act of 2009,³³ flavored tobacco products are still available to teens in the form of LCCs from ILVs. Online tobacco products have historically been a cause for concern because of extremely high rates of sales to minors,^34–36 fraud,³² and because they avoid taxes through smuggling, contraband, and untaxed products.^32,37–43

On May 5, 2016, the FDA’s Deeming regulations extended its authority to regulate cigars and cigarillos.⁴⁴ As a result, ILVs were required to comply with FDA regulations by August 8, 2016, including conducting adequate age verification. The regulations do not specify details for verifying age, but the FDA has indicated that any vendor who sells to minors (regardless of age verification methods employed) is in violation of the law, indicating that poorly-implemented age verification strategies are not an excuse for a lack of regulatory compliance with regard to selling to minors. This leaves the possibility that vendors unwittingly commit regulatory violations and face enforcement actions by selling to minors after employing a trusted but unproven and/or ineffective age verification strategy.

The deeming regulations also specified a deadline by which vendors must submit a premarket application for tobacco products, with that deadline advanced from its original date of February 2017 to 2021 for combustible and 2022 for non-combustible products (unless they are substantially equivalent to a product on the market before the predicate date).^44,45 After lawsuits forced the issue,⁴⁶ the deadline for all premarket applications for tobacco products has been set for May 11, 2020. The FDA also has announced its intent to enact regulations banning the sale of flavored LCCs, which could lead to decreased availability of products online or may increase availability as consumer demand increases and products disappear from local retail stores. Given our findings related to the regulation of online cigarette sales,^32,47 it is unlikely ILVs will broadly comply with regulatory requirements for age verification and banned sales of flavored products without active enforcement.

Our findings underscore the failure of nearly all ILVs’ age verification strategies, thereby allowing for high rates of underage teens’ access to LCCs via online sales. Given the public health concern about underage teens’ access to tobacco products such as LCCs, this study has demonstrated that, although not commonly used, potentially effective strategies exist to prevent their online access, and that many of these strategies were ineffective in practice. Our results provide support for regulations requiring specific, effective age verification strategies as well as specific guidance about their implementation for online sales. Simply prohibiting sales to minors without providing vendors with guidance about specific, effective age verification is a disservice to vendors and consumers. More research is needed to determine which forms of age verification are highly effective at blocking youth access when implemented properly.

IMPLICATIONS FOR TOBACCO REGULATION

With the FSPTCA banning flavored cigarette sales, youth have turned increasingly to flavored LCCs^21,23 and other tobacco and nicotine products.^18–20,48 Although the FDA Deeming Regulations, PACT Act, and FSPTCA are all important tobacco control regulations, PACT and FSPTCA are cigarette-specific without addressing specific age verification strategies or flavor bans for online sales of LCCs. The FSPTCA defines a cigarette as a tobacco product that because of its appearance, packing, labeling, and marketing is sold to, and/or sought by, consumers seeking cigarettes, regardless of how it is labeled. Little cigars are often purchased by consumers seeking flavored and/or less expensive tobacco products similar to cigarettes,^5,15,16 and should be treated the same as cigarettes under existing federal law, subject to the same sales restrictions. Easy access to LCCs online by underage teens underscores an immediate imperative to address this issue. Our study suggests a need to expand the reach of tobacco control regulations to include other flavored tobacco products and to address lax age verification strategies for online orders. Easy access to LCCs online suggests an ongoing need to test strategies that allow and prohibit access to these products. Although this study was completed before the 2016 FDA Deeming Regulations²⁷ requiring age verification for online tobacco product sales, our findings provide a baseline against which to compare the effects of the deeming regulations on ILV age verification strategies. Future studies should assess the extent of vendor compliance with these regulations as well as their enforcement and their success at preventing youth access.