Influence of Tobacco Displays and Ads on Youth:A Virtual Store Experiment

WHAT’S KNOWN ON THIS SUBJECT: Youth exposure to retailtobacco advertisements and displays is associated with smokinginitiation. The Family Smoking Prevention and Tobacco Control Actof 2009 gives states and local governments legal authority toregulate the time, place, and manner of tobacco advertising.

WHAT THIS STUDY ADDS: This is the first experimental studyusing a virtual store environment to provide evidence thata policy banning tobacco product displays at the point of sale maydeter youth from attempting to purchase tobacco products atretail stores.

abstractOBJECTIVE: To examine the potential impact of banning tobacco dis-plays and ads at the point of sale (POS) on youth outcomes.

METHODS: An interactive virtual convenience store was created withscenarios in which the tobacco product display at the POS was eitheropenly visible (status quo) or enclosed behind a cabinet (display ban),and tobacco ads in the store were either present or absent. A nationalconvenience sample of 1216 youth aged 13 to 17 who were either smok-ers or nonsmokers susceptible to smoking participated in the study.Youth were randomized to 1 of 6 virtual store conditions and givena shopping task to complete in the virtual store. During the shoppingtask, we tracked youth’s attempts to purchase tobacco products. Sub-sequently, youth completed a survey that assessed their perceptionsabout the virtual store and perceptions about the ease of buyingcigarettes from the virtual store.

RESULTS: Compared with youth in the status quo condition, youth in thedisplay ban condition were less aware that tobacco products were forsale (32.0% vs 85.2%) and significantly less likely to try purchasingtobacco products in the virtual store (odds ratio = 0.30, 95% confidenceinterval = 0.1320.67, P , .001). Banning ads had minimal impact onyouth’s purchase attempts.

CONCLUSIONS: Policies that ban tobacco product displays at the POSmay help reduce youth smoking by deterring youth from purchasingtobacco products at retail stores. Pediatrics 2013;131:1–8

AUTHORS: Annice E. Kim, PhD, MPH,a James M.Nonnemaker, PhD,a Brett R. Loomis, MS,a Asma Baig, MS,a

Edward Hill, BS,b John W. Holloway, BS,c Matthew C.Farrelly, PhD,a and Paul R. Shafer, MAa

aPublic Health Policy Research Program and cSurvey ResearchDivision, RTI International, Research Triangle Park, NorthCarolina; and bTarheel Technologies, Durham, North Carolina

KEY WORDSadolescence, tobacco advertising, tobacco control policy, tobaccouse/smoking

ABBREVIATIONPOS—point of sale

Drs Kim and Nonnemaker, Mr Loomis and Ms Baig contributedto study design, data collection, data analysis, and writing of themanuscript; Mr Hill and Mr Holloway contributed to studydesign, data collection, and review of the manuscript; Dr Farrellycontributed to study design and review of the manuscript; andMr Shafer contributed to revising of the manuscript andaddressing reviewer comments.

www.pediatrics.org/cgi/doi/10.1542/peds.2012-0197

doi:10.1542/peds.2012-0197

Accepted for publication Aug 30, 2012

Address correspondence to Annice Kim, PhD, MPH, RTIInternational, 3040 Cornwallis Rd, PO Box 12194, ResearchTriangle Park, NC 27709. E-mail: akim@rti.org

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).

Copyright © 2013 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE: The authors have indicated they haveno financial relationships relevant to this article to disclose.

FUNDING: Funded by the New York State Department of Health.

PEDIATRICS Volume 131, Number 1, January 2013 1

ARTICLE

by guest on May 26, 2018http://pediatrics.aappublications.org/Downloaded from

Despite significant progress in reducingyouth smoking in the United States,the rates of decline have stalled in thepast decade.1 States have experienceddramatic budget cuts for tobacco pre-vention programs, but the tobacco in-dustry continues to market its productsaggressively with nearly 90% of its $10billion marketing expenditure spent onretailer incentives and price promotionsat the point of sale (POS),2 making retailstores the most important advertisingchannel for the tobacco industry. Thelevel of retail cigarette advertisementsand promotions has increased overtime,3,4 and cigarette products areprominently placed on shelves behindcheckout counters, exposing all storecustomers, including youth, to tobaccoproducts. Studies show that youth arehighly aware of tobacco ads in stores,5

and such exposure influences youthperceptions regarding ease of accessto cigarettes, smoking prevalence, andpeer approval of smoking.6 Youth whoare exposed to retail tobaccomarketingare also more likely to experiment withsmoking and to become smokers.7–16

The Family Smoking Prevention andTobacco Control Act of 2009 gives statesand local governments legal authorityto regulate the time, place, andmannerof tobacco advertising. As states con-sider policies such as banning thedisplay of tobacco products, empiricalstudies are needed to determine thepotential impact of these regulations.Casestudies in IrelandandCanadashowthat youth and adult awareness ofPOS tobacco marketing dropped signif-icantly after a ban on tobacco displaysand ads.17,18 Although these case studiessuggest that banning POS tobacco dis-plays and ads could help deter youthsmoking, few studies have systemati-cally examined the potential impact ofthese policies on smoking outcomes.

In an experimental study, Wakefield andcolleagues19 examined the potential im-pact of regulating POS tobacco product

displays and ads by showing Australianyouth photos in which cigarette packswere displayed with ads (status quo inretail stores), without ads, and withneither ads nor packs displayed. Theyfound that exposure to tobacco productdisplays with and without ads wasassociated with increased perceptionsthat tobacco products were easy topurchase at the store compared withthose who viewed the no cigarette dis-play condition. In addition, they foundthat, compared with those in the nocigarette display condition, studentswho were exposed to the cigarette ad-vertising condition perceived it wouldbe less likely they would be asked forproof of age and that a greater numberof stores would sell them cigarettes. Apotential limitation of this study is thatthe stimuli were photographs shown ina classroom setting, which does not re-flect real-world exposure to POS displays.

We extend the Wakefield et al study byusing a virtual store to simulate amoreinteractive exposure to POS tobaccoproduct displays and ad bans. Virtualreality applications are intended tosimulate features of the real-worldenvironment, which enables research-ers to immerse participants into a hy-pothetical context and study theirbehavioral responses to environmentalcues that may be difficult to assess ina real-life setting. Virtual reality envi-ronments have been used to studya wide range of behaviors20 includingbehaviors among youth21,22 and to testthe effect of smoking cues on smok-ers’ cravings for cigarettes in youngadults23,24 and adults.25 However, todate, no studies have used virtualenvironments to study the impact oftobacco displays and ads on youth. Wedesigned a virtual store and gave youtha specific shopping task to assesswhether youth exposed to the displayban condition are less likely to attemptbuying tobacco products in the virtualstore and perceive that it is difficult to

buy cigarettes at the virtual store. Asecondary aim was to examine whetherthe presence of tobacco ads in the storemoderates these relationships.

METHODS

Study Design

We designed a 3 3 2 experimentalstudy with 3 variations of the POS to-bacco product displays (open, enclosed,enclosed with ads on cabinet) and 2variations of tobacco ads in store(present, absent) for a total of 6 con-ditions (defined in Table 1 and exhibitedin Figs 1 and 2).

An off-the-shelf model of a conveniencestore was purchased and extensivelycustomized for this study by using Unity3Dinteractivegamingsoftware.Theinitialbeta version of the virtual store was pilottestedwith focusgroupsof 12 youthaged14 through17whowerecurrent smokersor nonsmokers susceptible to smoking.The virtual store was updated on thebasis of feedback from the focus groups(eg, improved navigation speed, ex-panded product selections).

Participants and Virtual ShoppingTask

Anational conveniencesampleof currentsmoker or nonsmoker susceptible tosmoking youth aged 13 to 17 were re-cruited from Research Now’s e-Rewardsonline youth panel and via parents intheir adult panel who indicated havingchildren aged 13 to 17. Current smokerswere defined as smoking at least 1 cig-arette per day in the past 30 days. Non-smokers susceptible to smoking weredefined as those who had not smoked inthe past 30 days and answered “defi-nitely yes, probably yes, or probably not”to any 1 of these 3 questions: “Do youthink you will smoke a cigarette anytimeduring the next year?” “Do you think youwill try a cigarette soon?” and “If one ofyour best friends offered you a cigarette,would you smoke it?”

2 KIM et al

by guest on May 26, 2018http://pediatrics.aappublications.org/Downloaded from

E-mail invitations were sent, and 4189panelists consented to participate andmet the eligibility criteria (Fig 3). Par-ticipants initiated the study by clickingon a link that randomized them to 1 of 6conditions. If participants already hadthe Unity 3D player installed on theircomputer, the virtual store appli-cation loaded, and they began the vir-tual shopping task. If participants didnot have the Unity player, they wereprompted to download the player.

For the virtual shopping task, partic-ipants were instructed to select 4 itemsfor purchase: a snack from the aisles,a drink from the coolers, and 2 itemsof their choice from the checkoutcounter. Participants were not specifi-cally instructed to purchase tobacco.Participants had a total of 10minutes tocomplete their shopping task. As par-ticipants navigated through the store,they were exposed to one of two ver-sions of the store environment in which

tobacco ads were either present orabsent (see Fig 2). When the partici-pant moved to the checkout counter topurchase the last 2 items, he or she wasexposed to 1 of the tobacco displayconditions (Fig 1). At the checkoutcounter area, if the participant attemp-ted to purchase tobacco by clicking onthe tobacco product display, the ani-mated cashier asked, “Do you want tobuy tobacco products?” If the participantclicked “No,” the cashier responded “OK”and resumed talking on his phone. If theparticipant clicked “Yes,” the cashier re-sponded “Sorry, you are not old enoughto purchase this product” and resumedtalking on his phone. Once the final 2items from the checkout counter hadbeen selected, the participant was ex-posed to the display for another 10secondsbefore the shopping taskendedand the participant was directed to thesurvey.

One thousand two hundred sixteen par-ticipants completed the study, and mostwere recruited from the parent panel(62%). There were no meaningful socio-demographic differences among youthwho qualified (n = 4189) versus com-pleted (n = 1216) the study. Youth re-ceived a $6.50 e-Rewards dollar incentiveto be redeemed for products/servicesupon study completion. This study wasapproved by Institutional Review Boardsat RTI International and the New YorkState Department of Health.

Measures

The 2 key study outcomeswerewhetheryouth perceived cigarettes as easyto purchase in the virtual store andwhether youth tried to click and pur-chase tobacco in the virtual store.Perceivedeaseofaccesswasmeasuredwith the following item: “Imagine thata real store like this virtual store wasnear where you live; how easy or hardwould it be for you to buy cigarettesfrom this store?” Youth were definedas attempting to purchase tobacco ifduring their shopping task they clicked

TABLE 1 Study Conditions

Tobacco Product Display at POS Tobacco Ads in Store

Present Absent

Open Display: Open, fully visible current status quo situation C1 C4Enclosed Display: Not visible, enclosed behind opaque cabinet

with black and white signs noting “Cigarette Center” and“Chewing Tobacco & Cigars” at the top of the display cabinet a

C2 C5

Enclosed display + ads on cabinet: Not visible, enclosed behindopaque cabinet with “Cigarette Center” sign at top + ads oncabinet doors b

C3 C6

a Enclosed display condition was modeled after an actual policy adopted by stores in New York State to ban the display oftobacco products.b We created this scenario as a potential unintended consequence of the display ban.

FIGURE 1Screenshots of tobacco product display conditions in virtual retail store.

ARTICLE

PEDIATRICS Volume 131, Number 1, January 2013 3

by guest on May 26, 2018http://pediatrics.aappublications.org/Downloaded from

on the tobacco product display andconfirmed they wanted to purchase to-bacco. This information was capturedby the Unity 3D software.

Wealsoassessedperceptionsabout thevirtual store, including whether youthwere aware that tobacco productswereforsale in thevirtual storeandtheextent

to which they thought the virtual storeresembled a real convenience store. Wealso measured the following covariates:smoking behavior, youth’s usual sourceof cigarettes (retail or nonretail), socialinfluence to smoke (whether youth liveswith smoker, number of best friendswho smoke), sensation-seeking behav-ior (additive scale constructed from 4items that assessed the extent to whichyouth like to explore new places, like todo frightening things, like new and ex-citing experiences, and prefer friendswho are exciting and unpredictable),and key demographics (age, gender,race/ethnicity).

Analysis

Process data from the virtual storeshopping tasks were linked to thesurvey data via unique identifiers foreach study participant. We tested forpotential differences in each outcomeby condition. Logistic regression wasused because the main outcomes weredichotomized. In general form, we esti-matedthe followingregressionequation:

FIGURE 2Screenshots of tobacco ad conditions in virtual retail store. Note: in the conditions in which tobacco adswere present in the store (C4–C6), there were 2 ads on the exterior door along with 2 branded “pull”signs, 2 ads above the checkout counter, and 2 ads below the checkout counter. In the conditions inwhich tobacco ads were absent (C1–C3), there were no ads on the exterior door and ads for soft drinkand lottery tickets above and below the checkout counter.

FIGURE 3Study sample. Note: “Did not install Unity player plug-in” indicates participants who did not already have the Unity player and were unwilling to download it.“Did not complete shopping task” indicates participantswho did not have complete data from the virtual retail store. These participantsmay have ended theirvirtual shopping task early and thus did not have survey data. “Did not complete survey” indicates participantswho completed the virtual retail store data butdid not complete the survey.

4 KIM et al

by guest on May 26, 2018http://pediatrics.aappublications.org/Downloaded from

logitðoutcomeÞ  ¼   a  þ   b2C2   þ   b3C3  

þ   b4C4   þ   b5C5   þ   b6C6  

þ  +k

j¼7bj Xj

Ci are the indicators for each conditionwith condition 1 omitted as the referentcategory (ie, we compared each con-dition to the open display with ads inthe store condition because this is thecurrent practice in stores). Xj refers tothe set of covariates delineated in themeasures section. The perceived easeof access model excludes participantswho attempted to purchase tobacco inthe virtual store. All descriptive statis-tics and regression models were esti-mated by using Stata 11.0 (Stata Corp,College Station, TX).

RESULTS

Sample characteristics are summari-zed in Table 2. The distribution of de-mographics was similar across the 6conditions. The majority of study par-ticipants were aged 15 to 17, male, andwhite. More than half of participantsvisit convenience storesmore than onceper week and 17.4% of current youthsmokers report usually getting theircigarettes in a retail store. On average,participants took 172.3 seconds tocomplete the virtual store shoppingtask with ∼32.1 seconds spent at thecheckout counter. More than 80% ofrespondents agreed that the virtual storeresembled a real convenience store.

The proportion of participants whoreported seeing tobacco products forsale was highest in the open displaycondition with ads (C1, 85.2%) andlowest for the enclosed display condi-tion with no ads (C5, 32.0%). When to-bacco advertisements are present inthe store, 24.3% of study participantsclicked to purchase tobacco in the opendisplaycondition (C1)comparedwithonly9.0% in the enclosed display condition(C2).Whennotobaccoadswerepresent inthe store, 16.4% of youth clicked to pur-

chase tobacco in the open display con-dition (C4), compared with 10.8% in theenclosed display condition (C5).

The proportion of youth who perceivedthat it would be hard or very hard to buycigarettes from the retail store washigher for youth in the open displaycondition with tobacco ads in store(51.5% for C1) than in the encloseddisplay condition (48% for C2). Thispattern holds when tobacco advertisingwas removed from the store, with 47.6%in the opendisplay condition (C4) versus40.5% in the enclosed condition (C5).

Table 3 presents adjusted odds ratiosfor the 6 experimental conditions forthe key outcomes of perceived ease ofaccess to cigarettes and attempting topurchase tobacco. Current smokersdid not show significant differencesacross conditions regarding perceivedease of access to cigarettes. Amongnonsmokers susceptible to smoking,youth in condition C5 (enclosed display,no ads in store) were significantly lesslikely to believe that it would be hard topurchase cigarettes in the virtual con-venience store than youth in conditionC1 (open display; odds ratio = 0.49, 95%confidence interval = 0.28–0.85, P ,.05). Current smokers in condition C2(enclosed display, ads in store) weresignificantly more likely to say it wouldbe hard to buy cigarettes from the vir-tual store than current smokers incondition C3 (enclosed display, ads instore, ads on cabinet; P , .05). Non-smokers in condition C4 (open display,no ads in store) were significantly morelikely to say it would be hard to buycigarettes from the virtual store thanthose in condition C5 (enclosed display,no ads in store; P , .05).

After controlling for the covariates, allyouth who were exposed to any of theenclosed display conditions (exceptopen-to-smoking youth in C5) were sig-nificantly less likely to try purchasingtobacco in the virtual store than youthwho were exposed to the open display

condition (C1; see Table 3). Currentsmokers in conditions C5 (enclosed dis-play, no ads in store) and C6 (encloseddisplay, no ads in store + ads on cabinet)were less likely than current smokers incondition C4 (open display, no ads instore) to attempt purchasing tobacco.

DISCUSSION

In summary, we found that enclosingtobacco product displays significantlylowers the likelihood that youth will tryto purchase tobacco in the virtual storebut inconsistent results on whethertheyperceive cigarettes tobedifficult topurchase. Whereas Wakefield and col-leagues19 found that youth who wereexposed to no display or ads weremore likely to perceive tobacco as dif-ficult to purchase from the store, wefound opposite results among non-smokers. One possible explanation forthis may be that youth in enclosedconditions may perceive cigarettes tobe more difficult to access and thus aremore likely to try and purchase becausethey see it as a challenge. Policies thatrestrict access to tobacco productsmayhave the unintended consequences ofyouth perceiving these products as“forbidden fruit” that are more desir-able because they are associated withbeing an adult.26 Tobacco industrydocuments reveal that tobacco adver-tising has explicitly attempted to conveythe message that smoking cigarettes issynonymous with being an adult. In ourexperiment, although tobacco productswere not visible in the enclosed con-ditions, the cabinets were labeled withthe word “cigarettes,” which indicatedcigarette products may be availablefor sale. Enclosing the display in thismanner may have the unintended effectof emphasizing the message that to-bacco is for adults, thereby increasingyouth’s desire to use those products. Weincluded this signage in our design be-cause it resembled how stores in NewYork State and in other countries27 hadimplemented tobacco product display

ARTICLE

PEDIATRICS Volume 131, Number 1, January 2013 5

by guest on May 26, 2018http://pediatrics.aappublications.org/Downloaded from

bans and therefore represents a likelyscenario of how such policies may beadopted in the United States.

Although we found no evidence thatenclosing the display decreased youth’sperceived ease of access, we foundthat youth who were exposed to theenclosed display conditions were lesslikely to try purchasing tobacco in the

virtual store. We believe that an ob-servational measure of youth clickingon the tobacco product display andaffirming that they want to purchasetobacco products may be a more validmeasure of youth’s desire and intentionto buy cigarettes than a self-reportedmeasure about how easy it would be topurchase cigarettes from a pictured

store. It is better to observe what youthwill actually do in a situation in whichtobacco product displays are enclosedthan to simply ask them about whatthey might do. Additionally, the virtualstore may be a more realistic way toexpose participants to potential displayban conditions than static images andto engage themwith the environment by

TABLE 2 Characteristics of Study Sample

Characteristic Total(N = 1216)

C1 C2 C3 C4 C5 C6Open Display,Ads in Store(N = 202)

EnclosedDisplay,

Ads in Store(N = 201)

Enclosed Display + Ads inStore, Adson Cabinet(N = 203)

Open Display, NoAds in Store(N = 207)

Enclosed Display,No Adsin Store(N = 203)

EnclosedDisplay + NoAds in Store,Ads on Cabinet

(N = 200)

Smoking status, %Open to smoking 60.5 59.9 60.2 60.6 61.8 60.6 60.0Current smoker 39.5 40.1 39.8 39.4 38.2 39.4 40.0

Age, y (%)13 4.8 6.4 4.0 5.9 4.8 4.9 2.514 13.8 11.9 14.4 14.8 14.5 8.9 18.515 21.1 23.8 17.9 22.2 19.8 22.2 20.516 30.1 33.2 26.9 31.5 30.9 31.0 27.017 30.3 24.8 36.8 25.6 30.0 33.0 31.5

Male, % 53.5 52.0 56.2 51.7 56.0 50.7 54.5Race, %White, non-Hispanic 72.4 68.3 72.6 76.4 69.1 72.9 75.0Nonwhite, non-Hispanic 14.6 15.4 12.9 13.8 20.3 15.3 10.0Hispanic 13.0 16.3 13.8 9.9 10.6 11.8 15.0

Sensation scale: 0 (lowestsensationseeking) → 20 (highestsensationseeking), avg (SD)

13.6 (3.8) 13.5 (3.6) 13.6 (3.2) 13.6 (3.3) 13.5 (3.4) 13.3 (3.4) 13.9 (3.3)

Current smokers who purchasedfrom a convenience store,supermarket,or gas station, %

17.4 19.7 16.9 14.3 17.7 18.2 18.1

Live with smoker, % 32.3 31.7 32.0 36.5 29.6 34.2 30.0Said they have 2–5 best friendswho smoke, %a

38.2 38.3 39.5 40.6 34.3 40.4 36.4

Time spent in shopping aisles, s,avg (SD)

172.3 (91.0) 171.1 (87.7) 169.5 (81.4) 173.7 (83.5) 176.2 (109.80) 166.1 (87.6) 177.1 (93.0)

Time spent at counter, s, avg (SD) 32.1 (16.7) 32.4 (15.9) 30.8 (16.8) 31.0 (15.9) 33.9 (18.3) 32.3 (16.1) 32.1 (16.7)Said the virtual store resemblesreal-world convenience storesa lot, very, or somewhat, %b

84.8 83.1 89.1 80.8 89.4 82.3 84.0

Saw tobacco for sale, % 60.3 85.2 54.2 62.1 79.2 32.0 48.5Saw tobacco advertisements, % 57.1 75.3 59.2 69.5 49.8 29.6 59.5Said they go to convenience storesat least a few times a week, %c

54.8 58.7 53.7 52.7 58.0 53.7 52.0

Tried to purchase tobacco, % 13.5 24.3 9.0 10.3 16.4 10.8 10.0Said it would be hard or veryhard to buy cigarettes in virtualstore, %d

45.0 51.5 48.0 40.9 47.6 40.5 41.7

avg, average.a Dichotomized variable: reference category is 0–1 best friends who smoke.b Dichotomized variable: reference category is the virtual store resembles real-world stores a little or not at all.c Dichotomized variable: reference category is those who go to convenience stores approximately once a week, once a month, or hardly ever.d Dichotomized variable: reference category is those who said it would be very easy, easy, neither easy nor hard to buy cigarettes from the virtual store.

6 KIM et al

by guest on May 26, 2018http://pediatrics.aappublications.org/Downloaded from

giving them a specific shopping task.Indeed, previous studies in the field ofexperimental psychology confer thebenefits of using interactive virtualenvironments to simulate and test theimpact of environments or situationsthat are difficult to study.28–30 In addition,we also confirmed that youth thoughtour virtual store was a realistic repre-sentation of convenience stores thatthey frequent. Despite these strengths,our study has some potential limita-tions. First, youth’s exposure to theconditions may not have been sufficient.We attempted to provide substantialexposure to the display conditions byhaving youth select multiple products atthe checkout counter and having theanimated retailer purposely keep theyouth waiting by talking on his cellphone, but youth only spent an averageof 3.4 minutes in the virtual store, whichis considerably less than the averageof 16 minutes teenagers typically spendwhen visiting convenience stores.31

However, a previous eye-tracking study32

found differences in outcomes when

youth viewed tobacco ads for only 9 to 17seconds, suggesting that the 32 secondsour study participants spent on averageat the POS may have been adequate ex-posure to the display conditions. Sec-ond, it is possible that participants mayhave been aware that the virtual taskwas related to smoking. To select theright participants, the screener ques-tions asked about smoking and, per in-stitutional review board guidelines, theconsent formmentioned that the surveywould assess perceptions about smok-ing. This may have primed the re-spondent that the shopping task wasabout smoking. However, the shoppingtask itself did not mention anythingabout smoking and participants werenot directed to purchase cigarettes.Furthermore, if priming occurred, itshould affect participants across allconditions. Finally, our results have lim-ited generalizability because ResearchNow’s panelists are a conveniencesample recruited via non-probability-based online methods. However, be-cause the purpose of our study was to

test the potential impact of banningtobacco displays and ads, our primaryconcern was to minimize potentialthreats to internal validity at the cost ofpotentially limiting external validity. Usingan online panel like Research Now pro-vided the benefits of accessing a largenumber of youth who could be systemat-ically screened and randomized to thevirtual store conditions via the Internet.

CONCLUSIONS

We found that enclosing tobaccoproduct displays could deter youthfromattempting to purchase tobacco inretail stores. These results providesupport for policies that would ban thedisplay of tobacco products at the POS.The virtual store application can beused to test the impact of other POSpolicies, such as whether antismokingmessages at the POS could mitigate theimpact of open tobacco displays orpotentially strengthen the impact ofenclosed tobacco displays on smoking-related outcomes for youth and adults.

REFERENCES

1. Centers for Disease Control and Prevention(CDC). Tobacco use among middle and highschool students—United States, 2000–2009. MMWR Morb Mortal Wkly Rep. 2010;59(33):1063–1068

2. Federal Trade Commission. Cigarette re-port for 2007 and 2008. 2011. Availableat: www.ftc.gov/os/2011/07/110729ciga-rettereport.pdf. Accessed August 30,2011

3. Feighery EC, Schleicher NC, Boley Cruz T,Unger JB. An examination of trends inamount and type of cigarette advertisingand sales promotions in California stores,2002–2005. Tob Control. 2008;17(2):93–98

TABLE 3 Adjusted Odds Ratios (95% CI) for Perceived Ease of Access to Cigarettes and Attempting to Purchase Tobacco in the Virtual Store

Perceived Ease of Access to Cigarettes Attempting to Purchase Tobacco

Current Smokers Open to Smoking Current Smokers Open to Smoking

(N = 293) (N = 681) (N = 409) (N = 723)

C1: Open display, ads in store 1.00 1.00 1.00 1.00C2: Enclosed display, ads in store 1.20a (0.45–3.22) 0.78 (0.45–1.35) 0.30** (0.13–0.67) 0.28* (0.08–0.96)C3: Enclosed display, ads in store + ads on cabinet 0.38a (0.14–1.06) 0.68 (0.40–1.18) 0.39* (0.18–0.86) 0.22* (0.06–0.85)C4: Open display, no ads in store 0.65 (0.23–1.86) 0.96b (0.56–1.65) 0.84c,d (0.40–1.74) 0.45 (0.15–1.34)C5: Enclosed display, no ads in store 1.19 (0.45–3.16) 0.49*b (0.28–0.85) 0.27**c (0.12–0.63) 0.46 (0.16–1.33)C6: Enclosed display, no ads in store + ads on cabinet 0.68 (0.25–1.83) 0.70 (0.41–1.21) 0.32**d (0.14–0.69) 0.22* (0.06–0.86)

All odds ratios adjusted for age, race, gender, sensation-seeking scale, number of best friends who smoke, living with a smoker, frequency of going to a convenience store, and virtual storeresembles a real convenience store. Current smoker models also control for source of cigarettes (retail versus social source). Perceived ease of access to cigarettes model excludes thosewho attempted to purchase tobacco in the virtual store.a Significant difference (P , .05) between C2 and C3 in current smoker model.b Significant difference (P , .05) between C4 and C5 in open to smoking model.c Significant difference (P , .05) between C4 and C5 in current smoker model.d Significant difference (P , .05) between C4 and C6 in current smoker model.* P , .05;** P , .01.

ARTICLE

PEDIATRICS Volume 131, Number 1, January 2013 7

by guest on May 26, 2018http://pediatrics.aappublications.org/Downloaded from

4. Girlando M, Loomis B, Watson K, Farrelly M.Retail advertising and promotions for ciga-rettes in New York. Research Triangle Park,NC: RTI International; 2007. Available at: www.health.state.ny.us/prevention/tobacco_control/docs/cigarette_ads_and_promotion_report.pdf. Accessed August 30, 2011

5. Centers for Disease Control and Pre-vention. Smoking and tobacco use: NationalYouth Tobacco Survey. 2009. Available at:www.cdc.gov/tobacco/data_statistics/sur-veys/nyts. Accessed September 2, 2011

6. Henriksen L, Flora JA, Feighery E, FortmannSP. Effects on youth of exposure to retailtobacco advertising. J Appl Soc Psychol.2002;32(9):1771–1789

7. Paynter J, Edwards R. The impact of tobaccopromotion at the point of sale: a systematicreview. Nicotine Tob Res. 2009;11(1):25–35

8. Schooler C, Feighery E, Flora JA. Seventhgraders’ self-reported exposure to ciga-rette marketing and its relationship totheir smoking behavior. Am J Public Health.1996;86(9):1216–1221

9. Feighery EC, Henriksen L, Wang Y, SchleicherNC, Fortmann SP. An evaluation of fourmeasures of adolescents’ exposure to ciga-rette marketing in stores. Nicotine Tob Res.2006;8(6):751–759

10. Henriksen L, Feighery EC, Wang Y, FortmannSP. Association of retail tobacco marketingwith adolescent smoking. Am J PublicHealth. 2004;94(12):2081–2083

11. Lovato CY, Hsu HCH, Sabiston CM, Hadd V,Nykiforuk CIJ. Tobacco point-of-purchasemarketing in school neighbourhoods andschool smoking prevalence: a descriptivestudy. Can J Public Health. 2007;98(4):265–270

12. Slater SJ, Chaloupka FJ, Wakefield M,Johnston LD, O’Malley PM. The impact ofretail cigarette marketing practices onyouth smoking uptake. Arch Pediatr Ado-lesc Med. 2007;161(5):440–445

13. Weiss JW, Cen S, Schuster DV, et al. Longi-tudinal effects of pro-tobacco and anti-tobacco messages on adolescent smoking

susceptibility. Nicotine Tob Res. 2006;8(3):455–465

14. MacKintosh AM, Moodie C, Hasting G. Theassociation between point-of-sale displaysand youth smoking susceptibility. NicotineTob Res. September 8, 2011. Available at:http://ntr.oxfordjournals.org/content/early/2011/09/08/ntr.ntr185.short?rss=1. AccessedSeptember 27, 2011

15. Henriksen L, Feighery EC, Schleicher NC,Cowling DW, Kline RS, Fortmann SP. Is ad-olescent smoking related to the densityand proximity of tobacco outlets and retailcigarette advertising near schools? PrevMed. 2008;47(2):210–214

16. Henriksen L, Schleicher NC, Feighery EC,Fortmann SP. A longitudinal study of expo-sure to retail cigarette advertising andsmoking initiation. Pediatrics. 2010;126(2):232–238

17. McNeill A, Lewis S, Quinn C, et al. Evaluation ofthe removal of point-of-sale tobacco displaysin Ireland. Tob Control. 2011;20(2):137–143

18. Cohen JE, Planinac L, Lavack A, Robinson D,O’Connor S, DiNardo J. Changes in retailtobacco promotions in a cohort of storesbefore, during, and after a tobacco productdisplay ban. Am J Public Health. 2011;101(10):1879–1881

19. Wakefield M, Germain D, Durkin S, Henrik-sen L. An experimental study of effects onschoolchildren of exposure to point-of-salecigarette advertising and pack displays.Health Educ Res. 2006;21(3):338–347

20. Waterlander WE, Scarpa M, Lentz D, SteenhuisIHM. The virtual supermarket: an innovativeresearch tool to study consumer foodpurchasing behaviour. BMC Public Health.2011;11:589

21. Schwebel DC, Gaines J, Severson J. Valida-tion of virtual reality as a tool to un-derstand and prevent child pedestrianinjury. Accid Anal Prev. 2008;40(4):1394–1400

22. Nemire K, Beil J, Swan RW. Preventing teensmoking with virtual reality. CyberpsycholBehav. 1999;2(1):35–47

23. Traylor AC, Bordnick PS, Carter BL.Assessing craving in young adult smokersusing virtual reality. Am J Addict. 2008;17(5):436–440

24. Traylor AC, Bordnick PS, Carter BL. Usingvirtual reality to assess young adultsmokers’ attention to cues. CyberpsycholBehav. 2009;12(4):373–378

25. Baumann SB, Sayette MA. Smoking cues ina virtual world provoke craving in cigarettesmokers. Psychol Addict Behav. 2006;20(4):484–489

26. Reid DJ, McNeill AD, Glynn TJ. Reducing theprevalence of smoking in youth in Westerncountries: an international review. TobControl. 1995;4(3):266–277

27. Hegarty R. Tobacco display ban “to costretailers £15.6m.” The Grocer. Available at:www.thegrocer.co.uk/topics/health/tobacco-display-ban-to-cost-retailers-156m/228093.article. Accessed July 11, 2012

28. Hubal RC, Fishbein DH, Sheppard MS,Paschall MJ, Eldreth DL, Hyde CT. How dovaried populations interact with embodiedconversational agents? Findings from inner-city adolescents and prisoners. ComputHuman Behav. 2008;24(3):1104–1138

29. Hubal RC, Bonner MJ, Hardy KK, FitzGeraldDP, Willard VW, Allen TM. Technical aspectsand testing of a program to assess deficitsin facial expression recognition in child-hood cancer survivors. J Cyber TherRehabil. 2011;4(3):363–369

30. Shute VJ, Ventura M, Bauer M, Zapata-RiveraD. Melding the power of serious games andembedded assessment to monitor and fos-ter learning: flow and grow. In: Ritterfeld U,Cody M, Vorderer P, eds. Serious Games:Mechanisms and Effects. New York, NY:Routledge, Taylor and Francis; 2009:295–321

31. Chanil D. Profile of the convenience storecustomer. Convenience Store News. Febru-ary 11, 2002:54–70

32. Fox RJ, Krugman DM, Fletcher JE, FischerPM. Adolescents’ attention to beer andcigarette print ads and associated productwarnings. J Advert. 1998;27(3):57–68

8 KIM et al

by guest on May 26, 2018http://pediatrics.aappublications.org/Downloaded from

originally published online December 3, 2012; Pediatrics John W. Holloway, Matthew C. Farrelly and Paul R. Shafer

Annice E. Kim, James M. Nonnemaker, Brett R. Loomis, Asma Baig, Edward Hill,Influence of Tobacco Displays and Ads on Youth: A Virtual Store Experiment

ServicesUpdated Information &

012-0197http://pediatrics.aappublications.org/content/early/2012/11/27/peds.2including high resolution figures, can be found at:

Permissions & Licensing

https://shop.aap.org/licensing-permissions/in its entirety can be found online at: Information about reproducing this article in parts (figures, tables) or

Reprintshttp://classic.pediatrics.aappublications.org/content/reprintsInformation about ordering reprints can be found online:

ISSN: . 60007. Copyright © 2012 by the American Academy of Pediatrics. All rights reserved. Print American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois,has been published continuously since . Pediatrics is owned, published, and trademarked by the Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it

by guest on May 26, 2018http://pediatrics.aappublications.org/Downloaded from

originally published online December 3, 2012; Pediatrics John W. Holloway, Matthew C. Farrelly and Paul R. Shafer

Annice E. Kim, James M. Nonnemaker, Brett R. Loomis, Asma Baig, Edward Hill,Influence of Tobacco Displays and Ads on Youth: A Virtual Store Experiment

http://pediatrics.aappublications.org/content/early/2012/11/27/peds.2012-0197located on the World Wide Web at:

The online version of this article, along with updated information and services, is

ISSN: . 60007. Copyright © 2012 by the American Academy of Pediatrics. All rights reserved. Print American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois,has been published continuously since . Pediatrics is owned, published, and trademarked by the Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it

by guest on May 26, 2018http://pediatrics.aappublications.org/Downloaded from