apakah perokok ingin mengetahui tentang rokok

8/12/2019 Apakah Perokok Ingin Mengetahui Tentang Rokok

1/15


2/15

290 RESULTS FROM THE EDUCATE STUDY

images portrayed in cigarette marketing (Pollay &Dewhirst. 2002; Shiffman et al., 2001b). Many of thebrand features designed into cigarettes and commu-nicated to smokers are illusory in that they are aspectsof the brand image created to engender a moreaffective response in people. Unfortunately, smokers"misperceptions about the benefits of product features

such as low tar and filter efficacy may help immunizethem to messages about the health dangers of smoking(Cummings et al., 2000; Cummings et al.. 2004:Shiffman et al.. 2001b).

Only a handful of studies have examined the impactof interventions designed to address smokers' mis-conceptions abont product design features on theirreported intentions to stop smoking and on smokingbehavior {Kozlowski et al.. 1999; Kozlowski &Pillitteri. 2001 ; Shiffman. Pillitteri. Bu rton . Ro hay .& Gitchell, 2001a). Kozlowski et al. (1999) testedsimulated radio messages and found that "light"smokers who heard that "li ght" cigarettes were just asdangerous as "regular" cigarettes expressed anincreased desire to stop smoking. Many "light"cigarette smokers use lower-tar products becauseIhey believe they are reducing their health risks bydoing so.

Shiffman et al. (2001a) tested the impact of threehealth messages on beliefs and intention to quit. Thefirst message focused on cigarette vent holes, thesecond on sensory effects and beliefs about lights andultralights, and the third on health consequences ofsmoking. They used a national sample of 2,120 dailysmo kers of regu lar (n = 975, 46%), light (H = 8 2 7 39%),and ultralight (// = 3I8 , 15%) cigarettes. Th eir studywas a random-digit-dialed telephone survey withsubjects randomized to hear one of the three messages.The effects of the messages were tested by assessmentof beliefs and quitting intentions before and afterdelivery of the message. The message focusing onsmokers" sensory perceptions of light and ultralightcigarettes resulted in the most positive change inbeliefs about safety, delivery, and intent to quit.Related marketing research has shown that smokers"perception of the "feel"" of cigarette smoke as eitherharsh or mild (also light and airy vs. hot and

irritating) is often misinterpreted by smokers asmeaning that these cigarettes have reduced healthrisks (e.g., harsher smoke is more harmful)(Kozlowski & O'Con nor, 2002; Shiffman et al..2001b). Ironically, cigarette smoke that is less irritat-ing may be more harmful to smokers because it iseasier to inhale and can be held longer in the airways.ShifTman et al. (2001a) eoncluded that addressingsmokers* sensory perceptions about light and ultra-light cigarettes would be a promising strategy forincreasing cessation.

The studies by Shiffman et ai. (2001a) andKoziowski et al. (1999) suggest that smokers willrespond to messages concerning product features and

that well-crafted media messages substantially increasmokers' interest in quitting. Development of targetebrand-specific messages is a natural extension of thework and has potential as an effective public healtintervention.

The EDUCATE (EnD Use of CigArettes ThrougEducation) study was designed to assess smoker

receptivity to receiving information about producfeatures of their cigarette brands and to test the valuof using targeted, brand-specific messages to educatsmokers about their cigarettes and to motivate themto stop smoking. The present study tested twhypotheses; (a) Smokers who receive a producinfonnation brochure targeted to their particulacigarette brand or brand category will be morlikely to read and recall information than smokewho receive information not targeted to their cigaretbrand, and (b) smokers who receive informatioaboul cigarette design features and nicotine medications will be more knowledgeable about the produfeatures of their cigarette brand and will be morlikely to make a serious effort to stop smokingcompared with those who do not receive thinformation.

Method

To test interest in the product information moduleand the value of designing targeted brochures fosmokers, we conducted a study whereby smokers whcalled the New York State Smokers" Quit Line wer

asked if they were interested in receiving informatioon the product features of the cigarette brand thecurrently smo ked. Those who answered affirmativewere randomized to one of three experimental groupto test the value of targeting the brochure to thsmokers" specific brand.

Design of the brochure

Based on our previous research investigating whsmokers know and believe about cigarette produfeatures, a brochure was developed that containe

inform ation on five different prod uct features; (a) ligand ultralight cigarettes, (b) filters, (c) mentho(d) natural or additive-free, and (e) nicotine annicotine medications (Bansal et al., 2004; Cumminet al., 2004). Each module was one page in lengthwith text written to a sixth-grade reading level, aneaeh used visual images to illustrate concepts whenever possible.

The content of each module was intended taddress sm oke rs' mispe rceptions as identified fromprevious research (Bansal et al.. 2004; Cummings et a2004; Shiffman et al., 2001b). For example, thmodule on light and ultralight cigarettes wamesmokers not to rely on their sensory perception o


3/15

NICOTINE & TOBACCO RESEARCH 2 9 1

the smoke as less harsh, mild, or smoother as a way tojudge the harm caused by the product (Shiffman et al.,2()0lb). This low-tar module also included picturesillustrating filter ventilation holes (Kozlowski et a .,1996; Kozlowski et al., 1998). The module on filtersinformed smokers that the filter does not necessarilymake smoking less dangerous because the filter makes

it easier for the smoker to inhale smaller tar particlesdeep inlo the lungs. The filter module includedinformation about what filters are made of alongwith pictures depicting how loose cigarette filter fiberscan dislodge during smoking and be inhaled into theairways (Pauiy. Allaart. Rodriguez. & Streck, 1995;Pauly et al.. 1998; Pauly. Mepani. Lesses, Cummings,& Streck. 2002; Pauly, Stegmeier. Mayer, Lesses, &Streck, 1997). The module on menthol providedinformation on what menthol is, why it is added tocigarettes, and how it may mask the harshness ofcigarette smoke, thereby facilitating deeper inhalationof smoke (Hyland, Garten, Giovino, & Cummings,2002). The module on natural or additive-freeproducts addressed what is meant by such productclaims and warned smokers not to believe that suchcigarettes are less hazardous than other cigarettes(Arnett & Terha nia n, 1998). Finally, the mod ule onnicotine attempted to address the common mispercep-tion held by smokers that nicotine in cigarettes is whatcauses health problems and reassured readers aboutthe safety of using nicotine medications as a stop-smokLng aide (Bansal et al., 2004; Etter & Perneger,2001).

Several iterations of each module were developed

based on comments received from smokers who wereshown the brochure prior to its formal testing in thepresent study. Figure 1 shows the five inform ationmodules tested.

Targeting of the brochure

Our intent was to target the stop-smoking brochuressuch that more attention would be paid to theinformational content. We took advantage of theconditioned loyalty, reinforced through marketing,that many smokers have toward their cigarette brandand particular brand features to deliver science-basedinformation about the brand features. The informa-tion in the modules was not designed to be specific toa particular cigarette brand, because product featuresare common to many different brands. Therefore, lopersonalize a brochure to a smoker's particular brand,we placed colored pictures of the smoker's cigarettebrand on the cover of the brochure, specific to thetype of cigarette smoked (regular full-fiavor, light,ultralight, or menthol). We designed targeted bro-chur e covers for the 15 mo st comm on cigarettebrands, based on market share. In the presentstudy, this meant that 56% of smokers received a

targeted broehure. For the remaining subjects, we

created a cover showing a cigarette pack labeled"Your Brand." Figure 2 illustrates the three brochurecovers (i.e., nontargeted, brand-specific, and "YourBrand") .

Study population and study design

Participation in the study was restricted to adultcurrent cigarette smokers (aged 18 years or older) whocontacted the quit line seeking stop-smoking assis-tance for themselves and who granted permission tobe contacted again by the quit line. As part of thestandard counseling protocol followed by the quitline, counse lors routinely ask for and record infon na-tion about each smoker's current smoking habits andcigarette brand. Recruitment into the study wascarried out in February and March 200?. Duringthis time period, the quit line received calls from 1,381smokers who were eligible for the study.

At the end of each call, smokers were asked if theywere interested in receiving information on theproduct features of the cigarette brand they currentlysmoked. Overall. 982 {1VV) smokers answered affir-matively and were randomized lo one of threeexperim ental group s; (a) con trol grou p (/i = 329).(b) nontargeted brochure (?) = 328). and (c) targetedbrochure (n^325). All participants received standardtelephone counseling and were mailed the quit line'sstandard cessation guide (Break Lt)ose). whichincludes some facts about ingredients found incigarettes. Participants assigned to Group 2 (non-targeted brochure) also received a separate mailingthat included a broehure entitled Secrets of YourCigarettes, wliich included the modules on the fivedifferent produet features: (a) light and ultralightcigarettes, (b) filters, (c) menthol, (d) natural oradditive-free, and (e) nicotine and nicotine medica-tions. Participants assigned to Group 3 (targetedbroehure) received the same brochure as subjects inGro up 2 except that the cover to the b rochureincluded a picture of their cigarette brand (e.g..Marlboro Lights) or a generie brand labeled "YourBrand." In addition to receiving quit materials in themail, all participants were sent a letter thanking them

for calling the quit line and indicating that someonefrom the quit line would be calling them in 6-8 weeksto check on their progress in stopping smoking and toask them a few questions about the materials that hadbeen mailed to them.

Follow-up

Study subjects were contacted by telephone 6-8 weeksafter calling the quit line to confirm receipt ofmaterials sent to them in the mail, and to collectinformation about their smoking behavior and beliefs

about cigarette design features (i.e.. filters, low tar.


4/15

2 9 2 RESULTS FROM TH E EDUCATE STUDY

Y o u r L ijih t C i g a r e t t e sMenthol Cigarettes^.NotSoCool>,,

i ti MavM In

u n f l T U R R l O ie R S O T T B So s o Nicotine...Friend or Foe?

n w H U H I F TM Tunen n M H W I Wnrmnmiwmmna roDtniF

H M M Mv

tMvMnii

A W l ( t ^ M * 1 H IaHx ^ n ^11 Ik M> M mr > iwa i ^

(d) (e)

Figure 1. Five product information modu les tested in the study: (a) light and ultralight cigarettes, (b) filters(c) menthol, (d) natural or no-additive cigarette claims, and (e) safety of nicotine and nicotine medications.

menthol, no additive, and nicotine medications). A6- lo 8-week follow-up period was selected lo permitsufficient time for callers to receive and read thematerials and to facilitate recall of the informationmailed to them. The follow-up interview wa.s com-pleted with 506 of the 982 participants recruited intothe study, yielding an overall response rate of 52 A^.The main reasons participants were lost to follow-upwere failure to reach participants after five callbackattempts (311/476, 66%), incorrect or disconnected

telephone numbers (88/476. 18%). and refusals (77 7 6 16%). An analysis comparing the characteristicof responders and nonresponders revealed that nonresponders were more likely to be male, younger ia g e and non-White and were more likely lo repornever quitting before.

The response rate was similar between the controgroup and the two experimental groups. Traineinterviewers who were blinded as to which studgroup participants were assigned conducted follow-u


5/15

' ' ' I

NICOTINE Si TOBACCO RESEARCH 2 9 3

o f your i g r e t t e s

y u r

i g r e t t e sof your

cigarettes

U H R B R N D

\ l i i r l l t t ini

(a) (b)

Figure 2. Three brochure covers used in the study: (a) nontargeted cover (Group 2), (b) targeted cover with thesmoker's specific cigarette brand displayed (Group 3), and (c) plain pack cover ( Your Brand ) used with smokerswho smoked a low-market-share brand (Group 3).

interviews. The telep hone survey included 122 questionsdesigned to assess receipt and attention paid to themailed miiterials, beliefs about the cigarette designfeatures, and smoking behavior.

Outcome measuresReceipt and attention paid to the materials. Allstudy participants were asked if they recalled receiv-ing the Break Loose stop-smoking guide sent tothem. Those who answered anirmatively were askedto indicate whether they had read all of it. someof it, or none of it. Subjects also were asked ifthey recalled receiving the Secrets of Your Cigarettesbrochure sent to them. Those who answeredaffirmatively were asked to indicate whether theyhad read all of it. some of it, or none of it. Subjectsin groups 2 and 3 who reported having read some

of or the entire Secrets of Your Cigarettes brochurealso were asked to recall specific informa tion abo uteach of the five modules in the brochure. Unaidedrecall was assessed by asking subjects to name thespecific subject areas covered in the brochure. Apositive response to each of the five topic areas wasscored as unaided recall. Subjects who could notrecall any specific topics were prompted to recalleach topic area (aided recall) by first informingthem about each module (e.g., we told them therewas a section in the brochure describing light andultralight cigarettes) and then asking them to indi-cate whether they remembered seeing this section of

the brochure. Overall recall of each of the fivetopics covered in the brochure was determined bycombining unaided and aided recall.

I

Beliefs ahout product design features. Respondents

were asked to assess the truth of a series of state-ments related to the health risks and delivery fea-tures of the following cigarette design features: lightcigarettes, filtered cigarettes, menthol cigarettes, no-additive cigarettes, as well as the safety of nicotineand nicotine medications. Respondents were askedto indicate whether they agreed or disagreed witheach statement. If they agreed or disagreed with astatement they were asked to indicate the strengthof their belief (i.e., either strongly or somewhat).Respondents who were unsure were assigned a scorein the midpoint of the scale. Scores on each beliefitem ranged from 0 to 4; higher scores reflectedgreater awareness that the product feature madethe cigarette as dangerous to health as a regularcigarette. Table 1 displays the ind ividual item s and areliability coefficient for each index.

Light cigarettes. Subjects were asked to respond tothe following series of statements about lightcigarettes (i.e., strongly agree to strongly disagree):"Light cigarettes make it easier to quit;" "Lightcigarettes are less likely to cause cancer compared toregular cigarettes;" "The reduction of tar has madecigarettes less dangerous to smokers;" "Lightcigarettes are smoother on your throat:" "Smokers


6/15

2 9 4 RESULTS FROM THE EDUCATE STUDY

RE ffl

O 03

iT3

w cu o 2

111Q. tu OCO o i :O) ffi .O

ra:2 I

5

o

; t t

1- o CO COc i

CM00CM

Ocvj

0)

oT3tu

S

^ (T ^ ?: olo ro t o '1' t :g -o >

i^ *" c o> m ro o a, 5 o .g> -

O (D W Q d) P .^3 ra Q) o = Q> So 2^..c E c :^ x;

' >- U I 3 U- O


7/15

_to o

= ro

o oO . 9

o o ^

a8

S -0 5

O

2

NICOTINE TOBACCO RESEARCH 2 9 5

COi n

oiCMi ncsi

0 0CMCM

tCM 5

cto

a.3O

r-NCMCO

CT)CMCO

COinCO

i nCO

COi nCO

CO

cvi

CMtncsi

cnCO i nCM

CDCOCM

toc\CM

CMCOOJ

IT )

CM

COi nCO

i nCO

COCO

r--toCM

r

COoCOc^

ocy

C3JCvjCM

i nCM

t DCMCM

COCMCM

a)Ero

Uv

COCMCO

cnCMCO

i ni nCO CO

i nCO

CD

CM

CJ)tnCM

COc\i

COT f

c\i

i nCOCM

COCMCM CM

coCM

COi no i

oCO

ina.CJ

COCO

d

to

le

3O> ,

c

tl )

(0to 0)c to

ra

cra

o

6z

CE

ToCO0)

to

l _

sO

o**ffi

f

o

bra

CC

ae

to

ffra

>

T 3to

o

>Cl)

IP CT> to

-: Q). 7 . ; *- *

ro (0

CO0

jd

v

_c O) CO. 0

>

te d

0) ^ra O) 00 2

ie

tu

.i=c

to"5

ae

CO

oicuCOO )

5

>

d^ 3ro0

to

id

v

to

CO

e

t u

0

cto

6

C3>

c0

fS

(A03

O )

0

0

ffraDro

t5Q_

E

te

0

.CT0

Ql

tv

a0cra

CD

1

to501

4

>

0

E

o o c f ra ra

^ p O EZ

. rao to

1 1ro CO mto Q) a) ^ o o)s; to jco ff oc

:= 0) 5 3

E O QJ *

EP o

Q.O

CD CO It3CD CO CO

0)

ra0

ra

M

to

3

g

ir iqVa

O) t- m

to ^

,;^ro oO)ca0

lo0

ra.c0)m

Cg

to

.9

to

to 5to

ae

5cco5

ffi

.2c

CO

cro01in

a.0

0V)

-0 c0

S

0

roO )

dn

ro

re

TJ

raff3

JZ

0

( U

D )

"5

ee

c

Wh

cra

O 0)

f

to R o .

fill-

ll t

S8 8 Sw ^

o ^


8/15


of light cigarettes lake in less tar than smokers ofregular cigarettes:"" "Light cigarettes feel easier onyour chest than a regular cigarette:"' and "Youcough less smoking light cigarettes compared toregular cigarettes."' A single composite variablereflecting the respondents' beliefs about lhe benefitsof light cigarettes was constructed based on these

items. Scores on the index ranged from 0 to 4;higher scores reflected greater awareness that lightand ultralight cigarettes are as dangerous to healthas regular cigarettes. To assess respondents" aware-ness of how the tar and nicotine levels ofa cigaretteare influenced by how a person smokes the cigarette(i.e., deeper puffs, coverage of vent holes in thefilter), we asked the following two true-falsequestions: (a) "The way a smoker pufl"s on acigarette can affect the amount of tar and nicotine asmoker takes into their bodies,"" and (b) "The waya smoker holds a cigarette can affect the amount oftar and nicotine a smoker takes into their bodies."'A composite variable, reflecting the respondents"beliefs about how a person's smoking behavior caninfluence tar and nicotine delivery, was constructedby combining the responses to these two items(index labeled "behavior"').

Filtered cigarelte.s. Respo ndents w ere asked t oindicate their level of agreement (strongly agree tostrongly disagree) with the following series ofstatements about filtered cigarettes: "Filtered cigar-ettes are cleaner than unfiltered cigarettes:"" "Fil-tered cigarettes are safer than unfiltered cigarettes:""Filtered cigarettes are easier to quit smoking than

unfiltered cigarettes:"" "Filtered cigarettes give youless tar than unfiltered cigarettes;"' "You cough lesssmoking a filtered cigarette compared to anunfiltered cigarette;" "The smoke from a filteredcigarette feels smoother on your throat than anunfiltered cigarette;" and "Filtered cigarette smokefeels easier on your chest compared to an unfilteredcigarette." A composite variable reflecting therespondents' beliefs about the benefits of filteredcigarettes was constructed from these items. Scoreson the index ranged from 0 to 4: higher scoresreflected greater awareness that filtered cigarettes

are as dangerous to health as unfiUered cigarettes. Menthol cigarettes. Re.spondents were asked toindicate their level of agreement (strongly agree tostrongly disagree) with the following series ofstatements about menthol cigarettes: "Mentholcigarettes give you less tar than regular cigarettes;""Menthol cigarettes are safer than regular cigar-ettes:"" "Menthol cigarettes are cleaner than regularcigarettes;"' "Menthol cigarettes are easier to quitsmoking than regular cigarettes;" "Menthol cigar-ettes are smoother on your throat than regularcigarettes;"' and "Menthol cigarettes feel easier onyotir chest compared to regular cigarettes." Acomposite variable reflecting the respondents'

beliefs aboul health risks of menthol cigaretteswas constructed based on these items. Scores on thindex ranged from 0 to 4; higher scores rcfiectcdgreater awareness that menthol cigarettes are adangerous to health as regular cigarettes.

No-additive cigarettes. Respondents were asked toindicate their level of agreement (strongly agree to

strongly disagree) with the following series ofstatements about no-additive cigarettes: "Noadditive cigarettes are cleaner than cigarettes withadditives:"" "No-additive cigarettes are safer Ihancigarettes with additives;"" "No-additive cigarettegive you less tar than cigarettes wilh additives;"You cough less smoking a no-additive cigarettecompared to a cigarette with additives;'" "Thesmoke from a no-additive cigarette feels smootheon throat than the smoke from a cigarette withadditives;'" and "No-additivc cigarelles are easier toquit smoking than cigarettes with additives.' ' Acomposile variable measuring the respondentsbeliefs about the benefits of no-additive cigarettewas constructed from these items. Scores on theindex ranged from 0 to 4; higher scores reflectedgreater awareness that no additive cigarettes are adangerous to health as cigarettes with additives.

Nicotine. To measure respondents" beliefs about thesafety of nicotine medications, they were asked texpress their level of agreement (strongly agree tstrongly disagree) with the following statement"Stop-smoking medications may harm youhealth." To measure respondents' beliefs abouthe etTectiveness of nicotine medications, respondents were asked to indicate their level of agreemen(strongly agree to strongly disagree) to the followinstatement: "If you decided you wanted to quit, stosmoking medications would make it easier.'"

Smoking behavior and quit methods. All respond entswere asked if they had made an attempt to stop smoking since calling the quit line 6-8 weeks earlier, Thoswho indicated they had made a qtiit attempt werasked if they had managed to stay off cigarettes for aday or longer and if they had used nicotine medications to help them stop smoking. Respondents alswere asked lo indicate if they currently smoke everyday. on some days, or not at all. Respondents whoself-reported themselves as not smoking at all werasked if they had smoked in the past 7 days. Thoswho reported they were not smoking currently and hadnot had a cigarette in the past 7 days were classified ahaving quit smoking.

Data analysis

Descriptive statistics, such as simple percentages anmeans, were used to display respondents" interest ireceiving information about the product features o


9/15

NICOTINE TOBACCO RESEARCH 2 9 7

Table 2 , Com parability of baseline characteristics of responden ts assigned to the three experim ental groups^.

Characteristic

Subsample sizeResponse rateGender

Male

FemaleAge (years)Mean

18^2425-3435-4445-54a 55

RaceWhite, non-HispanicBlack, non-HispanicHispanicOther. non-Hispanic

Tar levelRegular, full flavorLightUltralight

MentholNonmentholMenthol

Health insuranceNoYe s

Cigarettes smoked per day1-9 (2 packs)

Number of years reported smoking1-5

6-1011-1516-2021 +

Previous quit attemptsNoYes

n

164

71

93

466

224 74536

1152512

6

7 66119

10452

23131

19385222

69

10

4

102018

103

10136

Group 1

%

32.4%49.8%

43.3%

56.7%

3.8%14.1%30.1%28.8%23.1%

72.8%15.8%

7.6%3.8%

48.7%39.1%12.2%

66.7%33.3%

14.9%85.1%

12.2%24.4%33.3%14 .1%

3.8%5.8%6.4%

2.6%

6.5%12.9%11.6%66.5%

6.8%93.2%

a

17 4

82

92

448

37414340

1222412

8

836216

10458

28134

20415122

819

3

6

131918

109

1414 4

Group 2

%

34.4%53,0%

4 7 . 1 %

52.9%

4.7%21.9%24.3%25.4%23.7%

73.5%14.5%

7.2%4.8%

51.6%38.5%

9.9%

64.2%35.8%

17.3%82.7%

12.2%25.0%31 .1%13.4%

4.9%11.6%

1.8%

3.6%

7.9%11.5%10.9%66.1%

8.9%91.1%

n

168

64

104

441428403943

11827

86

855816

9463

14137

18464818

416

8

9

18181698

17135

Group 3

%

33.2%51.7%

38.1%

61.9%

8.5%17 .1%24.4%23.8%26.2%

74.2%17.0%

5.0%3.8%

53.5%36.5%10.1%

59.9%4 0 . 1 %

9.3%90.7%

11.4%29.1%30.4%11.4%

2.5%10.1%

5.1%

5.7%

11.3%11.3%10.1%61.6%

11.2%88.8%

Total

n

506

217

289

452887

128127119

35 57 63220

24 4181

51

30 2173

65402

57125151

62184421

19

415752

31 0

41415

sample

%

100.0%52.0%

42.9%

5 7 . 1 %

5.7%17.8%26.2%26.0%24.3%

75.2%14.4%

8.7%1.6%

51.3%38.0%10.7%

65.8%34.2%

13.9%86.1%

11.9%26.2%31.6%13.0%

3.8%9.2%4 .4%

4.0%

8.6%11.9%10.9%64.7%

9.0%91.0%

Note. ^No statistically significant differences were found w ithin the tobacco use characteristics between respondents in the three study arms.

their cigarette brand and beliefs about productfeatures and nicotine medications. To evaluate theequivalency of the control group and the tvt-o

experimental groups on demographic and smokinghistory variables assessed at baseline, either the chi-square statistic for 2 x 2 contingency table analysesor tests for homogeneity of variance with analysis ofvariance were used to test for group differences. Theresults of these analyses are shown in Table 2 an dreveal that randomization was successful in creat-ing equivalent groups, making statistical adjustmentfor potential confounders unnecessary. The Mann-Whitney U-test was used to compare scores on beliefindices to test for differences between respondentswho did and did not receive the product information

brochures. We also tested for interactions between

groups on responses to belief indices by type ofcigarette smoked (i.e., light or ultralight, filtered,menthol, no additive) under the assumption that these

baseline features could moderate the impact of theproduct informatioti brochure provided to respondents.

Results

A total of 1,381 adult smokers called the quit lineduring the study period, 982 (71%) of whom agreed tobe enrolled in the study. Of the 399 smokers whochose not to be enrolled in the study, all did sobecause they did not want to be called back. No onerefused the offer to receive infonnation about his or

her cigarette brand.


10/15


Receipt and attention paid to the materials

Table 3 shows responses to questions abou t readingand recall ofthe materials by experimental condition.All respondents were sent the Break Loose stop-smoking guide, and about 86% of subjects recalledreceiving this guide. Among those who recalled gettingthe guide, 91 % reported having read all or some of it.As expected, there were no marked dilVerencesbetween the three groups in recal of the BreakLoose guide.

Targeting the brochure did not make a difference interms of the amount of the guide that was read or inthe recall of specific information in the guide. Overall,81% of subjects sent the Secrets of Your Cigarettesbrochure reported reading all or some of it. Unaidedrecall of the specific content areas of the brochureamong those who reported having read some or ail ofit was very low (less than 5% for most sections). Aidedrecall measures revealed somewhat higher levels of

awareness of the content of the brochure, but nomarked difTerences were seen between subjects whoreceived the brochure with a cover customized to theircigarette brand and those who received the brochurewith the basic cover (cigarette filters module; 71.8%recall for G rou p 2 vs. 84.1% recall for G rou p 3,p = .Q65). Overall awareness of the c ontent of thebrochure was highest for the section on cigarette filtersand low-tar (light) cigarettes (cigarette filters module:76.1% recall for G rou p 2 vs. 87.8% recall for Gr ou p 3./7=.O57; low-tar module: 69 /,, for Gr ou p 2 vs. 79.3%for Group 3, ^= .14 7) .

Beliefs ahout product features

As shown in Table 1, mean belief scores did not differsignificantly between the three experimental groups.However, among respondents who were sent theinformation brochure (Groups 2 and 3), mean beliefscores for light cigarettes, filters, and no-additives weresignificantly higher among respondents who reportedreading all or some ofthe brochure, compared with thosewho reported not reading the brochure. Stratification bywhether or not the respondent currently smoked a lightcigarette brand, had ever smoked a menthol brand, hadever smoked a no-additive cigarette brand, or had everused nicotine m edications did not alter the findings (datanot shown: tables available on request from thecorresponding author).

Smoking behavior

Table 4 displays the percentage of respondents whoreported quitting for a day or longer, the percentageof respondents who reported using nicotine medica-tions to help them stop smoking since calling the quitUne, and the percentage that were not smoking at thetime of the follow-up call, by experimental group. No

differences between experimental groups were statistically significant at a p value of less than .05. Overall,46% of respondents reported having quit for a day orlonger, 43% reported having used nicotine m edicationin their quit attem pt, and 15% reported not smokingfor at least 7 days prio r to the foUow-up interview.

Discussion

In the present study, we attempted to test twohypotheses, one focused on whether providing smokers with information about product features caninfluence iheir beliefs and behaviors and the otherfocused on whether such information is acceptableand interesting to smokers who call a quit line. For avariety of reasons, the study turned out to be a weaketest of the first hypothesis than the second. We feeconfident in our conclusion that smokers are receptiveto receiving information about the features of the

cigarettes they smoke. However, the impact ofthe product information brochures we tested on theknowledge, beliefs, and behavior of smokers is lesclear-cut, because of several limitations in the studydesign. The text-based product information sheets wetested were not especially memorable to respondentand had no measurable impact on their beliefs abouproduct features or smoking behavior. Some evidencindicates that beliefs were altered among those whoreported reading the information sheets. Thus, iappears that the challenge is to produce messages towhich smokers will attend. Targeting lhe cover of theproduct infonnation brochure to smokers' cigarettebrands increa.sed recall of the brochure slightly bunot enough to alter beliefs about product features osmoking behavior.

The nature of the population we studied could beone explanation for our failure to observe a markedeffect of the product information sheets. Smokers whocall a quit line are already highly motivated to stopsmoking by virtue of the fact that they have calledasking for help in quitting smoking. Logically speaking, information about product characteristics mighbe expected to be a source of motivation for smokerto quit rather than a practical quitting tool. Thereforethe subjects in the present study might not have beenthe ideal audience for the information provided onproduct characteristics, given that they already had ahigh level of motivation to make a quit attemptFuture studies should consider testing the value ogiving product information to smokers who may onlybe contemplating quitting. One possible venue fosuch a study could be offering such information tosmokers when they purchase cigarettes.

Another limitation of the study is the 6-weekaverage lag time between sending the productinformation sheets to smokers and assessment oftheir knowledge and beliefs. Recall of the materia


11/15


Table 3 .

Question

R e ce ip t a n d r ec all o t in te rv en ti on b ro ch ur e.

Group

n

1

%

Group

n

2

%

Group

n

3

%Chi-square statistic

p value

Sample sizeDo you recall receiving the Break Looseguide that we sent you in the mail?

Ye s

No/not SureDid you read all, some, or none ot the BreakLoose guide we sent you in the mail?

AllSomeNone

Do you recall receiving the brochure on the Secretsof Your Cigarettes, which had specific informationon different product characteristics of cigarettes?

Ye sNo/not sure

Did you read all, some, or none of the Secretsof Your Cigarettes we sent you in the mail?

AliSomeNone/Not Sure

Unaided recallThe brochure had separate sections givinginformation about different characteristics ofcigarettes, pictures describing these differentcharacteristics, and quotes from the tobaccoindustry. Do you remember the topics of anyof these specific sections?n = 1 5 3Cigarette filters- Your cigarettefilter may be killing youLow-tar cigarettes-'The darkside of light cigarettesMenthol cigarettes- Menthol-not so coolNatural, no-additive cigarettes- Naturalcigarettes, not soNicotine safety/stop-smokingmedications- Nicotine, friend or foeAided recallThere was a section in the brochuredescribing [product characteristic].Do you remember seeing this section?n = 1 5 3Cigarette filters- Your cigarettefilter may be killing youLow-tar cigarettes- The dark sideof light cigarettesMenthol cigarettes- Menthol-not so coolNatural, no-additive ciQarettes- Naturalcigarettes, not soNicotine safety/stop-smokingmedications- Nicotine, friend or foeOverall recallThe brochure had separate sections givinginformation about different characteristicsof cigarettes, pictures describing thesedifferent characteristics, and quotes fromthe tobacco industry. Do you rememberthe topics of any of these specific sections?n = 1 5 3Cigarette fiifers- You r cigarette ,filter may be killing youLow-tar cigarettes-'The dark sideof light cigarettesMenthol cigarettes- Menthol-not so coolNatural, no-additive cigarettes- Naturalcigarettes, not soNicotine safety/stop-smokingmedications- Nicotine, friend or foe

164 32.4% 174 34.4% 168 33.2%

137 83.5%

27 16.5%

71 51.8%53 38.7%13 9.5%

N/AN/A

N/AN/AN/A

N/AN/A

N/AN/AN/A

N/A N/A

147

27

84.5%

15.5%

153

15

91.1%

8.9%

78 53.1% 79 51.6%55 37.4% 61 39.9%14 9.5% 13 8.5%

8886

50.6%49.4%

10068

59.5%40.5%

50 56.8% 50 50.0%21 23.9% 32 32.0%17 19.3% 18 18.0%

0.0% 0.0%

.09

.99

.10

.460

N/A

N/A

N/AN/A

N/A

N/A

N/AN/A

3

1

13

4.2%

1.4%

1.4%4 .2%

3

1

21

3.7%

1.2%

2.4%1.2%

.86

.92

.65

.25

.00

N/A

N/A

N/AN/A

N/A

N/A

N/A

N/AN/A

N/A

51

48

4235

46

71.8%

67.6%

59.2%49.3%

64.8%

69

64

4435

60

84 .1%

78.0%

53.7%42 .7%

73 .2%

.07

.15

.49

.4 1

.26

N/A

N/A

N/AN/A

N/A

N/A

N/A

N/AN/A

N/A

54

49

4338

46

76 .1%

69.0%

60.6%53.5%

64.8%

7 2

65

4636

60

87.8%

79.3%

56.1%43.9%

73.2%

.057

.147

.577

.235

.262

Note. N/A, not applicable.


12/15


Table 4 . Tob acco use characteristics of participants who compteted the 6-week follow-up survey^.

Group 1 Group 2

Respondents whodid not receive or

Group 3 read none of brochure

Respondents whoread all or some

of brochure

Question n

Sample sizeHave you been off cigarettes

for at least a day or longer?YesRespondents who report theyare currently not smoking andhave not smoked even a puffin the past 7 days

Not currently smoking and nocigarette in the past 7 days

Since we last spoke with you,have you used any nicotine medication?

Yes

164 32.4% 174 34.4% 168 33.2% 306

73 44.5% 83 47.7% 76 45.2% 130

25 15.2% 25 14.4% 27 16.1 % 41

55 33.5% 72 41.4% 61 36.3% 103

60.5% 200 39.5%

42.5% 102 51.0%

13.4% 36 18.0%

33.7% 85 42.5%

Note. ^No statistically significant differences w ere found within the tobacco use characteristics betw een respondents in the three studyarms; respondents who read some or alt vs. none of the brochure were significantly more likely to report using nicotine medication,

sent in the mail was low, making assessment of effectson knowledge and beliefs difficult and likely to beunderestimated. In retrospect, tbe study would havebeen .strengthened had we done an assessment within aday or two of respondents receiving the productinformation sheets in the mail.

Until recently, most efforts to educate smokers havefocused on providing information about health risksof smoking. We believe tbis is the first study to bavetested the effectiveness of printed product infonnationbrochure on smokers" beliefs and behavior. We choseto test a print communication because it was

inexpensive and most closely represented wbat mighteventually be used to educate smokers about productfeatures (i.e.. as a mandated product insert). Severaltobacco companies bave started including informationinserts in certain brands and brand styles of theircigarettes. For example. Philip Morris began includingan insert on packs of its light and ultralight cigaretteswarning smokers not to assume that productsdescribed as ultralight, light, medium, and mild areless harmful than full-flavor cigarettes. Philip Morrisalso has distributed product information to consumersin muitipage newspaper inserts.

The negative results from the present study suggestthat recent efforts by cigarette manufacturers toaddress educational deficits of their customers byproviding text-laden product information inserts incigarette packages or in newspaper inserts are unlikelyto be effective. While disappointing, tbe results of thestudy are not especially surprising. First., it is ourimpression tbat reliance on text-oriented print mes-sages to convey product information to smokers islargely ineffective In comparison witb tbe visualimages and colors used by cigarette marketers tocreate and reinforce illusions about these sameproduct features (Cummings.. Morley, & Hyland,

2002; Pollay & Dewhirst. 2002: Wakefield. Morley,

Horan. & Cummings, 2002). The images and colorused in cigarette packaging and advertising areintended to convey a positive feeling about brandfeatures, which influences how a F>erson evaluates andthus chooses cigarette brands and brand styles. Wordsalso are important to consider, but eliminating termssuch as light, mild, an d smooth from cigaretteadvertising will not prevent cigarette marketers fromcontinuing to use images and colors to communicatinformation about product features in their brandadvertising. Tobacco control efforts intended locounteract misleading product advertising also need

to rely more on tbe use of images to communicatehealth information to smokers. For example, evidenceis now emerging that graphic health warnings oncigarette packs are more effective tban text-basedwarnings {Fong et al.. 2004; Straban et al.. 2002). Onereason for this is tbat grapbic depictions of diseaseslead viewers to imagine the disease more easily, whichleads them to perceive the health threat as more likelyGraphic images also are likely to capture viewersattention, thus enhancing persuasion by focusing theviewer on the meaning of the message.

Another reason that the product informationbrochure tested in the present study failed to influencebeliefs and behavior was that it did not register withtbe respondents. This observation is not surprisinggiven that we relied on a single effort, a phenomenonborne out in other studies (Flay. 1987). In retrospectthe odds of having respondents pay attention to tbeinformation sheets probably would have improvedhad the sheets been mailed to participants on multiploccasions. Better targeted imagery and inlbrmationabout cigarette brands could have a greater impacand thus be better remembered. Recall of th einformation sheets may have been enhanced badinformation sheets been mailed out individually ratbe

than bundling them into a single brochure. Frequen


13/15


exposure to the messages conveyed in cigaretteadvertising is one reason that sucb advertising iseffective (Warner. 1986). Finally, a one-time interven-tion, no matter how well designed, is likely to beincapable of overcoming decades of misinformationthat smokers have been exposed to through cigarettebrand marketing.

Requiring cigarette companies to include graphicpackage warnings and informational inserts thataddress specific areas in which smokers lack knowl-edge has many advantages from a public healthstandpoint. Package warnings or inserts will confrontsmokers at least as often as advertising in tbe media.An individual who smokes one pack per day, forexample, is potentially exposed to the bealth warning7.300 times in a single year. Second, tbe fact that theinformation would be included with the smokers"cigarette brand could help counteract some of themisleading perceptions about product features con-veyed in the brand's marketing. Few. if any, health

promotional messages are ever delivered in such anadvantageous situation or position, and none aredelivered witb such frequency.

The primary aim of the EDUCATE study was totest whether smokers wbo call a telephone quit linewould be interested in and benefit from receivinginformation about the features of the cigarette brandthey smoke. We found that smokers are receptive toreceiving such information; however, the productinformation brochure tested in tbe study did nothave a sizable effect on smokers' beliefs about productfeatures. To address areas in which smokers are

misinformed about product features (e.g., low tar,filter vents, no-additives claims), governments shouldrequire cigarette manufacturers to place graphicwarnings on packs and to include informationalinserts in packages designed to educate smokersabout product features. In addition., all cigarettebrand advertising should be prescreened by anindependent group of marketing experts to preventthe use of claims and imagery that may misleadconsumers about tbe safety and benefits of a product.

Acknowledgments

This project was supported partially by grants from the Robert WotidJohnson Foundation, the American legacy Foundalion. and ihcNalional Cancer Inslitule Cancer Center Support Grant (CAtft(l56).Special (hanks are due lo Anthony Brown, who helped design iheproduct informalion sheets, and to Paula Celesiino and the staff ofthe New York Slate Smokers' Quii Line, who were respon.siblc forthe recruicment of subjects into this study.

References

Arnett. J. J.. & Terhanian. G. (1998). Adolescents' responses tocigarette advertisements: Links between exposure, liking, and theappeal of smoking. Tohucco Conirol. 7. 129-133.

Bansal. M. A.. Cummings. K. M.. Hyland. A., & Giovino, G. A.

(2004). Slop-smoking mediaitiotis; Who uses them, who misuses

them, and who is misinformed about them? Nicotine & TobaccoRi-smnh. rt(Sup|>l. 3), S289 S302.

Cunimings. K. M.. Hastrup. J. L., Swedrock, T., Hyland, A., Perla, J.,& Pau ly, J. L. (2(H)0). Con sum er percep tion of risk associate d w ithtiiters contaniinaled with glass fibers [comment]. Cancer Epidemiol-ogy. Biomarkers A Prevcriiion. V. 977-979.

Cummings, K, M.. Hyland. A., Giovino, G. A., Hastnip. J., Bauer, J.,& Bansal, M. A. (2(H)4). Are smokers adequately informed aboutthe health risks of ,imoking and medicinal nicotine? NicoiitK dTobacco Research. fi(Suppl. 3), S333-S34(t.

Cumm ings, K. M., Hytand, A ., Pechacck, T. F., Orlandi, M ., &Lynn, W. R, (1997). Comparison of recent trends in adolcsccnl andaduli cigarette smoking behavior and brand preference. TobaccoConirol. rt(Suppl. 2), S3I-S37.

Cummings, K, M.. Morley, C. P.. & Hyland, A, (2002). Failedpromises of the cigarette industry and ils elTect on consumermisperceptions about the health risks of smoking. Tobacco Control.;/(S up pl. 1), 1110-1116.

Eller, J. F., & Perneger, T. V. (2001). Allittidcs toward nicotinereplacement therapy in smokers and ex-smokers in the generalpublic. C linical Pbarmacohgy and Therapeutics. 6V, 175-183.

Flay, B. R. (19S7). Selling the smokeless society: 56 Evaluated massmedia progranjs and ranipaign.v iwrtitxvide. Paper presented al theAmerican Public Health Association, Washington, DC.

Fong, G. T., Hammond. D., Borland, R., Hastings. G., & Cummings,K. M. (2004. February). Quasi-experimental evaluation of theenhancement of warning labels in the VK . Paper presented at theAnnual Meeting of the Society for Research on Nicotine andTobacco, Phoeni.i, AZ.

Hamiiton. W. L., DiSiefatio Norton, G.. Oucllctte, T. K.. Rhodes,W. M., & Connolly. G. C. (2004). Consumer responses toadverlisements for ptitcntial reduced-exposure products, Niiotim-


14/15


{2()01a). EfTect of health messages about "lig ht" an d "ultr a packag e warning labels: A social psychologicat perspective. Tobaccolight"' cigarcUes on beliefs and quitting inienl. Tobacco Control. Control, il. 183-190./mSuppt. I). 124-132. Wakefield, M . Morley. C . Hora n. J.. & Cumm ings. K. M. (2002

Shiflman, S.. Pillitteri. J. L., Burton, S. L., Rohay. J. M.. & Gitchell. The cigarette pack as image: New evidence from tobacco induslrJ. G. (2(K)lb). Smokers' beliefs about "lighl" and "ultra light" documents. Tobacco Control. //(Suppi, I). 162-172.cigareltes. Tohacru Control. /(Suppl, 1). 117-123, Wa rner. K. E. (1986), Selling smoki'.- Cigarette advertising an d public

Strahan, E. J.. White. K., Pong, G. T,. Fabrigar. L. R., Zanna, M. P., health. Paper presented LU ihe American Public Health Association,& Cam eron . R. (2002). Enhancing the etTectiveness of lobacco Was hington. DC.


15/15

apakah perokok ingin mengetahui tentang rokok

Documents