What is the truth an application of the

Thus, the EPPM cautions health campaign-ers against communicating the threat without highlighting the recommended response Witte, 1992.. For exam-ple, messages modeling the advocated beh

On: 02 Sept em ber 2013, At : 19: 19

Publisher : Rout ledge

I nfor m a Lt d Regist er ed in England and Wales Regist er ed Num ber : 1072954 Regist er ed office: Mor t im er House, 37- 41 Mor t im er St r eet , London W1T 3JH, UK

Health Communication

Publ icat ion det ail s, incl uding inst ruct ions f or aut hors and subscript ion inf ormat ion:

ht t p: / / www t andf onl ine com/ l oi/ hht h20

What Is the Truth? An Application of the Extended Parallel Process Model to Televised Truth® Ads

Nicol e R LaVoie a & Brian L Quick a a

Depart ment of Communicat ion, Universit y of Il l inois at Urbana–Champaign Publ ished onl ine: 18 Jan 2013

To cite this article: Nicol e R LaVoie & Brian L Quick (2013) What Is t he Trut h? An Appl icat ion of t he Ext ended Paral l el

Process Model t o Tel evised Trut h® Ads, Heal t h Communicat ion, 28: 1, 53-62, DOI: 10 1080/ 10410236 2012 728467

To link to this article: ht t p: / / dx doi org/ 10 1080/ 10410236 2012 728467

PLEASE SCROLL DOWN FOR ARTI CLE

Taylor & Francis m akes ever y effor t t o ensur e t he accuracy of all t he infor m at ion ( t he “ Cont ent ” ) cont ained

in t he publicat ions on our plat for m How ever, Taylor & Francis, our agent s, and our licensor s m ake no

r epr esent at ions or war rant ies w hat soever as t o t he accuracy, com plet eness, or suit abilit y for any pur pose of t he Cont ent Any opinions and view s expr essed in t his publicat ion ar e t he opinions and view s of t he aut hor s, and

ar e not t he view s of or endor sed by Taylor & Francis The accuracy of t he Cont ent should not be r elied upon and should be independent ly ver ified w it h pr im ar y sour ces of infor m at ion Taylor and Francis shall not be liable for any losses, act ions, claim s, pr oceedings, dem ands, cost s, expenses, dam ages, and ot her liabilit ies w hat soever

or how soever caused ar ising dir ect ly or indir ect ly in connect ion w it h, in r elat ion t o or ar ising out of t he use of

t he Cont ent

This ar t icle m ay be used for r esear ch, t eaching, and pr ivat e st udy pur poses Any subst ant ial or syst em at ic

r epr oduct ion, r edist r ibut ion, r eselling, loan, sub- licensing, syst em at ic supply, or dist r ibut ion in any

for m t o anyone is expr essly for bidden Ter m s & Condit ions of access and use can be found at ht t p: / /

w w w.t andfonline.com / page/ t er m s- and- condit ions

Health Communication, 28: 53–62, 2013

Copyright © Taylor & Francis Group, LLC

ISSN: 1041-0236 print / 1532-7027 online

DOI: 10.1080/10410236.2012.728467

What Is the Truth? An Application of the Extended Parallel

Nicole R LaVoie and Brian L Quick

Department of Communication University of Illinois at Urbana–Champaign

The purpose of this study was to analyze television ads in the truth® campaign using the

Extended Parallel Process Model (EPPM) as a framework Among the ads (n = 86)

ana-lyzed, results revealed a heavy reliance on severity messages, modest attention to susceptibility messages, and no inclusion of recommended response messages in the form of self-efficacy and response efficacy The reliance on emphasizing the health threat, without incorporating recommended response messages, is discussed with respect to the likelihood of galvaniz-ing maladaptive responses such as psychological reactance, denial, and defensive avoidance resulting from exposure to these ads Additionally, the unintended outcomes for secondary audiences including but not limited to stigma are considered Implications and suggestions for practitioners and theorists are explored

Tobacco smoking is a costly health behavior, both for

indi-viduals who smoke and for the broader community who pays

for tobacco-related disease According to the World Health

Organization (WHO, 2011a), smoking is the number one

contributor to preventable death in the world Further, for the

United States alone, the financial cost associated with these

preventable diseases is estimated at $193 billion per year

(Centers for Disease Control and Prevention [CDC], 2011)

Despite these rising health costs, however, 26.4% of U.S

adults above the age of 15 years currently smoke, and 15.7%

of adults report daily cigarette smoking (WHO, 2011b) On a

positive note, smoking prevalence has decreased over the

last decade or so Among ninth to 12th graders, for

exam-ple, smoking decreased from 35% in 1999 to 20% in 2009

(CDC, 2011; National Center for Health Statistics [NCHS],

2011) The percentage of adult smokers declined, as well,

from 1999 to 2004, although smoking prevalence among this

age group has since begun to increase again (CDC, 2011;

NCHS, 2011)

Many factors could be, and often have been, cited as

contributing to the overall decline in smoking in the United

States, including education, better understanding of

Correspondence should be addressed to Brian L Quick, Department of

Communication, University of Illinois at Urbana–Champaign, 244 Lincoln

Hall, MC-456, 702 S Wright Street, Urbana, IL 61801 E-mail:

bquick@uiuc.edu

addiction, antismoking policies, and tax increases Additionally, the contributions of antismoking cam-paigns and media efforts are certainly important and are our primary interest here Numerous studies have been conducted on the effectiveness of a variety of smoking prevention and cessation campaigns and advertisements (e.g., Davis, Farrelly, Messeri, & Duke, 2009; Davis, Nonnemaker, & Farrelly, 2007; Flay, 1987; Mudde &

De Vries, 1999; Niederdeppe, Fiore, Baker, & Smith, 2008; Siegel & Biener, 2000; Wakefield et al., 2006) However, as far as antismoking campaigns are concerned, one in particular has received the lion’s share of attention The truth® campaign, launched in 2000, is a national tobacco prevention effort that educates the public with facts about smoking consequences and “Big Tobacco.” The campaign has received much praise, and numerous studies have been conducted to analyze its effectiveness (e.g., Cowell, Farrelly, Chou, & Vallone, 2009; Farrelly, Davis, Duke, & Messeri, 2008; Farrelly, Nonnemaker, Davis, & Hussin, 2009; Richardson, Green, Xiao, Sokol,

& Vallone, 2010) However, surprisingly few studies have systematically examined the content embedded within these specific ads

Previous analysis of smoking prevention and cessa-tion message features and theories has included numer-ous theoretical frameworks, such as the stages of change model (DiClemente & Prochaska, 1985), prospect theory

(e.g., Toll et al., 2007), the activation model of information

exposure and sensation seeking (e.g., Helme, Donohew,

Baier, Zittleman, 2007; Niederdeppe, Davis, Farrelly, &

Yarsevich, 2007), message tailoring (e.g., Rimer & Kreuter,

2006), and the limited capacity model of motivated media

message processing (Leshner, Vultee, Bolls, & Moore,

2010) Given that the truth®ads present the uncensored truth

of the health costs associated with smoking using

unpar-alleled vivid language and images, however, we elected to

employ the Extended Parallel Process Model (EPPM) (Witte,

1992) to better understand the content within these

tele-visions ads In understanding what makes a fear appeal

effective, we turn to the EPPM (Witte, 1992)

EXTENDED PARALLEL PROCESS MODEL

With respect to the various discrete emotions to arouse

when creating health ads, few appear as often as fear

appeals In examining the effectiveness of fear appeal

mes-sages, Witte’s (1992) model represents the most

sophis-ticated framework as it draws from Leventhal’s (1970)

parallel process model and Rogers’s (1975) protection

moti-vation theory Together, Witte (1992) asserts that viewers

will strive to control their fear or the danger presented in

a fear-appeal message depending on which messages are

communicated (Witte, 1994) In the following, we briefly

review Witte’s model by identifying the key message

fea-tures included in fear appeals, the central hypotheses to

the EPPM, and the literature using Witte’s model as a

framework

A successful fear-appeal message will communicate both

the health threat and the recommended response (Roberto,

Murray-Johnson, & Witte, 2011) The health threat

con-veys the danger to which the message seeks to admonish

Fear appeals failing to communicate the threat will be

dismissed by most viewers as nonsignificant The threat

component of the EPPM can be bifurcated into two

mes-sage features: severity and susceptibility Severity refers to

the threat’s magnitude, whereas the latter communicates the

likelihood of experiencing the health costs associated with

the admonished behavior (Witte, 1992) Susceptibility, on

the other hand, emphasizes that the health threat could

hap-pen to a person who engages in the behavior Together,

the combination of severity and susceptibility creates an

overall assessment of the magnitude of the threat, thereby

arousing fear

Once message recipients appraise the threat, they

pro-ceed to consider the recommended response (Witte, 1992)

The second step of message appraisal is of extreme

impor-tance in the mediation of an audience’s ultimate reaction

(Witte, 1992) Thus, the EPPM cautions health

campaign-ers against communicating the threat without highlighting

the recommended response (Witte, 1992) Like the health

threat, the recommended response features two specific

message features—self-efficacy and response efficacy Self-efficacy refers to the message recipient’s perception that the recommendation is something that he or she is capa-ble of carrying out (Stephenson & Witte, 2001; Witte, 1992) Response efficacy, then, is the recipient’s per-ception that the recommended response will effectively reduce or eliminate the threat (Witte, 1992) For exam-ple, messages modeling the advocated behavior communi-cate self-efficacy, whereas attention to the positive health outcomes of not smoking would communicate response efficacy

Following exposure to a fear-appeal message, three dis-tinct types of message processing can occur (Roberto et al., 2011) First, if the message is evaluated as being of lit-tle or no personal threat, message processing ceases, and results in no motivation to change a behavior However, if the message is appraised as communicating a health threat,

a recipient is likely to experience fear, which results in fear-control processing (Stephenson & Witte, 2001; Witte, 1992) Fear-control processing results in a number of mal-adaptive responses such as defensive avoidance, denial, perceived manipulation, and psychological reactance Third, messages communicating both the health threat and the rec-ommended response will result in danger-control processing (Stephenson & Witte, 2001; Witte, 1992) Danger-control processing leads to a protection motivation and results in beliefs, attitudes, intentions, and behaviors consistent with the advocated message Danger-control processing is con-siderably more likely to occur when perceived efficacy exceeds the perceived health threat (Witte & Roberto, 2009) Research spanning more than two decades supports EPPM predictions (for a systematic review of this literature, see Roberto et al., 2011; Witte & Allen 2000); therefore, prac-titioners should be cognizant of the importance of commu-nicating both the health threat and recommended response

to ensure that perceived efficacy outweighs the perceived threat in order to elicit danger-control processing (Witte & Roberto, 2009)

The EPPM has been examined in a variety of health con-texts, such as messages regarding HIV/AIDS (e.g., Witte,

1994, 1998; Witte & Morrison, 2000), meningitis (Gore

& Bracken, 2005), cardiovascular disease (Rimal, 2001), sexually transmitted diseases (STDs) and pregnancy (e.g., Roberto, Zimmerman, Carlyle, & Abner, 2007), hearing loss for coal miners (Murray-Johnson et al., 2004), and various cancers (Hubbell, 2006; Kline & Mattson, 2000; Morman, 2000; Stephenson & Witte, 1998) Additionally, the EPPM has been used as a framework for other preventative behav-iors including gun safety (Roberto, Meyer, Johnson, & Atkin, 2000) and self-defense (Morrison, 2005) The EPPM,

as a design strategy for fear appeals, seems to be widely applicable to a number of dangerous behaviors but, interest-ingly, its utility for the theoretical analysis of message fea-tures and mechanisms has been rarely applied to antitobacco campaigns

APPLICATION OF THE EPPM TO TELEVISED TRUTH ADS 55 There has been a long history of criticism and resistance

to using fear appeals to persuade health behavior change,

especially within the context of smoking (Hill, Chapman,

& Donovan, 1998) Despite this, however, many smoking

prevention and cessation ads do arouse fear, with the

inten-tion of increasing smokers’ sense of urgency to quit (Hill

et al., 1998) Scare tactics, thus, have made their way

into the norm of mainstream health messages, especially

with respect to tobacco use Although fear appeals seem

to abound within the antismoking movement, few studies

examine these ads within the framework of the EPPM Smith

and Stuts (2003) examined the effectiveness of emphasizing

long- and short-term costs of smoking among adolescents

and found smoking behavior declined following message

exposure, regardless of message type More recently, Wong

and Cappella (2009) used the EPPM to predict smokers’

intentions to seek help In addition to discovering a threat

by efficacy interaction consistent with the EPPM predictions,

they discovered a three-way threat by efficacy by readiness

to quit interaction, such that efficacy was the most important

dimension for those high in readiness, but that both threat

and efficacy were equally important for those low in

readi-ness (Wong & Cappella, 2009) Although these few studies

are informative from a message design standpoint, little has

been done by way of analyzing the message features unique

to the EPPM within existing televised antismoking ads

Other scholars have examined antismoking message

features such as emotional content (Davis, Nonnemaker,

Farrelly, & Niederdeppe, 2011; Leshner et al., 2010),

stylistic features such as cuts and vividness (Niederdeppe

et al., 2007), and perceived vulnerability (Shadel et al.,

2006) However, the message components of the EPPM have

not been systematically analyzed Moreover, considering the

threatening consequences of smoking, it is interesting that

the EPPM has not enjoyed greater popularity as a

theo-retical lens through which to assess antitobacco messages

One such movement that offers a plethora of televised

mes-sages communicating the dangers of smoking is the truth®

campaign

The truth® campaign is the first and only national

antismoking campaign which is not sponsored by the

tobacco industry Introduced more than a decade ago,

the truth® campaign has become increasingly

recogniz-able, not only as a campaign, but as a brand Designed

by the American Legacy Foundation as a

countermar-keting effort, truth® is a prevention campaign aimed at

American adolescents between the ages of 12 and 17 years

(American Legacy Foundation [ALF], 2012) According to

the American Legacy Foundation, truth® “exposes the

tac-tics of the tobacco industry, the truth about addiction, and

the health effects and social consequences of smoking it

works by allowing teens to make their own informed choices about tobacco use” (ALF, 2012) The campaign has included television ads, print ads, interactive websites, and apps for social networking sites Further, it has had variously themed

“subcampaigns” over the years, such as “Infect truth,” “The Sunny Side of truth,” and “Crazyland” (truth®, 2012) Researchers and public health organizations, alike, have lauded the success of the truth® campaign According to several studies, truth®has been successful in lowering teen intentions to smoke, increasing positive antitobacco atti-tudes and beliefs, and even reducing peer smoking preva-lence perceptions (Cowell et al., 2009; Davis et al., 2009; Davis et al., 2007; Farrelly et al., 2008) Furthermore, some research cites this nationwide campaign as contribut-ing to reductions in teen smokcontribut-ing For example, Farrelly

et al (2005) found that a 2-year exposure to truth®resulted

in a 22% decrease in teen smoking from 1999 to 2002

A more recent study by Farrelly, Nonnemaker, Davis, and Hussin (2009) supports earlier findings, citing that exposure

to the truth® campaign had an independent effect on the reduction of smoking initiation among teens, beyond many other individual-, family-, and community-level influences Other supporting health agencies including the Institute of Medicine (2007) have also proffered their support of the truth®campaign, while other research has found truth®to be cost-effective and to have dramatic economic impacts when campaign cost is weighed against savings in health care expenditures (Holtgrave, Wunderink, Vallone, & Healton, 2009) However, it is important to note that although the truth®campaign, or others like it, may promote decreases in smoking among adolescents, it is not the only contributing factor

The truth®campaign undeniably takes a unique approach

to promoting smoking prevention, especially in contrast to tobacco sponsored prevention messages such as the Phillip Morris campaign “Think Don’t Smoke,” which has been shown to potentially lead to boomerang effects among youth (Davis et al., 2009; Farrelly et al., 2008) The ALF claims that the aim of truth® is not to judge anyone or to be

“antismoker” or “antismoking,” but to provide facts about smoking and the tobacco industry (ALF, 2012; truth®, 2012) Its various subcampaigns each have a different “feel” and strategy for delivering the facts about smoking However,

by nature of the potentially severe consequences of tobacco use, audiences may feel fearful when exposed to these mes-sages, regardless of the subcampaign or the intention of the campaign designers Further, although the campaign is intended to promote prevention among youths, its audi-ence is broad, by nature of its nationally televised ads, and often includes older viewers and smokers (Richardson et al., 2010) Following the guidance of the EPPM, this study is designed to analyze the specific message features presented

in truth®ads for television That is, are truth®ads consistent with EPPM maxims in communicating both the health threat and recommended response?

METHOD Sampling and Procedures

The current investigation examined truth® ads for

tele-vision These ads were obtained via extensive Internet

searches Specifically, because the desired ads were

tele-vision commercials, YouTube and Google Video were

the two primary search engines used In both of the

search engines, the following phrases and combinations of

phrases were entered: “truth campaign,” “American Legacy

Foundation,” “truth,” “smoking,” “antismoking,”

“antito-bacco,” “big to“antito-bacco,” “commercials,” “television,”

“adver-tisements,” and “videos.” Because videos are often posted

and reposted online, each video was reviewed by one of the

authors to confirm its originating organization, truth®or the

ALF Advertisements were included for analysis if (a) they

were produced by the ALF or truth®, (b) were complete (e.g.,

were not cut off), and (c) were of adequate visual and audio

quality to code All truth®commercials meeting these

crite-ria were coded, regardless of the subcampaign or production

year As ads were located and verified, they were compiled

into a document and each video’s working title, length, and

URL were recorded This process was repeated until each

subsequent search in Google or YouTube failed to produce

any new advertisements, regardless of the combination of

search terms.1

As with any content analysis project, an important

ques-tion pertains to the unit of analysis Neuendorf (2002) refers

to a unit of analysis as “an identifiable message or message

component, (a) which serves as the basis for identifying the

population and drawing a sample, (b) on which variables

are measured, or (c) which serves as the basis for

report-ing analysis” (p 71) Each ad represented a unit of analysis

for this investigation The final sample consisted of 86

differ-ent truth®television advertisements ranging in length from

30 seconds to 2 minutes

Two coders were extensively trained by the authors on

the four primary message components of the EPPM

includ-ing both the health threat (severity and susceptibility) and

recommended response (self-efficacy and response efficacy)

After adequate training of these four message components

was achieved, the coders independently coded

approxi-mately 25% of the messages (Lombard, Snyder-Dutch, &

Bracken, 2002) The intercoder reliabilities were

calcu-lated prior to resolving disagreements among the coders

Reliability between coders was calculated using simple

agreement as well as Brennan and Prediger’s (1981) kappa

Brennan and Prediger’s kappa was selected over Cohen’s

kappa because Cohen’s kappa penalizes coders for

main-taining higher level of simple agreement, whereas Brennan

and Prediger’s kappa corrects for this by assuming chance

1 The authors will provide a complete list of the final sample of

commercials upon request.

agreement is determined by the number of categories in the coding scheme For each variable coded, both simple agree-ment (SA) among the coders and Brennan and Prediger’s kappa (KB & P) are provided

Categories

Health threat Following the EPPM, the health threat

contains the severity of the threat and the susceptibility of experiencing the consequences of the threat (Witte, 1992) Each message was coded to determine the presence of threat severity and susceptibility overall as well as bifurcating each message by audio and video components to more precisely capture the magnitude of threat severity and susceptibility All primary categories were examined and coded dichoto-mously, since the scope of the current study was to determine what features of the EPPM were being employed in this cam-paign Additionally, message features such as vivid language and demographic similarity were also useful to examine (Witte, Meyer, & Martell, 2001) Therefore, these two cat-egories were also considered as subthemes within severity and susceptibility, respectively

Severity Message severity refers to the health costs

associated with performing a specific behavior (Witte, 1992) For the present study, severity was captured in commu-nicating the negative consequences of tobacco use (e.g., smoking causes cancer, debilitating diseases) or by showcas-ing the chemical components of cigarettes (e.g., cigarettes contain rat poison) Overall, the coders achieved acceptable intercoder reliability for threat severity (SA = 95, KB & P= 90) With respect to audio (SA = 1.0, KB & P = 1.0) and visual (SA = 85, KB & P= 70) components of the messages, severity received respectable intercoder reliability among the trained coders, as well Within the severity message fea-ture, we also coded for the magnitude of perceived message vividness overall as well as with respect to audio and visual components (Witte et al., 2001) In doing so, the coders

indi-cated the degree of vividness on a 1 = low vividness to 4 =

high vividnessscale With respect to message vividness for the audio (α = 93) and visual (α = 84) components, the coders achieved acceptable intercoder reliability

Susceptibility Message susceptibility is the likelihood

that an individual will experience the health costs of per-forming a specific behavior (Witte, 1992) Susceptibility was presented in the form of the probability that there would

be negative consequences due to smoking (e.g., tobacco kills 1 out of 3 of its users, tobacco kills 1,200 people a day) Intercoder reliability was established for susceptibil-ity among the coders overall (SA = 95, KB & P = 90), as well as for the audio (SA = 1.0, KB & P = 1.0) and visual (SA = 90, KB & P= 80) components of the messages For susceptibility, we also coded for the magnitude of perceived demographic similarity overall, as well as with respect to audio and visual components (Witte et al., 2001) In doing

APPLICATION OF THE EPPM TO TELEVISED TRUTH ADS 57

so, the coders indicated the degree of similarity on a 1 = low

similarity to 4 = high similarity scale For demographic

sim-ilarity, the coders achieved adequate intercoder reliability for

the audio (α = 1.0) and visual (α = 71) components

Recommended response The recommended

res-ponse refers to self-efficacy and resres-ponse efficacy (Witte,

1992) Similar to the health threat, each ad was coded for

the presence of self-efficacy and response efficacy overall,

as well as by examining the audio and video content of

each message to more precisely capture the magnitude of

self-efficacy and response efficacy

Self-efficacy Self-efficacy captures images and texts

reinforcing how to perform an advocated behavior in the

hopes of increasing confidence to follow through and

per-form the recommended act (Witte, 1992) In the current

study, phrases such as “You do not need tobacco”

cap-ture self-efficacy Overall, the coders achieved excellent

intercoder reliability for self-efficacy (SA = 1.0, KB & P =

1.0) Identical intercoder reliabilities were observed across

both audio (SA = 1.0, KB & P= 1.0) and visual (SA = 1.0,

KB & P= 1.0) components of the messages

Response efficacy Response efficacy suggests that

by performing the recommended behavior, an individual

will avert the health costs from occurring (Witte, 1992)

Phrases such as “By not smoking I can live longer”

exem-plify the response-efficacy construct in the current

investiga-tion Like self-efficacy, overall, the coders achieved perfect

intercoder reliability for response efficacy overall (SA = 1.0,

KB & P = 1.0) and across both audio (SA = 1.0, KB & P =

1.0) and visual (SA = 1.0, KB & P= 1.0) components of the

messages

Data Analysis Strategy

With an individual ad as the unit of analysis, it was

rea-sonable for an ad to contain more than one of the coded

message components After all, each category represented a

nonindependent dichotomous nominal variable For this

rea-son, our data analysis strategy took this non-independence

into account when selecting an approach to analyze the data

Specifically, Cochran’s Q tests were performed to ascertain if

certain message features emerged more frequently than

oth-ers Following a significant Cochran’s Q test, McNemar tests

were conducted to determine where significant differences

were present To minimize Type I error from occurring when

interpreting the McNemar tests, the decision was made to

adjust the alpha level by making a Bonferroni-type

correc-tion by dividing the convencorrec-tional alpha level (α = 05) by the

number of pairwise comparisons (6), resulting in an alpha

level of 008 In addition to presenting the statistical

anal-yses that follow, exemplars of each message feature were

extracted from the television ad to illustrate how each was

presented

RESULTS The overarching objective of the current project was to deter-mine the frequency with which truth®ads communicate the health threat and the recommended response A Cochran’s

Qtest showed that the ads differed in how often they com-municated health severity, susceptibility, self-efficacy, and

response efficacy across the ads, Q(3, N = 86) = 169.69,

p < 001 Among the four message features, severity was

presented most often (n = 72), followed by susceptibility (n = 31), self-efficacy (n = 0), and response efficacy (n = 0).

McNemar tests revealed that severity was communicated

more often than susceptibility (p < 001), self-efficacy (p < 001), and response efficacy (p < 001).

With respect to the audio components of the ads, a

Cochran’s Q test determined that the ads differed in com-municating severity (n = 54), susceptibility (n = 17), self-efficacy (n = 0), and response efficacy (n = 0),

Q (3, N = 86)= 128.93, p < 001 Specifically, McNemar

tests revealed severity was communicated more often than

susceptibility, self-efficacy, and response efficacy at p < 001 For the video components of the ads, a Cochran’s Q test found differences between severity (n = 48), suscepti-bility (n = 12), self-efficacy (n = 0), and response efficacy (n = 0), Q(3, N = 86) = 119.08, p < 001 McNemar

tests again revealed that severity was communicated more than susceptibility, self-efficacy, and response efficacy at

p <.001

Along with examining differences between severity, sus-ceptibility, self-efficacy, and response efficacy, we tested for variation in ad vividness communicating the severity

between audio and video components, t(86) = 2.378, p < 05 Results from the paired-sample t-test revealed that video (M = 1.90, SD = 1.02) content was perceived as being more vivid than audio content (M = 1.59, SD = 82).

The majority of the ads communicated severity by empha-sizing the health costs associated with smoking For

exam-ple, in the ad entitled The Real Marlboro Man: You Don’t

Always Die from Tobacco, viewers watch a man dressed as

a cowboy riding through the streets of New York The cow-boy sets up camp complete with a guitar-playing friend and campfire in the middle of the street He then takes off the bandana around his neck to reveal a quarter-sized hole where his larynx used to be, and proceeds to sing, with the aid of

a mechanical larynx, about the consequences of living with

smoking-initiated health costs In the ad entitled Rat Poison,

viewers see a person holding a digital number counter that rolls and then stops at 101 The accompanying onscreen writ-ing states, “Cigarette smoke contains 101 poisons That’s a hundred more than rat poison.”

Although not appearing as often as threat severity, suscep-tibility was portrayed with some regularity throughout the truth®ads Overall, across the ads, McNemar tests revealed that susceptibility was communicated more often than

self-efficacy and response self-efficacy at p < 001 Similarly,

susceptibility was communicated more than self-efficacy and

response efficacy with respect to both audio and visual

mes-sage features at p < 001 Differences in perceived similarity

were examined across audio (M = 1.23, SD = 76) and video

(M = 1.17, SD = 54) components of the ads with results

suggesting no statistical difference between them, t(86) =

.55, p = 58.

Several ads were designed to make viewers feel

suscep-tible to the health costs linked to smoking For example,

in the ad entitled 1,200, 1,200 people wear t-shirts labeled

with the numbers 1 through 1,200 as they stand outside a

major tobacco company Simultaneously, all 1,200 people

“fall down dead” in the street A man remains standing with

a sign stating, “Tobacco kills 1,200 people a day.” In another

ad entitled Career Seminar, a lecture hall is filled with

col-lege students energized to learn about an exciting career

opportunity The Big Tobacco executive asks the students,

“Who here would want to work in a high-paying recession

proof multi-billion-dollar industry once they graduate?” As

most hands go up, the executive concludes with the

ques-tion, “Who wants to work in an industry whose products

could potentially be responsible for 1 billion deaths in the

21st century?” Every hand in the audience goes down and

the speaker says, “Nobody?”

McNemar tests to determine differences between

self-efficacy and response self-efficacy were not conducted because

both variables contained the same values for the ads overall

Similarly, no differences could be detected with respect to

the audio and video components of the ads as these features

were not present

DISCUSSION According to the EPPM, a successful fear appeal must elicit

both perceived efficacy and perceived threat (Roberto et al.,

2011; Witte, 1992, 1994) Our findings reveal that the truth®

campaign relies heavily on communicating the health threat,

particularly with respect to video content, but offers no

rec-ommended response by way of self-efficacy and response

efficacy to viewers This result is consistent throughout the

ads and extends across visual and audio components as

well In particular, our results demonstrated greater

vivid-ness among the video components of these ads than audio

features Although many studies have claimed that the truth®

campaign has been a factor in the reduction of adolescent

smoking prevalence, there still lie several concerns

regard-ing the potential reactions to these advertisements, accordregard-ing

to Witte’s model In the following we discuss the theoretical

and practical implications of these findings

Theoretically speaking, if an audience is presented with

a threat but no recommended response, the audience’s

pro-cessing becomes one of fear control (Witte, 1992); that

is, the audience focuses on alleviating the emotion of fear

over eliminating the real danger Interestingly, one of the

potential results of fear control is psychological reactance (Brehm, 1966), which is an outcome that the truth® cam-paign fervently attempts to avoid (truth®, 2012) According

to Brehm (1966), reactance, operationalized as an amalga-mation of anger and negative cognitions (Dillard & Shen, 2005), follows a threat to or an elimination of an estab-lished freedom and motivates an individual to reestablish the threatened or removed freedom (Quick & Stephenson, 2008) Designers of truth® ads undeniably understand the importance of minimizing reactance, especially for their ado-lescent audience, who maintain a proclivity to experience reactance following freedom-threatening tobacco messages (see Miller, Burgoon, Grandpre, & Alvaro, 2006) In fact, the campaign arguably uses psychological reactance theory by tapping into adolescents’ natural tendency toward rebellion

to galvanize reactance against Big Tobacco (truth®, 2012) However, for all of these efforts, there is still one loose end: Reactance could become problematic for teens processing the truth®television ads provided the reliance on communi-cating the health threat without the recommended response (Roberto et al., 2011), especially for reactant prone teens (Miller et al., 2006) This is an empirical question deserving

of future research

Another potential fear-control process for adolescents is that teens may simply avoid the message (Witte, 1992)

In this case, ads highlighting the severity or likelihood of smoking consequences without any recommended response for how to avoid the social pressure to smoke may result

in teens agreeing with the danger, but initiating smoking anyway There is some concern, in fact, that the reduc-tion of smoking prevalence among adolescents has flat-lined (truth®, 2012) The authors acknowledge the myriad possible explanations for this leveling, including desensi-tization to the campaign However, another possibility to consider is the campaign’s lack of recommended response messages The truth® campaign provides the why to the argument against smoking, but it doesn’t provide the how with respect to strategies against temptation or social pres-sure As research has shown, the antidrug campaigns of the 1980s and 1990s showcased the severity of drug use while not adequately identifying the recommended response and

as a result were not particularly effective (e.g., Fishbein, Hall-Jamieson, Zimmer, von Haeften, & Nabi, 2002) Reactance and defensive avoidance are not the only pos-sible maladaptive outcomes of fear-control processing war-ranting empirical investigation in future studies, however Denial, message derogation, and perceived message manipu-lation may also be forms of fear-control processing (Roberto

et al., 2011; Stephenson & Witte, 2001; Witte, 1994) Although these outcomes have not been examined within a smoking prevention campaign, they deserve future consid-eration in this context For example, if teens do not know how to effectively avoid or resist the temptation to smoke, they may choose “not to think about it” as a defensively avoidant response to the real dangers of smoking Another

APPLICATION OF THE EPPM TO TELEVISED TRUTH ADS 59 outcome, denial, may also be problematic, especially among

teens, who, by and large, tend to hold beliefs of

invinci-bility (see Wickman, Anderson, & Smith-Greenberg, 2008)

In sum, although the truth®campaign has had some success

with smoking prevention among teenagers, the

overrepresen-tation of severity and susceptibility messages coupled with

the nonexistence of corresponding efficacy messages points

to potentially problematic outcomes, according to the EPPM

Scholars and practitioners alike would benefit from future

research exploring the conditions under which each of these

undesirable outcomes is employed as reactions to fear in

antismoking campaigns

In addition to potential maladaptive responses, there are

broader unintended consequences of the campaign Because

of the multimediated format of these ads, and especially

because of their television ads, the truth® campaign is

viewed by a much more diverse audience that

nonsmok-ing adolescents Accordnonsmok-ing to Richardson et al (2010), the

awareness of the truth®campaign is as high as 68% among

viewers between the ages of 18 and 24 years Because this

viewership reaches beyond adolescents, the primary strategy

of the truth®campaign (rebellion) may not be appealing or

understood by this unintended audience, whether 18–24 or

older Indeed, the truth®website states, “[Truth] is hard for

someone over the age of 30 to understand The adult word

[truth] is very different from the teens’ The values are

dif-ferent The goals are difdif-ferent If adults don’t get what we’re

saying and how we choose to say it, then it’s probably

okay” (truth®, 2012) By creating an “us versus them”

men-tality (teens versus adults), the campaign earns ethos with

adolescents and promotes healthy rebellion

However, there are several potential drawbacks with this

philosophy, as well First, the campaign mentions adults

(defined as 30 years and older) and teens, but what about

people between 18 and 29 who, as we already established, do

view the truth®ads (Richardson et al., 2010)? Second, it may

be undesirable for adults (or anyone else) unable to relate

to the campaign If a viewer doesn’t understand the primary

strategy (rebellion) or doesn’t identify with that strategy, it

leaves the viewer with an understanding of only the default

message, which is a fear appeal communicating the health

threat without any reference to a recommended response

One could argue that this may not be a bad thing; after all,

the audience is getting the message about the consequences

of tobacco However, the lack of identification, coupled with

the lack of efficacy messages, could be problematic for older

audiences, resulting in consequences such as low message

attendance or maladaptive fear-control processes

In addition, this campaign may act as a double-edged

sword, producing both positive and negative unintended

out-comes Awareness of truth® was actually higher among

smokers than nonsmokers (Richardson et al., 2010), and

sig-nificant correlations emerged between truth®antitobacco ads

and smoking cessation intentions (Richardson et al., 2010)

In other words, although the truth®campaign’s primary goal

is prevention, it may affect cessation efforts as a by-product Despite its prevention mission, it is possible that ads regard-ing the dangers of tobacco are more salient to those who already smoke, explaining the greater awareness of these ads among smokers On its face, intentions to quit smoking may seem like an inadvertent positive outcome of a campaign that is supposed to be aimed at prevention However, if this

is the case, this unintended benefit does not come without

a corresponding cost The imbalance of threat and recom-mended response may be of even greater concern for the smoking audience than for those who do not smoke Answers

to questions such as the ones raised here would be welcomed additions to the literature

Some of the unintended consequences for smokers are undeniably undesirable Ideally, a fear appeal guided by the EPPM should result in danger-control processes (not smok-ing for nonsmokers, and smoksmok-ing cessation for smokers)

if an adequate recommended response is included in the

ad However, for smokers who view the truth® ads, follow-ing EPPM predictions, it is theoretically more likely that they would resort to fear-control processes given the heavy reliance on threat and lack of attention to the recommended response The EPPM may showcase the health threat of smoking, but without any perceived self-efficacy or response

efficacy communicated, smokers may not feel like they can

quit For example, smokers may engage in denial to reduce their fear Many people with a friend or family member who smokes have likely heard a version of the following argu-ment: “Most people who die of lung cancer were smokers, but most smokers don’t die of lung cancer.” These types

of counterarguments, while they may be completely illogi-cal, may also act as self-soothing or reassuring responses to fear-arousing messages lacking efficacy, and aid in creating denial of susceptibility The potential feeling of “helpless-ness” (e.g., “Smoking is scary but I just don’t think I can quit

or know how”) is likely more exacerbated by particular mes-sages, such as several truth®ads that feature people living (or dying) with smoking-related diseases who share that they got these diseases long after they quit smoking Unfortunately, this may encourage smokers to adopt a “what’s the point?” attitude and further engage in fear control such as denial, since the perception may be that there is nothing they can

do to help themselves (no response efficacy even if they possessed self-efficacy)

Finally, although not the intended audience of the truth®

ads, smokers may feel stigmatized by fear appeals that don’t contain recommended response messages The graphic imagery and hard-hitting statistics used in truth® ads high-light the dangers of smoking but do not acknowledge how difficult it is to quit Although the campaign website states that it is not “antismoker,” it may be difficult for nonsmok-ers to grasp the true struggle with addiction, which could enhance negative perceptions of those who smoke The dis-crimination experienced by those with addiction can lead to social isolation and segregation from society (Smith, 2007;

Strauser, Ciftci, & O’Sullivan, 2009) This marginalization

can lead to self-stigmatizing beliefs (internalization), and has

been correlated with lower self-esteem, increased substance

use, poor physical and mental health, and less satisfaction

with life (Corrigan, Markowitz, Watson, Rowan, & Kubiak,

2003; Smith, 2007) In essence, a cycle is created and

neg-ative perceptions are continually reinforced by both public

attitude and the reception of that attitude by those living with

addiction

In all, the truth® campaign at best may help some who

already smoke But at worst, it may contribute to myriad

maladaptive and unintended responses that add to the

exist-ing problem It stands to reason that a campaign can have

a primary goal, such as prevention, but should also find

ways to promote secondary goals such as cessation Future

research should determine exactly how smokers and “adults”

(as defined by the truth® organization) perceive the

cam-paign, as well as how they react to it Additionally, studies

should investigate how antismoking ads contribute to

unin-tended effects by examining potential mediating factors such

as similarity and identification

The goal of our study was to determine whether the

com-ponents from the EPPM were present in televised truth®ads

However, this was only a first step in examining this

cam-paign Because we chose to initially look for the presence or

absence of each message feature of the EPPM framework,

our conclusions are constrained by a dichotomous coding

scheme, which is often the case in content analysis projects

(see Quick et al., 2011; Silk et al., 2003) Future researchers

may choose to utilize a more nuanced coding scheme to

further investigate message feature strategies Although we

have established the heavy reliance on threat throughout

these ads, understanding how threat is generated and what

message features might make a threat maximally effective

would be a worthy undertaking Determining how

consis-tently specific message strategies (e.g., color, cuts, graphic

imagery) are employed and to what degree they are

repli-cated may shed additional light on what conditions may

exacerbate the possible unintended consequences or enhance

the intended persuasiveness of this campaign Our study

found that vividness was more pronounced in the visual

elements than in the audio elements in these ads

Further, to achieve a sufficiently large sample, the authors

grouped all subcampaigns together for analysis Admittedly,

the truth®campaign employs multiple subcampaigns, which

offer a variety of distinct approaches in their thematic

pre-sentations However, most truth®ads relay the same ultimate

messages regarding the scariness of tobacco-use

conse-quences, even when the threat is embedded in a larger

dominant strategy Future research should consider each

subcampaign separately to tease out the various strategies

used in conjunction with threat, as well as determining

whether audiences perceive these subcampaigns differently

Additionally, establishing whether viewers perceive some

ads as more persuasive than others, along with the reasons

behind such perceptions, would be a welcomed addition to the literature

Although several studies have been conducted on the effectiveness or use of Witte’s EPPM, yet to be answered

is the question of proportions within the model In other words, what is the ideal proportion of each of the compo-nents within the EPPM? Although the EPPM’s effectiveness

is necessitated by ideal balance of threat and efficacy com-ponents, scholars have yet to thoroughly examine what the ideal balance might be and whether this balance is audience

or context specific Further research should explore these inquiries, and scholars should look to advance the EPPM

by refining and specifying the balance needed between components

The goal of this investigation was to apply the EPPM to the truth®campaign Our results suggest that these television ads do rely on fear appeals The commercials use severity messages frequently and susceptibility to a moderate degree However, no self- or response-efficacy messages were found

in the ads According to the EPPM, the campaign message will not be successful under these conditions Although there

is support for the effectiveness of truth®, it is difficult to claim that these results are strongly causally related to the campaign because of many other potential contributing fac-tors such as policy changes or taxes increases, among others Despite support for the campaign, recent evidence indicates that the reduction of teen smoking has leveled (truth®, 2012) Our application of the EPPM illuminates potential reasons why this campaign may be decreasing in effectiveness or sustainability for prevention, and concerns about the possible negative outcomes for cessation Although truth®should be applauded for its unique and innovative antismoking strate-gies, from an EPPM standpoint, it would benefit from the addition of recommended response messages to complement the health threat in its ads without sacrificing its strategy of youthful rebellion

REFERENCES

American Legacy Foundation (2012) truth Retrieved from http://www.

legacyforhealth.org

Brehm, J W (1966) A theory of psychological reactance New York, NY:

Academic Press.

Brennan, R L., & Prediger, D J (1981) Coefficient kappa: Some uses,

misuses, and alternatives Educational and Psychological Measurement,

41, 687–699 doi:10.1177/001316448104100307

Centers for Disease Control and Prevention (2011) Economic facts about

U.S tobacco production and use.Retrieved from http://www.cdc.gov/ tobacco/data_statistics/fact_sheets/economics/econ_facts/index.htm Corrigan, P W., Markowitz, F E., Watson, A C., Rowan, D., & Kubiak, M.

A (2003) An attribution model of public discrimination towards

per-sons with mental illness Journal of Health and Social Behavior, 44,

162–179.

Cowell, A J., Farrelly, M C., Chou, R., & Vallone, D M (2009) Assessing the impact of the national ‘truth’ anti-smoking campaign on beliefs,

atti-tudes, and intent to smoke by race/ethnicity Ethnicity and Health, 14,

75–91 doi:10.1080/13557850802257715

APPLICATION OF THE EPPM TO TELEVISED TRUTH ADS 61 Davis, K C., Farrelly, M C., Messeri, P., & Duke, J (2009) The

impact of national smoking prevention campaigns on tobacco-related

beliefs, intentions to smoke and smoking initiation: Results from a

lon-gitudinal survey of youth in the United States International Journal

of Environmental Research and Public Health , 2009; 6, 720–740.

doi:10.3390/ijerph6020722

Davis, K C., Nonnemaker, J M., & Farrelly, M C (2007) Association

between national smoking prevention campaigns and perceived

smok-ing prevalence among youth in the United States Journal of Adolescent

Health , 41, 430–436 doi:10.1016/j.jadohealth.2007.05.008

Davis, K C., Nonnemaker, J M., Farrelly, M C., & Niederdeppe, J.

(2011) Exploring differences in smokers’ perceptions of the

effec-tiveness of cessation media messages Tobacco Control, 20, 26–33.

doi:10.1136/tc.2009.035568

DiClemente, C C., & Prochaska, J O (1985) Processes and stages

of change: Coping and competence in smoking behavior change In

S Shiffman & T A Wills (Eds.)., Coping and substance abuse

(pp 319–343) New York, NY: Academic Press.

Dillard, J P., & Shen, L (2005) On the nature of reactance and its role

in persuasive health communication Communication Monographs, 72,

144–168 doi:10.1080/03637750500111815

Farrelly, M C., Davis, K C., Duke, J., & Messeri, P (2008) Sustaining

‘truth’: Changes in youth tobacco attitudes and smoking intentions after

3 years of a national anti-smoking campaign Health Education Research,

24, 42–48 doi:10.1093/her/cym087

Farrelly, M C., Nonnemaker, J M., Davis, K C., & Hussin, A.

(2009) The influence of the national ‘truth’ campaign on smoking

initiation American Journal of Preventative Medicine, 36, 379–384.

doi:10.1016/j.amepre.2009.01.019

Fishbein, M., Hall-Jamieson, K., Zimmer, E., von Haeften, I., &

Nabi, R (2002) Avoiding the boomerang: Testing the relative

effectiveness of antidrug public service announcements before a

national campaign American Journal of Public Health, 92, 238–245.

doi:10.2105/AJPH.92.2.238

Flay, B R (1987) Mass media and smoking cessation: A

crit-ical review American Journal of Public Health, 77, 153–160.

doi:10.2105/AJPH.77.2.153

Gore, T D., & Bracken, C C (2005) Testing the theoretical design of

a health risk message: Reexamining the major tenets of the Extended

Parallel Process Model Health Education and Behavior, 37, 27–41.

doi:10.1177/1090198104266901

Helme, D W., Donohew, R L., Baier, M., & Zittleman, L (2007).

A classroom-administered simulation of a television campaign on

adolescent smoking: Testing an Activation Model of Information

Exposure. Journal of Health Communication, 12, 399–415.

doi:10.1080/10810730701325947

Hill, D., Chapman, S., & Donovan, R (1998) The return of scare tactics.

Tobacco Control , 7, 5–8 doi:10.1136/tc.7.1.5

Holtgrave, D R., Wunderink, K A., Vallone, D M., & Healton, C G.

(2009) Cost-utility analysis of the national ‘truth’ campaign to

pre-vent youth smoking American Journal of Prepre-ventative Medicine, 36,

385–388 doi:10.1016/j.amepre.2009.01.020

Hubbell, A P (2006) Mexican American women in a rural area and barriers

to their ability to enact protective behaviors against breast cancer Health

Communication , 20, 35–44 doi:10.1207/s15327027hc2001_4

Institute of Medicine (2007) Ending the tobacco problem: A blueprint for

the nation Washington, DC: National Academies Press.

Kline, K N., & Mattson, M (2000) Breast self-examination

pam-phlets: A content analysis grounded in fear appeal research Health

Communication , 12, 1–21 doi:10.1207/S15327027HC1201_01

Leshner, G., Vultee, F., Bolls, P D., & Moore, J (2010) When a fear

appeal isn’t just a fear appeal: The effects of graphic anti-tobacco

messages Journal of Broadcasting & Electronic Media, 54, 485–507.

doi:10.1080/08838151.2010.498850

Leventhal, H (1970) Findings and theory in the study of fear

communica-tions In L Berkowitz (Ed.), Advances in experimental social psychology

(Vol 5, pp 119–186) New York, NY: Academic Press.

Lombard, M., Snyder-Duch, J., Bracken, C C (2002) Content analysis

in mass communication: Assessment and reporting intercoder reliability.

Human Communication Research , 28, 587–604

doi:10.1111/j.1468-2958.2002.tb00826.x Miller, C H., Burgoon, M., Grandpre, J R., & Alvaro, E M (2006) Identifying principal risk factors for the initiation of adolescent

smok-ing behaviors: The significance of psychological reactance Health

Communication , 19, 241–252 doi:10.1207/s15327027hc1903_6

Morman, M T (2000) The influence of fear appeals, message design, and masculinity on men’s motivation to perform the testicular

self-exam Journal of Applied Communication Research, 28, 91–116.

doi:10.1080/00909880009365558 Morrison, K (2005) Motivating women and men to take protective action against rape: Examining direct and indirect persuasive fear appeals.

Health Communication , 18, 237–256 doi:10.1207/s15327027hc1803_3

Mudde, A N., & De Vries, H (1999) The reach and effectiveness of a national mass media-led smoking cessation campaign in the Netherlands.

American Journal of Public Health , 89, 346–350.

Murray-Johnson, L., Witte, K., Patel, D., Orrego, V., Zuckerman, C., Maxfield, A M., & Thimons, E D (2004) Using the extended par-allel process model to prevent noise-induced hearing loss among coal

miners in Appalachia Health Education and Behavior, 31, 741–755.

doi:10.1177/1090198104263396

National Center for Health Statistics (2011) Health, United States, 2010:

With special feature on death and dying Retrieved from www.cdc.gov

Neuendorf, K A (2002) The content analysis guidebook Thousand Oaks,

CA: Sage.

Niederdeppe, J., Davis, K C., Farrelly, M C., & Yarsevich, J (2007) Stylistic features, need for sensation, and confirmed recall of national

smoking prevention advertisements Journal of Communication, 57,

272–292 doi:10.1111/j.1460-2466.2007.00343.x Niderdeppe, J., Fiore, M C., Baker, T B., Smith, S S (2008) Smoking-cessation media campaigns and their effectiveness among

socioeconom-ically advantaged and disadvantaged populations American Journal of

Public Health , 98, 916–924 doi:10.2105/AJPH.2007.117499

Quick, B L., Fiese, B H., Anderson, B., Koester, B D., & Marlin, D W (2011) A formative evaluation of shared family mealtime for parents

of toddlers and young children Health Communication, 26, 656–666.

doi:10.1080/10410236.2011.561920 Quick, B L & Stephenson, M T (2008) Examining the role of trait reactance and sensation seeking on reactance-inducing messages,

reactance, and reactance restoration Human Communication Research,

34, 448–476 doi:10.1111/j.1468-2958.2008.00328.x Richardson, A K., Green, M., Xiao, H., Sokol, N., & Vallone, D (2010) Evidence for truth ® : The young adult response to a youth-focused

anti-smoking media campaign American Journal of Preventative Medicine,

39, 500–506 doi:10.1016/j.amepre.2010.08.007 Rimal, R N (2001) Perceived risk and self-efficacy as motiva-tors: Understanding individuals’ long-term use of health

informa-tion Journal of Communication, 51, 633–654

doi:10.1111/j.1460-2466.2001.tb02900.x Rimer, B K., & Kreuter, M W (2006) Advancing tailored health

com-munication: A persuasion and message effects perspective Journal of

Communication , 56, 184–201 doi:10.1111/j.1460-2466.2006.00289.x

Roberto, A J., Meyer, G., Johnson, A J., & Atkin, C K (2000) Using the parallel process model to prevent firearm injury and death: Field

exper-iment results of a video-based intervention Journal of Communication,

50, 157–175 doi:10.1111/j.1460-2466.2000.tb02867.x Roberto, A J., Murray-Johnson, L., & Witte, K (2011) International health communication campaigns in developing countries In T L Thompson,

R Parrott, & J F Nussbaum (Eds.), The Routledge handbook of health