Efficacy of Anti-Tobacco Mass

Evidence-Based Media Campaigns on Adolescent

Practice Tobacco Use
Bernadette Mazurek Melnyk, PhD,
RN-CS, PNP
Michelle Zibisky Silver

important for health care professionals to take an active role

Y
outh tobacco consumption is currently the most
Important public heaith issue because it is the iead- in preventing tobacco use. The importance of tobacco use
ing cause of preventable death in the Gnited States prevention and cessation programs for youth was addressed
(Lamkin, Davis, & Kamen, 1998; Tuakli, Smith, & in Healthy People 2000, and new goals have been established
Heaton, 1990). Eighty percent of tobacco users begin smok- for Healthy People 2010. Limited progress has been made in
ing before they reach adulthood. More than 3,000 children reducing the proportion of people aged 20-24 who have
begin to smoke each day and, according to the CJ.S. Food and begun to smoke cigarettes, an approximate measure of youth
Drug Administration (FDA), at least 1,000 of those children initiation. According to the CDC, the smoking rate in young
will eventually die from a tobacco-related illness (Lamkin et adults dropped from 30% in 1987 to 27% in 1993 (CDC,
al., 1998; Peracchio, 1998). Smoking substantially increases 1995). Non-profit organizations and government agencies
the risk of cardiovascular disease, cancer, and cJironic lung are increasingly turning to the media to devise advertise-
diseases. If adolescents can be kept tobacco-free, most will ments with messages aimed to prevent adolescents from
not start smoking as adults (Elders, Perry, & Erickson, 1994). using tobacco.
Public health experts agree that tobacco control efforts must Over the last 30 years, the American public has been
focus on preventing tobacco use and addiction among chil- heavily influenced by the power of television and its ability to
dren. After decades of public health education about the long- dictate societal norms (DeJong & Winsten, 1990). Millions of
term negative effects of tobacco use, children and adoles- dollars from tobacco settlement lawsuits have been recently
cents are the only age groups who continue to initiate smok- earmarked for preventive interventions targeted at youth.
ing in large numbers, resulting in the FDA classifying smok- Specifically, large sums of money have facilitated the devel-
ing as a "pediatric disease" (Peracchio, 1998). opment and use of mass media anti-tobacco messages.
The Centers for Disease Control (CDC) conducted the Clinical Question
1999 National Youth Tobacco Survey, and results indicate
that about one in eight (12.8%) middle school students and Are media-based anti-tobacco campaigns targeted for
more than one third (34.8%) of high school students reported teenagers effective as a community health intervention in pre-
using some form of tobacco in the previous month (Tobacco venting youth tobacco use? Medline, FroQuest, and
Use, 1999). The CDC also analyzed data from the 1997 Youth Academic Index databases were searched using the key-
Risk Behavior Survey (YRBS) and found that, among U.S. words adolescents, tobacco, smoking prevention, anti-smok-
high school students in 1997, 70.2% had tried cigarette smok- ing campaigns, mass media, and health promotion. Members
ing. Among those high school students who had tried ciga- of the Duval County Health Department in Jacksonville,
rette smoking, 35.8% went on to smoke daily (CDC, 1998). Florida, who were knowledgeable in accessing tobacco-relat-
Given the high rate of smoking initiation among the young ed research, assisted in locating articles. Reference lists of
and the adverse health effects of smoking, it is increasingly data-based articles were used to locate other studies. Studies
were excluded from this review if they included adult subjects
or if they were not primary studies.

The Evidence
Study #1. Siegel and Biener (2000) examined the impact
of a statewide anti-smoking media campaign on smoking
Michelle Zibisky Silver, BSN, RN, is a graduate nurse practitioner, incidence in adolescents who were not established smokers at
(Jniversity of North Florida, Jacksonvilie, FL. baseline. In 1992, Massachusetts voters increased the ciga-
rette excise tax and established a comprehensive anti-smok-
Acknowledgment: The author would like to thank Doreen ing intervention that included a media campaign. Aspects of
Radjenovic, PhD, ARNP, for her encouragement, enthusiasm, and the media campaign targeted at young people were almost
editorial review. Without her valuable contribution, the completion completely restricted to billboard, radio, and television adver-
of this manuscript would not have been possible. tisements. Of the $735,000 expended on advertisements dur-
ing the first 6 months of the media campaign (when the base-
line survey took place), 80% was allocated for television
1 1 advertisements, 14 % for radio, 5% for billboards, and 1% for
jThe Euidence-Based Practice section focuses on the search for andj newspaper advertisements.
i critique of the "best evidence" to answer challenging clinical ques-|
itions so that the highest quality, up-to-date care can be provided to In this 4-year longitudinal study of 592 Massachusetts
; children and their families. To submit questions or obtain author youth, with baseline data collected in 1993, effects of expo-
iguidelines, contact Bernadette Mazurek Melnyk, PhD, RN-CS, PNP;i sure to television, radio, and outdoor anti-smoking advertis-
iSection Editor; University of Rochester School of Nursing; 6011 ing on the progression to established smoking habits was
I Elmwood Avenue; Box SON; Rochester, NY 14642; (716) 275- assessed. The sample was based on a probability sample of
|8903; [email protected] Massachusetts households drawn by random digit dialing.
Based on initial interviews with adult household informants, a

PEDIATRIC NURSING/May-June 2001/Vol. 27/No. 3

sample of youths was selected and interviewed over the jects lived in smoking households, and 59 lived in nonsmok-
phone. Four years later, a portion of this sample of youths was ing households. Respondents were divided into four groups
availahle for follow-up phone interview surveys. Exposure to based on their gender and household smoking behavior. Four
the three major channels of the media campaign (i.e., bill- focus groups of each type were conducted for a total of six-
boards, television, or radio advertisements) was measured by teen focus groups. When the youngest students were asked to
asking which one the respondent could recall. interpret the sample ads, they had difficulty understanding
The researchers used multiple logistic regression to test complex analogies representing the effects that smoking
separate and combined effects of each type of advertising on might have on their bodies. This research suggests that if the
progression to established smoking, controlling for: age, sex, long-term health risks of smoking are to be conveyed effec-
race, baseline smoking status, smoking by parents, friends tively to both future smokers and nonsmokers, the campaign
and siblings, television viewing, and exposure to anti-smok- must be created in a developmentally appropriate form. It was
ing messages not related to the media campaign. Results recommended that testing of anti-tobacco media advertising
indicated that younger adolescents (aged 12-13 years) campaigns is necessary to ensure that children can compre-
reporting baseline exposure to television anti-smoking adver- hend the advertising message.
tisements were significantly less likely to progress to estab- Study #3. In order to test the ability of a targeted mass
lished smoking than those who had not reported such expo- media intervention to enhance the effectiveness of a school-
sure. Exposure to television anti-smoking advertisements based smoking prevention program, Worden et al. (1996)
was not significantly associated with progression to estab- developed and tested a mass media smoking prevention
lished smoking among older adolescents (aged 14-15 years). intervention targeted toward adolescent girls at increased risk
No significant effect of exposure to radio or outdoor anti- for smoking. The study design included two treatment groups
smoking advertisements was found. with two geographically different matched pairs. One pair was
Study #2. Peracchio (1998) developed and tested an age- located in the northeastern U.S., and one pair in Montana.
appropriate print advertising campaign to discourage smok- Within each matched pair, one group included students who
ing initiation among children and teens. The campaign was only received school smoking interventions and the other
based on young smokers' and nonsmokers' perceptions of group included students who received the school smoking
tobacco and the emergence of children's analogical reason- interventions and the targeted mass media intervention. This
ing ability. In this two-phase study, a focus group was initially cohort study included 5,458 students surveyed at baseline in
used to assess teenagers' attitudes and beliefs toward smok- the fourth through sixth grades every year for 4 years.
ing. The convenience sample consisted of 48 male and 58 The school smoking prevention program included grade
female students, 15 to 18 years of age, in a large, public high specific curricular material, annual teacher training, and mon-
school in the Midwestern U.S. Respondents were invited to a itoring of program implementation. Educational objectives for
classroom furnished with sofas to participate in the focus the mass media intervention program based on social learn-
groups during a class period. Fifty-five students were smok- ing and change theories encouraged adolescent girls to have
ers and 51 were nonsmokers. a negative view of smoking, a positive view of nonsmoking,
Students completed a screening questionnaire that asked skills for refusing cigarettes, and the perception that most
about smoking behavior. Subjects were divided into four people their age do not smoke. With these objectives in mind,
groups based on gender and smoking behavior. Twenty focus media producers created 36 television and 17 radio mes-
groups, five of each type, were conducted. Differences in atti- sages that especially appeal to teenagers.
tudes and beliefs about smoking by gender and smoking sta- Mass media interventions targeted at eighth, ninth, and
tus were examined. Results suggested that by age 13, both tenth grade girls resulted in a 40 % lower weekly smoking rate
smokers and nonsmokers had already made long-term for those who received the media and school intervention
smoking decisions and no longer listened to anti-tobacco compared to the school intervention alone. Lower weekly
messages. Young smokers had difficulty conceptualizing the smoking rates in girls who received the mass media interven-
negative effects of smoking on their bodies. Thus, it was con- tions persisted through the tenth, eleventh, and twelfth grades.
cluded that an advertising campaign to discourage smoking Results indicate that mass media interventions targeting spe-
initiation should target young children from 7 to 11 years of cific audiences can reduce tobacco use for those populations.
age with the health risks communicated in a concrete man-
ner. Critique of the Evidence
In the second phase of this study, Perrachio (1998) Three studies involving mass media anti-tobacco cam-
described the development of an age-appropriate advertising paigns were reviewed that indicate the following major find-
campaign designed to discourage smoking initiation among ings: (a) anti-tobacco mass media advertising campaigns,
children. Based on the understanding of youth tobacco per- especially television targeted to younger children, are effec-
ceptions garnered in the first phase of this research, a print tive community health interventions in preventing youth
advertising campaign was designed by the American Cancer tobacco usage; (b) advertisements that target school-age
Society and BVK McDonald advertising agency. This cam- children and take into account the audiences' psychological
paign used age-appropriate analogies between negative maturity (analogical reasoning ability) are effective; and (c)
smoking effects and health consequences of smoking. For anti-tobacco mass media advertising campaigns may be
example, the simplest analogy made was between a dirty most effective in reducing tobacco use in children when used
sweat sock and an ashtray full of cigarette butts. In phase two as part of a multifaceted approach, including community
of this research, Peracchio tested the campaign with children health education in schools.
7 to 11 years of age in order to assess the age appropriate- All of the studies reviewed were school-based intervention
ness of tfie analogical messages developed. The ad campaign studies. Studies #2 and #3 used inductive descriptive meth-
was then implemented after testing. Peracchio's research did ods to develop interventions by gathering information from
not report an analysis of how the campaign impacted smok- surveys and focus groups with children. Sample sizes were
ing outcomes or how being in a smoking household affected large enough to obtain significant results in all three studies.
findings. The convenience sample included 50 male and 54 Cross contamination was not discussed in any of these stud-
female grade school students who attended a large public ies, but because the sample groups were in the same school
grammar school in the Midwestern U.S. Forty-five of the sub- where the intervention and control programs were adminis-

PEDIATRIC NURSING/May-June 2001 Atol. 27/No. 3

tered, there is a chance that cross contamination occurred. Glossary of Terms
Despite the presence of any cross contamination, significant
differences between the groups were obtained.
Cohort study
Studies #2 and #3 are based on theoretical frameworks. In A study of a group of subjects who share common char-
Study # 2, children's analogical reasoning theories were used acteristics or experiences over a specific time period.
to create advertising messages that communicated the health
risks of smoking. In Study #3, common educational objec-
Convenience sample
tives for the school and mass media programs were based on
social learning theory and related behavior change theories. A nonprobability sample of individuals readily avail-
able for participation in a study.
In the 4-year longitudinal youth study by Siegel and Biener
(2000), the authors were the first to examine the relationship Extraneous variable
between exposure to a state wide anti-tobacco media cam-
The quantity of variance in the dependent variable
paign and changes in smoking status among youth by using
caused by forces other than the independent variable.
a cohort design. Using multiple logistic regression, the
authors controlled for multiple variables that could have con-
founded the results (e.g., age, sex, race, baseline smoking Focus group
status, smoking by parents, friends, and siblings). Another A group of participants gathered to discuss, provide
strength of the study was random sampling methods from expertise or information about, or work on a selected
11,4630 households, which strengthens external validity of topic.
the study. One weakness of this study was that approximate-
ly half of the original baseline sample was lost to follow-up. Internal validity
Although common, subject attrition is problematic in longitu- The extent to which the independent variable, not
dinal research and may threaten the internal validity of a extraneous variables, accounts for the effects on the
study. dependent variable.
According to Siegel and Biener (2000), there were sever- Longitudinal study
al limitations to their study. First, the exposure measure that A study with data collection from the same sample at
was used assessed recall, not actual exposure to advertise- repeated points in time.
ments. Because the methodology is retrospective, extraneous
variables may have had an effect on the dependent variables
Multiple logistic regression
and, therefore, threatened the internal validity of this study.
Second, the baseline survey was administered at the same A statistical method for computing the direction and
time as the opening of the media campaign. Thus, it is possi- strength of the relationship between two or more inde-
ble that some early respondents who said they were not pendent variables and one nominal dependent vari-
exposed did not have a chance to be exposed. The authors able.
speculate that less intense or less expensive campaigns may
not have the same effect. Probability sample
A sample selected using random sampling procedures.
Few studies have explored how to develop effective adver-
tising message content to discourage youth smoking initia-
tion. TTierefore, in Study #2, Peracchio (1998) and her asso- Note: Adapted from Doordan, A.M. (1998). Research survival
ciates examined young smokers' and nonsmokers' attitudes guide. Philadelphia: Lippincott-Raven Publishers.
toward smoking and assessed children's ability to compre-
hend age-appropriate analogies, which allowed them to
develop an advertising campaign that would discourage chil-
dren from beginning to smoke. Peracchio's study makes sev- many users begin before adulthood, it is important for health
eral contributions to knowledge about smoking initiation. One care practitioners to be active in attempting to change this
contribution is the development of an anti-tobacco advertis- dangerous course of action. The three studies reviewed pro-
ing campaign to discourage youth smoking initiation. The vide data that suggest mass media anti-tobacco campaigns
advertisement campaign focuses on communicating the targeted at children are effective in preventing tobacco use,
long-term negative health effects of smoking in a concrete yet each study reveals something different about the effec-
way by creating analogies between the effects of smoking and tiveness of anti-tobacco mass media campaigns.
things with which children are familiar. Second, the research As the rate of tobacco use among children escalates, it is
clarifies some of the attitudinal differences between teenage increasingly important for health care professionals to take an
smokers and nonsmokers. Results also indicate that teen active role in preventing tobacco use. Preventing children
smokers do not have concrete internal images of the negative from using tobacco requires nurses to be active in many dif-
health effects of smoking, but nonsmokers do seem to have ferent ways. Tobacco related clinical interventions should be
such images. routinely incorporated into all office visits with youth (LaSala
A strength of Study #3 (Worden et al., 1996) is that it & Todd, 2000).
included a 4-year long cohort study of adolescents from four Despite evidence that it works, health care professionals
different communities. Two matched pairs of Standard often do not use opportunities for tobacco use intervention
Metropolitan Statistical Areas (SMSAs) were selected using (Danis & Seaton, 1997; Merrill, 1995). Nurses have valuable
U.S. Census data; one pair of communities was located in the opportunities for face-to-face meetings with children and
northeastern Gnited States and the other pair was in Montana. teens. Every patient should be asked about smoking status at
Another strength included providing a clear definition of a each visit and this information should be documented.
smoker and collecting saliva samples to increase accuracy of Documentation of smoking status on the chart serves as a
self-reports. reminder to members of the health care team to counsel the
child during each office visit (Larson, 1997). Nurses must
Implications for Practice work in collaboration with other clinical professionals and
Tobacco use is a significant health care issue. Because educators to counsel, educate, and motivate young children

PEDIATRIC NURSING/May-June 2001/Vol. 27/No. 3

to refrain from or stop smoking. Based on results from these Centers for Disease Control and Prevention (CDC). (1999). Tobacco
studies, addressing tobacco use prevention must begin with use among middle and high school students - Florida, 1998-
school-age chiidren and continue with adolescents. 1999. Morbidity and Mortaiity Weei<ly Report. 4S(12), 248.
Nurses also must be actively involved in the community, Danis, P.G., & Seaton, T L (1997). Helping your patients to quit
smoking. American Family Physician. 55,1207-1214.
directing resources for public awareness and tobacco preven- DeJong, W., & Winsten, J.A. (1990, Summer). The use of mass media
tion programs by assisting in formulating public policy to con- in substance abuse prevention. Health Affairs. 9, 30-46.
trol youth tobacco use. Nurses are in a unique position to con- Doordan, A.M. (1998). Research survival guide. Philadelphia:
tribute to a multifaceted approach to the anti-tobacco move- Lippincott-Raven Publishers.
ment that includes clinic-, community-, school-, and mass Elders, M.J., Perry, C.L, & Erikson, M.P. (1994). The report of the
media-based programs. Tobacco control programs need to Surgeon General: Preventing tobacco use among young peo-
be systematically implemented and evaluated to determine ple. American Journal of Public Health. 84(3), 543.
the most effective methods to reach children. Future research Givel, M.S., & Glantz, S.A. (2000). Failure to defend a successful
should focus on sustaining the effects of anti-smoking mass state tobacco control program: Policy lessons from Florida.
American Journal of Public Health. 90(5), 762-767.
media messages in all age groups. In Florida, kicking butts. (1999, April 24). The Economist (US). 351
Nurses need to be active on state policy boards and task (8116), 28.
forces, making decisions on how tobacco settlement dollars Lamkin, L., Davis, B., & Kamen, A. (1998). Rationale for tobacco ces-
granted to states will be spent. Funds need to be used for sation interventions for youth. Preventive Medicine. 27. A3-A8.
tobacco prevention and control activities, cessation pro- Larson, D. (1997). Smoking cessation: Counseling your patients.
grams, and for health care issues related to tobacco use. In Clinical Reviews. 7(6), 57-80.
order to make a powerful difference, nurses will need to be LaSala, K.B., & Todd, S.T. (2000). Preventing youth use of tobacco
intimately involved in the political process. It is not enough to products: The role of nursing. Pediatric Nursing. 26(2), 143-147.
write and/or call legislators; nurses will have to work their way Merrill, E. (1995). Preventing tobacco use in young people:
Strategies for the nurse practitioner. Nurse Practitioner Forum.
into the political process as public health representatives 6(1), 34-39.
(LaSala & Todd, 2000). Peracchio, LA. (1998). The development of an advertising cam-
In November of 1998, the attorneys general of 46 states paign to discourage smoking initiation among children and
signed a settlement agreement with the tobacco industry that youth. Journal of Advertising. 27(3), 49-56.
would fund a national tobacco control foundation. Payments Siegel, M., & Biener, L. (2000). The impact of an anti-smoking media
to the states will total approximately $206 billion through the campaign on progression to established smoking: Results of a
first 25 years. Settlement agreements are a result of the initial longitudinal youth study. American Journal of Public Health.
actions taken by the state of Florida in 1995. Florida sued the 20(3), 380-386.
tobacco industry to recoup costs for treating Florida's Tobacco use among middle and high school students—National
Youth Tobacco Survey Fact Sheet (1999). [On-line]. Available:
Medicaid patients suffering from tobacco-induced illnesses, http://www.ced.govAobacco/mmwr4903fs.htm. Atlanta: CDC.
fund smoking cessation, restrict tobacco marketing, and fund Tuakli, N., Smith, M.A., & Heaton, C. (1990, October). Smoking in
an anti-tobacco education campaign. Rorida started an anti- adolescence: Methods for health education and smoking ces-
smoking campaign not only aimed at teenagers, but also sation. Journal ofFamiiy Practice. 31, 369-374.
developed by them. According to the literature, it has been so Worden, J.K., Flynn, B.S., Solomon, L.J., Seeker-Walker, R.H.,
successful that in 1998, the year that the campaign received Badger, G.J., & Carpenter, J.H. (1996). Using mass media to
the most funding in Florida, the teenage smoking prevalence prevent cigarette smoking among adolescent girls. Health
rate dropped 10%, and the number of 12-14-year-olds who Education Quarteriy. 23(4), 453-468.
smoke dropped 19%. These figures made it the most effective
such campaign in the country {In Rorida, kicking butts,
1999; CDC, 1999). Unfortunately, some political leaders were
not convinced that the program was working and funding was CaJtifor PoMuty Matters
recently diminished (Givel & Qlantz, 2000). Section Articles
Evidence shows that statewide, state-funded, anti-tobacco
mass media campaigns are effective as additions to commu- The Family Matters section focuses on issues,
nity efforts to decrease tobacco use among the target popula- information, and strategies relevant to working
tion. The type of advertisements that appeal to children can be with families of pediatric patients. To suggest top-
expensive to produce, and researching effective messages is ics, obtain author guidelines, or to submit queries
costly. Therefore, it is important for nurses and health care pro- or manuscripts, contact Elizabeth Ahmann, ScD,
fessionals to be aware of these types of interventions and their
effectiveness, and support their funding by state legislation. RN; Section Editor; Pediatric Nursing; East Holly
Nurses and other heaith care professionals need to learn how Avenue Box 56; Pitman, NJ 08071-0056; (856)
to support those health advocates who are active lobbyists and 256-2300 or FAX (856) 256-2345.
become professional organization members and lobbyists for
health care issues that are important. At the very least, health
care professionals should be aware of the types of effective
CaJll for Pra^tice^ A ^ nsof
interventions and messages that children are currently being Articles
exposed to and continue to reinforce anti-tobacco messages. The Practice Applications of Research section pre-
sents reports of research that are clinically
References focused and discuss the nursing application of the
Centers for Disease Control and Prevention (CDC). (1995). Healthy findings. If you are interested in author guidelines
people 2000 midcourse review and 1995 revisions. [On-Line]. and/or assistance, contact Janice S. Hayes, PhD,
Available: http://www.cdc.govAobacco/hpl2000.htm RN; Section Editor; Pediatric Nursing; East Holly
Centers for Disease Control and Prevention (CDC). (1998). Selected
cigarette smoking initiation and quitting behaviors annong Avenue Box 56; Pitman, NJ 08071-0056; (850)
high school students - United States, 1997. Morbidity and 256-2300 or FAX (856) 256-2345.
Mortality Weekly Report, 47(19), 386-389.

PEDIATRIC NURSING/May-June 2001/Vol. 27/No. 3