The Health Belief Model and smoking cessation behaviours

Articles

Enhancing Health Knowledge, Health Beliefs, and Health Behavior in Poland through a Health

Promoting Television Program Series

FIONA CHEW

S.I. Newhouse School of Public Communications

Syracuse University

Syracuse, New York, USA

SUSHMA PALMER

Center for Communications

Health and the Environment

Washington , DC, USA

ZOFIA SLONSKA

Department of Health Promotion

National Institut e of Cardiology

Warsaw, Poland

KALYANI SUBBIAH

S.I. Newhouse School of Public Communications

Syracuse University

Syracuse, New York, USA

This study examined the impact of a health promoting television program series on health knowledge and the key factors of the health belief model (HBM) that have led people to engage in healthy behavior (exercising, losing weight, changing eating habits, and not smoking=quitting smoking). Using data from a posttest comparison � eld study with 151 viewers and 146 nonviewers in Poland, we found that hierarchical regression analysis showed stronger support for the HBM factors of ef� cacy, sus- ceptibility, seriousness, and salience in their contribution toward health behavior among television viewers compared with nonviewers. Cues to action variables

The authors thank the Center for Communications, Health and the Environment, Washington, DC for funding this project.

Address correspondenc e to Fiona Chew, S.I. Newhouse School of Public Communications, 215 University Place, Syracuse University, Syracuse, NY 13244-2100 , USA. E-mail: cmrfchew@ syr.edu

Journal of Health Communication, Volume 7, pp. 179±196, 2002 Copyright # 2002 Taylor & Francis 1081-0730 /02 $12.00 + .00

DOI: 10.1080 /1081073029008807 6

179

(including television viewing) and health knowledge boosted ef� cacy among viewers. Without the advantage of receiving health information from the television series, nonviewers relied on their basic disease fears on one hand, and interest in good health on the other to take steps toward becoming healthier. A health promoting television series can increase health knowledge and enhance health beliefs, which in turn contribute to healthy behaviors.

In the past three decades, health promotio n entities in the United States and around the

globe have increasingly recognized the importance of lifestyle factors in the etiology and

prevention of chronic diseases (American Heart Association, 1996; American Institute

for Cancer Research=World Cancer Research Fund, 1997; Healthy People 2000, 1992; National Academy of Sciences, 1989; U.S. Department of Health and Human Services,

1988; World Health Organization, 1997; 1998). Key lifestyle risk factors included a poor

diet, alcohol consumption , tobacco use, and lack of physical exercise. Concurrently, these

organizations have proposed dietary and other lifestyle modi®cations to the general

public , and in some cases to subgroups of the populatio n at high risk, to lower their risk of

chronic diseases such as cancer, coronary heart disease (CHD), non-insuli n dependent

diabetes, hypertension , and obesity, among others. The general consensus among these

scientists and public health authoritie s is that smoking cessation, increased consumptio n

of plant foods, reduction in total fats (especially saturated fats), avoidance of excessive

alcohol consumption , and regular physical activity could signi®cantly curtail premature

deaths and disabilitie s as well as societal costs associated with unhealthy behaviors

(World Health Organization, 1997; 1998).

In examining global epidemiologica l trends, the scienti®c and public health com-

munities have highlighte d the following : Mortality from lifestyle-related diseases in

Western and other industrialize d nations has in general been declining in recent decades

while the converse appears to be true in developing nations and in various Newly In-

dependent States of the Former Soviet Union (Murray & Lopez, 1996). In the latter

societies as well as Former Communist countries, the transition to a market economy

without the necessary social support mechanisms has caused the health care systems to

collapse in many countries, exacerbating the trend toward poor lifestyle and a high

prevalence of CHD, cancer, and other chronic diseases.

The situation in Poland is complex. Since the 1990 s, total mortality attributed

to cardiovascular disease has decreased in keeping with declines in Western and in-

dustrialize d nations (La Vecchia, Levi, Lucchini, & Negri, 1998; Zatonski, McMichael,

& Powles, 1998). However, cancer mortality has increased and the total cancer mortality

among middle-aged males is one of the highest ever registered in Europe (Levi, Lucchini,

Negri, Boyle, & La Vecchia, 1999). Also, a rapid increase in alcohol consumption has

led to a correspondin g increase in alcohol-relate d problems (Swiatkiewicz, 1997). Con-

sequently, steps are needed to ameliorate various lifestyle-relate d diseases.

The World Health Organization, local governments, and local and internationa l

nonpro®t organizations have attempted various approaches to address the health crisis in

Central and Eastern Europe. Among these, a Washington-base d nonpro®t organization

promulgatin g communications, health, and the environment has engaged in health pro-

motion programs in Russia, Hungary, Poland, and the Czech Republic. One major focus

of this group’s activities is the use of mass media, particularly in the form of a series of

short or long television programs intended to inform the public about self-help steps to

improve their own health.

This study assesses the impact of one such television series, A Family Year, on targeted health-related behaviors of television audiences. It applies the health belief

180 F. Chew et al.

model (HBM) as a framework of analysis and assesses the contributio n of various HBM

factors on health behavior among television viewers and nonviewers.

Health Belief Model

The HBM has been used to investigat e a plethora of health promotio n behaviorsÐre-

ceiving immunizations, using preventive dental and health services, disease screenings,

and diagnosti c tests; assessing risk behaviors; and complying with medical advice. Its

development was derived from health assessments of risk in relation to health behaviors

made by U.S. public health service professionals in the 1950s, and its explanatory power

is predicated on two conditions . First, the bene®ts of a recommended action outweigh its

barriers, and, second, disease susceptibility and severity are high (Rosenstock, 1974). An

internal or external stimulus comprising a cue-to-action prompting health behavior has

been studied as a component of the HBM (Janz & Becker, 1984; Rosenstock, 1974), and

health motivation has often been included as a supplementary component (Gochman,

1971; Ogionwo, 1973; Rosenstock, 1974).

The application of the HBM as a framework to assess risk factors and promote

speci®c health behaviors has been well documented and has been a focus in the study of

CHD and its prevention (Mirotznik, Feldman, & Stein, 1995; Troein, Rastam, & Se-

lander, 1997). Its helpfulness in discriminatin g exercise behavior and participatio n in

physical activity has been demonstrated (Dolman & Chase, 1996; Sorensen, 1997; Swift,

Armstrong, Beerman, Campbell, & Pond-Smith , 1995; Taggart & Connor, 1995). Its use

in understanding and improvin g dietary behaviors has been increasing (Chew, Palmer, &

Kim, 1998; Hayes & Ross, 1987; Hollis et al., 1984; Kloeben, 1999; Kristal et al., 1995;

Schafer, Keith, & Schafer, 1995), and it has been used to understand lifestyle behaviors

such as smoking cessation and alcohol consumptio n (Conrad, Campbell, Edington , Faust,

& Vilnius, 1996; Minugh, Rice, & Young, 1998). Apart from one study focused on

general health (Cheng et al., 1996), little has been done to investigate its application in

improvin g overall healthy behavior.

In the HBM, ®ve basic factors in¯uence disease prevention behaviors (Janz &

Becker, 1984; Maiman & Becker, 1974; Rosenstock, 1974). Perceived susceptibilit y is

the ®rst factor. This refers to a person’s beliefs about the possibilit y of getting the disease

or being harmed by the condition. The second factor is perceived seriousness of the

consequences of the disease or health condition such as a disability or mortality. Per-

ceived bene®ts of performing the recommended behavior is the third factor and includes

feeling healthier or living longer. Perceived barriers to the suggested actions is the fourth

factor and may include cost, time, or inconvenience. Finally, cues to action complete the

model and may constitut e a physician’s advice, print or electronic advertisement, or

television program that elicits readiness to apply preventive health behaviors. Previous

researchers have also included a motivationa l factor (Gochman, 1971; Ogionwo, 1973;

Rosenstock, 1974), which has been incorporated in subsequent studies (Chew et al. 1998;

Hayes & Rass, 1987; Mirotznik, Feldman, & Stein, 1995; Schafer, Schafer, Bultena, &

Hoiberg, 1993).

The HBM states that preventive health behavior results when readiness to act

(presence of perceived susceptibilit y and perceived seriousness) and ef®cacy of the re-

commended response (perceived bene®ts outweigh perceived barriers) work cumula-

tively. Different factors or combination s of factors have produced major portion s of the

variation in taking preventive health steps depending on the disease, type of health be-

havior promoted, and=or individual’ s situation (Rosenstock, 1974). Two meta-analyses point to perceived barriers as the strongest predictor of health behavior (Janz & Becker,

Enhancing Health Knowledge 181

1984; Zimmerman & Vernberg, 1994). They also found that perceived susceptibility was

a strong predictor.

Cues to action have been examined in relation to health beliefs and these have

resulted in improved nutritio n behavior (Chew et al., 1998) and smoking reduction

(Mirotznik et al., 1995). In the ®rst study, researchers found that as a result of viewing a

television program that promoted good nutrition , viewers reported consuming plant foods

more often and foods high in fats less often. In the second study, working in a health

promotin g environment led to lower levels of smoking.

Health decision-makin g comprises a series of steps. At each step, interactions with

individual s or events may in¯uence the likelihoo d of a particular response (Rosenstock,

1974). When researchers analyzed speci®c factors as mediating variables in physical

safety behavior, these pointe d to a better understandin g of the paths of action (Witte,

Stokols , Ituarte, & Schneider, 1993). The case has also been made in dietary behavior

(Chew et al., 1998). Therefore, analyzing the impact of other factors as mediating

variables and the contributio n of the latter on preventive health behaviors may locate and

identify opportunitie s for more successful intervention approaches.

In promotin g healthy behavior, two additiona l mediators seem appropriate in the

context of a recently democratized societyÐhealth motivation and salience. These

variables provide a measure of an individual’ s sense of participatio n in and responsibilit y

for his or her own health. Health motivation refers to a general predispositio n toward

health such that if good health is valued, health motivation becomes an important

modi®er in the HBM (Kegeles, 1969). It is also considered an important organizer of

health beliefs and intentions (Rosenstock, 1974) and a key component of the inclinatio n

toward healthful diets (Kristal et al., 1995). It assesses the degree of involvement in

health issues. Previous research has measured health motivation as scaled measures for

concern with appearance and with health (Hayes & Ross, 1987), rating the importance of

food to health, and eating low-fat foods (Kristal et al., 1995) and the frequency of reading

food labels (Chew et al., 1998). When treated as a modifyin g variable, it has related

signi®cantly with health- and safety-conscious behavior (Hayes et al., 1987) as well as

dietary behavior (Chew et al., 1998). Consequently, it is worthwhile to investigate the

extent to which it mediates perceived readiness to act and perceived ef®cacy of re-

commended healthy behaviors.

Salience characterizes how close an individua l feels to the health condition and has

been shown to modify behavior (Cotugna, Subar, & Heimendinger, 1992). In a society

where information has been centralized and tightly controlled, health programming

provide d by governmental broadcasting stations and agencies would be scrutinized

carefully by individuals . How an individua l feels toward health bene®ts and barriers,

susceptibility , and seriousness is sometimes determined by its source. Under these cir-

cumstances, salience toward health re¯ects the psychologica l distance perceived by an

individua l between good health and him/herself. In previous communication research

salience has been assessed by using a scaled measure to indicate the degree of closeness

or distance from an event (Chew & Palmer, 1994; Ettema, Brown, & Luepker, 1983),

strength of feelings (Hanneman & Greenberg, 1979), and level of interest in a topic

(Becker & Preston, 1969; Bogart, 1957; Funkhouse r & McCombs, 1971). This analysis

propose s to assess the impact of salience as a component of the HBM. In addition it

propose s to assess the role of motivation in the HBM.

Some studies have shown that demographic and socioeconomic variables such as

age, gender, education, and income have modi®ed disease prevention behaviors (Chew

et al., 1998; Hayes & Ross, 1987; Rosenstock, 1974; Sensiba & Stewart, 1995; Taggart &

Connor, 1995). This study includes sociodemographi c variables in its analyses.

182 F. Chew et al.

Influence of Televised Messages on Health Behavior

Television has supplante d print media as the most frequently cited source of health in-

formation (Kaiser Family Foundation /Harvard School of Public Health, 2001, 2002; Dan,

1987; Signorelli, 1993; Wade & Schramm, 1969). A study showed that, in general,

mediated sources of informatio n (television , radio, newspapers, books, and magazines)

were used more frequently than personal sources of information (doctors, nurses, nutri-

tionists , dietitians, and home economists) as resources for nutritio n knowledge (Chew,

Palmer, & Kim, 1995). Television is able to reach large, diverse populations (Nielsen,

1998) and to stimulate healthy behavior (Atkin, 1981; Belicha & McGrath, 1990; Brown

& Einsiedel, 1990; Warner, 1987), but few efforts have been successful in promotin g

comprehensive healthy behavior.

Research shows that mass media have succeeded in increasing factual knowledge

and public awareness of health issues although they are less successful in changing

established attitude s and behavior (Rogers & Storey, 1987). Regular viewing of an

entertainment television program series resulted in new knowledge of as well as action

regarding a chronic disease (Sharf, Freimuth, Greenspon, & Plotnick, 1996). Nested

messages in entertainment programs and comprehensive public service announcement

(PSA) campaigns instilled the acceptance of the social norm of nondrinkin g or ``desig-

nated’’ drivers (Dejong & Winsten, 1990; National Highway Traf®c Safety Adminis-

tration, 1995; 1996; Winsten, 1994). Mass media messages improved knowledge of

cardiovascular health but were not suf®cient to sustain behavior changes (Farquhar et al.,

1990; Maccoby, Farquhar, Wood, & Alexander, 1977; Flora, 2001). A meta-analysis of

48 health communication campaigns showed that media exposure accounted for a 7% to 10% behavioral change, which was not homogeneous across campaigns, thereby pointin g to differences in moderator variables (Snyder, 2001).

Similarly, the evidence from non-U.S. contexts demonstrates the effectiveness

of televised campaigns on various health promotio n behaviors. Among others, these

include smoking cessation in Britain (McVey & Stapleton, 2000) and Finland (Korhonen

et al., 1992; Puska et al., 1987), child safety promotio n in Sweden (Sundelin , Rasmussen,

Berfenstam, & Troedsson, 1996), HIV testing in the United Kingdom (Ross & Scott,

1993), and family planning in Nigeria (Piotro w et al., 1990) and India (Singha l & Rogers,

1989).

Disease prevention campaigns are often considered less effective since the bene®ts

are attained in the future rather than in the present (Rogers & Storey, 1987). However,

many have succeeded, and the elements of a successful media campaign include the

following : speci®c attainable objectives (Mendelssohn, 1947), recommended behavioral

changes that are easy to perform (Bettinghaus, 1986), and message accessibility and

appeal (Brown & Einsiedel, 1990). Building on these lessons, we can develop messages

for a television series that can provide simple-to-follo w self-help tips for health. Such

media health messages are assuming a more signi®cant role in communities where

individuals must begin to take more responsibilit y for their own health and welfare

as public health services downsize. Such was the case in Central and Eastern Europe in

the 1990s.

This study assesses and compares the contributio n of the key HBM factors or health

beliefs toward healthy behaviors among television program viewers and nonviewers in

order to determine the impact of the television series. It focuses on whether health

knowledge increases as a result of viewing the series and how knowledge, in turn,

functions as a mediating variable. Finally, it identi®es the paths of action among health

knowledge, health motivation, and cues to actions on health beliefs.

Enhancing Health Knowledge 183

Method

Television Program Series

A Family Year, a series of ®ve half-hour television programs, was commissioned by a Washington-base d nonpro®t organization focused on communications, health, and the

environment and produced by a London-base d productio n company, with input from

health promotio n experts from the nonpro®t group and its partners in Russia, Hungary,

the Czech Republic, and Poland. The series was based on material derived from several

scienti®c reports. These included the global perspective issued by the American Institute

for Cancer Research=World Cancer Research Fund (1997), the Dietary Guidelines for Healthy American Adults (American Heart Association, 1996), the Surgeon General’s Report on Nutrition and Health (U.S. Department of Health and Human Services, 1988), the National Academy of Sciences report, Diet and Health (1989), and formative research conducted to assess the health needs of adult population s in Russia, Hungary, the Czech

Republic, and Poland (Center for Communications, Health and the Environment, 1995).

The series featured the scienti®c basis for the connection between diet, exercise,

lifestyle , and disease; the debates about these relationships , and the health situation in

Central and Eastern Europe. It differentiate d fact from fallacy and suggested practical

ways to lower the risk of heart disease, certain cancers, and other major lifestyle-relate d

disabilities . It also included steps to reduce the adverse effects of environmenta l pollu-

tion, approaches to eating a healthy diet, and tips on shoppin g and cooking, increasing

physical activity, quittin g smoking, avoiding excess alcohol, and living healthily. It

featured four families, one from each country, tracked them through the cycle of a year,

and participated in their efforts to live healthier livesÐexercising, quittin g smoking,

avoiding excess alcohol, shopping for and eating more healthy foods, and taking steps to

live in a cleaner environment.

The television series was transmitte d via several broadcast entities in these four

Central and Eastern European countries between November 1996 and January 1998. Data

assessing the impact of the television series were collected in ®ve metropolita n areasÐ

Moscow, Budapest, Debrecen, Prague, and WarsawÐwith the participatio n of local re-

searchers from November 1996 through July 1998. Only the Polish data will be reported

in this study. An estimated viewership of approximately 141,360 Polish household s tuned

in each week to view the series (Center for Communications, Health and the Environ-

ment, 1998).

Sample

A two-stage longitudina l posttest comparison ®eld study was conducted among 151

viewers and 146 nonviewers aged 18 and above. Viewers living in Warsaw were ran-

domly selected to participate in the study in two ways. First, they were invited by tele-

phone to view the television series at a central location. Two screenings were organized

over a 5-day period in order to view one program a day, starting on February 26, 1998,

and March 6, 1998. Viewers were interviewed in person at home directly after the last

program was screened and 6 weeks to 3 months later. A third of the viewers participated

in this way. Second, the remaining two thirds of the viewers received a videocassette of

the series. They were interviewed 6 weeks to 3 months after they had viewed at least one

of the ®ve programs in the series. Any individua l who had seen any broadcast program of

the television series was excluded from the sample. Viewers watched an average of 3.36

programs out of the ®ve in the series. About half of them (50.7%) watched four or more programs, while the rest watched three or fewer shows. The sample re¯ected a quota

184 F. Chew et al.

distributio n of 50% females and males with 33% young (18±34), middle-aged (35±54), and older (55±65). In addition, education was equally divided among primary school-,

high school-, and college-educated viewers while work status was similarly split among

the employed, retired, and those not working.

A demographically matched group of nonviewers was randomly recruited and inter-

viewed at home twice, once during the time the viewers were interviewed after viewing the

television series and the second time about 6 weeks to 3 months later. About 80% of the origina l sample participated in the second phase of the study. Since healthy behavior is

the focus of this study, only these data will be focused on as the outcome. Technically, the

study design was not a true experiment because of the lack of respondent randomization

between viewer and nonviewer groups . However, this quasi-experimental design yielded

similar demographic characteristics between the two groups and provided insightfu l

comparative data. Most of the viewing and nonviewing groups were 45 years and older

(55% vs. 56%). Females composed 52.3% of viewers and 48.6% of nonviewers, while those who earned a college degree consisted of 24.5% and 30.8% of the respective groups.

Measures

Five sets of variables derived from the HBM were measured: (1) ef®cacy (index of bene®ts

and barriers), (2) readiness (susceptibilit y separate from seriousness) , (3) health motiva-

tion, (4) salience, and (5) cues to action. Health behavior was the dependent variable.

Efficacy

Five Likert-scale items measuring perceived bene®ts and perceived barriers were indexed

to obtain ef®cacy. Respondents were asked how strongly they agreed or disagreed with

two ``bene®ts’’ (healthy diet, cancer=heart disease prevention) and three ``barriers’’ (time consuming, costly, give up favorite foods) statements. The 5 items were scored and

summed (range of 5 to 25) so that high ef®cacy would be represented by higher scores.

Cronbach’s alpha reliability coef®cient for ef®cacy was .75.

Readiness

Two measures were used. Perceived susceptibilit y assessed respondents’ belief that their

health was at risk. The question asked whether respondents were currently considering

or were on a special diet for health reasons (yes=no). Perceived seriousness was assessed by summing 8 items (yes=no), which asked whether respondents had a speci®c health condition . These 8 condition s included high blood cholesterol, heart disease, high blood

pressure, respiratory problems, obesity=being overweight, liver damage, colon cancer, and lung cancer. A higher score (scale of 0 to 8) would indicate a higher level of per-

ceived seriousness of the diseases=health contraindication s resulting from not pursuing a healthy lifestyle .

Health Motivation

This general measure assessed whether respondents felt motivated to do something about

their health and comprised two Likert-scale items that asked whether respondents felt

they could in¯uence their own health and whether they felt they could do something

about their pollute d environment. The two items did not correlate with each other, in-

dicating that they were independent measures of motivation . They were scored and

summed so that higher scores (2 to 10) re¯ected a higher health motivation .

Enhancing Health Knowledge 185

Salience

This was measured using a single question assessing closeness to the topic of health. The

question was, ``How concerned are you about health and ®tness?’’ on a scale of 1 (not at

all concerned) to 4 (very concerned). Higher scores re¯ected higher salience.

Cues to Action=Media Message

Media messages function as cues to stimulate action, and previous studies have shown

that the populatio n can be alerted by newspapers, ®lms, television spots, and other

messages (Flora, 2001; Levy & Stokes, 1987; Rosenstock, 1974; Witte, Stokols , Ituarte,

& Schneider, 1993). In the United States a plethora of health-related informatio n is

readily available to the public in newspapers, magazines, television, and advertising

(Chew et al., 1995; Goldberg, 1992; Lunin, 1987; Squires, 1986). In contrast, health

informatio n in Poland is circumscribed by ®nancial access, motivation, and dif®cult

living conditions . Thus, this analysis focused on respondents’ con®dence in assessing the

quality of dietary informatio n (con-®dence in information ) and the availability of reliable

informatio n (reliable information). Two Likert-scale items (1 to 5) tapping these di-

mensions were, ``There are so many recommendations about healthy ways to eat, I don’t

know what is good or bad,’’ and ``If there were reliable information and easy steps to

follow, people would change their diets to lower their chances of getting heart disease

and cancer.’’ A high (disagree) score on the ®rst item re¯ected respondents’ con®dence in

their nutritio n knowledge base, and a high (disagree) score on the second indicated a lack

of reliable informatio n for change to follow. Exposur e to the television program series

was also considered a cue for action.

Health Knowledge

This measure consisted of 6 Likert-scale items (1 to 5) tapping knowledge about the

relationshi p between eating animal fats and heart disease, consuming fruits=vegetables and cancer, regular exercise and health, heavy drinkin g and liver damage, environmental

pollutio n and health, and smoking and lung cancer. Scores were recoded where relevant

and summed (6 to 30), so that higher scores re¯ected higher health knowledge levels.

Cronbach’s alpha reliability coef®cient for health knowledge was .67.

Health Behavior

This measure comprised four questions (yes=no) focused on healthy practices and in- cluded the following : exercising in the past month, changing eating habits, trying to lose

weight, and not smoking or trying to quit smoking. Scores were summed so that a higher

score would re¯ect healthier behavior on a scale of 0 to 4.

Sociodemographi c Variables

These included categories of age, education, income, and gender.

Analysis

First, the mean scores comparing viewers and nonviewers on the HBM items and

mediators were examined using t tests. Second, percentages related to HBM items for viewers and nonviewers were assessed. Finally, the determinants of health behavior were

analyzed using hierarchical regression to estimate their respective path coef®cients. This

statistical procedure was used to predict health behavior from perceived susceptibilit y

186 F. Chew et al.

and seriousness of getting diseases, perceived ef®cacy, health motivation , salience, and

sociodemographic variables. Additiona l regression analyses were run to assess the re-

lationshi p among cues to action, health knowledge, and the health beliefs, which were

signi®cant predictors of health behavior. Sociodemographi c variables were also included.

Tests were set at the probability level of p < .05.

Results

Statistically signi®cant differences were found between viewers and nonviewers for ef®-

cacy, susceptibility , seriousness, health motivation , and the cue to actionÐlack of reli-

able information, health knowledge, and health behavior (p < .01 to .05). See Table 1. Viewers tended to score higher on these items compared with nonviewers. Salience was similar

for the two groups (2.85 vs. 2.86), as was the cue to actionÐcon®dence in informatio n

(3.28 vs. 3.10). Health knowledge was higher among viewers than nonviewers (p < .001). Tables 2 and 3 compare the results of the regression equations predicting health

behavior from health beliefs, salience, motivation, and cues to action for viewers and

nonviewers, respectively. They list the semipartial correlation (sr2) or unique proportio n of variance explained by a speci®c independent variable, its standardized beta weight

(Beta), the correlation (R), and variance (R2) for each of the ®nal equations. In the ®rst regression, four variables were found to signi®cantly predict health be-

havior among viewers (r2 ˆ .31). In step 1, ef®cacy was entered with the largest beta (.42, p < .001), followed by susceptibilit y (beta ˆ .19, p < .001), salience (beta ˆ .20, p < .05), and seriousness (beta ˆ .19, p < .025). These HBM variables were direct and substantia l determinants of health behavior and accounted for 31% of variance. Health motivation did not contribute directly to health behavior and as a result was not examined as a

mediating variable. None of the sociodemographi c variables had any effect on health

behavior among viewers.

In the second regression predicting ef®cacy among viewers, three variables were

found to be signi®cant. In all, 50% of the variance was accounted for. The cue to ac- tionÐcon®dence in information Ðwas selected ®rst with the highest beta (.49, p < .001), followed by lack of reliable informatio n (beta ˆ .31, p < .001). Health knowledge was the ®nal factor contributin g toward ef®cacy (beta ˆ .16, p < .025).

Age was found to be a factor mediating health beliefs. Older viewers were more

likely to consider the seriousness of poor health and becoming ill (beta ˆ .16, p < .001, variance of 15%). In addition, they were more likely to consider themselves susceptible to a medical condition and therefore were on a special diet. Those with more health

knowledge were also more likely to consider themselves susceptible to a medical con-

dition (betas ˆ .30 to .32, p < .001, variance of 17%). Finally, more knowledgeable viewers rated health and ®tness as more salient compared with their less-knowledgeable

counterparts (beta ˆ .26, p < .01, variance of 6%). See Table 2. Among nonviewers, the regression picture has only slight similarity. Only two health

belief variables were signi®cant in predicting health behaviorÐsusceptibilit y was the

stronger variable (beta ˆ .27, p < .001), followed by salience (beta ˆ .24, p < .01). Per- ceived seriousness did not enter the equation as it did for viewers’ health behavior. Neither

did ef®cacy or health motivation . A total variance of 12% was accounted for by these two health belief model variables. In further regression analyses, no signi®cant strong single or

combination of variables emerged to optimally predict susceptibility or salience. See

Table 3.

As a cue to action, viewing the television program series enhanced the application

of the HBM in promotin g health behavior as the health belief variables among viewers

Enhancing Health Knowledge 187

T A

B L

E 1

C o m

p a ri

so n

o f

H e a lt

h B

e li

e f

M o d e l

V a ri

a b le

M e a n

S c o re

s, H

e a lt

h K

n o w

le d g e

a n d

H e a lt

h B

e h a v io

r M

e a n

S c o re

s, a n d

P e rc

e n ta

g e s

A m

o n g

V ie

w e rs

a n d

N o n v ie

w e rs

M ea

n sc

o re

s It

em (s

ca le

) V

ie w

er S D

N o

n v ie

w er

SD t

d f

E f®

c a c y

(5 ± 2 5 )

1 5 .6

2 4 .0

5 1 4 .6

6 3 .4

6 2 .1

0 *

2 6 8

S e ri

o u sn

e ss

(0 ± 8 )

1 .0

8 1 .2

8 0 .7

2 1 .0

7 2 .5

7 *

2 9 0

S a li

e n c e

(1 ± 4 )

2 .8

5 0 .6

5 2 .8

6 0 .5

5 0 .1

1 2 9 4

M o ti

v a ti

o n

(2 ± 1 0 )

6 .1

2 1 .4

6 5 .6

7 1 .3

2 2 .7

7 * *

2 9 0

L a c k

re li

a b le

in fo

(1 ± 5 )

2 .4

1 1 .0

7 2 .1

5 0 .8

5 1 .5

7 *

2 8 7

C o n ®

d e n c e

(1 ± 5 )

3 .2

8 1 .0

8 3 .1

0 0 .9

0 2 .3

1 2 9 3

H e a lt

h k n o w

le d g e

(6 ± 3 0 )

2 6 .5

2 2 .3

9 2 5 .4

0 2 .3

9 3 .9

4 * * *

2 8 5

H e a lt

h b e h a v io

r (0

± 4 )

2 .0

3 1 .1

6 1 .7

6 1 .0

4 2 .1

4 *

2 9 6

P e rc

en ta

g es

re sp

o n d

in g

`` y es

’’ It

em V

ie w

er (%

) N

o n

v ie

w er

(% )

P h i

d f

S u sc

e p ti

b il

it y

1 9 .5

1 1 .0

.1 1 8 *

1

* p

< .0

5 ;

* * p

< .0

1 ;

* * * p

< .0

0 1 .

188

TABLE 2 Hierarchical Regression Predicting Health Behavior, Ef®cacy, Seriousness, Salience, and Susceptibilit y Among Viewers (n ˆ 151)

Health Behavior Predictor variables sr2 Beta p

1. Ef®cacy .18 .42 .001

2. Susceptibilit y .10 .19 .001

3. Salience .02 .20 .05

4. Seriousness .03 .19 .025

R2 ˆ .31 R ˆ .57*

Ef®cacy

1. Con®dence in info. .42 .49 .001

2. Lack of reliable info. .07 .31 .001

3. Health knowledge .02 .16 .025

R2 ˆ .50 R ˆ .72*

Susceptibility

1. Age .09 .32 .001

2. Health knowledge .09 .30 .001

R2 ˆ .17 R ˆ .42***

Salience

1. Health knowledge .06 .26 0.1

R2 ˆ .06 R ˆ .26**

Seriousness

1. Age .16 .40 .001

R2 ˆ .15 R ˆ .40***

*p < .05; **p < .01; ***p < .001.

TABLE 3 Hierarchical Regression Coef®cients Predicting Health Behavior Among Nonviewers (n ˆ 146)

Health behavior Predictor variables sr2 Beta p

1. Susceptibilit y .073 .27 .001

2. Salience .057 .24 .01

R2 ˆ .12 R ˆ .36***

***p < .001.

Enhancing Health Knowledge 189

accounted for more than twice the health behavior variance compared with that for

nonviewers (31% variance for viewers vs. 12% among nonviewers). Four health belief variables were signi®cant predictors of health behavior among television viewers

compared with only two among the nonviewers. Ef®cacy (the assessment of bene®ts as

outweighing the barriers of actions taken) was a strong predictor of health behaviors

among viewers. This suggested that program viewing helped to provide a clear picture

of the bene®ts resulting from and barriers preventing health behavior. Program viewing

also boosted other cues to action, particularly the perception of a lack of reliable in-

formation about heart disease and cancer prevention. This variable and its counterpart

cue to action (con®dence in information ) combined with health knowledge to produc e

greater ef®cacy. Television viewing led to higher levels of health knowledge, which in

turn increased health salience and susceptibility .

Among nonviewers, susceptibilit y and salience led to the practice of health beha-

viors. Without the advantage of viewing the television series, these respondents had to

rely on their intrinsi c interest in health and ®tness and their own sense of the likelihood of

getting a disease or medical condition .

Discussion

This study examined the contributio n of various HBM factors toward the practice of

health behaviors in the presence or absence of viewing a health promotin g television

program series. The HBM factors included susceptibility , seriousness , ef®cacy, cues

to action, health motivation , and salience. In addition, health knowledge and socio-

demographic variables were analyzed as mediating variables. The ®ndings suggest that

viewing the television program series, A Family Year, improved ef®cacy, seriousness, susceptibility , health motivation and health knowledge, the perception of a lack of reli-

able information , and the practice of health behaviors. Age was linked to disease sus-

ceptibilit y and seriousness.

Overall, in the context of health behaviors such as losing weight, changing eating

habits, exercising, and smoking cessation, health decision-makin g seemed to occur in the

following way. Among recipients of health information, in this case television viewers of

the health promoting series, A Family Year, the ®rst step involve s a greater awareness of ef®cacy, susceptibility , seriousness, and salience catalyzed by the factors health knowl-

edge, con®dence in assessing dietary information, and perceiving a lack of reliable in-

formation. This seems logical, because when respondents are presented with scienti®c

health information , they not only become more knowledgeable, they are also con®dent in

their knowledge about health guideline s and recognize the bene®ts of and barriers to

practicing health behaviors. In addition, they realize the dearth of reliable health

informatio n around them. Consequently, they acknowledge that they are susceptible to

negative unhealthy behavior consequences and perceive the seriousness of negative

health condition s such as high blood cholesterol, high blood pressure, obesity, and the

like. Good health behavior thus becomes an immediate focus in their lives.

Older viewers in Poland are more likely to perceive the seriousness of the con-

sequences of medical condition s because these will constrain their abilities and com-

promise their quality of life. Compared with younger viewers, their aging bodies will

require a longer recovery period. In addition, older viewers are more susceptible to

contracting a chronic disease by virtue of their longevity. Thus they are more likely to be

on a special diet. It is also possible that viewing the television series stimulated viewers to

become more aware of the status of their health so that the seriousness of their health

conditio n is acknowledged and realized. This needs to be con®rmed in a future study.

190 F. Chew et al.

Among nonviewers who do not have the advantage of receiving health information ,

susceptibilit y and salience foster good health behavior. It is interesting that perceived

ef®cacy and seriousness did not signi®cantly contribut e to health behavior among non-

viewers. This could have been attributed to the lack of available health informatio n

apprising them of the symptoms of health contraindication s and the bene®ts of and

barriers to good health.

Overall, nonviewers were not as well informed about health compared with viewers.

When respondents feel a closeness to the topic of health and ®tness, this translates into a

general concern about health, which is expressed in taking steps toward health. The HBM

factor of perceived susceptibilit y also in¯uences the practice of healthy behaviors. When

respondents feel their health is susceptible and that they highly value health, they will

practice healthy behaviors. This applies to both viewers and nonviewers.

In comparison, viewers appear to engage in healthy behaviors when they have

con®dence in the correctness of their health knowledge. This knowledge base was con-

ceivably supplemented and strengthene d by viewing the television program series about

health and ®tness. Arguably, program exposure promoted healthful behavior via the

various mediating variables. These steps are represented in Figure 1, which displays the

factors predicting healthful decision-making.

Four implications are proposed regarding the steps involved in healthy behavior.

First, one strategy to promote healthy behavior is to increase perceptions of ef®cacy,

susceptibility , seriousness, and salience. Second, a television program series focused on

improvin g health through disease prevention can increase health knowledge and in so

doing boost viewers’ ef®cacy and salience. Third, the optimal opportunit y for stimulating

ef®cacy is to produce messages that provide respondents with information that enhances

their con®dence in their ability to differentiate accurate from inaccurate reports as well

FIGURE 1 Determinants* of health behavior.

Enhancing Health Knowledge 191

as help them recognize the dearth of reliable information around them. Both of these

factors enhance ef®cacy. Finally, the television program series, A Family Year, appeared to improve healthy behavior practices by enhancing viewer con®dence in assessing health

informatio n and pointin g out the lack of reliable information . The televisio n series also

sensitize d older viewers to their susceptibilit y to diseases and the seriousness of the

consequences of these conditions .

Helping consumers learn and understand disease prevention steps to improve their

health is a tried and tested approach. Viewers scored higher on perception of a lack of

reliable health information compared with nonviewers, who tended to perceive that there

was reliable information on health. This could have been a function of the comparatively

centralized system of informatio n in Poland. When viewers were exposed to science-

based health information in the television series, they may have become more sensitized

to the dearth of reliable health informatio n in their society. In contrast, it is interesting to

note that a U.S. study (Chew et al., 1998) found the opposite result, that viewers of a

science-based health promotio n television program were more likely to perceive the

presence of reliable information compared with nonviewers. Therefore, more programs or

a series of programs providin g evidence-based health information would likely enhance

health behavior. The effectiveness of the television series, A Family Year, could be at- tribute d to its clearly stated objectives, the presentation of easy health promotin g steps

(e.g., eating more fruits and vegetables and exercising regularly) and the appeal of fo-

cusing on and following a family with whom viewers could identify.

A limitation of analyzing health beliefs is that these beliefs are constantly evolving.

Previous health belief researchers acknowledged the dynamic nature of health beliefs that

were modi®ed by interactions with events and individual s (Rosenstock, 1974). Another

limitation focuses on the variable nature of HBM factors measured across studies.

Concepts of perceived susceptibility , ef®cacy, cues to action, health motivation , and

salience may be de®ned differently and consequently lack comparability. Until research

studies that are focused on similar concepts and use similar measurements accumulate,

these results need con®rmation.

Social desirabilit y response bias and memory dependence are limitation s of self-

reported measures. In addition, self-selection bias may be present among the program

viewers who chose to watch the shows. Since the results are derived from a quasi-

experiment where demographically comparable viewers and nonviewers are surveyed at

the same time, the nonviewer group functions as a control group in this regard and any

increase or difference represents a genuine shift. The sample size of 151 viewers and 146

nonviewers is not considered a major limitation in view of the fact that this is a ®eld study

where 30 respondents per group would provide usable data. For perspective, it is re-

assuring to note that other HBM studies have been publishe d with smaller sample sizes,

for example, 57 CHD patients in the following : Mirotznik and colleagues (1995), 82

smokers and 228 nonsmokers in Conrad and colleagues (1996), and 113 female college

students in Taggart and Connor (1995).

Finally, while the results of this analysis may not be generalizable to the general

Polish population , they are re¯ective of a sample of television viewers and nonviewers

with more upscale demographics, including a disproportionat e number of middle-aged

and older persons. The analyses provide a pro®le of a more upscale populatio n that

yielded useful information about the HBM, television program exposure, health knowl-

edge, and health behaviors. Further research should focus on sources of information and

health behaviors among other populations and assess how speci®c televized messages

in¯uence health beliefs, health knowledge, and health behaviors and whether the current

®ndings are reinforced.

192 F. Chew et al.

The study’s main contributio n has been its application of the HBM with its entire

array of health belief factors in assessing the in¯uence of a television program series on

health behaviors. The positive results support the model’s heuristic in explaining and

predicting disease prevention behavior. Consequently, the study emphasizes that the

HBM variables of ef®cacy regarding the bene®ts and barriers to healthy behavior, sus-

ceptibility , seriousness, salience, and cues to action are important dimensions in pro-

moting healthy behavior. Viewing a health promotin g television series was shown to

enhance health knowledge, health beliefs, and ultimately health behavior.

References

American Heart Association. (1996). Dietary guidelines for healthy Americans. Dallas, TX: Author. American Institute for Cancer Research=World Cancer Research Fund. (1997). Food, nutrition and

the prevention of cancer: A global perspective. Washington DC: Author. Atkin, C. K. (1981). Mass communication research principles for healthy education. In M. Meyer

(Ed.), Health education by television and radio. Sauer: New York. Becker, J. D., & Preston, I. L. (1969). Media usage and political activity. Journalism Quarterly, 46,

129±134.

Belicha, T., & McGrath, J. (1990). Mass media approaches to reducing cardiovascular disease risk.

Public Health Reports, 105, 245±252. Bettinghaus, E. P. (1986). Health promotion and the knowledge±attitude±behavior continuum.

Preventive Medicine, 15(5), 475±491. Bogart, L. (1957). Measuring the effectiveness of an overseas information campaign: A case his-

tory. Public Opinion Quarterly, 21, 475±498. Brown, J. D., & Einsiedel, E. F. (1990). Public health campaigns. In E. B. Ray & L. Donohe w

(Eds.), Communication and health: Systems and applications (pp. 153±170). Hillsdale, NJ: Erlbaum.

Center for Communications, Health and the Environment. (1995). CECHE’S TV series, ``A Family Year:’’ Formative research. Washington, DC: Author.

Center for Communications, Health and the Environment. (1998). CECHE’S TV series, ``A Family Year:’’ Evaluation of impact. Washington, DC: Author.

Cheng, T. L., Savageau, J. A., Bigelow, C., Charney, E., Kumar, S., & DeWitt, T. G. (1996).

Assessing mothers’ attitudes about the physician’s role in child health promotion. American Journal of Public Health, 86(12), 1809±1812.

Chew, F., & Palmer, S. (1994). Interest, the knowledge gap and television programming. Journal of Broadcasting and Electronic Media, 38(3), 271±287.

Chew, F., Palmer, S., & Kim, S. (1995). Sources of information and knowledg e about health and

nutrition: Can viewing one television program make a difference? Public Understanding of Science, 4, 17±29.

Chew, F., Palmer, S., & Kim, S. (1998). Testing the in¯uence of the Health Belief Model and

a television program on nutrition behavior. Health Communication, 10(3), 227±245. Conrad, K. M., Campbell, R. T., Edington, D. W., Faust, H. S., & Vilnius, D. (1996). The worksite

environment as a cue to smoking reduction. Research on Nursing and Health, 19(1), 21±31. Cotugna, N., Subar, A. F., & Heimendinger, J. (1992). Nutrition and cancer prevention knowledge,

beliefs, attitudes, and practices: The 1987 National Health Interview Survey. Journal of the American Dietetic Association, 92(8), 963±968.

Dan, B. (1987). One minute medicine. Journal of the American Medical Association, 257, 2798. Dejong, W., & Winsten, J. A. (1990). The Harvard Alcohol Project: A demonstration project to

promote the use of the ``designated driver.’’ In M. W. B. Perrine (Ed.), Proceedings of the 11th

International Conference on Alcohol, Drugs and Traf� c Safety. (pp. 456±460). Chicago: National Safety Council.

Dolman, M., & Chase, J. (1996). Comparison between the Health Belief Model and Subjective

Expected Utility Theory: Predicting incontinence prevention behaviou r in post-partum

women. Journal of Evaluation of Clinical Practice, 2(3), 217±222.

Enhancing Health Knowledge 193

Ettema, J. S., Brown, J. W., & Luepker, R. V. (1983). Knowledge gap effects in a health infor-

mation campaign. Public Opinion Quarterly, 47, 516±527. Farquhar, J. W., Fortmann, S. P., Flora, J. A., Taylor, C. B., Haskell, W. L., Williams, P. T.,

Maccoby , N., & Wood, P. D. (1990). Effects of community-wide education on cardiovascular

disease risk factors: The Stanford Five-City Project. Journal of the American Medical Asso- ciation, 264(3), 359±365.

Flora, J. A., Maccoby, N., & Farquhar, J. (1989). Communication campaigns to prevent cardio-

vascular disease: The Stanford Community Studies. In R. E. Rice & C. K. Atkin (Eds.), Public information campaigns (pp. 233±252). Newbury Park, CA: Sage.

Flora, J. A. (2001). The Stanford Community Studies: Campaigns to reduce cardiovascular disease.

In R. E. Rice & C. K. Atkin (Eds.), Public information campaigns (pp. 193±213). Newbury Park, CA: Sage.

Funkhouser , G. R., & McCombs, M. E. (1971). The rise and fall of news diffusion. Public Opinion Quarterly, 35, 107±113.

Gochman , D. S. (1971). Some correlates of children’s health beliefs and potential health behavior .

Journal of Health and Social Behavior, 12, 148±154. Goldberg, H. P. (1992). Nutrition and health communication: The message and the media over half

a century. Nutrition Reviews, 50(3), 71±77. Hanneman , G. J., & Greenberg, B. S. (1979). Relevance and diffusion of news of major and minor

events. Journalism Quarterly, 50, 433±437. Hayes, D., & Ross, C. E. (1987). Concern with appearance , health belief and eating habits. Journal

of Health and Social Behavior, 28, 120±130. Healthy People 2000. (1992). Boston: Jones and Bartlett. Hollis, J. F., Sexton, G., Connor, S. L., Calvin, L., Pereira, C., & Matarazzo, J. D. (1984). The

family heart dietary intervention program: Community response and characteristics of joining

and nonjoining families. Preventive Medicine, 13(3), 276±285. Janz, N. K., & Becker , M. H. (1984). The health belief model: A decade later. Health Education

Quarterly, 11, 1±47. Kaiser Family Foundation/Harvard School of Public Health (2001). Health News Index Poll

(January 2002) . Health Poll Search, kaisernet.org. http://www.ropercenter.uconn.edu/cgi-bin/ hsrun.exe/roperweb/HPOLL/StateId/STiUQeK8r90fcJkHpgnFRsezKi/HAHTpage /Summary Link?qstn id ˆ 459196 (retrieved, 5/10/02).

Kaiser Family Foundation/Harvard School of Public Health (2002). Health News Index Poll

(January 2002) . Health Poll Search, kaisernet.org. http://www.ropercenter.uconn.edu/cgi-bin/ hsrun.exe/roperweb/HPOLL/StateId/STiUQeK8r90fcJkHpgnFRsezKi/HAHTpage /Summary Link?qstn id ˆ 459196 (retrieved, 5/10/02).

Kegeles, S. S. (1969). A ®eld experimental attempt to change beliefs and behavior of women in an

urban ghetto. Journal of Health and Social Behavior, 10, 115±124. Kloeben, A. S. (1999). Folate knowledge , intake from forti®ed grain products and periconceptional

supplementation patterns of a sample of low-income pregnant women according to the Health

Belief Model. Journal of American Dietetics Association, 99(1), 33±38. Korhonen , H. J., Niemensivu, H., Piha, T., Koskela, K., Wiio, J., Johnson, C. A., & Puska, P.

(1992). National TV smoking cessation program and contest in Finland. Preventive Medicine, 21(1), 74±87.

Kristal, A. R., Patterson, R. E., Glanz, K., Heimendiner , J., Hebert, J. R., Feng, Z., & Probart, C.

(1995). Psychosocial correlates of healthful diets: Baseline results from the Working Well

Study. Preventive Medicine, 24, 221±228. La Vecchia, C., Levi, F., Lucchini, F., & Negri, E. (1998). Trends in mortality from major diseases

in Europe, 1980±1993. European Journal of Epidemiology, 14(1), 1±8. Levi, F., Lucchini , F., Negri, E., Boyle, P., & La Vecchia, C. (1999). Cancer mortality in Europe,

1990±1994, and an overview of trends from 1955 to 1994. European Journal of Cancer, 35(10), 1477±1516.

Levy, A. S., & Stokes, R. C. (1987). Effects of a health promotion advertising campaign on sales of

ready-to-eat cereals. Public Health Reports, 102, 398±403.

194 F. Chew et al.

Lunin, L. F. (1987). Where does the public get its health information? Bulletin of the N.Y. Academy of Medicine, 63, 923±938.

Maccoby, N., Farquhar, J. W., Wood, P., & Alexander, J. (1977). Reducing the risk of cardio-

vascular disease: Effects of a community-based campaign on knowledg e and behavior.

Journal of Community Health, 3, 100±114. Maiman, L. A., & Becker , M. H. (1974). The health belief model: Origins and correlates in psy-

chological theory. Health Education Monograph s, 2, 336±353. McVey, D., & Stapleton, J. (2000). Can anti-amoking television advertising affect smoking

behavior ? Controlled trial of the Health Education Authority for England’s anti-smoking TV

campaign. Tobacco Control, 9(3), 273±282. Mendelssohn, H. (1947). Some reasons why information campaigns can succeed. Public Opinion

Quarterly, 11, 412±423. Minugh, P. A., Rice, C., & Young, L. (1998). Gender, health beliefs, health behaviors , and alcohol

consumption. American Journal of Drug and Alcohol Abuse, 24(3), 483±497. Mirotznik, J., Feldman, L., & Stein, R. (1995). The health belief model and adherenc e with a

community center-based, supervised coronary heart disease exercise program. Journal of Community Health, 20(3), 233±247.

Murray, C. J. L., & Lopez, A. D. (Eds.). (1996). The global burden of disease: A compre- hensive assessment of mortality and disability from diseases, injuries, and risk factors in 1990 and projected to 2020. Cambridge, MA: Harvard School of Public Health on behalf of the World Health Organization and the World Bank (Global Burden of Disease and Injury Series, Vol. I).

National Academ y of Sciences. (1989). Diet and health: Implications for reducing chronic disease risk. Washington, DC: National Academ y Press.

National Highway Traf®c Safety Administration. (1995). National survey of drinking and driving attitudes and behavior: 1993. Washington, DC: Author.

National Highway Traf®c Safety Administration. (1996). National survey of drinking and driving attitudes and behavior: 1995. Washington, DC: Author.

Nielsen, A. C. (1998). Media research reports. New York: The A. C. Nielsen Company . Ogionwo, W. (1973). Socio-psychological factors in health behavior: An experimental study on

methods and attitude change [Supplement]. International Journal of Health Education, 16(2, Suppl.), 1±16.

Piotrow, P. T., Rimon, J. G., II, Winnard, K., Kincaid, D. L., Huntington, D., & Convisser, J.

(1990). Studies in Family Planning, 21(5), 265±274. Puska, P., McAlister, A., Niemensivu, H., Piha, T., Wiio, J., & Koskela, K. (1987). A televised

format for national health promotion: Finland’s ``Keys to Health.’’ Public Health Reports, 102(3), 263±269.

Rogers, E. M., & Storey, J. D. (1987). Communication campaigns . In C. R. Berger & S. H. Chaffee

(Eds.), Handboo k of communication science (pp. 817±846). Newbur y Park, CA: Sage. Rosenstock, I. M. (1974). The health belief model and preventive health behavior. Health Edu-

cation Monograph s, 2, 354±385. Ross, J. D., & Scott, G. R. (1993). The association between HIV media campaigns and number of

patients coming for HIV antibody testing. Genitourin Medicine, 19(3), 193±195. Schafer, R. B., Keith, P. M., & Schafer, E. (1995). Predicting fats in diets of marital partners using

the health belief model. Journal of Behavioral Medicine, 18(5), 419±433. Schafer, R. B., Schafer, E., Bultena, G. L., & Hoiberg, E. O. (1993). Food safety: An application of

the health belief model. Journal of Nutrition Education, 25, 17±23. Sensiba, M. E., & Stewart, D. S. (1995). Relationship of perceived barriers to breast self-exam-

ination in women of varying ages and levels of education. Oncology Nursing Forum, 22(8), 1265±1268.

Sharf, B. F., Freimuth, V. S., Greenspon, P., & Plotnick, C. (1996). Confronting cancer on thir- tysomething: Audience response to health content entertainment television. Journal of Health Communication, 1, 157±176.

Signorelli, N. (1993). Mass media images and impact on health. Westport, CT: Greenwood .

Enhancing Health Knowledge 195

Singhal, A., & Rogers, E. M. (1989). Prosocial television for developmen t in India. In R. E. Rice &

C. K. Atkin (Eds.), Public information campaigns (pp. 331±350). Newbury Park, CA: Sage. Sorensen, M. (1997). Maintenance of exercise behavior for individuals at risk for cardiovascular

disease. Perceptual and Motor Skills, 85(3 Pt. 1), 867±880. Snyder, L. (2001). How effective are mediated health campaigns ? In R. E. Rice & C. K. Atkin

(Eds.), Public communication campaigns (pp. 181±192). Thousand Oaks, CA: Sage. Squires, S. (1986, July 2). The gatekeeper s of medical news: Leading journals determine the

agenda. Washington Post Health, pp. 8±9. Sundelin, C., Rasmussen, F., Berfenstam R., & Troedsson, K. (1996). Information through tele-

vision: Does it promote child safety? Injury Prevention 2(1), 36±40. Swiatkiewicz, G. (1997). Regulating unregulated markets. Addiction, 92(Suppl.), S67±71. Swift, C. S., Armstrong, J. E., Beerman, K. A., Campbell, R. K., & Pond-Smith, D. (1995). Atti-

tudes and beliefs about exercise among persons with non-insulin-dependen t diabetes. Diabetes Education, 21(6), 533±540.

Taggart, H. M., & Connor, S. E. (1995). The relation of exercise habits to health beliefs

and knowledge about osteoporosis. Journal of the American College of Health, 44(3), 127±130.

Troein, M., Rastam, L., & Selander, S. (1997). Health beliefs and heart disease risk among middle-

aged Swedish men: Results from screening in an urban primary care district. Scandinavian Journal of Primary Health Care, 15(4), 198±202.

U.S. Department of Health and Human Services. (1988). Surgeon General’s report on nutrition and health. Washington, DC: Governmen t Printing Of®ce.

Wade, S., & Schramm, W. (1969). The mass media as sources of public affairs, science, and health

knowledge. Public Opinion Quarterly, 33, 197±209. Warner, K. E. (1987). Television and health education: Stay tuned. American Journal of Public

Health, 77(2), 140±142. Winsten, J. A. (1994). Promoting designated drivers: The Harvard Alcohol Project. American

Journal of Preventive Medicine, 10(Suppl. 1), 11±14. Witte, K., Stokols, D., Ituarte, P., & Schneider, M. (1993). Testing the health belief model in a ®eld

study to promote bicycle safety helmets. Communication Research, 20(4), 564±586. World Health Organization. (1997). Action Plan for a Tobacco-Free Europe and Tobacco and the

Media. The European Health Communications Network: CCEE=NIS Broadcaster Workshop. Kiev: Author.

World Health Organization. (1998). Opening address by WHO Director-General, Dr. Gro Harlem at the Seminar on Tobacco Industry Disclosures. Geneva: Author.

Zatonski, W. A., McMichael , A. J., & Powles, J. W. (1998). Ecological study of reasons for sharp

decline in mortality from ischaemic heart disease in Poland since 1991. British Medical Journal, 316(7137), 1047±1051.

Zimmerman, R. S., & Vernberg, D. (1994). Models of preventive health behavior : Comparison,

critique and meta-analysis. Advances in Medical Sociology, 4, 45±67.

196 F. Chew et al.