Assessing knowledge of cardiovascular health-related diet and exercise behaviors in anglo- and Mexican-Americans

PREVENTIVE

MEDICINE

16, 696-709 (1987)

Assessing Knowledge of Cardiovascular Health-Related Diet and Exercise Behaviors in Anglo- and Mexican-Americans1 WILLIAM A. VEGA, PH.D.,*,* JAMES E SALLIS, PH.D.,? THOMAS PATTERSON, PH.D. ,t JOAN RUPP, M.S., R.D. ,t CATHERINE ATKINS, PH.D.,? AND PHILIP R. NADER, M.D.-F *Hispanic

Community Research Project, San Diego State and fUniversity of California at San Diego,

University, La Jolla,

San Diego, California California 92093

92182,

This article describes the Adult and Child Behavior Knowledge Scales that were used as part of the San Diego Family Health Project to measure knowledge of health behaviors related to cardiovascular diseases in two ethnic groups: Anglo- and Mexican-Americans. The psychometric characteristics of these scales indicate acceptable reliabilities for assessing knowledge of dietary sodium, dietary fat, and exercise among both adults and children and differ from other health knowledge scales in that they focus on “behavioral capability” rather than on the link between behavior and disease. It is believed that the type of information measured by our scales is more closely related to behavior changes sought in contemporary cardiovascular disease prevention trials. Results of ANOVA used to test differences in knowledge by ethnicity and sex indicate strong main effects for ethnicity among both children and adults. However, sex was not consistently related to knowledge, except for the general tendency of males to be more knowledgeable about exercise. Step-wise and simultaneous-entry multiple regression were used to test a subset of variables, including sex, education, self-efficacy, acculturation (for Mexican-Americans), and parental health knowledge (for children) as determinants of health knowledge. Education was the strongest predictor for Anglo-American adults, and acculturation level was the strongest for Mexican-American adults. Among children, the only statistically significant variable was parental acculturation level for Mexican-Americans. The scales were found to be useful in measuring differences in knowledge across culturahlinguistic groups and to clearly identify marginally acculturated Mexican-Americans as being least aware of health-behavior knowledge. Implications are discussed. o 1987 Academic press, hc.

INTRODUCTION

Knowledge acquisition is believed to be one step in the process of health-behavior change (5), but the conceptualization and measurement of knowledge in health promotion research has been haphazard. Most health knowledge scales assess awareness of the link between various health behaviors and probable disease outcomes (3, 8, 16). While assessment of this type of knowledge may be important, different types of knowledge are required in order to make the behavior changes that are targets of most intervention programs. In order to successfully modify one’s health habits, it is necessary to have the “behavioral capa-

r Supported by Grant HL 30872, from the National Institutes of Health to Philip Nader, M.D., Principal Investigator, Division of General Pediatrics, University of California at San Diego. z To whom reprint requests should be addressed at Child and Family Health Studies, M-03lF, University of California at San Diego, La Jolla, CA 92093 696 0091-7435187 $3.00 Copyright 8 1987 by Academic Press, Inc. All rights of reproduction in any form reserved.

ASSESSING

KNOWLEDGE

OF CARDIOVASCULAR

HEALTH

697

bility.” Parcel and Baranowski (12) suggest that knowing what the behavior is is a precondition of behavior change. Thus, in a health promotion program, the most important aspects of knowledge should be those that are essential for, or at least directly related to, behavior change. In the present study knowledge scales that were designed to assess knowledge of health behaviors believed to be related to cardiovascular disease (CVD) risk were developed for children and adults. Since the scales are being used in a study of a family-based intervention aimed at changing dietary fat, dietary sodium, and exercise, knowledge of only those health behaviors was assessed. This article describes variations in levels of health knowledge among Anglo- and MexicanAmericans with data collected from a sample of parents and their fifth-/sixthgrade children participating in the San Diego Family Health Project (10). Since Mexican-Americans are the largest minority ethnic group in the Southwestern United States, it is surprising to find so few reports in the professional literature about their knowledge of health behavior. Several relationships between cardiovascular health knowledge and other variables have been found. For example, females typically have higher levels of health knowledge than males (16). Age, education, and income are positively related to knowledge as well (8, 9, 14, 16). Hazuda and associates (8), reporting results of a community study in Texas, found lower levels of coronary heart disease knowledge among Mexican-Americans when compared with Anglo-Americans, even when study respondents were stratified by neighborhood. Similarly, in a study of Texas schoolchildren (3), Mexican-Americans had lower heart health knowledge scores than Anglo-Americans. The present study had two major objectives: first, to develop scales to measure knowledge of health behaviors related to CVD among children and adults and to evaluate the psychometric qualities of the scales; and second, to investigate ethnic differences in health-behavior knowledge and to identify determinants of such knowledge. In addition to sex and education, self-efficacy was studied as a variable related to health knowledge. Self-efficacy, or confidence in one’s ability to perform a specific behavior, has consistently been related to actual behavior change (2), and it is hypothesized that persons with high levels of health-behavior knowledge will have more confidence, or self-efficacy, in their ability to perform the behavior. The variable of acculturation (cultural orientation) has not been directly measured in previous research with Mexican-Americans and is included in the present analyses to determine the strength of its relationship to healthbehavior knowledge. METHODS

Psychometric Characteristics of Adult Knowledge Scale Thirty-eight multiple-choice and true-false items were developed by the investigators or were taken from the Stanford Five-City Project (6). The draft scale was pretested at an elementary school in a low-middle-income neighborhood. The draft scale was sent home with all 191 third- and fourth-grade students. Sixty-four percent (n = 123) returned two surveys that were sent home at l-week intervals.

698

VEGA

ET AL.

The ethnic composition of the sample was as follows: whites, 36%; blacks, 17%; Mexican-Americans, 9%; and Asians, 38%. Items for the final scale were selected for the following reasons: (a) item difficulty (percentage of subjects answering scale items correctly), (b) item reliability (percentage correct or incorrect at both testings), and (c) item discriminability (correlation of item with total score). The best six items were selected for each of three subscales, i.e., dietary salt, dietary fat, and exercise, for a total of 18 items. Table 1 displays the mean, standard deviation, test-retest reliability, and alpha coefficient for each subscale and the total scale. The complete scale is shown in Appendix A. The final scale has adequate alpha reliability and is at an ideal difficulty level in that the mean score is about 50% of the possible score. The test-retest reliability reflects acceptable stability for the total score. The reliabilities of the subscales suggest that if one is cautious in his or her interpretation, subscale scores can be analyzed separately. Due to the nature of the subject sample, including the fact that English was a second language for many respondents and to the diverse socioeconomic composition, these results represent a conservative assessment of the psychometric qualities of this scale. Psychometric Characteristics of Child Knowledge Scale Three multiple-choice items sampled knowledge of behavior related to dietary salt, dietary fat, and exercise behaviors, for a total of nine items. The items were developed by the investigators and were taken from the Special Project in Nutrition being conducted in San Francisco by the Gladstone Foundation. The scale was tested on fifth and sixth graders (n = 68) in an elementary school located in a low-income area. Children completed the scale in class twice, and testings were separated by 1 week. The racial-ethnic composition of the sample was 12% white, 6% black, 58% Mexican-American, and 12% Asian. Table 2 presents the mean, standard deviation, test-retest reliability, and alpha coefficient for each subscale, as well as the total scale. The entire scale is shown in Appendix 2. The test-retest reliability and alpha coefficient of the total scale were acceptable, considering the age of the population, and indicate that this scale has moderate psychometric qualities. Test-retest reliabilities for child subscales were similar to values for the adult subscales. Alphas for the child scale were lower, TABLE

1

PSYCHOMETRIC CHARACTERISTICS OF THE ADULT HEALTH BEHAVIOR KNOWLEDGE SCALE (N = 123) Subscale

Mean

SD

Test-retest reliability

Alpha

Dietary salt (6 items) Dietary fat (6 items) Exercise (6 items) Total (18 items)

3.4 2.5

1.6 1.4

0.61 0.60

0.55 0.57

3.5 9.3

1.7 4.0

0.57 0.76

0.59 0.80

ASSESSING

KNOWLEDGE

OF CARDIOVASCULAR

HEALTH

TABLE 2 PSYCHOMETRICCHARACTERISTICSOFTHECHILDHEALTHBEHAVIORKNOWLEDGESCALE(N

699 = 68)

Subscale

Mean

SD

Test-retest reliability

Alpha

Dietary salt (3 items) Dietary fat (3 items) Exercise (3 items) Total (9 items)

1.1 1.8 1.4 4.3

0.92 0.83 0.95 1.91

0.66 0.58 0.58 0.73

0.29 0.27 0.26 0.51

but this was expected because of the small number of items in each subscale. Therefore, subscale scores on the Child Health Behavior Knowledge Scale should be considered with extreme caution. The mean score indicated ideal scale difficulty, in that most children get some correct, but there is much room for improvement. Both child and adult knowledge scales went through a rigorous translation process. After being translated into Spanish by two bilingual staff members, they were back translated by two bilingual community residents with limited educational attainment. Finally, a panel of bilingual research staff eliminated inconsistencies and produced the final version. Sample Characteristics The adult and child knowledge scales were administered to participants in the San Diego Family Health Project. This project is a trial of a family-based healtheducation program designed to alter health-related diet and exercise behaviors (10). Families were recruited through their fifth- and sixth-grade children from randomly selected grade schools located either in high-density Mexican-American or in Anglo-American neighborhoods. Results of baseline measurements were used for this article. Participating volunteers and nonparticipating families were compared on key variables, including socioeconomic status, nativity, language use, household composition, and smoking, diet, and exercise habits. No important differences were noted between participating and nonparticipating families within respective ethnic groups (1). The mean age of adult Anglo-American participants was 38.5 years and for Mexican-American respondents it was 36.6 years, with husbands in both groups being 2 years older than their spouses. However, socioeconomic status levels for Mexican-Americans were significantly lower than those for Anglo-Americans, and the majority of Mexican-Americans were monolingual Spanish speakers. Resource limitations precluded measuring more than two children in each family, and in the analyses that follow, only children between 10 and 13 years of age are included. Acculturation Acculturation was measured using a modified version of the Cuellar et al. (4) scale, which was also used in the Hispanic Health and Nutritional Examination Survey (11). This scale assesses cultural orientation of Mexican-Americans using an ordinal continuum of 1 to 5, with the lower end representing “very Mexican,”

700

VEGA ET AL.

the mid-point representing “bi-cultural,” and the high endpoint representing “fully Anglo acculturated.” The scale includes items that tap preferences in language use, friendship patterns, and media, as well as nativity status of respondent and immediate ancestors. Measurement of Health Self-Efficacy The self-efficacy scale required subjects to rate confidence in their ability to regularly engage in specific diet and exercise behaviors. Two questions on exercise (e.g., confidence that they can do vigorous activity such as jogging, running, and swimming laps for 20 minutes three or more times a week) have been shown to predict actual exercise changes (15). There were four questions on reducing fat intake (e.g., confidence that they can “use unsaturated vegetable oil instead of lard or shortening in cooking”). Two questions concerned confidence in ability to reduce salt intake (e.g., confidence that they can “reduce the salt you add both in cooking and at the table by half or more”). Each item was rated on a 5-point scale from “I know I cannot,” at the low end of the scale, to “I know I can,” at the upper end. The scale is available from the authors. RESULTS

Differences by Ethnicity and Gender Table 3 presents mean scores for the Adult Health Behavior Knowledge Scales and three subscales: exercise, fat, and salt. Scales were scored by giving 1 point for each item; 6 points were possible for each subscale and 18 points for the total scale. The mean score differences between Anglo- and Mexican-American adults were that Anglo-Americans scored consistently higher for all subscales and the total scale. Further, consistent and marked ethnic differences held, regardless of sex. On the total scale, ethnic differences averaged close to 5 points. Only on the fat subscale were there fairly similar knowledge levels across groups. Table 3 summarizes the two-way ANOVA for adults, by ethnicity and sex. Main effects for ethnicity were found for the total scale, as well as for exercise, TABLE 3 HEALTH-KNOWLEDGEMEANSCORESFORADIJLTS(N Mexican-Americans

Anglo-Americans Males (59)

Females

(88)

Males (42)

Females (105)

12.24 k 2.32

12.19 k 2.90

7.45 iz 2.28

6.96 r 3.54

Exercise

5.07 i 1.26

4.75 4 1.42

2.67 e 1.26

2.27 f 1.52

Fat

2.41 t 1.05

2.82 2 0.94

2.31 ‘- 1.11

2.17 + 1.22

Salt

4.76 k 1.07

4.63 k 1.31

2.48 t 1.44

2.52 +- 1.53

Subscale Total

= 294) 2-Way ANOVA source

sex eth sex eth sex eth sex eth

F

P<

0.44 203.53 4.23 223.66 1.38 12.83 0.97 180.78

0.507 0.000 0.041 0.000 0.241 0.000 0.757 0.000

Note. Interactions: fat (eth x sex) P < 0.05. Values are presented as means 2 standard deviation.

ASSESSING

KNOWLEDGE

OF CARDIOVASCULAR

701

HEALTH

fat, and salt, and these were significant at P < 0.001. The only significant main effect for sex (P < 0.05) was on the exercise subscale, where males scored higher on knowledge than did females in both ethnic groups. Only one interaction between ethnicity and sex was detected, and that was the fat subscale. Means from the Children’s Health Behavior Knowledge Scale are reported in Table 4. Scales were scored by giving one point for each correct item; three points were possible for each subscale and nine points for the total scale. As reported for the adults, the children’s mean scores were consistently higher for Anglo-Americans of both sexes than for Mexican-Americans. Only the fat subscale produced somewhat similar mean scores. The two-way ANOVA produced statistically significant main effects for ethnicity on the total scale (P < O.OOO), exercise (P < O.OOl), and salt (P < 0.000). There were no significant sex main effects, nor were there any interactions. Multiple Regression Anulyses

Since important ethnic effects were noted in the ANOVA for both adults and children, it was obvious that we were dealing with two different populations. Given this finding, and because we wanted to test the power of additional model variables known from previous research to be predictors of health knowledge, we conducted a series of multiple regression analyses separately for each ethnic group using both simultaneous entry for all variables and step-wise methods, with the total knowledge scores as dependent variable. The variables in the equations for adults and children were sex, education, and health self-efficacy, with acculturation added for Mexican-Americans. The variables selected were seen as the most parsimonious model, given anticipated intercorrelations for additional demographic variables in each ethnic group, which, in multiple regression analyses, serves to eliminate variables that explain common variance. It should be noted that the education variables for Anglo- and Mexican-Amer-

TABLE 4 HEALTH-KNOWLEDGE MEAN SCORES FOR CHILDREN (N = 270) Anglo-Americans Males

Mexican-Americans

Females (50)

Males

(65)

(72)

(83

Total

4.31 2 2.05

4.22 k 2.10

3.21 f 1.72

3.24 2 1.90

Exercise

1.65 f 1.01

1.58 f 0.95

1.21 k 0.96

1.22 t 0.92

Fat

1.00 f 0.87

1.00 * 0.97

0.97 k 0.82

0.83 f 0.84

Salt

1.66 2 0.99

1.64 2 0.83

1.03 ? 0.75

1.19 k 0.77

Subscale

2-Way ANOVA source

Females sex eth sex eth sex eth sex eth

F

P<

0.60 19.23 0.39 11.48 0.59 0.79 0.71 27.90

0.938 0.000 0.844 0.001 0.444 0.376 0.399 0.000

Note. Interactions: None at P < 0.05. Values are presented as mean 2 standard deviation.

702

VEGA

ET AL.

ican children and the acculturation variable for Mexican-American children were created using their parents’ mean scores for these variables. This was done because the children were very homogeneous in age, and therefore in education. In the case of Mexican-Americans, children were very similar in cultural orientation and generally were much more acculturated than their parents. An additional variable, “knowledge,” was used for children only and was included in order to test the value of parents’ health knowledge as a determinant of their children’s knowledge. Table 5 summarizes results of regression analyses for adults and children in the form of beta coefficients for each of the four subgroups. The simultaneous entry results are displayed as “Simultaneous” for all variables in the equation. Those variables that entered in the step-wise model, using the P < 0.05 criterion, are presented as “Stepwise” for each subgroup. There were only minor differences in beta coefficients using the two methods. For Anglo-American adults, education was the only statistically significant variable (P < O.OOl), and the total explained variance (R2) was 0.09 using all variables in the equation. For Mexican-American adults, only acculturation was statistically significant (P < 0.001). Although education was also a good predictor, it did not enter in the step-wise regression, presumably because of its high correlation (Y = 0.69) with acculturation. The total explained variance for Mexican-American adults was 0.17. For Anglo-American children, only betas for self-efficacy and parent knowledge were somewhat larger than other predictors, and even these variables failed to enter using the step-wise method. Total R2 was 0.05. For Mexican-American children, knowledge and acculturation were useful, but only acculturation entered using step-wise analysis and was significant at P < 0.01. Explained variance was 0.08. Perhaps the most puzzling finding was that health knowledge of parents was negatively correlated with their children’s knowledge in both ethnic groups, though the association was weak. DISCUSSION

Scales were developed to assess knowledge of health-related behaviors, and separate scales for children and adults were found to have adequate reliabilities. The scales are brief and have been used with subjects of widely varying educational levels. The Health Behavior Knowledge Scales have proven useful for assessing differences in knowledge levels among adults and children of differing cultural/language groups, Furthermore, the subscales of the adult measure also have value for more specific assessment of cardiovascular health knowledge. We are currently analyzing longitudinal data on knowledge changes between intervention and control groups and have found a large increase occurring in our intervention group as assessed by these scales. We strongly encourage replication of our research with other samples in order to generate normative data on cardiovascular knowledge and its relationship to other factors, such as attitudes, values, and behaviors, which are included regularly in descriptive and intervention research. While we believe that knowledge of health behaviors is an important precondi-

a NA, Not applicable. ** P < 0.01. *** P < 0.001.

R2

Sex Education Effkacy Acculturation Knowledge

0.07 0.31 0.09 NA” NA 0.09

Simultaneous

Simultaneous -0.02 0.17 0.05 0.26 NA 0.17

0.28*** NA NA 0.08

.39*** NA 0.15

Step-wise

Mexican-Americans

Step-wise

Anglo-Americans

Adults

- 0.05 0.04 0.18 NA -0.12 0.05

Simultaneous NA -

0.01 0.05 - 0.05 0.32 -0.17 0.08

0.23** 0.05

Step-wise

Mexican-Americans Simultaneous

Children

Step-wise

Anglo-Americans

TABLE 5 BETAS FOR SELECTED VARIABLES: ADULTS AND CHILDREN

Fl F 2

2 8 8 s E s %

i

2 E;; x

!2

704

VEGA

ET AL.

tion of behavior change, we do not want to foster a notion that knowledge 01 health behaviors is the most important determinant of actual behavior change The communication-behavior change model, developed by the Stanford Commu nity Studies investigators (5), suggests that awareness, knowledge, motivation and skills precede behavior change, though in reality the sequence is not in, variant. An individual may know which behaviors would promote health, bu effective motivation and proper skills are also required in most cases before tht behavior is performed. Thus, knowledge of health behavior is only one of severa important determinants of health behavior. Our results confirm earlier studies which have found Anglo-Americans to have higher levels of health knowledge than Mexican-Americans, though it was evi, dent from the multiple regression results that education mediated health knowl, edge in both ethnic groups (for adults). Our results differ from those of other studies in that sex was not consistently associated with knowledge in either ethnic group, except in the case of exercise, where knowledge among males wa: greater. This finding may be due to the type of knowledge that was being mea sured. Sources of information about health and diet may be similar, fragmentary and somewhat remote for both sexes, originating primarily from media sources On the other hand, information about exercise, as well as participation in exercise activities, may tend to be geared toward adult males in U.S. society. Another contribution of these analyses is the finding that the cultural orientation of Mexican-American parents is a powerful predictor of health knowledge foi themselves and their children and is a stronger predictor than educational attainment. We highly recommend inclusion of comparable acculturation indices in future health research with Mexican-American populations, including studies designed to examine a range of beliefs, values, and behaviors that could have heuristic value in accounting for knowledge differences within this ethnic group. It is of interest that the higher levels of health knowledge among parents were not reflected in higher knowledge scores in their children. Our only health attitude variable, health efficacy, is not very useful in predicting knowledge differences in our four subsamples, with a marginal exception being Anglo-American children. This finding could reflect actual differences between Anglo- and Mexican-Americans in parental modeling and in the practical experience of AngloAmerican children. On the other hand, it could also be the case that self-efficacy and knowledge are independent constructs, with differential value as behavioral determinants. An important outcome of our research is the finding that, despite rather low knowledge levels about cardiovascular risk indicators among Mexican-Americans, they are nevertheless responsive to primary prevention programming and participation in relatively intensive programs such as the Family Health Project. This discovery implies that interest in family health promotion may not be related to health knowledge among low-socioeconomic and marginally acculturated Mexican-Americans. Therefore, lack of knowledge in this population should not be construed as equivalent to disinterest in, or lack of receptivity to, cardiovascular interventions, especially when these are directed toward families. Paren-

ASSESSING

KNOWLEDGE

OF CARDIOVASCULAR

HEALTH

705

thetically, Ramirez et al. (13) report Mexican-Americans are likely to have many misconceptions about high blood pressure and to suffer from poor communication with their physicians, which aggravates the problem of receiving accurate cardiovascular health information. In conclusion, the Health Behavior Knowledge Scales were found to be useful in assessing levels of cardiovascular knowledge in two ethnic groups, Anglo- and Mexican-Americans. We believe that the “behavioral capability” orientation, as well as the technical performance of these scales, makes them well suited for use in other health-promotion studies. Our substantive results indicate that within the Mexican-American population, priority attention should be given to immigrants and other unacculturated families, as this is a rapidly increasing segment of the population that has a particularly low level of knowledge related to CVD prevention. The fact that these families are also least likely to have regular health screenings or to enjoy health insurance is further evidence of the need for carefully focused interventions. Certainly, Spanish-language television and radio stand out as primary vehicles for increasing knowledge about CVD within this subgroup of Mexican-Americans. Similarly, school systems should be encouraged to increase their educational programming in heart health by conducting CVD health interventions, by encouraging family participation in programs (lo), and by disseminating easily understood educational materials in Spanish and English. Although these steps may not address variables involved in the behavior change process (5), they do represent pragmatic and concrete methods of accessing this population, and they could ultimately be supplemented by carefully targeted community-based interventions (7).

APPENDIX

A

S.4MPLE

1

The best way to reduce blood pressure

2

0 1 0 2. 0 : 0 4. 0 5 0 8 Which

Reduce cholertcrol Intake Lose weight Rertnct rah I and 3 above 2 and 3 above Don’t knox of lhc lollormg roods IS highest m raturawd

0 0 0 0 0

Pcanu1r Brcf liver Frankiurters Roa.w Beef Don’1 know

1. 2 3 I 8

utthour

mrd,cauon

1a.t”

IS to

706

VEGA

ET AL.

3.

How are the mgredients

4.

0 0 0 0 0 A

5.

0 I. Use more cheese 0 2. Decrease the UM of vegetable oil 0 3. Cut down on bacon, sausage, and luncheon meat 0 1. Use fewer nuts and beans 0 8. Don“ know How long and how often do you need to exercise to improve 0 0 0 0 0

6

1 2. g. ,. 8.

time, once each tnne, each time, each tune,

1. 2. 2.. 4. 8.

I&els

1. 2. 3. I 8

from 2 brands of margarine,

check the one that ia brat for your heart.

that UC the best for preventing

heart disecuc are:

Short, energetic hard bursts 01 physxal exercise Physical activxty in whrch breathing pure air is m~portant Physical rctiwty which causes hard and rapid breathing for a rustamed Exerc~ss mvolving spec!ally dertgncd equipment Don’t know of cholesterol

m the blood

pencd

of time

is to

Avoid tension and stress Durease the .mounf of fat you eat Quit smoking end drink Ins &ohol Decrease thr amount of starchy foods in diet Don’t know

Hydrogenated

vcgc‘able

fats arc:

are either true or false. Please read each one and check (4 ‘True” if it is false. Please mark only one response fur each statement.

Soy IIUKC and steak sauce are lox in sodium. 0 1. True 0 2. False 0 8. Don’t

11.

the fitness of your heart and lungs?

Contains partially hydrogenated soybeu, and cottonseed oils Contains liquid safllower oil, partially hydrogenated soy oil Contains palm oil, partially hydrogenated soy oil Don’t know

0 1. Mainly saturated fats 0 2 Mainly polyunsaturated fats 0 2 Solid at mom temperature 0 a. 1 and 2 above 0 b. 2 and 2 above 0 8. Don’t know The following statements about health if the statement is true or check (\,j ‘False” 10.

on this matter

per week 2 times per week 3 times per week 6 times per week

The be,, may to reduce the amount 0 0 0 0 0

9

1. 2. 3. 8

Exercises 0 D C 0 3

8.

I hour each 20 minutes 20 minutes ,CI minutes Don’1 know

Below UC the ingrtdient 0 0 0 0

7.

listed an the label oi a food product?

1. In order of their nutritional content, from the most to the least nutritious 2 In order of their amount in the product, from the most to the least g. In order of how expensive the ingredients are, from mo.~t expensive to least expensive 1. There is no standard order of ingredient labeling, each manufacturer sets its own policy 8. Don’t know good way to reduce saturated fat intake is to:

Know

Mrchanical

dwices

0 I. True 0 2. False 0 8 Don’t

like sauna b&s

make it caster for you to develop

physical

fitness.

Know

12.

Exercising

for 2 hours on the weekend

12.

0 I. True 0 2. False 0 8. Don“ Know Seasoned salt, garlic salt and onion

14.

0 1. 0 2. 0 8. White

salt should

True False Don’t Know chnrc is lower in f&t than yellow

Cl 1. True 0 2. F&r 0 8. Don’t

Know

is just as good aa exercising

cheese-

be avoided

for SO minutes

on a low sodium

diet.

on 4 different

days.

ASSESSING

Answer 1. (5) 2. (2) 3. (2)

Key: 4. (3) 5. (3) 6. (2)

KNOWLEDGE

7. (3) 8. (2) 9. (4)

10. (2) 11.(2) 12. (2)

OF CARDIOVASCULAR

13. (I) 14. (2) 15.(l)

HEALTH

16.(l) 17. (2) 18.(2)

APPENDIX

B

SAMPLE A.

Walking

is Bood for your health

$( 1 True 0 2 False 0 8 Don’t know

1

A saturated 0 0 0 0

2.

Which 0 0 0 0

s.

1. 2. S. 8.

fat ,I

Butler Corn oil Walnuts I don’t knou of lherc types of cxcrcue

is good for your heart’

Four square i3wcball Aerobic dance I don’t know

A gored IOU. sodium snack 1s 0 0 0 0

4

I. 2. s. 8

I. 2 S. K.

A Inch 0 0 0 0

1. 2. S. 8.

Fresh fruit Pwklts Tortilla chips I don’t know that is healthy

for the heart u:

Bologna sandwich, milk, potato chips, cookies Peanut butter swtdrich, chew cubes, celery sticks, milk, apple Lowfat yogurt with fruit, carrot sticks, homemade banana nut bread. nonfat I don’t know

milk

VEGA ET AL.

708 To

help

your

Which

heut,

of thm

how

foods

many

are highest

0 0

1. 2.

Hamburger md Fresh vegetables

0 0

3. 8.

TV dinners and canned I don’t know

Some

foods

that

1. 2.

Three time Two timer

q

S.

Once

0

8

I don’t

Moat

pcoplr

should

“hidden”

fat

a week a week

a week know

C&I

Thirty

umn

08

1 don‘t

know

4.

(3)

5. 6.

(3) (3)

time?

arc:

03

Key

at l

tuna

Hall as much sodnum as they need Ar much rod,um a, they need

l.(l) 2. (3) 3. (1)

exercise

in salt?

01 0 2.

Answer

you

chicken and bruits

hsr

U 0

minutes

a

much

sodium

as they

need

: 7. 8. 9.

(2) (1) (3)

REFERENCES 1. Atkins, C., Patterson, T., Roppe, B., Sallis, J., and Nader, P. Health habits and the decision to volunteer for a family health promotion project. Amer. J. Prev. Med., 3, 87-94 (1987). 2. Bandura, A. “Social Learning Theory.” Prentice-Hall, NJ, 1977. 3. Burdine, J. N., Chen, M. S., Gottlieb, N. H., Peterson, E L., and Vacalis, D. T. The effects of ethnicity, sex and father’s occupation on heart health knowledge and nutrition behavior of school children: The Texas Youth Health Awareness Survey. J. Sch. Health 54, 87-90 (1984). 4. Cuellar, I., Harris, L., and Jasso, R. An acculturation scale for Mexican American normal and clinical populations. Hispanic J. Behav. Sci. 2, 199-217 (1980). 5. Farquhar, J. W., Maccoby, N., and Solomon D. S. Community applications of behavioral medicine, in “Handbook of Behavioral Medicine” (W. D. Gentry, Ed.). Guilford, New York, 1984. 6. Farquhar, J. W., Fortmann, S. P., Maccoby, N., Haskell, W. L., Williams, P. T., Flora, J. A., Taylor, C. B., Brown, B. W., Solomon, D. S., and Hulley, S. B. The Stanford Five-City Project: Design and methods. Amer. J. Epidemiol. 122, 323-334 (1985). 7. Fortmann, S. P., Williams, P T., Hulley, S. B., Maccoby, N., and Farquhar, J. W. Does dietary health education reach only the privileged? An answer from the Stanford Three Community Study. Circulation 66, 77-82 (1982). 8. Hazuda, H., Stern, M., Gaskill, S., Haffner, S., and Gardner, L. Ethnic differences in health knowledge and behaviors related to the prevention and treatment of coronary heart disease. Amer. J. Epidemiol. 117, 717-728 (1983). 9. Higgins, C. W., Price, J. O., and Dunn, J. D. A survey of health education in western Kentucky high schools. J. Sch. Health 52, 162-167 (1982).

ASSESSING

KNOWLEDGE

OF CARDIOVASCULAR

HEALTH

709

10. Nader, P. R., Sallis, J. F., Rupp, J., Atkins, C., Patterson, T., and Abramson, I. The San Diego Family Health Project: Reaching families through the schools. J. Sch. Health 56, 227-231 (1986). 11. National Center for Health Statistics. “Plan and Operation of the Hispanic Health and Nutrition Examination Survey, 1982-84.” Vital and Health Statistics Series 1, No. 19, DHHS Publ. No. (PHS) 851321. Public Health Service, U.S. Govt. Printing Oflice, Washington, DC, 1985. 12. Parcel, G., and Baranowski, T. Social learning theory and health education. Health Educ., 14- 18 (1981). 13. Ramirez, A. G., Herrick, K. L., and Weaver, F. J. El asesino silencioso: A methodology for alerting the Spanish speaking community, cited in Castro, F. G., Baezconde-Garbanti, L., and Beltran, H. Risk factors for coronary heart disease in Hispanic populations: A review. Hispanic J. Behav. Sci. 7, 153-175 (1985). 14. Riggs, R., and Nolan, M. Factors related to the health knowledge and health behavior of disadvantaged black youth. J. Sch. Health 54, 431-434 (1984). 15. Sallis, J. F., Haskell, W. L., Fortmann, S. P., Vranizan, K. M., Taylor, C. B., and Solomon, D. S. Predictors of adoption and maintenance of physical activity in a community sample. Prev. Med. 15, 331-341 (1986). 16. Williams, C. L., Carter, B. J., and Eng, A. The “Know Your Body” Program: A developmental approach to health education and disease prevention. Pro. Med. 9, 371-383 (1980).