Vous êtes sur la page 1sur 6

Journal of Cardiovascular Nursing

Vol. 23, No. 4, pp 332Y337 x Copyright B 2008 Wolters Kluwer Health | Lippincott Williams & Wilkins

Cardiovascular Disease Knowledge and

Risk Perception Among Underserved
Individuals at Increased Risk of
Cardiovascular Disease
Carol J. Homko, PhD, RN, CDE; William P. Santamore, PhD; Linda Zamora, BSN, RN;
Gail Shirk, MSN, RN; John Gaughan, PhD; Robert Cross, MD; Abul Kashem, MD, PhD;
Suni Petersen, PhD; Alfred A. Bove, MD, PhD, FACC

Background: Cardiovascular disease (CVD) risk factor awareness and knowledge are believed to be prerequisites
for adopting healthy lifestyle behaviors. The purpose of this study was to examine knowledge of CVD risk
factors and risk perception among individuals with high CVD risk. Methods: The sample consisted of inner city
and rural medically underserved patients at high risk of CVD. To be eligible for the trial, subjects were required to
have a 10% or greater CVD risk on the Framingham risk score. Knowledge of CVD was assessed with a 29-item
questionnaire created for this study. Subjects also rated their perception of risk as compared with individuals of
their own sex and age. Results: Data were collected from 465 subjects (mean [SD] age, 60.5 [10.1] years; mean
[SD] Framingham risk score, 17.3% [9.5%]). The mean (SD) CVD knowledge score was 63.7% (14.6%), and
mean (SD) level of risk perception was 0.35 (1.4). Men and women had similar Framingham risk scores, but
women perceived their risk to be significantly higher than that of their male counterparts. Women were also more
knowledgeable than men about CVD. Urban participants had significantly higher actual risks than did their rural
counterparts (18.2% [10.7%] vs 16.0% [8.9%], respectively; P = .01) but were significantly less knowledgeable
about heart disease and also perceived their risk to be lower. Conclusions: These results indicate a low perception
of risk and cardiovascular knowledge especially among men and inner city residents. Innovative educational
strategies are needed to increase risk factor knowledge and awareness among at-risk individuals.
KEY WORDS: cardiovascular risk factors, CVD knowledge, Framingham risk score, risk perception,
underserved individuals

H eart disease and stroke rank first and third,

respectively, as the leading causes of death
among Americans. 1 Although, there have been
(CHD) mortality since the 1960s and stroke mor-
tality since the 1920s, the rates of decline have
slowed over the past 15 years.2 Striking differ-
substantial declines in coronary heart disease ences in mortality rates and trends have been

Carol J. Homko, PhD, RN, CDE Abul Kashem, MD, PhD

Assistant Research Professor, Section of Cardiology, Department Research Fellow, Section of Cardiology, Department of Medicine,
of Medicine, Temple University School of Medicine, Philadelphia, Temple University School of Medicine, Philadelphia, Pennsylvania.
Pennsylvania. Suni Petersen, PhD
William P. Santamore, PhD Assistant Professor, Department of Psychological Studies in Education,
Professor, Section of Cardiology, Department of Medicine, Temple Temple University, Philadelphia, Pennsylvania.
University School of Medicine, Philadelphia, Pennsylvania. Alfred A. Bove, MD, PhD, FACC
Linda Zamora, BSN, RN Professor and Chief, Section of Cardiology, Department of
Research Nurse, Section of Cardiology, Department of Medicine, Medicine, Temple University School of Medicine, Philadelphia,
Temple University School of Medicine, Philadelphia, Pennsylvania. Pennsylvania.
Gail Shirk, MSN, RN This study was supported by a grant from the Commonwealth of
Nurse Practitioner, Section of Cardiology, Department of Medicine, Pennsylvania.
Geisinger Medical Center, Danville, Pennsylvania. The Pennsylvania Department of Health specifically disclaims
John Gaughan, PhD responsibility for any analyses, interpretations, or conclusions.
Associate Professor, Biostatistics Consulting Center, Temple University Presented at the American College of Cardiology Meetings on March
School of Medicine, Philadelphia, Pennsylvania. 11Y14, 2006, in Atlanta, Georgia.
Robert Cross, MD Corresponding author
Research Fellow, Section of Cardiology, Department of Medicine, Carol J. Homko, PhD, RN, CDE, Temple University Hospital, 3401 N Broad
Temple University School of Medicine, Philadelphia, Pennsylvania. Street, GCRC/4 West, Philadelphia, PA 19140 (homkoc@temple.edu).


Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
CVD Knowledge and Risk Perception 333

noted by race/ethnicity, socioeconomic status, and severe angina, significant cognitive deficits from
geography. stroke or dementia, end-stage renal disease on dialy-
Stroke mortality rates are higher among Blacks sis, living in nursing homes or boarding homes, and
in general, and CHD rates are particularly high in pregnancy. Subjects who were unable to understand
middle-aged black men relative to other groups. In the study protocol or who were not competent to sign
addition, mortality rates have declined more slowly the informed consent were excluded. All subjects were
in black men and women than in white individuals. instructed on the nature of the study and signed an
The decline in cardiovascular disease (CVD) mortal- informed consent. The subjects had an initial medi-
ity rates began later in women with a lower socio- cal history, physical examination, electrocardiogram,
economic status than in those of a higher status.2 and blood and urine analysis. A fasting blood sample
Available data demonstrate a graded relationship was obtained from each subject to determine serum
between income and heart disease mortality, and cholesterol, low- and high-density lipoprotein, and
there are data to suggest that this gap is widening.3Y8 triglyceride levels. All subjects underwent a formal
The relationship between lifestyle and CVD is assessment of CVD risk using the Framingham risk
well established. Obesity, a sedentary lifestyle, smok- assessment model. Subjects knowledge of CVD risk
ing, and a high dietary fat intake have all been factors and perception of risk were obtained at base-
identified as independent risk factors for the devel- line via questionnaires as described below.
opment of atherosclerosis.9,10 Alternatively, the adop-
tion of healthy lifestyle behaviors has been shown to
reduce CVD risk.11 Adequate knowledge of CVD risk
factors is believed to be a prerequisite for making CVD Risk Knowledge
sound decisions about health. According to health be- Knowledge was assessed using a questionnaire de-
havior models, knowledge of the negative health con- veloped for the current study (Cronbach " = .72). The
sequences of a behavior is a necessary condition for Knowledge Questionnaire contained a total of 29
behavior change.12Y14 However, knowledge alone is multiple-choice questions (5.1 Flesch-Kincaid Grade
not sufficient to promote behavior change.15Y17 It is level). The content areas included exercise, weight
generally thought that individuals who perceive them- loss, nutrition, hypertension, smoking, diabetes, and
selves to have an increased risk of CVD (the perceived cholesterol. There were 4 or more questions for each
susceptibility dimension of the Health Belief Model) content area, and questions were formatted so that at
are more likely to adopt behaviors that reduce their least 1 question required participants to be able to
risk, such as smoking cessation, weight loss, and medi- identify recommended targets or goals for that risk
cation compliance.17Y19 The purpose of this study was factor. At least 1 question required participants to be
to examine knowledge of CVD risk factors and risk able to recognize associated risks, for example, that
perception among a cohort of medically underserved hypertension is associated with an increased risk of
patients at high risk of CVD. stroke, myocardial infarction, and kidney problems,
and at least 1 question targeted desirable lifestyle
Methods or behavioral changes to reduce risk. Scoring is addi-
tive and calculated for each section as well as a total
Patient Selection
knowledge score as percentage correct answers (range
Subjects for this study included 465 inner city and rural 0% to 100%). Face validity of the questionnaire was
individuals at high risk of CVD. The study was con- established using the expert opinions of cardiologists,
ducted at Temple University Hospital and Geisinger and the questionnaire was piloted to access readabil-
Medical Center. Both institutions provide healthcare ity and comprehension.
in areas of Pennsylvania considered to be medically
underserved. Subjects were recruited from the general CVD Risk Perception
outpatient populations of both institutions as well as A scale developed by Schwarzer and Renner22 was
through flyers and presentations at local churches and used to measure the subjects perceptions of risk
community centers. (Cronbach " = .78). The scale has 3 questions fo-
To be eligible for participation, subjects were re- cusing on the likelihood of experiencing heart
quired to be between 18 and 85 years of age and disease, high blood pressure, and stroke in com-
have a 10% or greater 10-year risk of CVD as deter- parison with peers. Individuals rated themselves
mined by the Framingham risk prediction equations, on a 7-point Likert scale ranging from j3 (much
which use sex-specific criteria.20,21 All subjects were below average) to +3 (much above average), with
able to read and had access to a telephone. Exclusion 0 representing average risk (a risk that was equal
criteria included documented medical history of overt to other individuals of their same age and sex).
coronary artery disease, class 3 or 4 heart failure, The responses to these 3 questions were averaged

Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
334 Journal of Cardiovascular Nursing x July/August 2008

to provide the CVD risk perception score (which TABLE 2 Race/Ethnicity, Total Family Income,
ranged from j3 to +3). and Educational Level of the Study Populationa
Administration of both questionnaires was super-
Rural Urban
vised by graduate students, and participants could
choose to read and answer the questions themselves Race/Ethnicity
or have the surveys read to them. The 2 instruments White 98 (252) 13 (28)
African American 1 (2) 78 (165)
took approximately 15 to 20 minutes to complete. Latino/Hispanic 0 7 (14)
Other 0 2 (4)
Framingham Risk Assessment20,21 Total family income
The Framingham risk prediction equations are a tool G$15,000 4 (11) 49 (103)b
used for estimating risk of a cardiac event within $15,000Y$24,999 20 (51) 17 (35)
$25,000Y$34,999 30 (74) 15 (31)b
10 years. The variables used to calculate this risk 9$35,000 46 (117) 20 (42)b
include age, total cholesterol and high-density lipo- Education
protein levels, systolic blood pressure, sex, smoking, GHigh school 1 (4) 16 (34)b
and the presence of diabetes. High school graduate 46 (115) 35 (73)
9High school 53 (133) 48 (100)
Data Analysis Data are presented as % (no. of subjects).
P G .05 compared with rural group.
All data are reported as mean (SD). Student t tests
were used to compare means of continuous vari- or between CVD knowledge and either perceived or
ables, and the relationship between variables was actual CVD risk.
analyzed using Pearson product moment correlations
(SigmaStat Statistical Software, version 1.0). Multi- Subgroup Analysis
ple regression analysis was used to assess the impact Subgroup analysis revealed that women and men had
of age, race, sex, education, and income on risk per- similar Framingham risk scores (17.2% [9.7%] vs
ception and knowledge of CVD risk factors. 17.5% [9.4%]), but women perceived their risk to be
significantly higher (0.61 [1.3] vs 0.15 [1.4]; P G .01;
Results Figure 1) than men did. Women were also
Baseline data were available for 465 subjects. The more knowledgeable than men (65% [14%] vs
mean (SD) age of subjects was 60.5 (10.1) years, and 62% [14%]; P G .05) about CVD (Figure 2).
mean (SD) Framingham risk score was 17.3% Significant differences in knowledge were observed in
(9.5%). Fifty-five percent of the participants were the areas of weight reduction and diabetes between the
from rural areas (45% from the inner city), 44% 2 groups, with women receiving higher scores. When
were female (56% male), 45% had diabetes, and subjects from rural and urban areas were compared,
27% were smokers. Additional baseline data are inner city subjects had a significantly higher ac-
reported in Tables 1 and 2. Mean (SD) knowledge tual CVD risk (18.2% [10.7%] vs 16.0% [8.9%];
scores were 63.7% (14.7%), and mean (SD) levels of P = .001). In addition, subjects from the inner city were
risk perception were 0.35 (1.4). Participants scored significantly less knowledgeable about heart disease
highest in the areas of weight reduction and exercise (56% [13%] vs 70% [13%]; P G .001; Figure 3) and
and lowest in the areas of smoking and diabetes. No perceived their risk to be lower (0.18 [1.4] vs 0.50
association was found between perceived and actual [1.3]; P = .01) than their rural counterparts. Significant
CVD risk, as assessed by the Framingham risk score, differences in knowledge were observed across all
content areas (Figure 4) between the 2 groups, with
rural participants receiving higher scores.
TABLE 1 Demographic Characteristics of the
Study Population Multivariate Analysis
Rural (n = 254) Urban (n = 211)
We developed a multiple regression model to identify
Age, mean (SD), y 62.2 (9.6) 58.0 (10.1)a
the baseline characteristics that predicted increased risk
Body mass index, 31.7 (6.0) 32.1 (7.0)
mean (SD), kg/m2 perception (model 1) and knowledge of CVD risk fac-
Risk score, 16.4 (8.7) 18.5 (10.3)a tors (model 2). Predictors of increased risk percep-
mean (SD), % (median, 14.4) (median, 15.9) tion included presence of diabetes, white race, female
Sex, % male 56 55 gender, and advanced age (F = 13.16, R2 = 0.111,
Diabetes, % 37 55a
P G .001). In model 2, the same variables with the
Smoker, % 13 43a
addition of higher levels of income (F = 31.29, R2 =
P G .05 compared with rural group. 0.269, P G .001) predicted CVD knowledge.

Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
CVD Knowledge and Risk Perception 335

This study found that underserved individuals at high
risk of CVD demonstrated limited CVD risk factor
knowledge and a reduced perception of CVD risk de-
spite being assessed as high risk by the Framingham
model. Our survey is unique in the fact that it included
only individuals identified to be at high risk of CVD
(910% on the Framingham risk assessment) and a high
percentage of minority and lower socioeconomic status
individuals. Women, individuals from rural commu-
nities, and individuals with higher annual incomes had
increased awareness and knowledge of cardiovascular
risk factors. However, education in our sample was not
predictive of CVD knowledge. The fact that 34% of
FIGURE 2. Cardiovascular disease (CVD) knowledge scores
our patients were 65 years or older and retired may
(% correct answers). Left, men (black bar) and women (gray
help to explain the apparent discrepancy between in- bar); right, rural (black bar) and urban (gray bar) participants.
come and education, which are generally highly cor-
related. Income and socioeconomic status, however, the major causes of heart disease among older indi-
did not seem to affect perceptions of risk. However, the viduals. Like the present study, these studies found that
present study has limitations. Rural African Americans women tended to have a better knowledge of the fac-
were extremely underrepresented in our study popula- tors causing heart disease than men did26 and that
tion, with almost all African Americans coming from socioeconomic status was a strong and consistent pre-
the urban environment. It is possible that the rural dictor of risk factor knowledge.27 A large survey from
African Americans differ from urban African Ameri- the United Kingdom28 also found large differences in
cans both in actual risk factors and in knowledge, risk perception and attitudes between the 2 sexes. Obese
perception, and other psychosocial variables. In ad- women were more likely to believe that their diet was
dition, all participants were from medically under- harmful than were obese men. Similarly, more female
served areas and all were at high risk of CVD. There than male smokers perceived their habit to be harmful.
were no comparison groups of either individuals at In contrast, Frijling and others29 found that sex was not
low risk of CVD or from suburban populations. It is associated with level of perceived risk in subjects with
possible that similar gaps in risk perception and knowl- hypertension and diabetes but without overt CVD.
edge of CVD risk factors exist among these groups. A shift in awareness of heart disease as the leading
Regardless, our findings are consistent with other sur- killer of women has occurred since 1997. A tele-
vey data.23Y25 Data from the Canadian Heart Health phone survey performed in 200430 of 1,024 ran-
Surveys Initiative completed in older Canadians (ages domly sampled individuals found that awareness
55 to 74 years) examined awareness and knowledge of and knowledge of CVD risk are increasing among
women, although minority women still lag behind
their white counterparts. In comparison to previous
surveys performed in 1997 and 2000, the percentage
of women who identified heart disease as the leading
cause of death in women had increased from 30% in
1997 to 46% in the current survey. They reported
that black, Hispanic women younger than 45 years
had lower awareness of heart disease than did white
and older women. In the current study, age and race
also predicted CVD knowledge.
The National Conference on Cardiovascular Dis-
ease Prevention2 continues to find health disparities
based on race and ethnicity, socioeconomic status,
and geography. The CHD mortality rate is declin-
FIGURE 1. Cardiovascular disease (CVD) risk scores for men ing, but at a slower rate than previously reported, and
(black bar) and for women (gray bar). Left, 10-year CVD stroke mortality rates have changed very little in most
risk scores as assessed by the Framingham risk score; right,
perceived CVD risk scores. Potential scores range from populations. The mortality rates are especially high
j3 (much below average) to +3 (much above average), with in middle-aged black men relative to other race/sex
0 representing average risk. groups. Declines in CHD mortality rates among black

Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
336 Journal of Cardiovascular Nursing x July/August 2008

influence of family preferences) have a significant

impact on health behaviors related to CVD risk
reduction.31,32 Culture beliefs influence individuals
perceptions of health and illness. They can also in-
fluence how symptoms are recognized and interpreted
and impact how individuals access the healthcare sys-
tem and make healthcare decisions.
Our data indicate that the segments of our society
who demonstrate the highest CVD mortality rate2 are
the least knowledgeable about cardiovascular risk
factors and do not perceive themselves to be at risk.
Reduction of CVD risk has a strong behavioral com-
ponent, which emphasizes the adoption of healthy
FIGURE 3. Cardiovascular disease (CVD) risk scores for rural behaviors and/or the modification of risky ones.
(black bar) and urban (gray bar) participants. Left, 10-year Although knowledge alone is not sufficient to change
CVD risk scores as assessed by the Framingham risk score; behavior, knowledge is generally believed to be a
right, perceived CVD risk scores. Potential scores range from prerequisite for change. The Health Belief Model12,13
j3 (much below average) to +3 (much above average), with
0 representing average risk. postulates that individuals will take action to ward
off, to screen for, or to control ill health if they believe
women have also been slower than among white it to have potentially serious consequences or if they
women. Stroke mortality rates are higher in black believe it would be beneficial in reducing either sus-
individuals. The available data suggest that the gap in ceptibility or the severity of their condition. To adopt
CVD mortality between the poor and undereducated a new behavior, people must have knowledge about
versus the wealthy and well educated has not lessened the condition (CVD risk factor knowledge), perceive
and may well be widening. The National Conference themselves as susceptible to the disease (CVD risk
on Cardiovascular Disease Prevention2 clearly identi- perception), and believe that they are capable of doing
fied that these disparities account for the disappoint- something to prevent or cure the disease (self-efficacy).
ing trends in mortality, risk factors, and preventive Overall, our results indicate a low perception of risk
services despite community-wide public health cam- and cardiovascular knowledge for all participants,
paigns encouraging the adoption of healthy lifestyle especially pronounced among men and inner city resi-
interventions. There is also growing evidence that dents, despite the magnitude of their actual risk. These
cultural beliefs (ie, cultural food patterns and the data suggest that there are important opportunities

FIGURE 4. Comparison of individual risk factor scores on the CVD knowledge test between rural (black bars) and inner
city (gray bars) participants. aP G .05.

Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
CVD Knowledge and Risk Perception 337

to educate underserved disadvantaged individuals 13. Janz NK, Becker MH. The health belief model: a decade
about CVD risk and prevention. Innovative educa- later. Health Educ. 1984;11:1Y47.
14. Ajzen I. The theory of planned behavior. Organ Behav
tional strategies are needed to not only increase risk Hum Decis Process. 1991;50:179Y211.
factor knowledge but also enhance self-efficacy among 15. DeSalvo KB, Gregg J, Kleinpeter M, Pedersen BR, Stepter A,
these individuals. This includes the provision of edu- Peabody J. Cardiac risk underestimation in urban, black
cation and skill training for CVD prevention based women. J Gen Intern Med. 2005;20:1127Y1131.
16. Krummel DA, Humphries D, Tessaro I. Focus groups on
on a better understanding of the individuals cultural cardiovascular health in rural women: implications for
beliefs and life situations through the involvement of practice. J Nutr Educ Behav. 2002;34:38Y46.
the community and/or individuals being targeted. Also 17. Lynch EB, Liu K, Kiefe CI, Greenlan P. Cardiovascular
needed are interventions aimed at ensuring cultural disease risk factor knowledge in young adults and 10-year
competence among healthcare providers and efforts to change in risk factors: the Coronary Artery Risk Develop-
ment in Young Adults (CARDIA) study. Am J Epidemiol.
increase health literacy. Future research should exam- 2006;164:1171Y1179.
ine the relationship between knowledge, risk percep- 18. Segan CJ, Borland R, Greenwood KM. What is the right
tion, self-efficacy, and behavior change. thing at the right time? Interactions between stages and
processes of change among smokers who make a quit
attempt. Health Psychol. 2004;23:86Y93.
REFERENCES 19. Dishman RK, Motl RW, Saunders R, et al. Self-efficacy
1. Thom T, Haase N, Rosamond W, et al. Heart disease and partially mediates the effect of a school-based physical-
stroke statisticsV2006 update: a report from the Ameri- activity intervention among adolescent girls. Prev Med.
can Heart Association Statistics Committee and Stroke 2004;38:628Y636.
20. Third report of the National Cholesterol Education Program
Statistics Subcommittee. Circulation. 2006;105:e85Ye135.
(NCEP) Expert Panel on Detection, Evaluation, and Treat-
2. Cooper R, Culter J, Desvigne-Nickens P, et al. Trends and
ment of High Blood Cholesterol in Adults (Adult Treatment
disparities in coronary heart disease, stroke, and other
Panel III). Circulation. 2002;106:3143.
cardiovascular disease in the United States: findings of the
21. Wilson PW, DAgostino RB, Levy D, Belanger AM,
National Conference on Cardiovascular Disease Preven-
Silbershantz H, Kannel WB. Prediction of coronary heart
tion. Circulation. 2000;102:3137Y3147.
disease using risk factor categories. Circulation. 1998;97:
3. Cooper R, Casper M, Barnett E. Trends in heart disease by
race/ethnicity, socioeconomic status and geography. Paper
22. Schwarzer R, Renner B. Social-cognitive predictors of health
presented at: National Conference on CVD Prevention;
behavior: action self-efficacy and coping self-efficacy. Health
September 27Y29, 1999; Bethesda, MD.
Psychol. 2000;19:487Y495.
4. Feldman JJ, Makuc DM, Kleinman JC, Comoni-Huntley J. 23. Ruland S, Raman R, Chaturvedi S, Leurgans S, Gorelick PB.
National trends in educational differentials in mortality. Awareness, treatment, and control of vascular risk factors
Am J Epidemiol. 1989;129:919Y933. in African Americans with stroke. Neurology. 2003;60:
5. Pappas G, Queen S, Hadden W, Fisher G. The increasing dis- 64Y68.
parity in mortality between socioeconomic groups in the United 24. Avis NE, Smith KW, McKinlay JB. Accuracy of perceptions
States, 1960 and 1986. N Engl J Med. 1993;329:103Y109. of heart attack risk: what influences perceptions and can they
6. Langenberg C, Shipley MJ, Batty GD, Marmot MG. be changed? Am J Public Health. 1989;17:1608Y1612.
Adult socioeconomic position and the association be- 25. Avis NE, McKinlay JB, Smith KW. Is cardiovascular risk
tween height and coronary heart disease mortality: find- factor knowledge sufficient to influence behavior? Am J
ings from 33 years of follow-up in the Whitehall Study. Prev Med. 1990;6:137Y144.
Am J Public Health. 2005;95:628Y632. 26. Kirkland SA, MacLean DR, Langille DB, Joffres MR,
7. Manderbacka K, Hetemaa T, Keskimaki I, Luukkainen P, MacPherson KM, Andreou P. Knowledge and awareness of
Koskinen S, Reunanen A. Are there socioeconomic dif- risk factors for cardiovascular disease among Canadians 55
ferences in myocardial infarction event rates and fatality to 74 years of age: results from the Canadian Heart Health
among patients with angina pectoris? J Epidemiol Com- Surveys 1986-1992. JAMC. 2000;161(suppl 8):S10YS16.
munity Health. 2006;60:442Y427. 27. Potvin L, Richard L, Edwards AC. Knowledge of cardiovas-
8. Manson JE, Tosteson H, Ridker PM, et al. The primary cular disease risk factors among the Canadian population:
prevention of myocardial infarction. N Engl J Med. 1992; relationships with indicators for socioeconomic status.
326:1406Y1416. CMAJ Can Med Assoc J. 2000;162(suppl 9):S5YS11.
9. Pollitt RA, Rose KM, Kaufman JS. Evaluating the 28. Siglagy C, Muir J, Coulter A, Thorogood M, Roe L.
evidence for models of life course socioeconomic factors Cardiovascular risk and attitudes to lifestyle: what do
and cardiovascular outcomes: a systematic review. BMC patients think? Br Med J. 1993;306:1657Y1660.
Public Health. 2005;5:7Y20. 29. Frijling BD, Lobo CM, Keus IM, et al. Perceptions of
10. Anderson KM, Odell PM, Wilson PWF, Kannel WB. Cardio- cardiovascular risk among patients with hypertension or
vascular disease risk profiles. Am Heart J. 1990;121:293Y298. diabetes. Patient Educ Couns. 2004;52:47Y53.
11. Elmer PJ, Obarzanek E, Vollmer WM, et al. Effects of 30. Mosca L, Ferris A, Fabunmi R, Robertson RM. Tracking
comprehensive lifestyle modification on diet, weight, physi- womens awareness of heart disease: an American Heart
cal fitness and blood pressure control: 18 month results of a Association national study. Circulation. 2004;109:573Y579.
randomized trial. Ann Intern Med. 2006;144:485Y495. 31. Mullins CD, Blatt L, Gbarayor CM, Yang HW, Baquet C.
12. Rosenstock IM. The health belief model: explaining health Health disparities: a barrier to high-quality care. Am J
behavior through expectancies. In: Glanz K, Marcus Lewis F, Health Syst Pharm. 2005;62:1873Y1882.
Rimer BK, eds. Health Behavior and Health Education. San 32. Kernicki JG. A multicultural perspective of cardiovascular
Francisco, CA: Jossey-Bass Publishers; 1990:42. disease. J Cardiovasc Nurs. 1997;11:31Y40.

Copyright @ 2008 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.