ORIGINAL ARTICLE
CORONARY HEART DISEASE IN ETHNICALLY DIVERSE WOMEN
Coronary Heart Disease in Ethnically Diverse Women:
Risk Perception and Communication
ALLISON H. CHRISTIAN, EDD; HEIDI Y. MOCHARI, MPH, RD; AND LORI J. MOSCA, MD, PHD
OBJECTIVES: To assess perceived vs calculated risk of coronary
heart disease (CHD), preferred methods of communicating risk,
and the effect of brief educational intervention to improve accu-
rate perceptions of personal risk.
SUBJECTS AND METHODS: Of 1858 women who underwent
screening mammography between April and September 2003 at
the Columbia University Medical Center in New York, NY, we
assessed 125 women with no history of cardiovascular disease
who participated in a risk factor screening and education pro-
gram. Demographic variables were evaluated by interviewer-as-
sisted standardized questionnaires. Absolute 10-year CHD risk
was calculated using the Framingham global risk assessment.
Perceived 10-year risk and preferred method of communicating
risk were evaluated systematically.
RESULTS: Among 110 research participants who were eligible for
risk estimation, 59% had a 10-year risk of less than 10%. However,
only half of those women accurately perceived their risk as low.
After a brief educational intervention, the women’s ability to
correctly categorize their personal CHD risk improved signifi-
cantly. Preferred methods to communicate risk varied by level of
education and age. Older women (≥65 years) and those with a
high school education or less were more likely to prefer simple
methods of having CHD risk communicated compared with their
counterparts.
CONCLUSIONS: These data underscore the need to determine
preferences for providing risk information and to test various
formats for communicating CHD risk to improve awareness and
management of CHD risk factors, especially among women of
different age groups and education levels.
Mayo Clin Proc. 2005;80(12):1593-1599
BAC = breast arterial calcification; CHD = coronary heart disease;
CI = confidence interval; HBM = Health Belief Model; OR = odds ratio
C oronary heart disease (CHD), the leading cause of
death and premature disability among women in the
United States, disproportionately afflicts ethnic minority
groups including African American and Hispanic women.1-6
Despite this fact, numerous studies have shown suboptimal
rates of CHD risk awareness among women. In a national
survey of women with an oversampling of ethnic minorities
conducted by the American Heart Association, 13% of
women cited heart disease as their greatest health threat,
with less awareness and knowledge of risk factors noted
among ethnic minority women.7 Another important finding
of this survey was that women perceived breast cancer as
their greatest health concern.7 Mammography has become
a common and highly accepted screening tool for breast
cancer among women older than 40 years. The point of
service of mammography may represent a unique opportu-
Mayo Clin Proc. • December 2005;80(12):1593-1599
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
nity to screen women for CHD risk factors and to inform
them of their personal risk of CHD. This population is a
desirable target for intervention because women who un-
dergo routine mammography represent an aging population
in need of CHD risk factor evaluation and management.
Perceptions of CHD risk appear to be positively corre-
lated with a desire to make risk-reducing behavioral
changes and with actual behavioral change.8,9 Perceptions
of personal risk occupy a central role in theories of indi-
vidual health behavior such as the Health Belief Model
(HBM).10 The HBM suggests that perceptions of risk play a
critical role in a patient’s compliance with recommended
health behaviors. In a comprehensive review of the HBM
and of cardiovascular risk factor reduction behaviors, per-
ceived vulnerability was found to be a significant contribu-
tor in almost half the studies that assessed behaviors related
to CHD such as screening for hypertension.11
Effective primary prevention of CHD requires early
detection of risk factors and communication of risk to
patients. National standards to stratify women on the basis
of CHD risk and to guide preventive treatment are based on
the Framingham global risk.12 Assessment and communi-
cation of risk is necessary because it can assist patients in
developing a more realistic perceived risk of CHD that, in
turn, may motivate them to initiate and maintain heart
healthy behaviors at an appropriate target level.13 Optimally,
a patient’s perceived risk of developing CHD should match
the patient’s actual calculated risk. However, there is lack of
agreement concerning the best approach for evaluating risk
or for communicating CHD risk to patients. The purposes of
this study were (1) to assess perceived risk vs calculated risk
on the basis of the Framingham global risk score, (2) to
evaluate the effect of brief educational intervention to im-
prove knowledge of personal CHD risk, and (3) to determine
preferred methods of communicating risk among women
who undergo routine screening mammography.
From the Department of Medicine, Columbia University, New York, NY.
This study was supported by the Preventive Cardiology Program at NewYork-
Presbyterian Hospital.
Individual reprints of this article are not available. Address correspondence to
Lori J. Mosca, MD, PhD, Preventive Cardiology Program, NewYork-Presbyterian
Hospital, 622 W 168th St, PH10-203B, New York, NY 10032 (e-mail: LJM10
@columbia .edu).
© 2005 Mayo Foundation for Medical Education and Research
• www.mayoclinicproceedings.com 1593
CORONARY HEART DISEASE IN ETHNICALLY DIVERSE WOMEN
SUBJECTS AND METHODS
RESEARCH PARTICIPANTS AND DESIGN
This is a substudy of 125 women with no known history of
cardiovascular disease who participated in a clinical trial
conducted at the Columbia University Medical Center in
New York, NY, which evaluated CHD risk in those with
and without breast arterial calcification (BAC). This cross-
sectional substudy with a 1-month follow-up included
women aged 35 years and older who were eligible for
participation in the trial provided they had undergone rou-
tine screening mammography as an outpatient at the Co-
lumbia University Radiology Clinic between April and
September 2003 and either were found to have BAC on
mammography or were serving as the next consecutive
age-matched control (within 5 years) to a woman who had
BAC and was enrolled in this study.
Potential research participants were identified through a
systematic review of mammograms, and women were con-
tacted by telephone within 1 year of undergoing mammogra-
phy. Research participants agreed to participate in a 1-month
study that included a hospital-based CHD risk factor screen-
ing and brief educational intervention plus a follow-up tele-
phone survey. A trained research assistant conducted all base-
line screenings and 1-month follow-up surveys. Permission
from the research participant’s physician was not required
for participation. Inclusion criteria included the ability to
give written informed consent, speak English or Spanish,
and complete a 1-month follow-up questionnaire. Research
participants were excluded if they had a history of cardiovas-
cular disease including myocardial infarction, coronary
artery bypass grafting, angioplasty, stent, angina, stroke,
carotid artery disease, peripheral arterial disease, or abdomi-
nal aortic aneurysm or if they were disoriented to person,
place, or time. The Institutional Review Board at Columbia
University Medical Center approved the study, and all sub-
jects gave written informed consent.
Of the 1858 women who underwent routine screening
mammography between April and September 2003 at the
Columbia University Radiology Clinic, we attempted to
reach 1042 by telephone by using an age-matching sam-
pling procedure. Among those who underwent routine
mammography during the specified period, women with
BAC were contacted by telephone first. Once a woman was
enrolled in the study, a control was identified by selecting
the next consecutive woman without BAC from the same
list of women matched to the research participant by age
within 5 years. The control also was contacted by tele-
phone. If the woman without BAC was interested in partici-
pating, she was enrolled and served as the age-matched
control. If she did not want to participate, the same age-
matching procedure was followed until a control was en-
1594 Mayo Clin Proc. • December 2005;80(12):1593-1599
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
rolled. Neither research participants nor controls were
recruited by mail. Of those who qualified as research
participants or controls, we spoke to 398 women and in-
formed them about the study. Inclusion criteria for participa-
tion were met by 354 of those women, and 125 were enrolled
subsequently in the study (35% acceptance rate). Primary
reasons for nonparticipation included the following: (1) hav-
ing a health care provider who monitored their CHD risk
factors regularly (26%), (2) having no desire to participate in
research (23%), (3) being too busy with work or family
responsibilities (18%), (4) living too far from the medical
center (11%), and (5) being too old and/or ill (10%).
Once enrolled, all research participants received stan-
dardized CHD risk factor screening by a trained research
assistant and brief educational intervention. During the
educational intervention, which took approximately 1 hour,
the research assistant provided information on the preva-
lence of and the risk factors for CHD among women, the
participant’s risk factor screening results, current risk fac-
tor goals, and methods to decrease the risk of developing
CHD. At 1 month, all research participants were contacted
via telephone or mail (per their preference) to complete a
questionnaire that assessed change in knowledge and
awareness of CHD and its risk factors, risk factor status,
medical history, level of perceived risk, and preferred
methods for having risk communicated.
DEMOGRAPHICS
Demographic variables were evaluated at baseline through
self-report by use of standardized interviewer-assisted
questions.
CARDIOVASCULAR RISK FACTORS
A trained research assistant measured traditional CHD risk
factors at baseline with use of standardized procedures.
Blood pressure was assessed by manual aneroid sphygmo-
manometer by using standard National Institutes of Health
Joint National Committee 7 guidelines.14 Lipids, lipopro-
teins, and blood glucose levels were obtained by using a
lancet to obtain a drop of blood from the participant’s
fingertip. The blood measurements were analyzed by
fingerstick methods by using Cholestech technology
(Cholestech Corp, Hayward, Calif), which is commonly
used in screenings programs.15 Height was measured by a
precision wall-mounted standardized-height rod. Weight
was obtained by a standard physician’s scale with patients
in light clothing and without shoes. Body mass index was
calculated directly by the standard formula: Weight (kg)/
[Height (m)]2. Waist circumference was measured under
clothing using the guidelines provided in the Third National
Health and Nutrition Examination Survey (NHANES III)
protocol.16
• www.mayoclinicproceedings.com

Other traditional risk factors for CHD including family
history of premature CHD (age <65 years in women, <55
years in men), diabetes mellitus, and smoking were evalu-
ated via self-report using standardized interviewer-assisted
questionnaires.
The Framingham global risk score was calculated with
use of a National Heart, Lung, and Blood Institute work-
sheet and standard measurements of traditional risk factors
including total cholesterol, smoking status, high-density li-
poprotein, and systolic blood pressure (Appendix 1).17-19
This global risk estimate was used to determine each
participant’s absolute risk, defined as the probability of hav-
ing a heart attack or dying of CHD during a 10-year period.
In accordance with recent guidelines, if a research partici-
pant reported a family history of premature CHD, she was
classified at intermediate risk of CHD (10%-20%).12 Also,
because diabetes mellitus is considered a cardiovascular dis-
ease risk equivalent, any research participant with self-
reported diabetes mellitus and/or a fasting plasma glucose
level of 126 mg/dL or greater was considered as having
diabetes and thus categorized as high risk (>20%).12,17
PERCEIVED RISK
Various likelihood scales have been used to assess perceived
risk of disease among patient populations including percent-
age scales, 5- or 7-point verbally labeled scales, and visual/
graphic scales. However, there is no indication that any scale
is consistently superior to another for measuring risk percep-
tions. Percentage scales are recommended when comparing
risk perceptions with actual statistics.20 Therefore, perceived
risk was evaluated in this study by asking research partici-
pants, “In the next 10 years, what do you think your chances
are of having a heart attack or dying of heart disease com-
pared to a woman of similar age as you?” Perceived 10-year
risk was measured before the baseline screening, imme-
diately after the educational intervention, and at the 1-month
follow-up by using this question. Response options included
low (less than 10%), moderate (between 10% and 20%), and
high (more than 20%).
COMMUNICATION OF RISK
The preferred method of communicating risk was evalu-
ated after the baseline screening and intervention and at 1
month by a standardized interviewer-assisted questionnaire.
Research participants were asked to identify their single
preferred method of having CHD risk communicated from
the following list: (1) relative risk, (2) absolute risk, (3)
cardiovascular age equivalent, (4) number of traditional risk
factors, (5) risk categorized as low/moderate/high, (6) other,
and (7) do not know. The research assistant discussed each
method of communicating risk in detail with the research
participant by using a standardized script (Appendix 2).
Mayo Clin Proc. • December 2005;80(12):1593-1599
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
CORONARY HEART DISEASE IN ETHNICALLY DIVERSE WOMEN
STATISTICAL ANALYSES
Demographic variables included race (white or ethnic
minority [African American, Hispanic, other]), age (<65 or
≥65 years), education (≤high school or >high school edu-
cation), total annual household income (<$20,000 or
≥$20,000), and marital status (married or unmarried). Cut
points for age and education level were chosen on the
basis of sample means. Descriptive statistics (frequencies,
means, SDs, percentages) were carried out for demo-
graphic data. Prevalences of absolute risk, perceived risk,
and preferred methods of communicating risk are pre-
sented as counts and percentages. Continuous variables are
presented as mean ± SD after checking for normality.
Within-group changes in accurate risk perception from
preintervention to postintervention, from prescreening to 1
month, and from postscreening to 1 month were evaluated
by using the McNemar test for paired data. Univariate
predictors of accurate perceived risk and preferred risk
communication methods were evaluated using the χ2 test.
Logistic regression was used to evaluate the independence
of demographic factors to predict accurate perceived risk
and preferred methods of communicating risk. All
univariate predictors at P<.10 were included in the models.
Age, as a predictor of accurate perceived risk and preferred
risk communication method, was examined as both a cat-
egorical variable by use of the χ2 test and as a continuous
variable by use of the Student t test. Statistical significance
was accepted at the 95% confidence interval (CI) or P<.05.
The SAS analytic program, version 8e, (SAS Institute Inc,
Cary, NC) was used for all analyses.
RESULTS
The baseline characteristics of the study population are
presented in Table 1. Research participants ranged between
ages 38 and 86 years with a mean age of 62.9 (±SD 10.9).
No significant difference in age existed between white and
ethnic minority research participants. Among participants,
56 (44.8%) were Hispanic, 37 (29.6%) were white, 24
(19.2%) were African American, and 8 (6.4%) classified
themselves as “other.” Thirty-six participants (28.8%)
were married. Among participants, 60 (48.0%) had more
than a high school education, 36 (28.8%) had less than a
high school education, and 29 (23.2%) had a high school
diploma (or equivalent). Fifty-nine (54.6%) of the 108
women who responded to an income level question re-
ported being of low-income status with a total annual
household income of less than $20,000. Most research
participants (90 of 124 [73%]) were unemployed, with
48 (38.7%) retired, 15 (12.1%) on medical disability,
17 (13.7%) homemakers, 7 (5.7%) out of work, 2 (1.6%) in
school, and 1 (0.8%) other. Most (98; 78.4%) had Medicaid
• www.mayoclinicproceedings.com 1595
CORONARY HEART DISEASE IN ETHNICALLY DIVERSE WOMEN

TABLE 1. Characteristics of 125 Women Who Underwent risk, 22 (20%) had between a 10% and 20% 10-year risk,
Routine Screening Mammography*
and 23 (21%) had a greater than 20% 10-year risk of CHD.
Ethnic Paradoxically, only 52% (29 of 56 responders) of those
Overall White minority
Demographics (n=125) (n=37) (n=88) calculated to be at low risk correctly perceived their risk of
CHD as low (<10%) at baseline. Immediately after the
Mean ± SD age (y) 62.94±11 65.46±12 61.89±10
White 37 (30.0) NA NA educational intervention there was a significant increase in
>High school education† 60 (48.0) 31 (83.8) 29 (33.0) the proportion of women who correctly categorized their
Unemployed‡ 90 (72.6) 23 (62.2) 67 (77.0) risk as low (<10%) (67% [36 of 54 responders] vs 52%;
Married§ 36 (28.8) 16 (43.2) 20 (22.7)
Low income⁄⁄ 59 (54.6) 7 (24.1) 52 (65.8) P=.02). However, the improvement in accurate risk percep-
tion deteriorated after 1 month with only 28 of the 54
*All data presented as number (%). Significant difference found between
white and ethnic minority research participants. NA = not applicable. female responders (52%) correctly perceiving their risk as
†P<.001. low. Among the moderate and high-risk women (≥10%)
‡Of 124 women overall and 87 ethnic minority women who responded. there was a gradual decline in the number of women who
§P=.02.
⁄⁄Of 108 women who responded. accurately perceived their risk as moderate or high from
prescreening (71% [25 of the 35 responders]) to post-
screening (68% [23 of the 34 responders]) to 1-month
or Medicare as their primary form of health insurance. follow-up (64% [25 of the 39 responders]). Change in
Significant differences existed between white and ethnic accurate risk perception over time is presented in Figure 1.
minority research participants in education level (P<.001), Among women at low risk, no predictors of accurate
marital status (P=.02), and income level (P<.001). risk perception were found at prescreening, postscreening,
Of the 125 women who participated in the baseline or 1-month follow-up when evaluating demographic fac-
screening, 111 (89%) completed the 1-month follow-up tors including age, education level, ethnic minority status,
survey. Primary reasons for incomplete follow-up included and income. Among moderate- or high-risk women, no
an inability to reach the research participant (via telephone predictors of accurate risk perception were found at
and mail) or the participant reporting to be too busy. prescreening or postscreening. However, among moderate-
Among the 110 research participants eligible for or high-risk women at 1-month follow-up, in a multiple
Framingham global risk estimation (women aged 20-79 linear regression model controlling for confounders, age
years, inclusive), 65 (59%) had a less than 10% 10-year (≥65 years) was associated with accurate risk perception
(odds ratio [OR], 4.51; 95% CI, 0.97-21.00). No other
predictors of accurate risk perception were found when
100 analyzing family history of premature CHD, previous CHD
Prescreening
90 Postscreening risk factor (ie, blood pressure, cholesterol) screenings, hav-
1-Month follow-up
80 ing a basic knowledge of CHD (ie, CHD is the leading
70
cause of death for women in the nation), and feeling well
Percent correct

informed about CHD in women.

60
Overall, research participants preferred the simpler
50 methods of presenting CHD risk information. Among the
40 106 women who provided data on preferred methods of
30 risk communication, 36 (34%) preferred CHD risk ex-
20
pressed as number of risk factors, 29 (27%) preferred their
risk categorized as high, moderate, or low, 21 (20%) pre-
10
ferred absolute risk (ie, Framingham 10-year risk score),
0
Low or <10% risk Intermediate/high or ≥10% risk
6 (6%) preferred receiving a cardiovascular age equivalent,
(n=65) (n=55) 5 (5%) preferred other ways (ie, combined methods),
Calculated absolute 10-year risk 7 (6%) responded that they did not know, and only 2 (2%)
preferred relative risk. In a univariate model, at baseline,
women younger than 65 years were more likely to prefer
FIGURE 1. Prevalence and change in accurate risk perception.
Calculated absolute risk was based on results of the woman’s risk information expressed through detailed methods (rela-
Framingham global risk estimate. Percent correct includes women tive risk, absolute risk, or cardiovascular age equivalent)
who correctly assessed their personal risk of heart disease com- than as number of risk factors and risk factor categories
pared with their true calculated risk. Significant change was from
prescreening to postscreening (P=.02). Analysis was limited to 110 compared with women age 65 years or older (OR, 2.44;
research participants between ages 20 and 79 years, inclusive. 95% CI, 0.97-6.15). When examining age as a continuous

1596 Mayo Clin Proc. • December 2005;80(12):1593-1599 • www.mayoclinicproceedings.com

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

variable, younger women were also more likely to prefer
the more detailed methods (P=.04). A trend remained for
age as a predictor of preferred method of communicating
risk (OR, 2.40; 95% CI, 0.93-6.17) in a multivariate
analysis controlling for potential confounders (ethnicity,
age, education). This multivariate logistic regression
model also revealed that women with a high school edu-
cation or less were more likely to prefer simpler methods
of having risk presented (number of risk factors and risk
factor categories) than more detailed methods (relative
risk, absolute risk, or cardiovascular age equivalent) com-
pared with women with more than a high school educa-
tion (OR, 2.80; 95% CI, 0.98-7.99). Ethnic minority sta-
tus, income, marital status, family history of premature
cardiovascular disease, and calculated absolute risk were
not statistically significant predictors of preferred risk
communication methods in this multivariate analysis.
There was no relationship found between preferred
method of having risk communicated and accurate risk
perception before or after intervention.
DISCUSSION
This study documented that only about one half of women
at low risk and two thirds of women at moderate or high
risk of CHD accurately perceived their risk level. Most
women at low risk overestimated their risk level in this
study. Importantly, we found that young women calculated
to be at moderate or high risk underestimated their risk of
CHD, suggesting that education and prevention messages
need to be targeted to this subpopulation at increased risk
of poor clinical outcomes and future CHD events.
Our findings concur with those of Frijling et al,21 who
found that participants at moderate to high risk of cardio-
vascular disease (n=1194) overestimated their actual 10-
year risk by more than 20%, with men having higher accu-
racy of perceived absolute risk. These findings may explain
why our all-female sample showed inaccurate perceptions
of personal CHD risk. However, it is important to note that
the Frijling et al sample primarily consisted of white pa-
tients, and 42% were men.21 Therefore, although our find-
ings enhance existing data on risk perception, further data
are needed to evaluate risk perceptions among women and
ethnic minorities.
Previous literature also suggests that demographic vari-
ables such as age, sex, and education are related to an in-
creased perceived risk for certain diseases such as heart
attack.21-23 In this study, older age was found to be a signifi-
cant predictor of accurate high-risk perception at 1-month
follow-up. This may suggest that the educational interven-
tion was more effective among older women. This is not
unexpected because the educational intervention empha-
Mayo Clin Proc. • December 2005;80(12):1593-1599
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
CORONARY HEART DISEASE IN ETHNICALLY DIVERSE WOMEN
sized age as the driving factor in the Framingham global risk
assessment. However, our study identified no additional
predictors of perceived risk found in previous trials such as
family history of CHD and actual calculated absolute risk.21
This study found that a simple yet systematic CHD risk
factor screening and educational intervention for women
undergoing routine mammography was sufficient to im-
prove women’s ability to correctly categorize their per-
sonal risk. However, although immediate recall was good,
this improvement deteriorated over time, suggesting that
reinforcement and educational follow-up may be necessary
to sustain improvements in knowledge and awareness. The
long-term effectiveness of a brief screening and educa-
tional intervention on women’s accurate perceptions of
personal risk cannot be determined from this study, and
further evaluation is required.
An important finding of this study was that most women
(61%) preferred simpler methods of having risk communi-
cated. Our findings correspond with those of Fortin et al,24
who found that women preferred simpler formats for pre-
senting health risk information (ie, bar graph) and data
framed in absolute rather than relative terms. In the Fortin
et al descriptive pilot study, focus group discussions with
40 women (mean age, 51 years; 50% ethnic minority) were
conducted to identify preferred methods of having CHD,
hip fracture, and breast cancer risk communicated. The
women interviewed preferred simple bar graphs over line
graphs, thermometer graphs, 100 representative faces, and
survival curves. In addition, they reported a preference for
absolute risk over relative risk, concurring with the find-
ings of our study. Preference for having risk communicated
varied in our study by age, ethnicity, and education level.
However, these potential predictors need to be further
evaluated in clinical trials with larger sample sizes.
A limitation of the Framingham global risk assessment
is that calculated absolute risk may be inaccurate in ethnic/
racial minorities. Statistical models used to predict per-
sonal risk of death from CHD have been based on studies
among white populations such as the Framingham Heart
Study.18,19 Attempts to assess the generalizability of the
prediction model have yielded inconsistent results, in part
because of the limited cohort data available for women and
ethnic minority groups.25-27 A recent study provided evidence
that although sex-specific Framingham CHD risk prediction
functions perform well among white and African American
patients, recalibration may be necessary for other ethnic
groups, based on the varying prevalence of risk factors and
underlying rates of CHD events.28 Therefore, the use of
Framingham prediction scores in this study, with the high
participation rate of Hispanic women, may have led to inac-
curate estimations of risk. Nonetheless, the Framingham
score is the national standard to guide the intensity of CHD
• www.mayoclinicproceedings.com 1597
CORONARY HEART DISEASE IN ETHNICALLY DIVERSE WOMEN

preventive interventions regardless of ethnic status. This APPENDIX 1. Framingham Global Risk Estimate of
research suggests that although use of the Framingham glo- 10-Year Risk for Women*
bal risk assessment is recommended to modify the intensity Risk factor
of risk-reduction efforts, health care providers should use
Age (y) Points
caution in using it to motivate patients to adhere to risk- 20-34 –7
reduction therapies as previously suggested.13 35-39 –3
Major strengths of this study include greater than 50% 40-44 0
45-49 3
representation by ethnic minorities in the sample and high 50-54 6
follow-up (89%). However, the study has several limita- 55-59 8
tions. Its nonrandom sample primarily comprised low-risk 60-64 10
women. As such, results may not generalize to other popula- 65-69 12
70-74 14
tions, and further evaluation of preferences for having CHD 75-79 16
risk communicated among intermediate-risk and high-risk
women is needed. In addition, the absence of a control group Total
cholesterol (mg/dL) Points for age range (y)
makes it impossible to compare the results of those who
20-39 40-49 50-59 60-69 70-79
received the risk factor screening and intervention with a <160 0 0 0 0 0
representative group of women who did not. The validity of 160-199 4 3 2 1 1
the instrument used to assess methods of communicating 200-239 8 6 4 2 1
risk is unknown; therefore, further research is needed to 240-279 11 8 5 3 2
>280 13 10 7 4 2
determine the most valid, timely, and efficient method of
assessing preferred methods of communicating CHD risk, Points for age range (y)
particularly among samples of women and diverse ethnic
20-39 40-49 50-59 60-69 70-79
groups. Multiple statistical comparisons were made without Nonsmoker 0 0 0 0 0
adjustment. Nevertheless, this was an exploratory analysis, Smoker 9 7 4 2 1
and findings should be considered for future analyses rather
than as definitive associations. Finally, we were unable to HDL (mg/dL) Points
≥60 –1
correlate behavior change associated with change in risk 50-59 0
perception; thus, future studies should evaluate whether ac- 40-49 1
curate perception of personal risk leads to improved clinical <40 2
outcomes.
Systolic BP (mm Hg) If untreated If treated
<120 0 0
CONCLUSIONS 120-129 1 3
130-139 2 4
A substantial proportion of women in this study did not 140-159 3 5
>160 4 6
accurately perceive their CHD risk level, and a brief educa-
tional intervention significantly improved short-term risk Point total 10-year risk (%)
perception. Furthermore, our results suggest that methods to <9 <1
communicate personal coronary risk should be tailored to a 9 1
participant’s age and education level on the basis of pre- 10 1
11 1
ferred methods reported by women in this study, although 12 1
individualization of information by using the method that 13 2
most suits each patient is optimal. Various formats for com- 14 2
municating CHD risk should be tested, especially among 15 3
16 4
women of different ages and education levels, to determine 17 5
whether providing a preferred method results in improved 18 6
awareness and management of risk factors. 19 8
20 11
21 14
We thank the Cholestech Corporation of Hayward, Calif, for 22 17
donating the cholesterol measurement supplies. We acknowledge 23 22
24 27
Drs Suzanne Smith and Hongju Son for their assistance in review-
>25 >30
ing mammography films, as well as Dr Jhansi Reddy for her
assistance with data collection. *BP = blood pressure; HDL = high-density lipoprotein.

1598 Mayo Clin Proc. • December 2005;80(12):1593-1599 • www.mayoclinicproceedings.com

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

APPENDIX 2. Script for Assessing Preferred Method of Having
CHD Risk Communicated
There are different ways that health care professionals can communicate
your risk of heart disease to you. I’m going to go through some of those
different ways. After I get through explaining them, please let me know
how you would you like your heart disease risk to be described to you.
Please choose only one way, state “other” if you have a different way
you’d like, or “don’t know” if you cannot choose between the methods
discussed or you are unsure. If you don’t understand a specific method or
the question, in general, please feel free to stop me and I will explain
again.
(1) Relative risk is a measure of how much a particular risk factor (say
cigarette smoking) influences the risk of a specified outcome (say, devel-
oping CHD by age 65).
(2) Absolute risk is the observed or calculated probability of develop-
ing CHD. This is what was given to you today, the Framingham global
risk score we reviewed.
(3) Cardiovascular age equivalent is the age you become when you
add your CHD risk factors to your real age. For example, if you are 50
years old and your low-density lipoprotein (LDL) cholesterol level is
severely elevated and you smoke, your cardiovascular age equivalent may
be 62 years old.
(4) Number of traditional risk factors would simply be the number of
traditional CHD risk factors, modifiable and non-modifiable, you have.
For instance, if you were a smoker and were a woman older than 65 years
of age, you would have 2 risk factors.
(5) Risk categories mean that you will be told if you are at high (>20%
chance of developing CHD or having a heart attack in the next 10 years),
moderate (between 10% and 20% chance of developing CHD or having a
heart attack in the next 10 years), or low (<10% chance of developing
CHD or having a heart attack in the next 10 years) risk.
(6) Other means that you would prefer to have your risk of CHD
explained to you in another way (if so, what) or through a combination of
the methods offered above.
(7) Do not know means that you are unsure of how you prefer to have
your CHD risk communicated to you.
REFERENCES
1. American Heart Association. Heart Disease and Stroke Statistics—2005
Update. Available at: www.americanheart.org. Accessed December 2004.
2. Winkleby MA, Robinson TN, Sundquist J, Kraemer HC. Ethnic varia-
tion in cardiovascular disease risk factors among children and young adults:
findings from the Third National Health and Nutrition Examination Survey,
1988-1994. JAMA. 1999;281:1006-1013.
3. Harris MI, Flegal KM, Cowie CC, et al. Prevalence of diabetes, impaired
fasting glucose, and impaired glucose tolerance in U.S. adults: the Third
National Health and Nutrition Examination Survey, 1988-1994. Diabetes
Care. 1998;21:518-524.
4. King H, Aubert RE, Herman WH. Global burden of diabetes, 1995-2025:
prevalence, numerical estimates, and projections. Diabetes Care. 1998;21:
1414-1431.
5. Gillum RF. Secular trends in stroke mortality in African Americans: the role
of urbanization, diabetes and obesity. Neuroepidemiology. 1997;16:180-184.
6. Greenlund KJ, Giles WH, Keenan NL, Croft JB, Casper ML, Matson-
Koffman D. Prevalence of multiple cardiovascular disease risk factors among
women in the United States, 1992 and 1995: the Behavioral Risk Factor
Surveillance System. J Womens Health. 1998;7:1125-1133.
7. Mosca L, Ferris A, Fabunmi R, Robertson RM, American Heart Associa-
tion. Tracking women’s awareness of heart disease: an American Heart Asso-
ciation national study. Circulation. 2004;109:573-579.
Mayo Clin Proc. • December 2005;80(12):1593-1599
For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.
CORONARY HEART DISEASE IN ETHNICALLY DIVERSE WOMEN
8. Winkleby MA, Flora JA, Kraemer HC. A community-based heart dis-
ease intervention: predictors of change. Am J Public Health. 1994;84:767-772.
9. Silagy C, Muir J, Coulter A, Thorogood M, Roe L. Cardiovascular risk
and attitudes to lifestyle: what do patients think? BMJ. 1993;306:1657-1660.
10. Rosenstock IM. The health belief model and preventive health behavior.
Health Educ Monogr. 1974;2:354-386.
11. Janz NK. The Health Belief Model in understanding cardiovascular risk
factor reduction behaviors. Cardiovasc Nurs. 1988;24:39-41.
12. Mosca L, Appel LJ, Benjamin EJ, et al, American Heart Association.
Evidence-based guidelines for cardiovascular disease prevention in women.
Circulation. 2004;109:672-693.
13. Grundy SM, Pasternak R, Greenland P, Smith S Jr, Fuster V. Assessment
of cardiovascular risk by use of multiple-risk-factor assessment equations: a
statement for healthcare professionals from the American Heart Association and
the American College of Cardiology. J Am Coll Cardiol. 1999;34:1348-1359.
14. Chobanian AV, Bakris GL, Black HR, et al, National Heart, Lung, and
Blood Institute Joint National Committee on Prevention, Detection, Evalua-
tion, and Treatment of High Blood Pressure, National High Blood Pressure
Education Program Coordinating Committee. The Seventh Report of the Joint
National Committee on Prevention, Detection, Evaluation, and Treatment of
High Blood Pressure: the JNC 7 report [published correction appears in JAMA.
2003;290:197]. JAMA. 2003;289:2560-2572.
15. Cobbaert C, Boerma GJ, Lindemans J. Evaluation of the Cholestech
L.D.X. desktop analyser for cholesterol, HDL-cholesterol, and triacylglycerols
in heparinized venous blood. Eur J Clin Chem Clin Biochem. 1994;32:391-
394.
16. National Health And Nutrition Examination Survey III: Reference
Manuals and Report—1996. Hyattsville, MD: US Department of Health and
Human Services, Center for Disease Control and Prevention, National Center
for Health Statistics. Available at: www.cdc.gov/nchs. Accessed December
2004.
17. Third Report of the National Cholesterol Education Program (NCEP)
Expert Panel on Detection, Evaluation, and Treatment of High Blood Choles-
terol in Adults (Adult Treatment Panel III) final report. Circulation. 2002;
106:3143-3421.
18. Wilson PW, D’Agostino RB, Levy D, Belanger AM, Silbershatz H,
Kannel WB. Prediction of coronary heart disease using risk factor categories.
Circulation. 1998;97:1837-1847.
19. Kannel WB, McGee D, Gordon T. A general cardiovascular risk profile:
the Framingham Study. Am J Cardiol. 1976;38:46-51.
20. Weinstein ND, Diefenbach MA. Percentage and verbal category mea-
sures of risk likelihood. Health Educ Res. 1997;12:139-141.
21. Frijling BD, Lobo CM, Keus IM, et al. Perceptions of cardiovascular risk
among patients with hypertension or diabetes. Patient Educ Couns. 2004;52:
47-53.
22. Legato MJ, Padus E, Slaughter E. Women’s perceptions of their general
health, with special reference to their risk of coronary artery disease: results of
a national telephone survey. J Womens Health. 1997;6:189-198.
23. Pilote L, Hlatky MA. Attitudes of women toward hormone therapy and
prevention of heart disease [editorial]. Am Heart J. 1995;129:1237-1238.
24. Fortin JM, Hirota LK, Bond BE, O’Connor AM, Col NF. Identifying
patient preferences for communicating risk estimates: a descriptive pilot study.
BMC Med Inform Decis Mak. 2001;1:2-12.
25. Tyroler HA, Knowles MG, Wing SB, et al. Ischemic heart disease risk
factors and twenty-year mortality in middle-age Evans County black males.
Am Heart J. 1984;108(3, pt 2):738-746.
26. Keil JE, Sutherland SE, Knapp RG, Lackland DT, Gazes PC, Tyroler
HA. Mortality rates and risk factors for coronary heart disease in black as
compared with white men and women. N Engl J Med. 1993;329:73-78.
27. Cooper RS, Ford E. Comparability of risk factors for coronary heart
disease among blacks and whites in the NHANES-I Epidemiologic Follow-up
Study. Ann Epidemiol. 1992;2:637-645.
28. D’Agostino RB Sr, Grundy S, Sullivan LM, Wilson P, CHD Risk Predic-
tion Group. Validation of the Framingham coronary heart disease prediction
scores: results of a multiple ethnic groups investigation. JAMA. 2001;286:180-
187.
• www.mayoclinicproceedings.com 1599