Cardiovascular Risk Factor Knowledge among Monolingual Hispanics

Abstract

Hispanics in the United States have worse cardiovascular disease (CVD) risk factor profiles than non-Hispanic Whites. Cardiovascular health literacy is important for health promotion but is not well characterized among monolingual Spanish-speaking Hispanics outside of health care settings. We recruited Hispanic participants (N=235) from a community-based health fair in Denver, Colorado. A total of 182 participants (77%) completed a subsequent language-congruent telephone survey to assess CVD risk-factor knowledge. Of these, 174 self-identified as monolingual Spanish-speaking, and constituted the analysis cohort. Cardiovascular disease risk knowledge score was defined as the number of established risk factors an individual participant could name (out of 10 pre-specified), and multivariable regression analyses were conducted to determine factors independently associated with knowledge. The mean knowledge score for the cohort was 2.2 ± 1.1 out of 10. This suggests an unmet need for tailored educational interventions beyond simple screening events.

The U.S. Hispanic population increased from 47.8 million in 2008 to an estimated 59.9 million in 2018, with Hispanics accounting for more than half (52%) of all U.S. population growth during this period.¹ Hispanic individuals in the United States have worse cardiovascular disease (CVD) risk-factor profiles than non-Hispanic Whites, and in spite of relatively lower CVD morbidity and mortality, CVD continues to be a leading cause of death in this population.^2,3

Approximately 70% of all Hispanics 5 years of age or older living in the U.S. are proficient in English, but this proportion is much lower (<35%) among those who are foreign-born.⁴ Monolingual Spanish-speaking Hispanics are less likely to be screened for cardiovascular risk,⁵ and have higher systolic blood pressure, low-density lipoprotein (LDL) cholesterol, and serum glucose compared with English-speaking Hispanics,⁶ which may in part account for the increased CVD mortality among foreign-born versus U.S.-born Hispanic adults.⁷

Basic knowledge about CVD and its contributing risk factors (CVD health literacy) is an important prerequisite for individual health behavioral changes leading to CVD prevention.⁸ Compared with non-Hispanic Whites, Hispanic individuals have significantly lower CVD health literacy, lower awareness of their own CVD risk factors,^9,10 and lower rates of treatment and risk-factor control.¹¹ However, to date, CVD risk factor knowledge is poorly characterized among the subset of Hispanics who are Spanish monolingual, who are often excluded from CVD literacy studies that use only English-language surveys or interviews.^12-14

The present study aimed to address this research gap by assessing CVD risk knowledge in a sample of monolingual Spanish-speaking Hispanics in Colorado, which is home to more than 1.1 million Hispanics (21% of the state’s population).¹⁵

Methods

Study participants

Participants were identified at a no-cost neighborhood health event organized in coordination with local community leaders by the Colorado Prevention Center, an organization dedicated to creating innovative health improvement programs for underserved individuals. The health event served primarily the Denver Metro area’s uninsured Hispanic population, and included point-of-care cholesterol testing (Cholestech LDX System, Hayward, CA), blood pressure screening, and a bilingual interactive computer-based CVD risk assessment module.¹⁶ Each participant was provided with a language-appropriate one-page printout with their test results, 10-year global risk score, education regarding 10 major CVD risk factors, and recommendations for lowering CVD risk based upon national guidelines. Those presenting with uncontrolled risk factors or moderate to high 10-year risk were directed to a bilingual medical professional, generally physicians or nurses, who were dedicated to on-site, language-appropriate counseling and education, including discussion of health care access, medication compliance, and resources to support healthy lifestyle modification.

A bilingual study coordinator offered all health fair participants the opportunity to provide their name and phone number on a contact request form for the purpose of participating in a subsequent phone-based survey designed to assess knowledge of cardiovascular risk factors and prevention behaviors in the Hispanic community. Consenting individuals were informed that participation was voluntary and that the results of the survey would be used to help develop targeted educational programs about CVD awareness for the Hispanic community. There was no financial incentive to participate in the study. The Colorado Multiple Institutional Review Board (COMIRB) approved the study protocol.

Survey development and administration

The study survey was based on a previously validated English language instrument used nationally by the American Heart Association to assess CVD literacy, including CVD risk factor knowledge, awareness, and perception.¹⁴ Questions were translated into Spanish and adapted to meet the requirements for spoken survey implementation among a primarily Spanish-speaking audience with lower educational levels (short and clear sentences, using words that are simple and common, have as few syllables as possible, and do not have ambiguous or multiple meanings). The final survey was developed by Colorado Prevention Center staff including a preventive cardiologist, medical anthropologist, primary care physician, and public health staff fluent in both Spanish and English. The goal was to capture key information on risk factors that have been established to increase CVD events but to provide it in a way that was as clear and easy to understand as possible, as well as easy to deliver by Spanish-speaking survey personnel with limited medical training.

The survey instrument captured demographic characteristics and socioeconomic factors, and used an open-ended question format to assess knowledge of the 10 established CVD risk factors shared during the neighborhood health event. These risk factors included high cholesterol, high blood pressure, poor diet, overweight/obesity, lack of exercise, diabetes, smoking, family history of premature CVD, older age, and male gender. Individuals were given a knowledge score based upon how many of these 10 established CVD risk factors they could name. The instrument was first pilot-tested on 10 participants drawn from the sample population. Based upon feedback from the lead interviewer, changes were made to the wording of several questions to enhance clarity, delivery, and comprehension by monolingual participants. The final survey was also back-translated into English, to examine the equivalency of meaning with the original English language survey, and to help identify and eliminate any potential ambiguities or errors.

The survey was conducted two months after the health fair and consisted of telephone interviews administered in Spanish by trained bilingual interviewers. While the choice of two months was somewhat arbitrary, we deliberately sought to avoid conducting the survey too soon after the health fair, because our aim was not to test the participants’ immediate recall of health-related “key words” heard at the health fair, but to understand if lasting comprehension could be achieved via the on-site risk-factor testing and counseling format provided. To ensure consistency in our approach, each interviewer used a pre-determined script to engage the participant or to leave a message with either a family member or a phone messaging service regarding the nature of the call. The survey strategy consisted of up to three phone call attempts to reach the participant at the number provided on the contact request form.

Data entry and validation

All survey data were recorded by the bilingual interviewers on paper case-report forms. Survey information was entered into the study database using an independent double-data entry technique. Data were entered into a Microsoft Access database using an add-only entry form customized for the specific survey. To ensure data fidelity, 10% of entered records were randomly selected and each field was reviewed using a read-only interface. Data review suggested an entry error rate less than 1%; all corrections were performed prior to analysis.

Statistical analyses

The overall survey response rate was tabulated. Descriptive statistics for sociodemographic variables were then presented as number and percent of the overall population. Univariate analyses were conducted to evaluate associations of these characteristics with the 10-point knowledge scores. The knowledge scores were presented as mean ± standard deviation. Multiple linear regression analysis with knowledge score as the outcome variable was conducted to test for associations. All analyses were performed using SAS Version 9.2 (SAS Institute, Cary, NC, USA).

Results

Consent for the study was obtained from 235 out of 414 individuals participating in the health fair, and 182 consenting participants completed the subsequent telephone survey (77% response rate). Of those individuals, 174 self-identified as Spanish-monolingual, and constituted the pre-specified analysis cohort.

The study cohort had a mean age of 41 ± 11 years, and included 69% women, 61% participants with less than a middle school education; 46% with reported annual family income of less than $20,000; and 90% with no health insurance. The vast majority (97%) of participants reported a foreign country of birth, predominantly Mexico (Table 1).

Table 1.

CARDIOVASCULAR DISEASE RISK FACTOR KNOWLEDGE SCORE AND SOCIODEMOGRAPHIC CHARACTERISTICS OF THE STUDY POPULATION

Variable	N (%)	Knowledge Score	p-value
Gender
• Female	120 (69)	2.2 (1.2)	.11
• Male	54 (31)	2.0 (1.0)
Age group (years)
• <30 yrs	25 (14)	2.1 ± 1.1	.53
• 30–39 yrs	68 (39)	2.3 ±1.1
• 40–49 yrs	37 (21)	2.1 ±1.2
• 50+ yrs	44 (25)	2.0 ±1.0
Country of birth
• Mexico	144 (83)	2.2 ± 1.1
• United States	5 (3)	1.8 ± 1.3	.61
• Other	25 (14)	2.0 ± 1.0
Years in United States
• < 5	30 (17)	2.4 ± 1.2
• 5–15	88 (51)	2.1 ± 1.1
• > 15	50 (29)	2.1 ± 1.0	.20
• Born in United States	5 (3)	1.8 ± 1.3
• No response	1 (1)	0
Education
• < High school	106 (61)	2.0 ± 1.1	.004
• > High school	68 (39)	2.5 ± 1.0
Health Insurance
• Yes	17 (10)	2.1 ± 1.0	.86
• No	157 (90)	2.2 ± 1.1
Indigent Healthcare Assistance Program
• Yes	20 (13)	3.0 ± 0.9	.0002
• No	137 (87)	2.0 ± 1.1
Employed
• full/part time	113 (65)	2.1 ± 1.0	.10
• stay at home/not employed/retired	61 (35)	2.3 ± 1.2
Annual Income
• < $20,000	80 (46)	2.3 ± 1.1
• $20,000 +	68 (39)	2.1 ± 1.1	.48
• No response	26 (15)	2.0 ± 1.1
Married
• Yes	134 (77)	2.1 ± 1.1	.28
• No or no response	40 (23)	2.3 ± 1.0
How informed are you about CVD?
• Informed	113 (65)	2.3 (1.0)	.01
• Not at all	61 (35)	1.9 (1.1)
Have you ever talked with a health professional in the US about heart disease?
• Yes	30 (17)	2.2 (1.3)
• No	144 (83)	2.1 (1.0)	.69
Did you speak with a health professional about heart disease at this health fair?
• Yes	75 (43)	2.2 (1.1)
• No	99 (57)	2.1 (1.0)	.90

The proportion of participants citing each of the 10 established CVD risk factors is shown in Figure 1. High cholesterol was the most commonly cited risk factor (56%), followed by high blood pressure (34%), poor diet (32%) and overweight/obesity (30%). Only 16% correctly identified smoking and 17% identified diabetes as CVD risk factors. None of the participants identified male gender as a CVD risk factor and 10 of the 174 participants were unable to identify any CVD risk factors.

Figure 1.

Percent of study participants citing each of the 10 established cardiovascular disease risk factors.

The average CVD risk knowledge score for the entire cohort was 2.2 ± 1.1 out of a maximum score of 10. Table 1 shows CVD knowledge scores stratified by sociodemographic characteristics and selected survey questions. As shown in Table 1, having at least a high school education or enrollment in an indigent health care assistance program for uninsured individuals was associated with a higher knowledge score. Participants who considered themselves “informed” (n=113) had a higher knowledge score than those who considered themselves “not at all informed” (n=61), (2.3 ± 1.0 versus 1.9 ± 1.1, p = .01). Discussion about CVD with a health care professional prior to the screening event (2.2 ± 0.20 versus 2.15 ± 0.090, p = .69) or risk factor counseling at the event (2.2 ± 0.13 versus 2.2 ± 0.109, p = .90) were not significantly associated with an increased aggregate knowledge score. Multiple linear regression analysis confirmed that enrollment in an indigent health care assistance program and having at least a high school education was associated with higher knowledge scores (Table 2).

Table 2.

MULTIVARIABLE ASSOCIATIONS WITH CARDIOVASCULAR DISEASE RISK-FACTOR KNOWLEDGE

Variable	Estimate	Standard Error	p-value
Age	−0.003	0.008	.8
Female	0.09	0.20	.6
Born in US vs. Not Born in US	−0.05	0.65	.9
Years In US >15 years	−0.16	0.26	.5
Years In US 5–15 years	−0.25	0.22	.3
High school or greater education	0.44	0.17	.01a
Employed full time	−0.24	0.22	.3
Employed part time	−0.27	0.26	.3
No current employment/retired	0.13	0.34	.7
Household income > $20,000	0.23	0.29	.3
Household income < $20,000	0.39	0.27	.2
Married	−0.12	0.21	.6
No Health Insurance	0.006	0.28	.98
Indigent Healthcare Assistance Program	0.65	0.24	.007a

Note

^aStatistically significant variable associated with the primary outcome.

Table 3 shows the self-reported sources of CVD knowledge for study participants. Nearly 70% of participants received their information about heart disease from television, and nearly 60% from a health fair. Overall, less than a third of the study cohort received CVD education from a physician or another health care source.

Table 3.

PARTICIPANT SOURCES OF INFORMATION ABOUT CARDIOVASCULAR DISEASE

Source	N(%)
Television	121 (69.5)
Free health fair	104 (59.8)
Magazine or newspaper	58 (33.3)
Doctor	54 (31.0)
Family member	42 (24.1)
Radio	39 (22.4)
Friends	18 (10.3)
School/classes	11 (6.3)
Social/community gathering	7 (4.0)
Church	3 (1.7)
Other sources	9 (5.2)

Discussion

Monolingual Spanish-speaking individuals in this study had very low levels of CVD risk-factor knowledge, despite prior participation in a health fair that included CVD risk-factor screening, 10-year global CVD risk assessments, and language-appropriate risk-factor counseling and educational materials. Cardiovascular disease risk knowledge was particularly low among respondents with less than a high school education, and those with limited access to health care (no health insurance and no enrollment in an indigent health care assistance program). These findings suggest a critical unmet need with regard to CVD health education among monolingual Spanish-speaking Hispanics, as well as the potential for expanded access to care to mitigate this disparity.

Although obesity, diabetes, hypertension, and dyslipidemia are highly prevalent among Hispanics,^2,3,17 a high percentage of study participants were unable to name these as risk factors for developing CVD. Given that total and LDL-cholesterol levels may be the most amenable to reduction via medical therapy, it is somewhat encouraging that 56% of study participants were able to identify high serum cholesterol as a risk factor for CVD, although this was possibly cued by on-site point-of-care testing for hypercholesterolemia at the local screening event.

Enrollment in an indigent health care assistance program was associated with better knowledge of CVD risk factors in our study and highlights the importance of access to care for this largely uninsured population. This finding adds to a growing body of evidence suggesting that a dependable source of health care is associated with better health literacy in vulnerable populations, including higher awareness and treatment rates for diabetes, hypertension and dyslipidemia in the Hispanic community.^10,18

The U.S. is projected to become more ethnically and racially diverse over the coming decades, including anticipated increases in the proportion of Hispanics and in the proportion of foreign-born individuals of any ethnicity.¹⁹ In this context, addressing the paucity of linguistically- and culturally-tailored health education efforts targeting the population of monolingual Spanish-speaking Hispanics is imperative. Our study suggests that providing ‘one-off’ CVD risk-factor screening and counseling is inadequate in establishing a functional knowledge base for CVD risk factors and prevention behaviors in this vulnerable population. Improving CVD risk knowledge and promoting heart-healthy behaviors is likely to require novel interventions that are sustained, interactive, tailored to language and education level, and in compliance with the principles and ethics of cultural competence.²⁰ While interventions relying on trained community health workers (Promotores de Salud) have already shown great promise in reducing CVD risk factors among Hispanics,^21-23 a combination of community outreach and the use of novel technologies such as mobile health (mHealth)^24,25 may be required for an intervention that is both scalable to large populations and sustainable over extended periods of time. Further interventions should also focus on the development of culturally relevant and understandable materials for health education among Hispanic populations, developed in collaboration with the target populations, and used consistently across platforms (including health fairs and Spanish-language radio and television).²⁶

Our findings may have public health relevance regarding improving both minority and gender health equity since over two-thirds of our study population were monolingual Hispanic women. Lack of knowledge and misperceptions about CVD risk among women can lead to delays in seeking treatment, less favorable clinical outcomes, and increased mortality compared with men with similar burden of disease.^27,28 Furthermore, among women diagnosed with acute myocardial infarction at several Colorado hospitals, we found that significantly fewer Hispanic women than non-Hispanic White women presented within 24 hours of symptom onset.²⁹ Spanish-monolingual Hispanic women in that study had longer mean delay times compared with both English-speaking Hispanic and non-Hispanic women. This suggests that upstream programs aimed at promoting knowledge of both CVD symptoms and CVD risk factors have the potential to improve women’s health substantively in the Hispanic community.

An important strength of the present study is the use of a Spanish-language survey instrument with questions based on a validated, English-only national survey of CVD literacy.¹³ This approach addresses an important research gap arising from the exclusive use of English language interviews in prior surveys of this type.^12-14 Another strength of our study is the high response rate, consistent with systematic review data about survey performance across racial and ethnic populations,³⁰ and suggest that our results are a valid representation of the local community.

This study also has important limitations. First, participants represented a convenience sample recruited from a health fair environment at only one location. We cannot exclude the possibility that health fair participation, as well as consent for the present study, inevitably selected individuals with relatively higher health literacy than their peers who chose not to participate. In addition, given the heterogeneity of Hispanic communities across the United States, our findings in a Colorado Hispanic population of predominantly Mexican origin may not be representative of the broader U.S. Hispanic population. Another potential limitation is the requirement to provide a telephone number in order to be included in the study, which could have theoretically introduced sampling bias. However, this is unlikely to have meaningfully impacted our results, since 96% of Hispanic adults in the U.S. are estimated to own a mobile phone.³¹ Finally, inclusion of a concomitant control group of English-speaking Hispanics in the study would have provided valuable context and additional insights, but was not possible because only a small fraction of (8 out of 182) of consenting Hispanic participants recruited at the health fair were English speakers.

In summary, this study highlights important deficits in knowledge and awareness of CVD risk factors among monolingual Hispanics, particularly those with limited access to care and a lower education status. These findings underscore the need for novel CVD education and prevention programs that are linguistically and culturally tailored, scalable, and temporally sustainable, with the goal of promoting heart-healthy behaviors in this vulnerable population. Future prospective studies evaluating the impact of evidence-based, language-congruent interventions within the Hispanic community are warranted.