Skip to main content
NCBI home page
HHS Author Manuscripts logoLink to HHS Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 4.
Published in final edited form as: Am J Prev Med. 2014 Nov 10;48(1):58–69. doi: 10.1016/j.amepre.2014.09.014

Predicted 10-Year Risk of Developing Cardiovascular Disease at the State Level in the U.S.

Quanhe Yang 1, Yuna Zhong 1, Matthew Ritchey 1, Fleetwood Loustalot 1, Yuling Hong 1, Robert Merritt 1, Barbara A Bowman 1
PMCID: PMC10910565  NIHMSID: NIHMS1968223  PMID: 25450016

Abstract

Background:

Cardiovascular disease (CVD) is the leading cause of death in the U.S. State-specific predicted 10-year risk of developing CVD could provide useful information for state health planning and policy.

Purpose:

To estimate state-specific 10-year risk of developing CVD.

Methods:

Using the updated non–laboratory-based Framingham CVD Risk Score (RS), this study estimated 10-year risk of developing CVD; coronary heart disease (CHD); and stroke, stratified by demographic factors and by state among 2009 Behavioral Risk Factors Surveillance System participants aged 30–74 years. Data analysis was completed in June 2014.

Results:

The age-standardized mean CVD, CHD, and stroke RSs for adults aged 30–74 years were 14.6%, 10.4%, and 2.3% among men, respectively, and 7.5%, 4.5%, and 1.8% among women. RSs increased significantly with age and were highest among non-Hispanic blacks, those with less than high school education, and households with incomes <$35,000. State-specific age-standardized CVD, CHD, and stroke RS ranged, among men, from lows in Utah (13.2%, 9.6%, and 2.1%, respectively) to highs in Louisiana (16.2%, 11.7%, and 2.6%), and among women, from lows in Minnesota (6.3%, 3.8%, and 1.5%) to highs in Mississippi (8.7%, 5.3%, and 2.1%).

Conclusions:

The predicted 10-year risk of developing CVD varies significantly by age, gender, race/ethnicity, educational attainment, household income, and state of residence. These results support the development and implementation of targeted prevention programs by states to address the risk of developing CVD, CHD, and stroke among their populations.

Introduction

In the U.S., annual total cardiovascular disease (CVD) costs are $312.6 billion, including $192.1 billion in direct medical expenses and more than $120.5 in lost productivity from premature mortality.1 Nearly 800,000 people die each year from CVD, making it the leading cause of death in the U.S.1 Although CVD mortality has continued to decline during the past 40 years,2 the rates of decline vary significantly among the states.1,3,4 Across states, significant variation has been described in the prevalence of CVD and cardiovascular health5-7 and risk factors, including hypertension,8 smoking,9 high blood cholesterol,10 diabetes,11 overweight/obesity,12 physical inactivity,13,14 and limited consumption of vegetables and fruits.14

The updated 10-year CVD Framingham Risk Score (RS)15 was established to be a more comprehensive measure of an individual’s risk for developing any CVD-related condition over the next 10 years. The RS combines a number of components, including modifiable risk factors (i.e., systolic blood pressure [BP]; antihypertensive medication use; smoking status; and total and high-density lipoprotein [HDL] cholesterol levels) and non-modifiable risk factors (i.e., age and sex), into a CVD risk prediction algorithm. RS results can then be multiplied by gender-specific calibration factors to estimate coronary heart disease (CHD) or stroke-specific risk. The RS can be used to understand disparities in predicted CVD risk among populations. Although multiple studies have used data from the National Health and Nutrition Examination Survey (NHANES) to estimate distribution and trends of predicted 10-year CVD risk at the national level,16-18 no state-level estimates are available. Having these estimates will improve priority-setting efforts so that geographic disparities in CVD incidence and mortality can be better addressed.19 Overcoming geographic disparities is critical for meeting the USDHHS Million Hearts® initiative’s20 goal of preventing one million heart attacks and strokes by 2017 and the Healthy People 202021 objectives of reducing the CHD and stroke death rates by 20%. The objective of this study is to estimate, using RS and 2009 Behavioral Risk Factor Surveillance System (BRFSS) data, the state-specific predicted 10-year risk of developing CVD, CHD, or stroke.

Methods

Behavioral Risk Factor Surveillance System

The BRFSS is a state-based, random-digit–dialed landline telephone survey of the U.S. civilian, non-institutionalized population aged 18 years or older. It is the world’s largest ongoing telephone health surveillance system that tracks health conditions and health-related behaviors in all 50 states and Washington DC. A complete, detailed description of the BRFSS can be found at www.cdc.gov/brfss. This study used the information derived from the core BRFSS questions to calculate consistent and comparable estimates of predicted CVD risk across all states and the District of Columbia. Among 432,607 participants in BRFSS 2009, 94,686 participants aged < 30 or > 74 years were excluded to meet the recommended RS age standards.15 Subsequently, 31,685 participants with self-reported CHD or stroke at baseline, 1,427 pregnant women, and 3,296 participants with missing RS component data were excluded, leaving 297,910 participants for analysis.

Definition of Cardiovascular Disease Risk Factors Included in the Risk Score Calculation

The laboratory-based RS is determined using the following components: age, sex, total and HDL cholesterol, systolic BP, smoking, and diabetes status.15 Because the laboratory-based measurements of total or HDL cholesterol were not available in BRFSS, this study used the coefficients derived from the non–laboratory-based RS and replaced the cholesterol measures with the participant’s BMI.15 Participant age was self-reported, and self-reported weight and height were used to calculate BMI (weight [kg]/height [m2]). Current smoking was defined as having smoked at least 100 cigarettes in one’s lifetime and reporting smoking every day or some days at the time of survey. Diagnosed diabetes was defined as having ever been told by a doctor that respondents had diabetes, excluding gestational diabetes.

Estimate of Systolic Blood Pressure in the Behavioral Risk Factor Surveillance System

The non–laboratory-based RS includes the systolic BP in the risk estimation, but measured blood pressure was not available in BRFSS. This study used an approach similar to that developed by Ezzati and colleagues22 to estimate the systolic BP for BRFSS participants, but included a different set of predictors based on its model selection criterion. First, multivariable linear regression models were developed to predict participants’ systolic BP based on the information derived from NHANES 2005–2010. NHANES is a series of cross-sectional national, stratified, multistage probability surveys of the civilian, non-institutionalized U.S. population. Detailed descriptions of the plan and operation of each survey are published elsewhere.23,24 This study pooled three cycles of NHANES data (2005–2006, 2007–2008, and 2009–2010) to develop stable parameter estimates that overlapped with the time period of the 2009 BRFSS survey so the NHANES-derived parameters could be applied to BRFSS participants. The dependent variable of the prediction models was the individual’s systolic BP measurement. The independent variables included a set of health system and sociodemographic covariates chosen a priori that are related to systolic BP, including age; race/ethnicity; educational attainment; BMI; physical activity level (inactive, insufficiently active, or active); household income (<$35,000 or ≥$35,000); smoking status (never, former, or current); alcohol consumption (none, < 1 drink/day, or ≥1 drink/day); health insurance (yes or no); diabetes (yes or no); and antihypertensive medication use (yes or no). In addition to main effects of the independent variables, PROC GLMSELECT procedure (SAS, version 9.3) with stepwise selection and the Schwarz Bayesian Criterion25 were used to screen for the significant interactions among all covariates. This study stratified NHANES data by sex and hypertension status (yes or no). Hypertension was defined as systolic BP ≥ 140 mmHg, diastolic BP ≥ 90 mmHg, or reported use of antihypertensive medication. This study then developed four multivariable prediction regression models. The coefficients of each prediction model are presented in the online Appendix Tables 1 and 2. Then, these regression coefficients derived from NHANES 2005–2010 were applied to similar questions in BRFSS 2009 to estimate systolic BP for each BRFSS participant by each stratum.22

Statistical Analysis

This study calculated the predicted 10-year CVD risk using the gender-specific non–laboratory-based RS for each BRFSS 2009 participant and applied gender-specific calibration factors to estimate the 10-year risk of CHD and stroke.15 In addition, this study also calculated the age-standardized (to U.S. Census 2000 population) and weighted mean and 95% CI of CVD, CHD, or stroke RS, and the prevalence of CVD and CHD RS ≥ 20% (95% CI) by sex, race/ethnicity, educational attainment, and household income. Those with RS ≥ 20% are often identified as the group with greatest CVD risk; however, a stable estimate of RS ≥ 20% for stroke could not be obtained owing to the limited sample size. For the state-specific estimates, the age-standardized mean CVD, CHD, or stroke RS, and the prevalence of predicted CVD and CHD RS ≥ 20% were calculated. The survey weights were used to represent the target population, and age standardization was used to control for the possible confounding effect of difference in age structure by sociodemographic status and across state. All results are from the weighted analyses.

This study conducted multiple sensitivity analyses. First, this study compared the characteristics of participants aged 30–74 years by the selected sociodemographic variables in NHANES 2005–2010 with those of BRFSS 2009 participants (Appendix Table 3). Second, this study calculated the national predicted 10-year risk of developing CVD, CHD, or stroke by sex using the measured systolic BP in NHANES 2005–2010 and compared the results with those from the BRFSS 2009 estimates (Appendix Tables 4 and 5). Third, this study compared the distributions of the predicted 10-year risk of developing CVD by three approaches (laboratory-based, non–laboratory-based, and non–laboratory-based with predicted systolic BP using NHANES 2005–2010 data [Appendix Figure 1]). Fourth, this study presented the age-standardized (Appendix Tables 6 and 7) and crude (Appendix Tables 8 and 9) predicted 10-year risk of developing CVD, CHD, or stroke by state in BRFSS 2009. Finally, the bias associated with using a point-estimate of systolic BP in the CVD RS estimate was quantified (Appendix Table 10).

Data were analyzed using SAS, version 9.3, and SUDAAN, version 10, taking into account the complex sampling design and weights in NHANES and BRFSS. To account for multiple statistical testing, the significance level was set to control for a false discovery rate (FDR) at 5.0%, and adjusted p-values were calculated. The FDR represents the proportion of incorrectly rejected null hypotheses among all rejected null hypotheses.26 SAS PROC MULTTEST with the FDR option was used to calculate the FDR-adjusted p-values and p<0.05 was considered significant. This study presents the usual p-values and highlights those that were significant after controlling for FDR at 5.0%. All tests were two-sided. Data analysis was completed in June 2014.

Results

The characteristics of participants aged 30–74 years in BRFSS 2009 were largely comparable to those from NHANES 2005–2010. The prevalence of less than high school education, current smoking, and obesity (BMI ≥ 30) were lower in BRFSS 2009 than in NHANES 2005–2010 (Table 1 and Appendix Table 3).

Table 1.

Characteristics of adult participants 30–74 years of age, Behavioral Risk Factor Surveillance System, 2009, % (95% CI) unless otherwise noted.

Characteristics N Men N Women p-value
Mean age (years) 112,856 47.6 (47.5, 47.8) 185,054 48.7 (48.6, 48.8) <0.001
Race/ethnicity
 Non-Hispanic white 90,322 68.9 (68.3, 69.5) 144,542 70.1 (69.6, 70.5)
 Non-Hispanic black 7,397 9.3 (8.9, 9.7) 16,353 10.6 (10.3, 10.9)
 Hispanic 7,300 14.4 (13.9, 15.0) 13,224 13.7 (13.3, 14.0)
 Others 6,671 7.4 (7.0, 7.8) 9,597 5.7 (5.5, 5.9) < 0.001
Educational attainment
 <High school 8,008 9.2 (8.8, 9.6) 13,238 8.5 (8.2, 8.7)
 High school 31,037 25.7 (25.2, 26.2) 51,903 25.6 (25.3, 26.0)
 > High school 73,556 65.1 (64.5, 65.7) 119,474 65.9 (65.5, 66.3) 0.009
Household income ($)
 <35k 29,999 27.9 (27.3, 28.5) 60,519 32.9 (32.4, 33.3)
 ≥35k 73,646 72.1 (71.5, 72.7) 101,959 67.1 (66.7, 67.6) <0.001
Smoking status
 Current smoking 20,301 19.0 (18.5, 19.5) 30,628 16.1 (15.8, 16.4)
 Former 37,970 29.2 (28.7, 29.7) 47,446 22.4 (22.1, 22.8)
 Never 54,585 51.8 (51.2, 52.4) 106,980 61.5 (61.1, 61.9) <0.001
BMI a
 <25.0 26,770 23.3 (22.8, 23.8) 68,510 38.3 (37.9, 38.8)
 25.0–29.9 51,186 45.2 (44.6, 45.8) 54,730 28.8 (28.4, 29.2)
 ≥30.0 34,900 31.5 (30.9, 32.0) 61,814 32.9 (32.5, 33.3) <0.001
Diabetes
 Yes 11,878 8.9 (8.6, 9.3) 17,996 8.1 (7.9, 8.3)
 No 100,978 91.1 (90.7, 91.4) 167,058 91.9 (91.7, 92.1) <0.001
Hypertension
 No 73,028 70.1 (69.6, 70.6) 124,651 73.4 (73.1, 73.8)
 Yes 39,828 29.9 (29.4, 30.4) 60,403 26.6 (26.2, 26.9) <0.001
 Taking anti-hypertensive medication 30,975 71.9 (70.9, 72.9) 51,066 81.0 (80.3, 81.6) 0.001
Systolic blood pressure (mmHg; M) b 112,856 123.4 (123.3, 123.5) 185,054 120.0 (119.8, 120.0) 0.001
Open in a new tab

Note: Values within parentheses are 95% CIs. Boldface indicates significant p-value after controlling for the false discovery rate at 5.0% for multiple comparisons; comparisons were made between men and women based on t tests for the continuous variables and χ2 tests for the categorical variables; all tests were two-tailed.

a

BMI was based on self-reported weight and height.

b

Systolic blood pressure was estimated based on the multivariable regression models developed using NHANES 2005–2010 data.

Nationally, the absolute difference in age-standardized predicted 10-year CVD risk for the population aged 30–74 years was nearly twice as high in men compared with women (14.6% vs 7.5%; Tables 2 and 3). Prevalence of CVD RS ≥ 20% was almost threefold higher in men than in women (25.0% vs 8.7%). Similar patterns of gender differences were observed for CHD and stroke risk.

Table 2.

Distribution of predicted 10-year cardiovascular disease risk scores for men 30–74 years of age, Behavioral Risk Factor Surveillance System, 2009

CVD predicted risk scoresa
 CHD predicted risk scoresa
 Stroke predicted risk scoresa,b
Characteristics N M % Risk scores ≥20% M % Risk scores ≥20% M
Total 112,856 (52.5%)c 14.6 (14.5, 14.7) 25.0 (24.8, 25.3) 10.4 (10.4, 10.5) 14.9 (14.6, 15.1) 2.32 (2.30, 2.33)
Age (years)
 30–39 18,005 4.1 (4.0, 4.2) 0.3 (0.2, 0.4) 2.95 (2.90 – 2.99) 0.04 (0.02, 0.09) 0.65 (0.64, 0.66)
 40–49 26,636 9.3 (9.1, 9.4) 5.0 (4.5, 5.5) 6.64 (6.55, 6.74) 1.6 (1.4, 1.9) 1.47 (1.45, 1.49)
 50–59 32,689 18.6 (18.4, 18.8) 33.3 (32.4, 34.3) 13.3 (13.2, 13.5) 14.0 (13.3, 14.7) 2.95 (2.92, 2.99)
 60–74 35,526 33.2 (32.9, 33.5) 81.5 (80.7 – 82.3) 23.8 (23.6, 24.1) 56.8 (55.8, 57.7) 5.28 (5.24, 5.33)
Race/ethnicity
 Non-Hispanic white 90,322 14.1 (14.0, 14.2) 24.1 (23.9, 24.4) 10.13 (10.07-10.18) 13.9 (13.7, 14.2) 2.24 (2.23, 2.26)
 Non-Hispanic black 7,397 17.5 (17.2, 17.9) 31.5 (30.4, 32.7) 12.6 (12.3, 12.9) 22.2 (21.0, 23.4) 2.79 (2.73, 2.85)
 Hispanics 7,300 15.3 (14.9, 15.8) 25.9 (24.5, 27.4) 11.0 (10.7, 11.3) 15.7 (14.5, 17.0) 2.44 (2.37, 2.51)
 Others 6,671 14.0 (13.6, 14.3) 23.9 (22.6, 25.2) 10.0 (9.8, 10.3) 14.6 (13.5, 15.9) 2.22 (2.16, 2.28)
Educational attainment
 <High school 8,008 17.4 (17.0, 17.8) 31.1 (29.9, 32.3) 12.5 (12.2, 12.8) 20.6 (19.5, 21.7) 2.77 (2.71, 2.83)
 High school 31,037 16.0 (15.8, 16.2) 28.6 (28.0, 29.2) 11.5 (11.4, 11.6) 17.7 (17.2, 18.2) 2.54 (2.52, 2.57)
 >High school 73,556 13.5 (13.4, 13.6) 22.7 (22.4, 23.1) 9.7 (9.6, 9.8) 12.9 (12.6, 13.2) 2.15 (2.14, 2.17)
Household income ($)
 <35k 29,999 16.6 (16.4, 16.8) 30.0 (29.3, 30.7) 11.9 (11.8, 12.1) 18.7 (18.2, 19.3) 2.64 (2.61, 2.68)
 ≥35k 73,646 13.5 (13.4, 13.6) 22.7 (22.4, 23.1) 9.7 (9.6, 9.8) 12.8 (12.5, 13.1) 2.15 (2.14, 2.16)
Open in a new tab

Note: Values within parentheses are 95% CIs.

a

Age-standardized by the direct method to the US 2000 census population using the age groups 30–39, 40–49, 50–59, and 60–74 years. Predicted 10-year risk of developing cardiovascular disease were based on the predicted systolic blood pressure using multivariable linear regression models developed from NHANES 2005–2010.

b

Stable estimates of risk score prevalence ≥20% for stroke could not be obtained due to the limited sample size.

c

Percentage in parentheses is the survey response rate.

CHD, coronary heart disease; CVD, cardiovascular disease; NHANES, National Health and Nutrition Examination Survey.

Table 3.

Distribution of predicted 10-year cardiovascular disease for women participants 30–74 years of age, Behavioral Risk Factor Surveillance System, 2009

CVD predicted risk scoresa
 CHD predicted risk scoresa
 Stroke predicted risk scoresa,b
Characteristics n M % Risk scores ≥20% M % Risk scores ≥20% M
Total 185,054 (52.5%)c 7.5 (7.4, 7.5) 8.7 (8.6, 8.9) 4.53 (4.51, 4.56) 2.8 (2.7, 3.0) 1.78 (1.77, 1.79)
Age (years)
 30–39 29,337 1.68 (1.66, 1.71) 0.0 1.03 (1.01, 1.04) 0.0 0.40 (0.39, 0.41)
 40–49 41,054 4.26 (4.20, 4.32) 0.7 (0.6, 0.8) 2.59 (2.56, 2.63) 0.07 (0.04, 0.12) 1.02 (1.00, 1.03)
 50–59 52,346 9.0 (8.9, 9.2) 7.7 (7.2, 8.1) 5.5 (5.4, 5.6) 1.3 (1.2, 1.5) 2.16 (2.13, 2.18)
 60–74 62,317 18.8 (18.6, 19.0) 34.3 (33.7, 35.0) 11.4 (11.3, 11.5) 12.7 (12.2, 13.2) 4.48 (4.44, 4.52)
Race/ethnicity
 Non-Hispanic white 144,542 7.01 (6.97, 7.05) 7.6 (7.5, 7.8) 4.27 (4.24, 4.29) 2.3 (2.2, 2.4) 1.67 (1.66, 1.68)
 Non-Hispanic black 16,353 10.2 (10.0, 10.4) 15.1 (14.4, 15.8) 6.2 (6.1, 6.3) 6.6 (6.1, 7.2) 2.43 (2.38, 2.48)
 Hispanic 13,224 8.0 (7.8, 8.2) 10.8 (9.9, 11.6) 4.9 (4.7, 5.0) 3.6 (3.1, 4.2) 1.99 (1.97, 2.01)
 Others 9,597 7.4 (7.1, 7.7) 9.1 (8.2, 10.1) 4.5 (4.3, 4.7) 3.1 (2.6, 3.8) 1.77 (1.76, 1.79)
Educational attainment
 <High school 13,238 10.2 (10.0, 10.4) 15.6 (14.8, 16.4) 6.2 (6.1, 6.3) 6.1 (5.6, 6.6) 2.43 (2.38, 2.48)
 High school 51,903 8.34 (8.25, 8.42) 10.7 (10.3, 11.0) 5.08 (5.02, 5.13) 3.4 (3.2, 3.6) 1.99 (1.97, 2.01)
 >High school 119,474 6.64 (6.59, 6.69) 6.75 (6.5, 6.8) 4.04 (4.01, 4.07) 2.0 (1.9, 2.1) 1.58 (1.57, 1.60)
Household income ($)
 <35k 60,519 9.11 (9.01, 9.20) 12.8 (12.5, 13.1) 5.54 (5.48, 5.60) 4.6 (4.4, 4.8) 2.17 (2.15, 2.19)
 ≥35k 101,959 6.33 (6.28, 6.38) 5.7 (5.5, 5.9) 3.85 (3.82, 3.88) 1.4 (1.3, 1.5) 1.51 (1.50, 1.52)
Open in a new tab

Note: Values within parentheses are 95% CIs.

a

Age-standardized by the direct method to the U.S. 2000 census population using the age groups 30–39, 40–49, 50–59, and 60–74 years. Predicted 10-year risk of developing cardiovascular disease were based on the predicted systolic blood pressure using multivariable linear regression models developed from NHANES 2005–2010.

b

Stable estimates of risk score prevalence ≥20% for stroke could not be obtained because of the limited sample size.

c

Percentage in parentheses is the survey response rate.

CHD, coronary heart disease; CVD, cardiovascular disease; NHANES, National Health and Nutrition Examination Survey.

The predicted risk among men increased significantly with age, and was highest among non-Hispanic blacks, those who had less than high school education, and those with household income <$35,000 (Table 2). The pattern of the predicted risk was similar among women (Table 3). The pattern of the predicted national CVD risk using BRFSS 2009 was consistent with that of NHANES 2005–2010, which used measured BMI and systolic BP (Appendix Tables 4 and 5).

Among men, the mean age-standardized predicted state-level 10-year risk of developing CVD, CHD, or stroke ranged from lows in Utah (13.2% [12.9, 13.5], 9.5% [9.3, 9.7], and 2.1% [2.1, 2.1], respectively) to highs in Louisiana (16.2% [15.8, 16.7], 11.7% [11.3, 12.0], and 2.6% [2.5, 2.7], respectively) (Figure 1 Map A, Figure 3A, and Appendix Table 6). Prevalence of age-standardized CVD and CHD RS ≥ 20% ranged from lows in Colorado for both RS measures (21.6% [20.5, 22.7], 11.6% [10.7, 12.5], respectively) to highs for CVD in Mississippi (29.8% [28.3, 31.3]) and for CHD in Kentucky (19.4% [17.4, 21.5]) (Figure 2 Map A and Appendix Table 6). Among women, the mean age-standardized CVD, CHD, or stroke RS ranged from lows in Minnesota (6.3% [6.1, 6.5], 3.8% [3.7, 3.9], and 1.5% [1.5, 1.6], respectively) to highs in Mississippi (8.7% [8.5, 8.9], 5.3% [5.2, 5.4], and 2.1% [2.0, 2.1], respectively) (Figure 1 Map B, Figure 3B, and Appendix Table 7). Prevalence of RS ≥20% ranged from lows in Minnesota (5.5% [4.9, 6.2] for CVD and 1.4% [1.1, 1.8] for CHD) to highs for CVD in Mississippi (11.4% [10.6, 12.2]) and for CHD in Indiana (4.0% [3.5, 4.7]) (Figure 2 Map B and Appendix Table 7). In addition, the higher predicted 10-year CVD risk tended to cluster in the southeastern states and the lower risk in the northwestern states (Figures 1 and 2 Maps A and B).

Figure 1. Age-standardized mean predicted 10-year risk of developing cardiovascular disease (CVD) among men (A) and women (B) by state, Behavioral Risk Factor Surveillance System, 2009.

Figure 1.

Open in a new tab

Note: For A, the CVD risk for men was classified into five categories: ≤ 13.7%, 13.8%–14.1%, 14.2%–14.5%, 14.6%–15.0%, and ≥15.1%. For B, the CVD risk for women was classified into five categories: ≤6.9%, 7.0%–7.2%, 7.3%–7.4%, 7.5%–7.9%, and >7.9%.

Figure 3.

Figure 3.

Figure 3.

Open in a new tab

State-specific distribution of age standardized mean predicted 10-year risk of developing cardiovascular disease (CVD) among men (A) and women (B), Behavioral Risk Factor Surveillance System, 2009.

Figure 2. Age-standardized prevalence of predicted 10-year risk of developing cardiovascular disease (CVD) ≥20% among men (A) and women (B) by state, Behavioral Risk Factor Surveillance System, 2009.

Figure 2.

Open in a new tab

Note: For A, the CVD risk ≥20% for men was classified into five categories: ≤23.1%, 23.2%–24.1%, 24.2%–25.1%, 25.2%–26.7%, and >26.7%. For B, the CVD risk ≥20% for women was classified into five categories: ≤7.1%, 7.2%–7.8%, 7.9%–8.8%, 8.9%–9.8%, and ≥9.9%.

The sensitivity analysis using NHANES 2005–2010 data showed that the distribution of predicted 10-year risk was nearly identical using predicted versus measured systolic BP (Appendix Figure 1). The ranking and pattern of crude CVD, CHD, or stroke RS remained unchanged as compared to the age-standardized estimate (Appendix Tables 8 and 9). The bias associated with using the point-estimate of systolic BP on RS was quite small (Appendix Table 10).

Discussion

By applying the RS to BRFSS 2009 data for participants aged 30–74 years, this study estimates the predicted 10-year risk of developing CVD, CHD, or stroke by selected characteristics and by state of residence. The results indicate that the predicted 10-year risk varies significantly by sex, race/ethnicity, educational attainment, and household income, consistent with previous studies on disparities in prevalence of CVD risk factors, incidence, prevalence, and mortality.1,7,16,17 In addition, this study estimates, for the first time, CVD, CHD, and stroke RS at the state level, and documents significant variations by state. The patterns of differences in the predicted 10-year risk by selected groups at the national level in this study are consistent with the findings of other studies.17,18 However, the level of the predicted 10-year risk of CVD in this study is higher than that of other studies, which might be due to the use of different CVD prediction models.27 The predicted 10-year risks of developing CVD, CHD, or stroke at the national level by the selected characteristics in BRFSS 2009 are largely comparable to the predicted 10-year risk using NHANES 2005–2010 with measured BP (Tables 2 and 3 and Appendix Tables 4 and 5).

The clinically based prevention efforts of Million Hearts focuses on the “ABCs” of heart health (i.e., appropriate aspirin use, blood pressure control, cholesterol management, and smoking assessment and cessation treatment).28 The state-specific data on the distribution of CVD, CHD, and stroke RS provide essential information for targeting interventions to achieve effective reductions in CVD, CHD, and stroke morbidity and mortality. Although all states with high levels of predicted 10-year CVD risk are in need of interventions, the states with the highest predicted risk need even more interventions to effectively reduce the disparities in CVD morbidity and mortality. By better understanding the predicted 10-year CVD RS, efforts to address these disparities can be implemented.

Hypertension is a major risk factor for CVD and stroke, and it affects approximately one in three U.S. adults.1 Many studies have documented geographic variations in BP status, prevalence of self-reported hypertension, and antihypertensive medications use.8,22,29-31 A recent study indicated that, among 2009 BRFSS participants, the age-adjusted prevalence of self-reported hypertension among adults ranged from 20.9% in Minnesota to 35.9% in Mississippi. The proportion of those with hypertension who reported use of antihypertensive medications ranged from 52.3% in California to 74.1% in Tennessee.8 Our finding supports the Million Hearts goal of improving hypertension control to reduce CVD morbidity and mortality, especially among states with high hypertension prevalence and a low proportion of antihypertensive medication use among people with hypertension.

Diabetes mellitus is another established major risk factor for CVD and stroke.32 Previous studies examined geographic variation and trends in diabetes by state.33-36 Although the prevalence of diagnosed diabetes increased by 95% among all states from 1995 to 2010 (from 4.2% to 8.2%), the disparities of the relative increase in prevalence varied from 27.1% in Vermont to 226.7% in Oklahoma.36 U.S. incidence and prevalence of diabetes are projected to continue to increase until 2050, suggesting that diabetes might play an increasingly important role in predicted CVD risk in the future, especially among non-Hispanic blacks.37 Effective prevention strategies, such as increasing physical activity, having a healthy diet, and weight control, which target the high-risk groups as well as the whole population, are needed to reverse the trend of increasing diabetes prevalence and reduce the risk of developing CVD (www.diabetes.org/diabetes-basics/prevention/).

Smoking and obesity are important risk factors for CVD and stroke.38 The prevalence of smoking among U. S. adults has declined over the last 40 years,39 but an estimated 46.6 million adults (aged ≥18 years), or 20.6%, were current cigarette smokers in 2009.40 Many studies have documented the increased prevalence of obesity among U.S. adults, although the rate of increase has slowed since the mid-2000s.41-43 During 2009–2010, about 36% of U.S. adults, >78 million, were obese (BMI ≥ 30).43 Continued public health intervention programs aimed at improving lifestyle choices are central to reducing the prevalence of obesity as well as the risk of CVD (www.cdc.gov/obesity/strategies/index.html).

Limitations

First, this study uses the model-estimated systolic BP, instead of measured systolic BP, for BRFSS participants to calculate RS,22 and the accuracy of predicted 10-year risk of developing CVD using this method is unknown. However, mean predicted systolic BP in BRFSS participants is nearly identical to that of NHANES participants with measured SBP (Table 1 and Appendix Table 3). In addition, in the sensitivity analysis using NHANES 2005–2010 data, the distribution of predicted 10-year risk is nearly identical using predicted versus measured systolic BP (Appendix Figure 1). Second, this study uses the point-estimate of systolic BP for BRFSS participants to calculate the RS, which did not take into account the variability associated with such estimates. However, the results of the simulation study suggest that the potentially introduced bias is quite small (Appendix Table 10). Third, this study uses the non–laboratory-based RS to predict the CVD risk, which tends to overestimate CVD risk as compared to the laboratory-based RS (Appendix Figure 1).16 Different CVD prediction models are known to yield different predicted risks of developing CVD for the same population27,44; therefore, the present predicted risk estimates should be interpreted with some caution. Fourth, this study uses self-reported BMI and diagnosed diabetes by BRFSS participants to predict CVD risk, as measured BMI and diabetes were not available. Under-reporting of BMI is well documented in BRFSS, which might underestimate the CVD risk for some participants.45 However, studies indicate that self-reported diabetes status and those based on actual diagnoses are in substantial agreement in BRFSS.46 In addition, the estimated prevalence of diagnosed diabetes in BRFSS is in close agreement with other national estimates.47 Fifth, the RS predicts the 10-year CVD risk for people aged 30–74 years only; therefore, interpretation of these results should be mainly focused on that population. Finally, the 52.5% response rate to the 2009 BRFSS, though acceptable, could limit the generalizability of the results if non-respondents systematically differ from respondents in their answers. Despite these limitations, BRFSS data, collected via the largest ongoing state-based cross-sectional survey in the U.S., are reliable and valid as measured against other national population–based surveys and provide the opportunity to report health indicators at the state level.

In summary, this study demonstrates significant state-level variation in predicted 10-year risk of developing CVD, CHD, and stroke. The results of this analysis support the development and implementation of targeted prevention programs by states to address the risk of developing CVD, CHD, and stroke among their populations.

Supplementary Material

Supplemental Material

Acknowledgments

We thank Professor Majid Ezzati, Department of Epidemiology and Biostatistics, School of Public Health, Imperial College London, United Kingdom, for help with estimating systolic blood pressure for Behavioral Risk Factor Surveillance System participants using parameters derived from the National Health and Nutrition Examination Survey 2005–2010 data. We thank Drs. Mark Cobain, Unilever Research, Corporate Biology, United Kingdom, and Michael Pencina, Department of Mathematics and Statistics, Boston University, for their help in using the updated Framingham Risk Score to estimate the 10-year risk of developing cardiovascular disease. We thank Dr. Lawrence Barker, Division of Diabetes Translation, CDC, for his statistical advice regarding quantification of the bias associated with using point-estimates of systolic blood pressure in calculation of the Framingham Risk Score.

Footnotes

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of CDC.

No financial disclosures were reported by the authors of this paper.

Supplementary Data

Supplementary data associated with this article can be found at http://dx.doi.org/10.1016/j.amepre.2014.09.014.

References

  • 1.Go AS, Mozaffarian D, Roger VL, et al. Heart disease and stroke statistics—2013 update: a report from the American Heart Association. Circulation 2013;127(1):e6–e245. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Kochanek KD, Xu J, Murphy SL, Minino AM, Kung HC. Deaths: final data for 2009. Natl Vital Stat Rep 2011;60(3):1–116. [PubMed] [Google Scholar]
  • 3.Cooper R, Cutler J, Desvigne-Nickens P, et al. Trends and disparities in coronary heart disease, stroke, and other cardiovascular diseases in the United States: findings of the national conference on cardiovascular disease prevention. Circulation 2000;102(25):3137–47. [DOI] [PubMed] [Google Scholar]
  • 4.Gillum RF, Kwagyan JP, Obisesan TO. Ethnic and geographic variation in stroke mortality trends. Stroke 2011;42(11):3294–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Fang J, Shaw KM, Keenan NL. Prevalence of coronary heart disease—United States, 2006-2010. MMWR Morb Mortal Wkly Rep 2011;60(40):1377–81. [PubMed] [Google Scholar]
  • 6.Xu F, Town M, Balluz LS, et al. Surveillance for certain health behaviors among states and selected local areas—United States, 2010. MMWR Surveill Summ 2013;62(1):1–247. [PubMed] [Google Scholar]
  • 7.Mensah GA, Mokdad AH, Ford ES, Greenlund KJ, Croft JB. State of disparities in cardiovascular health in the United States. Circulation 2005;111(10):1233–41. [DOI] [PubMed] [Google Scholar]
  • 8.CDC. Self-reported hypertension and use of antihypertensive medication among adults—United States, 2005-2009. MMWR Morb Mortal Wkly Rep 2013;62(13):237–44. [PMC free article] [PubMed] [Google Scholar]
  • 9.CDC. State-specific prevalence of cigarette smoking and smokeless tobacco use among adults—United States, 2009. MMWR Morb Mortal Wkly Rep 2010;59(43):1400–6. [PubMed] [Google Scholar]
  • 10.CDC. Prevalence of cholesterol screening and high blood cholesterol among adults—United States, 2005, 2007, and 2009. MMWR Morb Mortal Wkly Rep 2012;61(35):697–702. [PubMed] [Google Scholar]
  • 11.Ramsey F, Ussery-Hall A, Garcia D, et al. Prevalence of selected risk behaviors and chronic diseases—Behavioral Risk Factor Surveillance System (BRFSS), 39 steps communities, United States, 2005. MMWR Surveill Summ 2008;57(11):1–20. [PubMed] [Google Scholar]
  • 12.CDC. Prevalence of overweight and obesity among adults with diagnosed diabetes—United States, 1988-1994 and 1999-2002. MMWR Morb Mortal Wkly Rep 2004;53(45):1066–8. [PubMed] [Google Scholar]
  • 13.CDC. Prevalence of regular physical activity among adults—United States, 2001 and 2005. MMWR Morb Mortal Wkly Rep 2007;56(46):1209–12. [PubMed] [Google Scholar]
  • 14.CDC. Prevalence of fruit and vegetable consumption and physical activity by race/ethnicity—United States, 2005. MMWR Morb Mortal Wkly Rep 2007;56(13):301–4. [PubMed] [Google Scholar]
  • 15.D'Agostino RB Sr, Vasan RS, Pencina MJ, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation 2008;117(6):743–53. [DOI] [PubMed] [Google Scholar]
  • 16.Ford ES. Trends in predicted 10-year risk of coronary heart disease and cardiovascular disease among U.S. adults from 1999 to 2010. J Am Coll Cardiol 2013;61(22):2249–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Lopez-Jimenez F, Batsis JA, Roger VL, Brekke L, Ting HH, Somers VK. Trends in 10-year predicted risk of cardiovascular disease in the United States, 1976 to 2004. Circ Cardiovasc Qual Outcomes 2009;2(5):443–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Marma AK, Berry JD, Ning H, Persell SD, Lloyd-Jones DM. Distribution of 10-year and lifetime predicted risks for cardiovascular disease in US adults: findings from the National Health and Nutrition Examination Survey 2003 to 2006. Circ Cardiovasc Qual Outcomes 2010;3(1):8–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Mensah GA, Dunbar SB. A framework for addressing disparities in cardiovascular health. J Cardiovasc Nurs 2006;21(6):451–6. [DOI] [PubMed] [Google Scholar]
  • 20.Frieden TR, Berwick DM. The "Million Hearts" initiative—preventing heart attacks and strokes. N Engl J Med 2011;365(13):e27. [DOI] [PubMed] [Google Scholar]
  • 21.USDHHS. Healthy People 2020. www.healthypeople.gov/2020/.
  • 22.Ezzati M, Oza S, Danaei G, Murray CJ. Trends and cardiovascular mortality effects of state-level blood pressure and uncontrolled hypertension in the United States. Circulation 2008;117(7):905–14. [DOI] [PubMed] [Google Scholar]
  • 23.National Center for Health Statistics, CDC. Plan, and operation of the Third National Health and Nutrition Examination Survey (NHANES III, 1988-94): reference manuals and reports. Weighting and estimation methodology report. Hyattsville MD: DHHS, Public Health Service, CDC, 1998. [Google Scholar]
  • 24.National Center for Health Statistics. National Health and Nutrition Examination Survey. Questionnaires, datasets, and related documentation. www.cdc.gov/nchs/nhanes/nhanes_questionnaires.htm. [Google Scholar]
  • 25.Schwarz G. Estimating the dimention of a model. Ann Stat 1978;6(2):461–4. [Google Scholar]
  • 26.Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Ser B 1995;57(1):289–300. [Google Scholar]
  • 27.Allan GM, Nouri F, Korownyk C, Kolber MR, Vandermeer B, McCormack J. Agreement among cardiovascular disease risk calculators. Circulation 2013;127(19):1948–56. [DOI] [PubMed] [Google Scholar]
  • 28.CDC. Million Hearts: strategies to reduce the prevalence of leading cardiovascular disease risk factors—United States, 2011. MMWR Morb Mortal Wkly Rep 2011;60(36):1248–51. [PubMed] [Google Scholar]
  • 29.Fang J, Yang Q, Hong Y, Loustalot F. Status of cardiovascular health among adult Americans in the 50 states and the District of Columbia, 2009. J Am Heart Assoc 2012;1(6):e005371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Howard G, Prineas R, Moy C, et al. Racial and geographic differences in awareness, treatment, and control of hypertension: the REasons for Geographic And Racial Differences in Stroke study. Stroke 2006;37(5):1171–8. [DOI] [PubMed] [Google Scholar]
  • 31.Olives C, Myerson R, Mokdad AH, Murray CJ, Lim SS. Prevalence, awareness, treatment, and control of hypertension in United States counties, 2001-2009. PLoS One 2013;8(4):e60308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Grundy SM, Benjamin IJ, Burke GL, et al. Diabetes and cardiovascular disease: a statement for healthcare professionals from the American Heart Association. Circulation 1999;100(10):1134–46. [DOI] [PubMed] [Google Scholar]
  • 33.Barker LE, Kirtland KA, Gregg EW, Geiss LS, Thompson TJ. Geographic distribution of diagnosed diabetes in the U.S.: a diabetes belt. Am J Prev Med 2011;40(4):434–9. [DOI] [PubMed] [Google Scholar]
  • 34.CDC. State-specific incidence of diabetes among adults—participating states, 1995-1997 and 2005-2007. MMWR Morb Mortal Wkly Rep 2008;57(43):1169–73. [PubMed] [Google Scholar]
  • 35.CDC. Estimated county-level prevalence of diabetes and obesity—United States, 2007. MMWR Morb Mortal Wkly Rep 2009;58(45):1259–63. [PubMed] [Google Scholar]
  • 36.CDC. Increasing prevalence of diagnosed diabetes—United States and Puerto Rico, 1995-2010. MMWR Morb Mortal Wkly Rep 2012;61(45):918–21. [PubMed] [Google Scholar]
  • 37.Boyle JP, Thompson TJ, Gregg EW, Barker LE, Williamson DF. Projection of the year 2050 burden of diabetes in the US adult population: dynamic modeling of incidence, mortality, and prediabetes prevalence. Popul Health Metr 2010;8:29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Go AS, Mozaffarian D, Roger VL, et al. Heart disease and stroke statistics—2014 update: a report from the American Heart Association. Circulation 2014;129(3):e28–e292. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Garrett BE, Dube SR, Trosclair A, Caraballo RS, Pechacek TF. Cigarette smoking—United States, 1965-2008. MMWR Surveill Summ 2011;60(S):109–13. [PubMed] [Google Scholar]
  • 40.CDC. Vital signs: current cigarette smoking among adults aged 4 or—18 years—United States, 2009. MMWR Morb Mortal Wkly Rep 2010;59(35):1135–40. [PubMed] [Google Scholar]
  • 41.Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA 2003;289(1):76–9. [DOI] [PubMed] [Google Scholar]
  • 42.Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999-2008. JAMA 2010;303(3):235–41. [DOI] [PubMed] [Google Scholar]
  • 43.Flegal KM, Carroll MD, Kit BK, Ogden CL. Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999-2010. JAMA 2012;307(5):491–7. [DOI] [PubMed] [Google Scholar]
  • 44.Siontis GC, Tzoulaki I, Siontis KC, Ioannidis JP. Comparisons of established risk prediction models for cardiovascular disease: systematic review. BMJ 2012;344:e3318. [DOI] [PubMed] [Google Scholar]
  • 45.Yun S, Zhu BP, Black W, Brownson RC. A comparison of national estimates of obesity prevalence from the behavioral risk factor surveillance system and the National Health and Nutrition Examination Survey. Int J Obes (London) 2006;30(1):164–70. [DOI] [PubMed] [Google Scholar]
  • 46.Bowlin SJ, Morrill BD, Nafziger AN, Jenkins PL, Lewis C, Pearson TA. Validity of cardiovascular disease risk factors assessed by telephone survey: the Behavioral Risk Factor Survey. J Clin Epidemiol 1993;46(6):561–71. [DOI] [PubMed] [Google Scholar]
  • 47.Li C, Balluz LS, Ford ES, Okoro CA, Zhao G, Pierannunzi C. A comparison of prevalence estimates for selected health indicators and chronic diseases or conditions from the Behavioral Risk Factor Surveillance System, the National Health Interview Survey, and the National Health and Nutrition Examination Survey, 2007-2008. Prev Med 2012;54(6):381–7. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Material
NIHMS1968223-supplement-Supplemental_Material.pdf (572.6KB, pdf)

RESOURCES