Normal BP is defined as systolic BP less than 120 mm Hg and diastolic BP less than 80 mm Hg; elevated BP, systolic BP from 120 to 129 mm Hg and diastolic BP less than 80 mm Hg; and hypertension, systolic BP at least 130 mm Hg and/or diastolic BP at least 80 mm Hg and/or antihypertensive medication use.
Data are presented as percentage of population-attributable risk (95% CI). The population-attributable risk for heart failure associated with elevated blood pressure calculated was negative and therefore was considered to be 0.
eTable 1. Prevalence and Standard Error of Blood Pressure Categories Among Non-Hispanic Blacks Without a History of Cardiovascular Disease in the National Health and Nutrition Examination Survey 2011-2014
eTable 2. Sex Stratified Incidence Rates, Adjusted Hazard Ratios, and Population Attributable Risk for Cardiovascular Events by BP Category Among the Pooled JHS and REGARDS Sample
eTable 3. Age Stratified Incidence Rates, Adjusted Hazard Ratios, and Population Attributable Risk for Cardiovascular Events by BP Category Among the Pooled JHS and REGARDS Sample
eTable 4. Incidence Rates, Adjusted Hazard Ratios, and Population Attributable Risk for Cardiovascular Disease, Coronary Heart Disease, Heart Failure, and Stroke by Blood Pressure Category in the JHS and REGARDS Study
Customize your JAMA Network experience by selecting one or more topics from the list below.
Clark D, Colantonio LD, Min Y, et al. Population-Attributable Risk for Cardiovascular Disease Associated With Hypertension in Black Adults. JAMA Cardiol. 2019;4(12):1194–1202. doi:10.1001/jamacardio.2019.3773
What proportion of cardiovascular disease among black adults is associated with hypertension?
In this cohort study of 12 497 black participants in the United States, the population-attributable risk of cardiovascular disease associated with hypertension was 32.5%.
These findings suggest that interventions to maintain normal blood pressure throughout the life course may be associated with reduced incidence of cardiovascular disease among black adults.
The prevalence of hypertension and the risk for hypertension-related cardiovascular disease (CVD) are high among black adults. The population-attributable risk (PAR) accounts for both prevalence and excess risk of disease associated with a risk factor.
To examine the PAR for CVD associated with hypertension among black adults.
Design, Setting, and Participants
This prospective cohort study used data on 12 497 black participants older than 21 years without CVD at baseline who were enrolled in the Jackson Heart Study (JHS) from September 26, 2000, through March 31, 2004, and cardiovascular events were adjudicated through December 31, 2015. The Reasons for Geographic and Racial Differences in Stroke (REGARDS) study participants were enrolled from July 1, 2003, through September 12, 2007, and cardiovascular events were adjudicated through March 31, 2016. Data analysis was performed from March 26, 2018, through July 10, 2019.
Normal blood pressure and hypertension were defined using the 2017 American College of Cardiology/American Heart Association blood pressure guideline thresholds.
Main Outcomes and Measures
The PAR for CVD associated with hypertension, calculated using multivariable-adjusted hazard ratios (HRs) for CVD, coronary heart disease, heart failure, and stroke associated with hypertension vs normal blood pressure. Prevalence of hypertension among non-Hispanic black US adults 21 years and older without CVD was calculated using data from the National Health and Nutrition Examination Survey, 2011-2014.
Of 12 497 participants, 1935 had normal blood pressure (638 [33.0%] male; mean [SD] age, 53.5 [12.4] years), 929 had elevated blood pressure (382 [41.1%] male; mean [SD] age, 58.6 [11.8] years), and 9633 had hypertension (3492 [36.3%] male; mean [SD] age, 62.0 [10.3] years). For a maximum 14.3 years of follow-up, 1235 JHS and REGARDS study participants (9.9%) experienced a CVD event. The multivariable-adjusted HR associated with hypertension was 1.91 (95% CI, 1.48-2.46) for CVD, 2.41 (95% CI, 1.59-3.66) for coronary heart disease, 1.52 (95% CI, 1.01-2.30) for heart failure, and 2.20 (95% CI, 1.44-3.36) for stroke. The prevalence of hypertension was 53.2% among non-Hispanic black individuals. The PAR associated with hypertension was 32.5% (95% CI, 20.5%-43.6%) for CVD, 42.7% (95% CI, 24.0%-58.4%) for coronary heart disease, 21.6% (95% CI, 0.6%-40.8%) for heart failure, and 38.9% (95% CI, 19.4%-55.6%) for stroke. The PAR was higher among those younger than 60 years (54.6% [95% CI, 37.2%-68.7%]) compared with those 60 years or older (32.0% [95% CI, 11.9%-48.1%]). No differences were present in subgroup analyses.
Conclusions and Relevance
These findings suggest that a substantial proportion of CVD cases among black individuals are associated with hypertension. Interventions to maintain normal blood pressure throughout the life course may reduce the incidence of CVD in this population.
Hypertension is the leading contributor to cardiovascular disease (CVD) in the United States and worldwide; the prevalence of hypertension is higher among black adults compared with other racial/ethnic groups.1 According to an analysis of the 2011-2014 National Health and Nutrition Examination Survey (NHANES),2 54.9% of non-Hispanic black adults in the United States have hypertension as defined by the 2017 American College of Cardiology/American Heart Association (ACC/AHA) blood pressure (BP) guideline (ie, systolic BP [SBP] ≥130 mm Hg, diastolic BP [DBP] ≥80 mm Hg, or antihypertensive medication use). For comparison, 47.3% of non-Hispanic white, 36.7% of non-Hispanic Asian, and 34.4% of Hispanic adults in the United States had hypertension.2
The population-attributable risk (PAR) represents the proportion of cases of a disease in a population that can be attributed to a risk factor. It accounts for both the prevalence of a risk factor and the excess risk of disease associated with the risk factor. The PAR can be used to inform policy and prioritize public health interventions by identifying modifiable risk factors that can have the greatest association with reducing disease risk. Given the high prevalence of hypertension and increased risk for CVD associated with hypertension among black adults, the PAR for CVD associated with hypertension may be substantial. The goal of the current study was to determine the PAR for CVD associated with hypertension among black adults.
We used data from 3 sources for this cohort study. The Jackson Heart Study (JHS)3 and the Reasons for Geographic and Racial Differences in Stroke (REGARDS) study4 were used to estimate the incidence rates and hazard ratios (HR) for CVD associated with hypertension. The NHANES 2011-2014 was used to estimate the prevalence of hypertension among non-Hispanic black adults in the United States. The JHS protocol was approved by the institutional review board of the University of Mississippi Medical Center in Jackson. The REGARDS study protocols were approved by the institutional review boards governing research in human subjects at the participating centers (University of Alabama at Birmingham, Birmingham; University of Vermont, Burlington; and Wake Forest University, Winston-Salem, North Carolina) and by an external observational study monitoring board appointed by the funding agency (National Institute of Neurological Disorders and Stroke). The NHANES protocol was approved by the National Center for Health Statistics research ethics review board, Hyattsville, Maryland. All participants provided written informed consent. Data analysis was performed from March 26, 2018, to July 10, 2019.
The JHS is a prospective community-based observational study designed to investigate risk factors for CVD among black adults. Details of the study design and recruitment have been previously published.3,5,6 In brief, the JHS recruited adults from 4 groups of residents of the Jackson, Mississippi, metropolitan area: (1) participants enrolled in the Atherosclerotic Risk in Communities study,7 (2) randomly selected community-dwelling residents, (3) family members of enrolled participants, and (4) volunteers. Overall, 5306 black adults 21 years or older were enrolled in the JHS between 2000 and 2004. The REGARDS study was designed to investigate reasons underlying the higher rate of stroke mortality among black adults compared with white adults and among residents of the southeastern United States compared with other US regions.4 A total of 30 239 black and white adults 45 years or older from all 48 contiguous US states and the District of Columbia were enrolled between January 1, 2003, and October 31, 2007, in the REGARDS study. The REGARDS study participants who had anomalies in their informed consent (n = 56) and who were white adults (n = 17 669) were excluded from the current analysis. We also excluded JHS and REGARDS study participants with prevalent CVD at baseline (n = 4499), with missing BP readings (n = 44) or covariate data (n = 406), and without follow-up information (n = 374). After these exclusions, data were available from 12 497 black participants (8455 REGARDS participants and 4042 JHS participants) for the present analysis. The NHANES is conducted by the National Center for Health Statistics of the Centers for Disease Control and Prevention with the goal of monitoring the health status of the US general population. The current analysis included 1896 non-Hispanic black participants 21 years or older without a history of CVD from NHANES 2011-2014.
Information on age, sex, education, cigarette smoking, physical activity, history of myocardial infarction and stroke, and self-reported use of antihypertensive and glucose-lowering medication was collected at baseline by self-administered questionnaires for JHS participants and by computer-assisted telephone interviews for REGARDS study participants. Height and weight were measured during each study’s baseline examination and used to calculate body mass index. Blood and urine samples collected at baseline were used to estimate total and high-density lipoprotein cholesterol levels. Diabetes was defined as a fasting serum glucose level of at least 126 mg/dL (to convert to mmol/L, multiply by 0.0555) or self-report of a previous diagnosis with the use of glucose-lowering medication. In addition, JHS participants with a glycated hemoglobin (hemoglobin A1c) proportion of at least 6.5% of total hemoglobin were also considered to have diabetes.3 Hemoglobin A1c was not measured at baseline in the REGARDS study. Electrocardiography was performed for all participants during their baseline visit. History of CVD, which was an exclusion criterion, was defined as self-reported stroke or myocardial infarction or evidence of a myocardial infarction reported on the study electrocardiogram. In addition, REGARDS study participants who self-reported a history of coronary revascularization were considered to have a history of CVD.
In both the JHS and the REGARDS study, SBP and DBP were measured twice after the participant had been seated for 5 minutes. Each study followed a BP measurement protocol with data monitored for quality control. Blood pressure was measured using an aneroid sphygmomanometer (American Diagnostic Corporation) in the REGARDS study and using a Hawksley random-zero sphygmomanometer (Hawksley and Sons Ltd) in the JHS. The random-zero BP measurements in the JHS were calibrated to an oscillometric device using robust regression as described previously.8,9
The primary outcome was incident CVD, which was composed of fatal and nonfatal coronary heart disease (CHD), fatal and nonfatal stroke, and heart failure. Methods for identifying CVD events in the REGARDS study and JHS have been detailed previously.10,11 In brief, study participants or their proxies were contacted to identify hospitalizations and possible CVD events at 6-month intervals in the REGARDS study and annually in the JHS. When potential events were identified, medical records were retrieved and events were adjudicated by trained physicians. Hospital discharge lists with specific diagnosis criteria were obtained from the Jackson tricounty area hospitals. Deaths were detected by reports from next of kin, vital records, or the National Death Index. Cause of death was adjudicated using information from death certificates, medical records, and autopsy reports. Participants were followed up from their baseline examination through an outcome event, the last contact by the study staff, or the end of study follow-up, whichever occurred first. In the JHS, follow-up for heart failure began on January 1, 2005. Follow-up for the current analysis was available through December 31, 2014, for both REGARDS study and JHS participants.
The NHANES data were collected through the administration of standardized questionnaires and a medical evaluation at a mobile examination clinic. The use of antihypertensive medication was self-reported. Blood pressure was measured 3 times by a trained physician using a mercury sphygmomanometer following a standardized protocol. The mean of the 3 BP measurements was used in the analyses.12
Blood pressure was categorized according to the 2017 ACC/AHA high BP guideline.13 Among participants not taking antihypertensive medication, normal BP was defined as an SBP less than 120 mm Hg and a DBP less than 80 mm Hg and elevated BP was defined as an SBP from 120 to 129 mm Hg and a DBP less than 80 mm Hg. Hypertension was defined as an SBP of at least 130 mm Hg, a DBP of at least 80 mm Hg, or self-reported antihypertensive medication use.
Participants from the JHS and REGARDS study were pooled together for all analyses. Summary statistics (mean [SD] or percentage) were calculated for JHS and REGARDS study participant characteristics by BP category: normal BP, elevated BP, and hypertension. Incidence rates were calculated for CVD, CHD, heart failure, and stroke by BP category in the pooled JHS and REGARDS study sample. The approach described by Fine and Gray14 was used to account for death as a competing risk to calculate the cumulative incidence and HRs for CVD, CHD, heart failure, and stroke among participants with elevated BP and hypertension vs those with normal BP. The HRs were adjusted for age, sex, diabetes, body mass index, total cholesterol level, high-density lipoprotein cholesterol level, cholesterol-lowering medication use, current smoking, physical activity, and educational level. The prevalence of normal BP, elevated BP, and hypertension among non-Hispanic black adults in the United States was calculated using NHANES 2011-2014 data. These calculations incorporated the complex sampling design used to select NHANES participants, and weights were applied to obtain nationally representative prevalence estimates. The PARs for CVD, CHD, heart failure, and stroke were calculated as pe × (HR − 1)/(1 + pe × [HR − 1]); where pe is the prevalence of the exposure (ie, elevated BP or hypertension) in NHANES 2011-2014 and HR represents the multivariable-adjusted association between elevated vs normal BP or hypertension vs normal BP in the pooled JHS and REGARDS data. The above analyses were repeated in subgroups defined by sex, age (<60 years or ≥60 years), and cohort (JHS or the REGARDS study). The subgroup specific prevalence of hypertension was used in subgroup analyses. We tested whether the multivariable-adjusted HRs between BP categories and CVD risk differed by sex, age, and cohort by comparing the log-likelihood from regression models that included the full population with and without multiplicative interaction terms (eg, sex × elevated BP and sex × hypertension). The 95% CIs for the PARs were calculated using a 2000-iteration bootstrap with bias correction.
The incidence rates and HRs were calculated for CVD, CHD, heart failure, and stroke associated with an SBP less than 120 mm Hg and a DBP less than 80 mm Hg, an SBP of 120 to 129 mm Hg and a DBP less than 80 mm Hg, and an SBP of at least 130 mm Hg or a DBP of at least 80 mm Hg further stratified by antihypertensive medication use. For this analysis, JHS and REGARDS study participants not receiving antihypertensive medication who had an SBP less than 120 mm Hg and a DBP less than 80 mm Hg served as the reference group. The HRs were adjusted for age, sex, diabetes, body mass index, total cholesterol level, high-density lipoprotein cholesterol level, cholesterol-lowering medication use, current smoking, physical activity, and educational level. The PARs associated with BP categories for participants not receiving and receiving antihypertensive medication were calculated separately, with 95% CIs estimated using 2000-iteration bias-corrected bootstraps. All statistical analyses were conducted using SAS, version 9.4 (SAS Institute) and plotted using R, version 3.5.0 (R Core Team).
Of 12 497 JHS and REGARDS study participants, 1935 had normal BP (638 [33.0%] male; mean [SD] age, 53.5 [12.4] years), 929 had elevated BP (382 [41.1%] male; mean [SD] age, 58.6 [11.8] years), and 9633 had hypertension (3492 [36.3%] male; mean [SD] age, 62.0 [10.3] years). Those with elevated BP and hypertension were older, more likely to be taking cholesterol-lowering medication (normal BP: 159 [8.5%]; elevated BP: 92 [9.9%]; and hypertension: 2066 [26.0%]), and more likely to have diabetes (normal BP: 185 [9.9%]; elevated BP: 137 [14.7%]; and hypertension: 2503 [26.0%]) compared with those with normal BP (Table 1). Participants with elevated BP and hypertension were less likely to have graduated high school (normal BP: 1735 [89.9%]; elevated BP: 784 [84.5%]; and hypertension: 7794 [81.0%]). The mean body mass index was higher among participants with elevated BP (29.6 [6.1%]) and hypertension (31.4 [6.7%]) compared with their counterparts with normal BP (28.8 [6.5%]).
For a maximum of 14.3 years of follow-up, 1235 (9.9%) of 12 497 JHS and REGARDS study participants experienced a CVD event. Hypertension was associated with an increased risk for CVD, CHD, heart failure, and stroke (Figure 1 and Table 2). According to the NHANES 2011-2014, the prevalence (SE) of normal BP was 34.8% (1.6%); elevated BP, 12.0% (0.6%); and hypertension, 53.2% (1.3%) among non-Hispanic black adults without a history of CVD (eTable 1 in the Supplement). The PAR associated with hypertension was 32.5% (95% CI, 20.5%-43.6%) for CVD, 42.7% (95% CI, 24.0%-58.4%) for CHD, 21.6% (95% CI, 0.6%-40.8%) for heart failure, and 38.9% (95% CI, 19.4%-55.6%) for stroke (Figure 2). The PAR for CVD associated with hypertension was 31.1% (95% CI, 15.0%-45.8%) for women and 33.9% (95% CI, 15.4%-45.8%) for men (eTable 2 in the Supplement). The PAR for CVD associated with hypertension was higher for those younger than 60 years (54.6% [95% CI, 37.2%-68.7%]) compared with those 60 years or older (32.0% [95% CI, 11.9%-48.1%]) (eTable 3 in the Supplement). The PAR for CVD associated with hypertension was 31.8% (95% CI, 12.2%-49.2%) for JHS participants and 32.1% (95% CI, 16.5%-46.2%) for REGARDS study participants (eTable 4 in the Supplement).
Compared with participants not receiving antihypertensive medication who had an SBP less than 120 mm Hg and a DBP less than 80 mm Hg, the incidence rates and adjusted HRs were higher among participants not receiving antihypertensive medication who had an SBP of at least 130 mm Hg or a DBP of at least 80 mm Hg and those receiving antihypertensive medication at all BP levels (Table 3). The highest PARs were for the groups of participants receiving and not receiving antihypertensive medication who had an SBP of at least 130 mm Hg or a DBP of at least 80 mm Hg (Table 3).
In the present analysis, using definitions from the 2017 ACC/AHA BP guideline, hypertension was associated with a 2-times higher risk for incident CVD compared with normal BP. Also, 53.2% of non-Hispanic black adults without a history of CVD in the United States had hypertension. Taken together, the PAR for hypertension associated with incident CVD was 32.5% (95% CI, 20.5%-43.6%). In addition, hypertension was associated with a substantial percentage of CHD, heart failure, and stroke events.
A major finding of this study was the association of hypertension with incident CVD among black adults. Using the 2017 ACC/AHA guideline hypertension definition, more than 50% of black adults in the United States are estimated to have hypertension.12 The JHS and REGARDS study represent 2 large prospective cohorts assessing BP and CVD among the black population. The PAR takes into consideration both the prevalence of a risk factor and its association with an outcome, highlighting that the high prevalence of hypertension may have a substantial association with CVD among black adults in the United States. The present study adds to previous work15,16 showing that hypertension was associated with a larger proportion of CVD compared with other traditional risk factors, including diabetes, hypercholesterolemia, obesity, and smoking. In the present study, the age-specific risk of CVD associated with hypertension was statistically significantly higher among those younger than 60 years vs those 60 years or older. These findings suggest the need for diagnosis and management of hypertension among younger adults, a population that may be undertreated and thought to have low CVD risk.17 However, a large PAR (>25%) was present across all subgroups investigated, emphasizing the importance of wide-reaching public health interventions aimed at the prevention of hypertension among black adults.
An analysis of the Atherosclerotic Risk in Communities study15 showed that the PAR for CVD associated with hypertension was higher for black adults (36%) compared with white adults (21%). Furthermore, among black adults, hypertension had the highest PAR compared with other traditional risk factors, including diabetes (13%), hypercholesterolemia (10%), obesity (5%), and smoking (12%).15 In a pooled analysis18 including more than 22 000 individuals from the Atherosclerotic Risk in Communities study, the Cardiovascular Health Study, the Framingham Offspring study, and the Multi-Ethnic Study of Atherosclerosis, the PAR for CHD associated with systolic BP of at least 130 mm Hg was 28% compared with 17% for non–high-density lipoprotein cholesterol level of at least 130 mg/dL (to convert to mmol/L, multiply by 0.0259). These findings were not stratified by race/ethnicity, and black participants represented only 18% of the overall cohort. The high PAR for hypertension in the present study is consistent with previous research. The present study expands on the previous findings by including a large sample of black participants, applying BP thresholds for hypertension from the 2017 ACC/AHA guideline, and weighting the results to produce hypertension PAR estimates for the US population.
Hypertension was independently associated with incident CHD, heart failure, and stroke in the present study. The most substantial finding, a PAR of 69% for stroke associated with hypertension, was present among those younger than 60 years. Previous work from the REGARDS study showed racial differences in the association of systolic BP with stroke risk. Specifically, for each 10-mm Hg higher systolic BP, black adults had a 3-times higher risk of stroke (24% increase in risk) compared with white adults (8% increase in risk), even after adjusting for traditional CVD risk factors.19 The results of the present study further support an association between hypertension and stroke among black adults. The incidence rates for CHD, heart failure, and stroke were each 3 to 4 times higher among participants with hypertension compared with those who had normal BP measurements. However, the HR for heart failure was lower compared with the HRs for CHD and stroke after multivariable adjustment, resulting in a smaller PAR. Although the factors included in the multivariable-adjusted model were associated with a substantial portion of the excess risk for heart failure, hypertension was still associated with more than 1 in 5 heart failure events.
Among adults receiving antihypertensive medication, the PAR was smaller among participants with lower BP levels, emphasizing the importance of effective hypertension management. Furthermore, within each BP level studied, the PAR was higher among participants receiving vs not receiving antihypertensive medication. These findings highlight the excess CVD risk among adults with hypertension that is not entirely mitigated by pharmacologic therapy.20,21 To achieve maximum population health benefit, the present results suggest that effective interventions are needed for primordial prevention of elevated BP and hypertension. Various nonpharmacologic interventions, including regular physical activity, weight control, adopting a heart healthy diet such the Dietary Approaches to Stop Hypertension eating plan, and minimizing alcohol intake, have been shown to help prevent the development of hypertension.22,23
In addition to individual measures to reduce BP, the present study supports broader approaches to reduce BP at the population level.24-26 Average sodium intake in the United States is well in excess of national dietary guideline recommendations, and more than 70% of sodium is processed into food before it reaches the table.27 Dietary sodium intake is associated with elevated BP readings, and even modest reductions in the food supply could substantially reduce the prevalence of hypertension and incidence of CVD.28 Experience from the United Kingdom shows that coordinated efforts by public health advocates and food manufacturers to reduce sodium consumption are associated with reduced CVD mortality. Specifically, the implementation of a nationwide salt-reduction program including active salt intake surveillance, consumer awareness campaigns, and collaboration with industry to reformulate foods was associated with a 40% decline in CVD mortality for 8 years.24 The present findings support further investigation and evaluation of policies to reduce BP at a population level.29,30
The present study has some strengths. The JHS and REGARDS study represent 2 population-based cohorts with broad representation of black men and women. Data collection, including BP measurements, used standardized protocols. Participants were followed up prospectively for up to 14.3 years, and CVD events were adjudicated following published recommendations by trained personnel.
This study has limitations. For risk factors with a high prevalence, such as hypertension, PAR estimates may be less precise and therefore should be interpreted with caution.31 We relied on BP from a single visit, and current guidelines recommend that the classification of BP be based on 2 or more BP measurements at 2 or more visits.22 The HRs for hypertension associated with CVD among JHS and REGARDS participants may not be the same for the entire population of black adults in the United States. We used hypertension prevalence estimates from NHANES 2011-2014 to maximize generalizability of these findings to the black population in the United States. However, NHANES does not have follow-up for CVD events, and we relied on HRs from the JHS and REGARDS study.
The present study found that 32.5% of CVD events in black adults were associated with hypertension, as defined by the 2017 ACC/AHA BP guideline. Hypertension was associated with a large proportion of CHD, heart failure, and stroke. These data suggest that interventions to maintain normal BP across the life course may reduce the incidence of CVD in this population.
Accepted for Publication: July 30, 2019.
Published Online: October 23, 2019. doi:10.1001/jamacardio.2019.3773
Correction: This article was corrected on April 1, 2020, to fix the equation for population-attributable risk in the Statistical Analysis subsection of the Methods section.
Corresponding Author: Donald Clark III, MD, MPH, Department of Medicine, University of Mississippi Medical Center, 2500 N State St, Jackson, MS 39216 (email@example.com).
Author Contributions: Drs Clark and Muntner had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Clark, Shimbo, Ogedegbe, Howard, Jones, Correa, Muntner.
Acquisition, analysis, or interpretation of data: Clark, Colantonio, Min, Hall, Zhao, Mentz, Howard, Levitan, Correa, Muntner.
Drafting of the manuscript: Clark, Min, Correa.
Critical revision of the manuscript for important intellectual content: Clark, Colantonio, Hall, Zhao, Mentz, Shimbo, Ogedegbe, Howard, Levitan, Jones, Correa, Muntner.
Statistical analysis: Colantonio, Min, Zhao, Levitan.
Obtained funding: Ogedegbe, Howard, Jones, Correa, Muntner.
Administrative, technical, or material support: Clark, Shimbo, Howard, Jones, Correa, Muntner.
Supervision: Clark, Hall, Mentz, Jones, Correa, Muntner.
Conflict of Interest Disclosures: Dr Min reported receiving other support from the National Heart, Lung, and Blood Institute during the conduct of the study. Dr Mentz reported receiving grants and personal fees from Amgen, AstraZeneca, Bayer, Merck, and Novartis during the conduct of the study; receiving research support from Akros, Amgen, AstraZeneca, Bayer, GlaxoSmithKline, Gilead, InnoLife, Luitpold/American Regent, Medtronic, Merck, Novartis, Otsuka, and ResMed; receiving grants from the National Institutes of Health; receiving honoraria from Abbott, Amgen Inc, AstraZeneca, Bayer AG, Boston Scientific Corporation, Janssen, Luitpold Pharmaceuticals Inc, Medtronic, Merck, Novartis, and ResMed; and serving on advisory boards for Amgen Inc, AstraZeneca, Boehringer Ingelheim, Luitpold Pharmaceuticals Inc, Merck, and Novartis. Dr Shimbo reported being a consultant for Abbott Vascular, Edward Lifesciences Corporation, Medtronic, and Tryton Medical Incorporated and conducting event ascertainment for trials (testing coronary and valvular disease interventions) outside the submitted work. Dr Levitan reported receiving grants and personal fees from Amgen and personal fees from Novartis outside the submitted work. Dr Muntner reported receiving grants from the National Institutes of Health and the American Heart Association during the conduct of the study and receiving grants from Amgen Inc outside the submitted work. No other disclosures were reported.
Funding/Support: The Reasons for Geographic and Racial Differences in Stroke (REGARDS) study was supported by award number 5U54GM115428 from the National Institute of General Medical Sciences of the National Institutes of Health; grants R01HL117323, K24 HL125704, R01 HL080477, 1K01HL133468, K24 HL125704, and K24 HL111154 from the National Heart, Lung, and Blood Institute; and grants 15SFRN2390002 and 19CDA34760232 from the American Heart Association. The Jackson Heart Study (JHS) was supported and conducted in collaboration with Jackson State University (HHSN268201800013I), Tougaloo College (HHSN268201800014I), the Mississippi State Department of Health (HHSN26820180015I) and the University of Mississippi Medical Center (HHSN268201800010I, HHSN268201800011I, and HHSN268201800012I) contracts from the National Heart. Lung, and Blood Institute and the National Institute on Minority Health and Health Disparities.
Role of the Funder/Sponsor: Representatives of the funding agency have been involved in the review of the manuscript but not directly involved in the collection, management, analysis or interpretation of the data.
Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Neurological Disorders and Stroke; the National Heart, Lung, and Blood Institute; the National Institutes of Health; or the US Department of Health and Human Services.
Additional Contributions: We thank the investigators, staff, and participants of the JHS and the REGARDS study for their valuable contributions.
Create a personal account or sign in to: