Abstract
Background:
Hypertension awareness, treatment and control programs were initiated in the United States (US) during the 1960s and 70s. While blood pressure (BP) control in the population and subsequent reduced hypertension-related disease risks have improved since the implementation of these interventions, it is unclear if these BP changes can be generalized to diverse and high risk populations. This report describes the four-decade change in BP levels for the population in a high disease risk southeastern region of the US. The objective is to determine the magnitude of the shift in systolic blood pressure (SBP) among blacks and whites from the Southeast between 1960 and 2005 with the assessment of the unique population cohorts.
Methods:
A multi-cohort study design compared BPs from the Charleston Heart Study (CHS) and Evans County Heart Study (EHS) in 1960, and the REasons for Geographic And Racial Differences in Stroke (REGARDS) four decades later. The analyses included participants 45 years and older from CHS (n=1323), ECS (n=1842), and REGARDS (n=6294) with the main outcome of SBP distribution.
Results:
Among whites aged 45-54 years, the median SBP was 18 mmHg (95% CI: 16 – 21 mmHg) lower in 2005 than 1960. This median shift 45 mmHg (95% CI: 37 – 51 mmHg) decline for those aged 75+ years. The shift was also larger for blacks with median declines of 38 mmHg (95% CI: 32 – 40 mmHg) at age 45-54, and 50 mmHg (95% CI: 33 – 60 mmHg) for ages 75+. The 95th percentile of SBP decreased 60 mmHg for whites and 70 mmHg for blacks.
Conclusions:
The results of these current analyses of the unique cohorts in the Southeast confirm the improvements in population SBP levels since 1960. This assessment provides new evidence of improvement in SBP suggesting that strategies and programs implemented to improve hypertension treatment and control have been extraordinarily successful for both blacks and whites residing in a high risk region of the US. Severe BP elevations commonly observed in the 1960s have been nearly eliminated, with the current 75th percentile of BP generally less than the 25th percentile of blood pressure in 1960.
Keywords: Hypertension, Population, Disparities, Blood Pressure
Introduction
The excess mortality associated with elevated BP has been recognized for nearly 100 years beginning with quantified risk reporting from actuaries in the 1920s.1 In the current era of intensive pharmacological treatment of hypertension, there is considerably less awareness of the extreme BPs that were prevalent in the past, such as the values of Franklin Roosevelt progressing from 162/98 mmHg at age 54 in 1938 to 300/190 mmHg in 1945.2 High BP detection and control efforts have shifted from these extreme values where ‘mild benign hypertension’ was defined as BP up to 200/100 mmHg, to the current definition of hypertension as levels above 130/80 mmHg.3,4 In the late 1960’s, a landmark clinical trial raised awareness of the need for diagnosis and treatment of elevated BP.5 In 1969 the Framingham Study provided epidemiological evidence that hypertension is a major contributor to CHD morbidity and mortality and is readily controlled by “hygienic and pharmacologic measures”.6 The benefits of hypertension detection, treatment, and control with subsequent risk reduction prompted the implementation of the High Blood Pressure Education Program in 1972.7
An analysis of “virtual birth cohorts” compiled from multiple National Health and Nutrition Examination Survey (NHANES) examinations spanning 1960 and 1994 indicated a decline in population distribution of SBP over the decades.8 A recent review by Fuchs and Whelton further identified elevated BP as the predominant risk factor for cardiovascular disease (CVD), and proposed that CVD is caused by the rightward shift in the distribution of BP above biologically normal levels.9 The authors concluded that pharmacologic interventions could have shifted the higher percentiles of the BP distribution downward, and that population-based changes could have contributed to a downward shift in the entire distribution, including lower percentiles, of BP. However, it remains unclear if the changes in blood pressure levels during this period are consistent in high-risk geographic areas, and can be generalized to multi-racial, economically disadvantage, and rural populations with high disease risks. The southeastern region of the US is one such area with an excess burden of hypertension-related outcomes including stroke.10,11,12 In 1960, Georgia and South Carolina had high risk populations of low socio-economic status with per capita income in the lower fifth of the nation, rural health characteristics including low access to health care, and high racial diversity with a substantial proportion of blacks.13 The goal of this report is to use data from the unique historical Charleston Heart Study (CHS) and Evans County Heart Study (EC) with the REasons for Geographic And Racial Differences in Stroke (REGARDS) study to compare the BP levels in the early 1960’s (immediately prior to the implementation of population hypertension control programs) with those in the 2000’s.
Methods
Data availability statement.
The data used in these analyses include potentially identifying participant information and therefore are not publicly available due to legal and ethical restrictions. Qualified investigators may request access from the University of Alabama at Birmingham (regardsadmin@uab.edu) to obtain de-identified data.
Study Cohorts
The Black Pooling Project included two population cohorts from the Southeast in 1960’s. The CHS included a random sample of 2181 black and white participants 35 years of age and older recruited in the early 1960’s. The study cohort represented an overall response rate of 84 percent to the population-based sampling plan. Details of the sampling procedure and other procedures have been published previously.14 As a summary, a cross-sectional population based sample was drawn using the 1950 census. The expected yield was 2,500 persons, aged 35 years or more; within this age group, which constituted a 3.825% sampling rate. A stratified sampling (with a shared sampling rate) was employed based on social density: city proper, the urban fringe, the open country, and rural settlements. These areas were then subdivided into units of 12 households each with units were selected by a random sample. Variables measured at baseline included the first BP measurement taken in the seated position (fifth phase), with use of a frequently standardized aneroid manometer with standard cuff; 99% of all readings taken by the same observer, on whom repeated audiometric examinations were made. To harmonize with age strata data available from the REGARDS study (that recruited those aged 45+), only 1,323 participants aged 45+ were included in these analyses.
The EC also recruited black and white participants in 1960-1962. All non-institutionalized residents of Evans County and five districts in adjacent Bulloch County, Georgia, aged 40 years or older and 50% of those 15-39 years of age were invited to participate. Details of the methodology have been reported previously.15 The study population for the EC was selected from the census list and consisted of the entire Evans County population aged 40-74 years, plus the five enumeration districts in Bulloch County population. As indicated, a randomly selected 50 percent sample of the Evans County population aged 15-39 years was also included. The initial BP measurement recorded during the 1960-1962 baseline assessment used mercury sphygmomanometers with standard cuffs, with subjects seated and assessment using the left arm. Participants were not medically treated for high BP at baseline. To harmonize with age data available from the REGARDS study, only 1,842 participants aged 45+ were included in these analyses.
REGARDS is an ongoing population-based national study that recruited 30,239 black and white participants aged 45+ between 2003 and 2007. Participants were recruited from the 48 contiguous US states, with oversampling of blacks and residents of the Stroke Belt (North Carolina, South Carolina, Georgia, Tennessee, Alabama, Mississippi, Louisiana and Arkansas). REGARDS used a stratified random sample based on 12 strata: three regions (buckle of stroke belt, rest of stroke belt, rest of the nation), two races (black and white), and sex (men and women). This report uses only data from within the stroke buckle and is stratified by race; hence, no adjustment is needed for these two factors. The recruitment goal by sex was 50:50 and 55:45 (Female:Male) was achieved. The adjustment for sex is also not needed given the distribution of sex is similar to that in the general population, and given that sex differences are quite small compared to the differences between cohorts. SBP was defined as the mean of 2 seated measures taken after a 5-minute rest. Details of study design and sample selection have been previously reported.16 To harmonize with CHS and EC for age and geography, only 6,294 participants from the southeastern coastal plain region (the region where CHS and EC were conducted) were included in these analyses. The geographic locations of the three cohorts are presented in Figure 1.
Figure 1:
Geographic locations of Evans County, Charleston County and REGARDS cohorts inset in the United States.
The protocol was approved by the relevant health authorities and institutional review boards. Written informed consent was required from all participants.
Statistical Analysis
The focus of statistical analysis was a description of the distribution of SBP as characterized by the 5th, 25th, 50th, 75th and 95th percentiles. The percentiles of SBP and their 95% confidence intervals were calculated using the UNIVARIATE procedure from SAS 13.1 (Cary, NC). The difference between the percentiles was calculated by simple subtraction of the CHS/EC percentiles from REGARDS percentiles, with the 95% confidence bounds for this difference calculated using bootstrap techniques with 5,000 replications. These percentiles were displayed graphically using box-and-whisker plots and also provided in tabular format (along with their 95% confidence bounds, and differences between studies/periods with their 95% confidence bounds).
Results
Table 1 presents the numbers of participants in each era (CHS/EC in 1960 versus REGARDS in 2005), race (black and white) and age stratum (45-54, 55-64, 65-74 and 75+). The analytical data base provides essential numbers for each race-age category for both study eras.
Table 1:
Number of participants in the analyses from the Evans County Study (EC), Charleston Heart Study (CHS), and REasons for Geographic And Racial Differences in Stroke (REGARDS) by age strata and race
| Age Group | White | Black | ||
|---|---|---|---|---|
| EC/CHS | REGARDS | EC/CHS | REGARDS | |
| 45-54 | 950 | 543 | 526 | 423 |
| 55-64 | 606 | 1467 | 360 | 909 |
| 65-74 | 366 | 1339 | 245 | 648 |
| 75+ | 66 | 708 | 41 | 257 |
The percentiles of SBP are shown graphically in Figure 2 and numerically (with 95% confidence bounds, and differences between eras with 95% confidence bounds) for whites and blacks in Table 2. Overall, the percentiles of SBP were strikingly lower in 2005 compared to 1960, with a number of notable patterns:
Figures 2:
Box and whisker plot showing the distribution of SBP by race (white left panel, black right panel), age strata (45-54, 55-64, 65- 74, and 75+ years), and study (C/E: Charleston Heart Study/Evans County Study circa 1960; and R: REGARDS circa 2005). The box represents the 25th, 50th and 75th percentiles of the distribution, the whiskers the 5th and 95th percentiles of the distribution, and the individual data points the observations below the 5th or about the 95th percentile.
Table 2:
Age-specific percentiles (95% confidence limits) of SBP for whites and blacks in the combined Charleston Heart Study (CHS) and Evans County Study (EC) conducted in the early 1960’s, and the REasons for Geographic and Racial Differences in Stroke (REGARDS) study conducted in the early 2000’s; with the difference (95% confidence interval) between the two time periods.
| Race | Age | Parameter | Percentile | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| P5 | P25 | P50 | P75 | P95 | ||||||||
| CHS/EC | REGARDS | CHS/EC | REGARDS | CHS/EC | REGARDS | CHS/EC | REGARDS | CHS/EC | REGARDS | |||
| White | 45-54 | Estimate | 110 (108-112) |
98 (92 – 99) |
122 (122-124) |
108 (106-110) |
136 (134-138) |
118 (116-119) |
152 (150-155) |
125 (123-127) |
180 (176-186) |
141 (139-144) |
| Difference | 12 (9-17) |
14 (12-17) |
18 (16-21) |
27 (25-31) |
39 (33-44) |
|||||||
| 55-64 | Estimate | 114 (111-118) |
100 (99-101) |
133 (130-136) |
113 (111-115) |
150 (146-152) |
121 (121-122) |
166 (164-170) |
131 (130-132) |
205 (196-213) |
148 (146-150) |
|
| Difference | 14 (12-18) |
20 (17-23) |
29 (25-31) |
35 (33-39) |
57 (47-64) |
|||||||
| 65-74 | Estimate | 118 (114-123) |
102 (100-104) |
144 (140-148) |
118 (116-119) |
162 (160-166) |
126 (125-127) |
184 (178-190) |
137 (136-138) |
220 (210-224) |
157 (152-160) |
|
| Difference | 16 (12-22) |
26 (22-31) |
36 (34-40) |
47 (41-53) |
63 (41-68) |
|||||||
| 75+ | Estimate | 125 (104-134) |
104 102-105) |
152 (140-162) |
119 (118-120) |
172 (164-178) |
127 (125-128) |
190 (178-210) |
139 (137-140) |
222 (215-276) |
160 (157-163) |
|
| Difference | 21 (6-34) |
33 (21-43) |
45 (37-51) |
51 (39-62) |
62 (49-75) |
|||||||
| Black | 45-54 | Estimate | 116 (114-120) |
99 (94-101) |
140 (136-142) |
116 (113-118) |
160 (154-162) |
122 (121-123) |
180 (175-184) |
132 (130-134) |
220 (214-230) |
159 (150-171) |
| Difference | 17 (14-23) |
24 (20-28) |
38 (32-40) |
48 (42-53) |
61 (49-75) |
|||||||
| 55-64 | Estimate | 122 (118-124) |
104 (102-107) |
148 (142-152) |
119 (118-120) |
170 (165-174) |
127 (125-128) |
198 (190-204) |
139 (138-140) |
239 (230-243) |
160 (158-163 |
|
| Difference | 18 (13-21) |
29 (23-34) |
43 (38-48) |
59 (51-65) |
79 (69-82) |
|||||||
| 65-74 | Estimate | 126 (120-135) |
104 (101-105) |
158 (150-160) |
119 (118-120) |
180 (173-184) |
129 (128-130) |
204 (200-212) |
140 (139-141) |
242 (236-260) |
165 (161-170) |
|
| Difference | 22 (17-34) |
39 (31-43) |
51 (43-55) |
64 (59-71) |
77 (67-85) |
|||||||
| 75+ | Estimate | 118 (102-128) |
107 (103-110) |
152 (128-166) |
121 3(119-122) |
181 (164-192) |
131 (128-135) |
204 (190-224) |
143 (141-148) |
234 (190-254) |
164 (160-180) |
|
| Difference | 11 (−7-36) |
31 (20-47) |
50 (33-60) |
61 (43-74) |
70 (46-93) |
|||||||
The magnitude of the difference in BP between eras was substantially larger for blacks than for whites. The differences between eras were approximately twice as large for blacks as whites between the ages of 45 to 54. For example, the difference in the 50th percentile was 18 mmHg (95% CI: 16 – 21 mmHg) for whites, but 38 mmHg (95% CI: 32 – 40 mmHg) for blacks. With increasing age, the black-white difference in the magnitude of the decrease became smaller.
The magnitude of the difference in BP between eras was larger at older ages for blacks and whites. For example, for blacks the median difference between eras increased from 18 mmHg (95% CI: 16 – 21 mmHg) for participants aged 45-54, to 29 mmHg (95% CI: 25 – 31mmHg) for those aged 55-64, to 36 mmHg (95% CI: 34 – 40 mmHg) for those aged 65-74, and finally to 45 mmHg (95% CI: 37 – 51mmHg) for those aged 75+. Similarly, for whites, the increase was from 38 mmHg (95% CI: 32 – 40 mmHg), to 43 mmHg (95% CI: 38 – 48mmHg), to 51 mmHg (95% CI: 43 – 55 mmHg), and to 50 mmHg (95% CI: 33 – 60 mmHg).
The magnitude of the difference between eras was substantially larger for the higher percentiles. For example, at age 65-74 for whites there was a 16 mmHg (95% CI: 12 – 22 mmHg) difference at the 5th percentile, a 26 mmHg (95% CI: 22 – 31 mmHg) difference at the 25th percentile, a 36 mmHg (95% CI: 34 – 40mmHg) difference at the 50th percentile, a 47 mmHg (95% CI: 41 – 53 mmHg) difference at the 75th percentile, and a 63 mmHg (95% CI: 41 – 68mmHg) difference at the 95th percentile.
Cumulatively, these changes have contributed to the dramatic shift in SBP levels shown in Figure 3. In 1960, the SBPs were generally greater than 150 mmHg for blacks and above 130 mmHg for whites, with few individuals having SBP levels below 120 mmHg or above 225 mmHg. In contrast, by 2005 the SBP levels for both blacks and whites were generally less than 130 mmHg, with SBP levels under 120 mmHg and above 180 mmHg occurring in only a few individuals.
Figure 3:
Histogram of systolic blood pressure for blacks and whites in 1960 (Charleston Heart Study and Evans County Study) and 2005 (REGARDS Study).
Discussion
In the years following the implementation of high BP detection, treatment, control and prevention programs, there has been an immense decline in the distribution of SBP in the general population. Improvement in hypertension treatment and control and subsequent shifts in BP have been recently reported from national population surveys17, as well as global pooled analyses from the Non Communicable Disease Risk Factor Collaboration Risk Factor Collaboration.18-19 While these reports provide consistent evidence of improvements in population BP levels, it remains unclear if the changes in SBP levels can be generalized to high risk and vulnerable populations. The CHS/EC and REGARDS data and analyses provide a unique resource to assess BP changes in a high risk geographic area with an economically disadvantaged and diverse population. Consistent with the higher cardiovascular and cerebrovascular disease rates, the BPs in CHS/EC were substantially higher in 1960 compared values in the Framingham Study 20,21 and the 1960 NHANES17. In the 1960’s, SBP levels of 160 to 220 mmHg were not unusual, falling between the 75th and 95th percentiles of SBP reported for that period in this high-risk population. After 4 decades, these severely elevated BPs have become much rarer, with even the 95th percentile values being lower than 160 mmHg. With the exception of only one race-age stratum in this analysis, the 75th percentile of SBP in 2005 was below the 25th percentile in 1960. The exception was whites aged 45 to 54 years where the 25th percentile in 1960 (122 mmHg) and the 75th percentile in 2005 (125 mmHg) were nearly identical. Between 1960 and 2000 there were dramatic reductions in both heart disease and stroke mortality, and the declines in SBP were a major contributor. 22
These results confirm previous reports of declines across the entire BP distribution.17-19 Clearly, introduction of pharmacological antihypertensive treatment is a major contributor to the temporal decline in the distribution of BP at the higher percentiles. However, the finding of decreases at the lower percentiles (even the 5th percentile), suggests that population-wide changes in lifestyle and other factors may have played a role. Thus, the current study provides new evidence suggesting both the medical treatment and control of hypertension, as well as lifestyle modifications have made impact in high risk populations.
In the previous report using “virtual cohorts”, there was an estimated change of 1.19 mmHg per decade at the 10th percentile, 2.40 mmHg per decade at the 50th percentile, and 4.62 mmHg per decade at the 90th percentile of the BP range.7 Hence, a 9.6 mmHg decline might have been expected over 40 years at the 50th percentile. The declines using the actual cohorts in the current report are dramatically larger. For example, there was a 36 mmHg (95% CI: 34 – 40 mmHg) decline in the 50th percentile for whites aged 65-74, and a 51 mmHg (95% CI: 43 – 55 mmHg) for blacks in the same age strata. This corresponds to decreases of approximately 10 mmHg per decade. This larger decline could reflect the temporal positioning of the cohorts used in this report, where the CHS/EC cohorts were nearly ideally positioned to collect data immediately before the growth in the recognition of the importance of BP control. Nonetheless, these declines in the median SBP levels of 10 mmHg per decade are remarkable population health achievements.
The larger decreases in SBP observed at higher percentiles of the blood pressure range are associated with a near elimination of critically high BPs requiring immediate medical attention. In the 1960’s, 25% of the black population aged 55 to 64 had SBPs above 198 mmHg, and 5% had SBP levels of 239 mmHg or higher. In the 2005, only 7% of the black population had values above 198 mmHg and 1% had values above the 239 mmHg. Hence, while very high levels of SBP persisted, the proportion of the population affected was much smaller.
Mortality from hypertension-related conditions, such as stroke, has been declining rapidly over the period from the 1960s to the 2000’s.22 This decline in mortality was more rapid in whites than blacks.17 Thus, the current finding that declines in hypertension and SBP levels were strikingly larger in blacks than their white counterparts are new and unpredicted. The black-white differences in the magnitude of the decline in SBP were larger at younger ages, with both blacks and whites having larger but approximately equal declines at older ages. Thus, the finding that racial differences were diminished at older ages where stroke events are more prevalent may underlie the smaller declines in stroke mortality in blacks.
There are several limitations that should be considered when interpreting the results. While the REGARDS 2005 comparison metrics might be considered dated, the analytical design identified substantial improvements in the SBP distributions in the high-risk region in the early phases of population high BP control. Data from current population assessments suggest that the improvements on SBP have been sustained.23 The measurements of BP were not standardized among the different cohorts, but were consistent with the protocols of the era. As BP measurement was relatively new for the CHS/EC participants in 1960, it is reasonable to consider a ‘white-coat’ effect.24 While the cohorts did not include indicators of white coat hypertension, the magnitude of the SBP values are much greater than the 10 mmHg reported with those with the white coat effect.24 Further, the SBP levels in the CHS/EC were considerably greater than other cohorts of the era. However, the high SBP values are consistent with the higher prevalence of hypertension-related outcomes for the geographic region suggesting validity of the measurements. In addition, because of the stronger association of SBP (relative to DBP) with health outcomes, these analyzes focus solely on SBP. The study also includes many strengths, the largest being the use of three large high-quality studies nearly perfectly temporally positioned to assess the change in BP levels from before to after the widespread acceptance of the need for control of BP. An additional strength is inclusion of participants that were all residents of the historically high-risk coastal plain region of the Southeast.
In conclusion, this paper provides new evidence that SBP levels were reduced over a 40-year period in high-risk populations geographically disadvantaged by social determinants of disease. Specifically, the analyses quantify and document the magnitude in the decline in the distribution of SBP in the Southeastern US over the past 40 years following widespread acceptance of the importance of BP control. These reductions were particularly evident at the higher end of the SBP distribution, with decreases of 50 mmHg or more in the 95th percentile of SBP. While the results confirm and reinforce the findings of other studies demonstrating the successful lowering of elevated BP in the US, the SBP reductions described were even larger in this high-risk population of the Southeast and blacks. The results provide an indication of the potential population impact of structured hypertension awareness, treatment, control, and prevention programs for all populations including areas of high risk such as areas with diverse demographics, rural, economically disadvantaged populations, and geographies with excess disease burden.
Clinical Perspective.
What Is New?
Analyses of data from two unique population-based cohorts from the Southeastern coastal plain of the US; the Charleston Heart Study and Evans County Heart Study in 1960, and the REasons for Geographic and Racial Differences in Stroke four decades later identified decreases in blood pressure levels for blacks and whites.
Although there were consistent and similar systolic blood pressure reductions, the changes were greater for blacks compared to whites at different age categories.
Shifts in blood pressure distributions were detected for both whites and blacks with the severe values common in 1960 practically eliminated over the 45-year period.
What Are the Clinical Implications?
The clinical strategies of high blood pressure detection, treatment and control implemented in the later part of the last century are effective in all patients including blacks and whites residing in high risk and socioeconomically disadvantaged geographic areas.
Primordial and primary prevention activities implemented at the clinical setting can have impact on blood pressure levels for blacks and whites in vulnerable areas.
Early detection of hypertension using proper blood pressure measurement and prompt appropriate treatment can be effective in high blood pressure control for all patients regardless of population social determinants.
Acknowledgments
Funding/Support: This research paper is supported by a cooperative agreement U01 NS041588 from Reasons for Geographic And Racial Differences in Stroke (REGARDS) Study from the National Institute of Neurological Disorders and Stroke, National Institutes of Health (NIH), Department of Health and Human Services, and also The Black Pooling Project (R01 HL072377) from the National Heart Lung and Blood Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views and positions of the National Institute of Neurological Disorders and Stroke (NINDS) or the National Institutes of Health. Representatives of the NINDS were involved in the review of the article before submission for publication.
Non-standard Abbreviations and Acronyms
- (BP)
blood pressure
- (CVD)
cardiovascular disease
- (CHS)
Charleston Heart Study
- (EHS)
Evans County Heart Study
- (NHANES)
National Health and Nutrition Examination Survey
- (REGARDS)
REasons for Geographic And Racial Differences in Stroke
- (SBP)
systolic blood pressure
- (US)
United States
Footnotes
Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr. Cushman reported in kind research support to her institution from SphingoTec GmbH. Dr. Safford reported funding from Amgen, Inc. Dr Oparil reported receiving grant support from Actelion, Novartis, and Bayer Healthcare Pharmaceuticals; receiving personal fees from Boehringer-Ingelheim, Lilly, George Clinical Pty Ltd/Actelion Clinical Research Inc, Lundbeck, Novo Nordisk, ROXMedical, and 98point6; and serving as editor in chief for Current Hypertension Reports. No other disclosures were reported.
Supplemental Material: None
References
- 1.Morton AP, Lew EA, November WJ, Webster AC, Hutchinson JJ. Build and blood pressure study. Transactions of Society of Actuaries. 1959; 11:987–997. [Google Scholar]
- 2.Moser M Historical perspectives on the management of hypertension. J Clin Hypertens (Greenwich). 2006;8: ( Suppl 2):15–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Whelton PK, Carey RM, Aronow WS, Casey DE, Collins KJ, Dennison Himmelfarb C, DePalma SM, Giddings S, Jamerson KA, Jones DWJ, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardio. 2018;71: e127–e248. [DOI] [PubMed] [Google Scholar]
- 4.Friedberg CK. Diseases of the heart. Philadelphia, PA: W. B. Saunders, Co; 1949. [Google Scholar]
- 5.Effects of treatment on morbidity in hypertension: results in patients with diastolic blood pressures averaging 115 Through 129 mm Hg. JAMA. 1967; 202:1028–1034. [PubMed] [Google Scholar]
- 6.Kannel WB, Schwartz MJ, McNamara PM. Blood pressure and risk of coronary heart disease: the Framingham study. Dis Chest. 1969; 56:43–52. [DOI] [PubMed] [Google Scholar]
- 7.Roccella EJ. Meeting the 1990 hypertension objectives for the nation--a progress report. Public Health Rep. 1985;100:652–656. [PMC free article] [PubMed] [Google Scholar]
- 8.Goff DC, Howard G, Russell GB, Labarthe DR. Birth cohort evidence of population influences on blood pressure in the United States, 1887-1994. Ann Epidemiol. 2001;11:271–279. [DOI] [PubMed] [Google Scholar]
- 9.Fuchs FD, Whelton PK. High blood pressure and cardiovascular disease. Hypertension. 2020;75:285–292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Borhani NO. Changes and geographic distribution of mortality from cerebrovascular disease. Am J Public Health. 1965; 55:673–681. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lanska DJ, Kuller LH. The geography of stroke mortality in the United States and the concept of a Stroke Belt. Stroke. 1995;26:1145–1149. [DOI] [PubMed] [Google Scholar]
- 12.Perry HM, Roccella EJ. Conference Report on Stroke Mortality in the Southeastern United States. Hypertension. 1998;31:1206–1215. [DOI] [PubMed] [Google Scholar]
- 13.Bureau of the Census. 1960 Census of Population, Supplement Reports per Capita and Median Family Income July 30, 1965. PC(81)–48. US Printing Office, Washington DC, 1965. [Google Scholar]
- 14.Boyle E Jr,. Biological pattern in hypertension by race, sex, body weight, and skin color. JAMA. 1970;213:1637–1643. [PubMed] [Google Scholar]
- 15.McDonough JR, Hames CG, Stulb SC, Garrison GE. Cardiovascular Disease Field Study in Evans County, Ga. Public Health Rep. 1963;78:1051–1059. [PMC free article] [PubMed] [Google Scholar]
- 16.Howard VJ, Cushman M, Pulley L, Gomez CR, Go RC, Prineas RJ, Graham A, Moy CS, Howard G. The reasons for geographic and racial differences in stroke study: objectives and design. Neuroepidemiology. 2005;25:135–143. [DOI] [PubMed] [Google Scholar]
- 17.Lackland DT, Beilin LJ, Campbell NRC, Jaffe MG, Orias M, Ram CV, Weber MA, Zhang XH. Global implications of blood pressure thresholds and targets. Hypertension. 2018;71:985–987. [DOI] [PubMed] [Google Scholar]
- 18.NCD Risk Factor Collaboration (NCD-RisC). Worldwide trends in blood pressure from 1975 to 2015: a pooled analysis of 1479 population-based measurement studies with 19·1 million participants. Lancet. 2017;389:37–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19,NCD Risk Factor Collaboration (NCD-RisC) . Contributions of mean and shape of blood pressure distribution to worldwide trends and variations in raised blood pressure: a pooled analysis of 1018 population-based measurement studies with 88.6 million participants. Int J Epidemiol. 2018;47:872–883i. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pencina MJ, D’Agostino RB, Larson MG, Massaro JM, Vasan RS. Predicting the thirty-year risk of cardiovascular disease: The Framingham Heart Study. Circulation. 2009; 119: 3078–3084. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kannel WB, Schwartz MJ, McNamara PM. Blood pressure and risk of coronary heart disease: The Framingham Study. Dis Chest. 1969; 56: 43–52. [DOI] [PubMed] [Google Scholar]
- 22.Lackland DT, Roccella EJ, Deutsch AF, Fornage M, George MG, Howard G, Kissela BM, Kittner SJ, Lichtman JH, Lisabeth LD, et al. Factors influencing the decline in stroke mortality: a statement from the American Heart Association/American Stroke Association. Stroke. 2014; 45:315–353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Virani SS, Alonso A, Benjamin EJ, Bittencourt MS, Callaway CW, Carson AP, Chamberlain HM, Chang HR, Cheng S, Delling FN, et al. Heart Disease and Stroke Statistics-2020 Update: A Report from the American Heart Association. Circulation. 2020;141: e139–e596. [DOI] [PubMed] [Google Scholar]
- 24.Franklin SF, Thijs L, Hansen TW, O’Brien E, Staessen JA. White-coat hypertension: new insights from recent studies. Hypertension. 2013; 62:982–998. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data used in these analyses include potentially identifying participant information and therefore are not publicly available due to legal and ethical restrictions. Qualified investigators may request access from the University of Alabama at Birmingham (regardsadmin@uab.edu) to obtain de-identified data.
Study Cohorts
The Black Pooling Project included two population cohorts from the Southeast in 1960’s. The CHS included a random sample of 2181 black and white participants 35 years of age and older recruited in the early 1960’s. The study cohort represented an overall response rate of 84 percent to the population-based sampling plan. Details of the sampling procedure and other procedures have been published previously.14 As a summary, a cross-sectional population based sample was drawn using the 1950 census. The expected yield was 2,500 persons, aged 35 years or more; within this age group, which constituted a 3.825% sampling rate. A stratified sampling (with a shared sampling rate) was employed based on social density: city proper, the urban fringe, the open country, and rural settlements. These areas were then subdivided into units of 12 households each with units were selected by a random sample. Variables measured at baseline included the first BP measurement taken in the seated position (fifth phase), with use of a frequently standardized aneroid manometer with standard cuff; 99% of all readings taken by the same observer, on whom repeated audiometric examinations were made. To harmonize with age strata data available from the REGARDS study (that recruited those aged 45+), only 1,323 participants aged 45+ were included in these analyses.
The EC also recruited black and white participants in 1960-1962. All non-institutionalized residents of Evans County and five districts in adjacent Bulloch County, Georgia, aged 40 years or older and 50% of those 15-39 years of age were invited to participate. Details of the methodology have been reported previously.15 The study population for the EC was selected from the census list and consisted of the entire Evans County population aged 40-74 years, plus the five enumeration districts in Bulloch County population. As indicated, a randomly selected 50 percent sample of the Evans County population aged 15-39 years was also included. The initial BP measurement recorded during the 1960-1962 baseline assessment used mercury sphygmomanometers with standard cuffs, with subjects seated and assessment using the left arm. Participants were not medically treated for high BP at baseline. To harmonize with age data available from the REGARDS study, only 1,842 participants aged 45+ were included in these analyses.
REGARDS is an ongoing population-based national study that recruited 30,239 black and white participants aged 45+ between 2003 and 2007. Participants were recruited from the 48 contiguous US states, with oversampling of blacks and residents of the Stroke Belt (North Carolina, South Carolina, Georgia, Tennessee, Alabama, Mississippi, Louisiana and Arkansas). REGARDS used a stratified random sample based on 12 strata: three regions (buckle of stroke belt, rest of stroke belt, rest of the nation), two races (black and white), and sex (men and women). This report uses only data from within the stroke buckle and is stratified by race; hence, no adjustment is needed for these two factors. The recruitment goal by sex was 50:50 and 55:45 (Female:Male) was achieved. The adjustment for sex is also not needed given the distribution of sex is similar to that in the general population, and given that sex differences are quite small compared to the differences between cohorts. SBP was defined as the mean of 2 seated measures taken after a 5-minute rest. Details of study design and sample selection have been previously reported.16 To harmonize with CHS and EC for age and geography, only 6,294 participants from the southeastern coastal plain region (the region where CHS and EC were conducted) were included in these analyses. The geographic locations of the three cohorts are presented in Figure 1.
Figure 1:
Geographic locations of Evans County, Charleston County and REGARDS cohorts inset in the United States.
The protocol was approved by the relevant health authorities and institutional review boards. Written informed consent was required from all participants.
Statistical Analysis
The focus of statistical analysis was a description of the distribution of SBP as characterized by the 5th, 25th, 50th, 75th and 95th percentiles. The percentiles of SBP and their 95% confidence intervals were calculated using the UNIVARIATE procedure from SAS 13.1 (Cary, NC). The difference between the percentiles was calculated by simple subtraction of the CHS/EC percentiles from REGARDS percentiles, with the 95% confidence bounds for this difference calculated using bootstrap techniques with 5,000 replications. These percentiles were displayed graphically using box-and-whisker plots and also provided in tabular format (along with their 95% confidence bounds, and differences between studies/periods with their 95% confidence bounds).