Menu
Jones E, Norris K, Jenkins B, Addison C, Payton M. The relationship between live births and CVD among African American women in Jackson heart study. HPHR. 2023;76. https://doi.org/10.54111/0001/XXX3
Previous studies have reported mixed results on the relationship between live births and cardiovascular disease among African American females, with few controlling for an inclusive list of major cardiovascular disease (CVD) risk factors and/or including a large population of African American female participants
To help clarify this issue we examined the relationship between live births and cardiovascular events, including major CVD risk factors in a well-studied, exclusively African American cohort. Data was from 3367 African American women, 21 and older, in the Jackson Heart Study. Myocardial Infraction (MI), Self-reported history of cardiac procedures, Coronary Heart Disease Status/History, Self-reported history of Carotid Angioplasty, Cardiovascular Disease history, and Heart Failure History were obtained from 2000 to 2004 interviews. Live births were self-reported. Chi-square test indicated a significant association between live births and the presence of CVD (p =.000). Logistic regression estimated the relationship between live births and cardiovascular events, adjusting for major cardiovascular risk factors.
After adjusting for hypertension status, current smoking status, cholesterol status, diabetes status, family history of heart diseases (mother), family history of heart disease (father), age, frequency of alcohol use in the past 12 months, and the average number of drinks per week in the past 12 months, there was no longer a significant relationship between live births and the risk of CVD (AOR 1.012, 95% CI, 1.009-1.015, p=.895).
The unadjusted significant relationship between live births and CVD events disappeared after adjusting for major CVD risk factors, which suggests that an association between live births and CVD is indirect and based on the distribution of CVD risk factors.
Therefore, we propose more research on the use of live births as a CVD indicator in African American/Black women to ensure that African American/Black women obtain optimum cardiovascular health and make informed reproductive decisions.
Cardiovascular Disease (CVD) is a significant problem within the United States.1 Unfortunately, CVD disproportionately impacts the African American community, especially African American women.2 According to the American Heart Association, CVD kills nearly 50,000 African American women each year.3 Additionally, 49% of African American women ages 20 and older suffer from CVD.4 Due to the physiological changes and toll of pregnancy, live births might contribute to high CVD rates among African American women. This might be plausible because African American women have a higher rate of live births than Caucasian women (1,775.5 per 1000 women vs. 1,610.5 per 1000 women), who also have lower rates of CVD.5 Although there are numerous studies examining the relationship between CVD and live births, these studies have yielded mixed findings with some studies finding an association between live births and CVD among women6-9 and other studies finding no association between live births and CVD among women.10-13 Moreover, many of these studies either lack a significant number of African American participants and/or do not control for a wide array of confounding CVD risk factors. To address the limitations of previous studies and better understand potential preventative efforts to reduce the impact of CVD incidence among African American/Black women, we examined the association of live births and CVD among African American women. Our null hypothesis was that there will be no relationship between live births and CVDs among African American/Black women after controlling for major CVD risk factors such as, age, hypertension status, diabetes status, smoking status, total cholesterol status, frequency of alcohol use in the past 12 months, average number of drinks per week in the past 12 months, and family history of heart disease- mother and father.
Cardiovascular Disease (CVD) is a significant problem within the United States.1 Unfortunately, CVD disproportionately impacts the African American community, especially African American women.2 According to the American Heart Association, CVD kills nearly 50,000 African American women each year.3 Additionally, 49% of African American women ages 20 and older suffer from CVD.4 Due to the physiological changes and toll of pregnancy, live births might contribute to high CVD rates among African American women. This might be plausible because African American women have a higher rate of live births than Caucasian women (1,775.5 per 1000 women vs. 1,610.5 per 1000 women), who also have lower rates of CVD.5 Although there are numerous studies examining the relationship between CVD and live births, these studies have yielded mixed findings with some studies finding an association between live births and CVD among women6-9 and other studies finding no association between live births and CVD among women.10-13 Moreover, many of these studies either lack a significant number of African American participants and/or do not control for a wide array of confounding CVD risk factors. To address the limitations of previous studies and better understand potential preventative efforts to reduce the impact of CVD incidence among African American/Black women, we examined the association of live births and CVD among African American women. Our null hypothesis was that there will be no relationship between live births and CVDs among African American/Black women after controlling for major CVD risk factors such as, age, hypertension status, diabetes status, smoking status, total cholesterol status, frequency of alcohol use in the past 12 months, average number of drinks per week in the past 12 months, and family history of heart disease- mother and father.
The study population was selected from the Jackson Heart study (JHS). The JHS is a large, community, observational study that consists of African American/Black participants from urban and rural areas of three counties (Hinds, Madison, and Rankin) in Mississippi. Briefly, 5,306 African American/Black men and women, 21 and older were recruited to participate in the study. Follow-up examinations were conducted in 2000-2004 (visit 1), 2005-2008 (visit 2), and 2009-2013 (visit 3). For this analysis, we included only women (n=3367), who had information on live births and cardiovascular events (presence of CVDs) (n=3077) from 2000-2004 (visit 1).
The study protocol was approved by JHS and the Institutional Review Board at Jackson State University.
The predictor variable for the study was live births. At visit 1, African American/Black women were asked to report the total number of live births in an administered questionnaire. We used live births to identify women with live births.
The covariates for the study were age, hypertension status, diabetes status, smoking status, total cholesterol status, frequency of alcohol use in the past 12 months, average number of drinks per week in the past 12 months, and family history of heart disease- mother and father. Women reported the covariates included in this study during an administered questionnaire at visit 1. At visit 1, plasma cholesterol, triglycerides, and A1C were measured from blood samples collected at visit 1 to determine cholesterol status, and diabetes status. Blood pressure was also measured during visit 1 to determine systolic blood pressure, diastolic blood pressure, and hypertension status.
The outcomes of interest were cardiovascular events (presence of CVD/CVDHx). Cardiovascular events (presence of CVD/CVDHx) are signified by a derived variable that combines clinical measures with self-reported data. The variables that comprise cardiovascular events (presence of CVD/CVDHx) are the following: Self-reported history of MI (MIHx), Self-reported history of cardiac procedures (CardiacProcHx), Coronary Heart Disease Status/History (CHDHx), Self-reported history of Carotid Angioplasty (CarotidAngioHx), Cardiovascular Disease history (CVDHx), and Heart Failure History (HFHx).
Missing data were imputed into the data set. Cases with missing values were excluded listwise, which means observations with missing values on any of the variables in the analysis were omitted from the analysis. For categorical variables, only valid percentages were included in the results of the study.
The statistical analyses included the summarization of categorical variables by number, valid percentage, and confidence intervals and the summarization of continuous variables by mean, minimum, maximum, range, outlier, and confidence intervals. Chi-square was used to assess an association between live births and cardiovascular events (presence of CVD/CVHDx). Logistic regression was used to assess a relationship between live births and cardiovascular events (presence of CVD/CVDHx) while controlling for the covariates included in the study. Pre-screening of the data was done to test for multicollinearity to identify collinearity between the independent variables.
The level of statistical significance was set at p<0.05, and p-value for all hypothesis tests were two-tailed. All statistical analyses were performed using Statistical Package for the Social Sciences, version 28.0 (SPSS).14
The analyses were based on 3,077 adult Black women in the central Mississippi area. The mean age of eligible respondents was 55.3 years; more than half (61.5%) had greater than a high school education, and less than a quarter (18.5%) were considered poor. Majority (90.1%) of the respondents also do not have CVDs, do not smoke (89.7%), do not have diabetes (75.3%), did not frequently drink alcohol in the past 12 months of the administration of the survey (61.5%), had no family history of heart disease- maternal (68%) or paternal (63.9%). Less than a quarter (15.3%) had poor health in terms of the American Heart Association’s standards for cholesterol status. More than half (58.9%) of eligible respondents do have hypertension. The mean of BMI for respondents was 32.8kg/m2. (Table 1) (Table 2)
Characteristic | N | % | ||
Race Black/African American |
3367
|
100
| ||
Sex Female |
3367 |
100 | ||
Education Level <High school graduate High School or equivalent graduate >High school graduate (Missing 12) |
603 690 2062 |
18.0 20.6 61.5
| ||
Income Poor Lower middle Upper middle Affluent (Missing 523) |
527 750 843 724
|
18.5 26.4 29.6 25.5 | ||
Presence of CVD Yes No (Missing 290) |
306 2771 |
9.9 90.1 | ||
Hypertension Yes No (Missing 2) |
1848 1229 |
60.1 39.9 | ||
Current Smoking Status Yes No (Missing 32) |
303 2744
|
9.9 90.1 | ||
Cholesterol Status (AHA) Poor Health Intermediate Health Ideal Health (Missing 583) |
428 1157 1199 |
15.4 41.6 43.1 | ||
Diabetes Status Yes No (Missing 330) |
760 2277
|
25.0 75.0 | ||
Frequency of alcohol use in the past 12 months Yes No (Missing 21) |
1287 2059
|
38.4 61.5
| ||
Pregnancy Categories No Pregnancy 1 Pregnancy 2-3 Pregnancies 4 or more Pregnancies (Missing 290) |
118 478 1454 1027 |
3.8 15.5 47.3 33.4 | ||
Family History of Heart Disease (Mother) Yes No Don’t Know (Missing 305) |
752 2065 245
|
24.6 67.4 8.0 | ||
Family History of Heart Disease (Father) Yes No Don’t Know (Missing 303) AHA, cholesterol status based on standards from the American Heart Association; |
516 1951 597 |
16.8 63.7 19.5 | ||
Table 2 (Sociodemographic and Health Characteristics of Mississippi black women, Jackson Heart Study Visit 1, 2000-2004)
Characteristic (Overall) | N | Mean | Minimum | Maximum | |
Age | 3367 | 55.3
| 20 | 91 | |
BMI (kg/m2)
| 3358 | 32.8 | 15 | 92 | |
Systolic Blood Pressure
Diastolic Blood Pressure
| 3351
3351 | 126.97
74.4 | 78
36 | 228
111 | |
The overall prevalence of CVD events was 9.9% (95% CI, 8.9-11.1). A chi-square test for independence indicated a significant association between live births and the presence of CVD (c2 (3, n=3077) = .113, p =.000, phi =.113). The chi-square test also indicated an almost medium effect size based on the Cramer’s V statistic of .113 (.01-small effect vs .03-medium effect). (Table 3)
Table 3 (CVDs among Mississippi AA/black women by number of pregnancies, JHS, 2000-2004)
Characteristic | Presence of CVD % (n=3077) |
|
p-value | |
|
Yes No 306 (9.9%) 2771 (90.1%) |
|
<.001
| |
Number of Pregnancies No Pregnancy 1 Pregnancy 2-3 Pregnancies 4 or more Pregnancies
| Yes No 17 (14.4%) 101 (85.6%) 34 (7.1%) 444 (92.9%) 108 (7.4%) 1346 (92.6%) 147 (14.3%) 880 (85.7%)
|
|
<.0001 | |
By contrast, the results of the logistic regression models including controls for hypertension status, current smoking status, cholesterol status, diabetes status, family history of heart diseases (mother), family history of heart disease (father), age, frequency of alcohol use in the past 12 months, and the average number of drinks per week in the past 12 months showed no significant relationship between live births and CVD events (AOR 1.012, 95% CI, 1.009-1.015, p=.895). (Table 4)
Table 4: (Association between Presence of CVD and number of pregnancies among AA/black women, JHS, 2000-2004)
Model | Characteristic | OR | 95% CI | p-value |
1 | Number of Live Born Children | 1.175 | 1.123-1.229 | <.001 |
2
| Number of Live Born Children Participants’ Age Income Status (1): Lower-middle Income Status (2): Upper-middle Income Status (3): Affluent Education Attainment (1): High School/GED Education Attainment (2): >High School Graduate | 1.010 1.059 .807 .567 .422 .591 .730 | .954-1.070 1.044-1.074 .572-1.138 .375-.858 .258-.688 .399-.876 .502-1.063 | .772 <.001 .222 .007 <.001 .009 .101 |
3
| Number of Live-born Children Participants’ Age Income Status (1): Lower-middle Income Status (2): Upper-middle Income Status (3): Affluent Education Attainment (1): High School/GED Education Attainment (2): >High School Graduate Alcohol drinking in the past 12 months (Y/N)(1):N Avg. # of drinks per week in the past 12 months Participants current cigarette smoking status (1):Y | 1.013 1.059 .807 .602 .482 .609 .767 .756 1.000 2.218 | .954-1.075 1.043-1.075 .567-1.149 .396-.916 .293-.794 .407-.911 .523-1.124 .530-1.077 .949-1.054 1.473-3.341 | .676 <.001 .234 .018 .004 .016 .173 .121 .999 <.001 |
4 | Number of Live-born Children Participants’ Age Income Status (1): Lower-middle Income Status (2): Upper-middle Income Status (3): Affluent Education Attainment (1): High School/GED Education Attainment (2): >High School Graduate Alcohol drinking in the past 12 months (Y/N)(1):N Avg. # of drinks per week in the past 12 months Participants current cigarette smoking status (1):Y Hypertension Status: Y Diabetes Status (ADA 2010): Y Total Cholesterol (AHA) Mother health history: N Father health history: N
ADA, diabetes status based on standards from the American Diabetes Association; AHA, cholesterol status based on standards from the American Heart Association;
| 1.006 1.052 .883 .768 .626 .617 .704 .934 1.006 2.192 2.293 1.621 1.009 1.091 .952 | .943-1.074 1.034-1.071 .593-1.315 .483-1.222 .360-1.088 .397-.959 .461-1.074 .639-1.365 .955-1.059 1.377-3.490 1.502-3.503 1.173-2.240 .813-1.252 .839-1.420 .743-1.219 | .856 <.001 .541 .265 .097 .032 .104 .723 .832 <.001 <.001 .003 .936 .515 .694 |
Multicollinearity testing was performed. Pre-screening of the data to test for multicollinearity revealed there was no statistically significant relationship between the variables.
In this analysis of the relationship between live births and CVD events in African American/Black women cohort in 3 central areas within Mississippi, we found a significant unadjusted relationship, but no significant relationship after controlling for major CVD risk factors. The lack of risk between CVD events associated with live births has been reported in previous studies, 15,16 although neither study adjusted for alcohol use, cholesterol, and family history of heart disease (maternal and paternal). Only one study adjusted for diabetes, hypertension, and smoking, and neither study involved a high percentage of African American/Black female participants. There are several mechanisms that may explain the lack of association between live births and CVD events in Black women. There is evidence that establishes that the risk of CVD events for prior pregnancy may be more associated with behavioral factors, such as stress rather than physical factors.17 The stress of childrearing, balancing work and family, and the financial responsibilities of children can be more of an indicator of possible CVD risk than solely considering live births. While stress was not studied in this study, it is very plausible based on other studies that stress maybe the source of CVD events for women with prior pregnancy. 17,18 It may also be attributable to the fact that women have a higher stress level than men, which may account for the difference in the rate of cardiovascular events between women and men.17 The lack of an association may also be based on the fact that the risk of CVD events in women may be more associated with lifestyle or socioeconomic factors, such as income, or employment.19 A recent study identified an increased risk of CVD events among pregnant and post-partum Black women than White women based on education and income.19 While this study did not include a highly inclusive list of socioeconomic factors in the logistic regression analysis, the study did assess several socioeconomic factors for participants involved in the study. However, the majority of the women included in this study were well educated with 61.5% having higher than a high school education, and less than 20% being considered poor based on income. These findings suggest that the women with prior pregnancies in this study may also be less at risk for CVD events due to their income stability and educational background.
Our results contrast with some previous studies that found an increased risk of CVD events for women with live births.20,21 This may be explained by the adjustment for major CVD risk factors and by characteristics of the cohort involved in the study. If CVD risk factors are not controlled, and the women involved in the study do not have high socioeconomic statuses, women- in this case African American women may have an increased risk of CVD outcomes.
There is an obvious need for accurate assessments of CVD disease risks in women, especially African American/ Black women who are impacted by CVD events at a higher rate than White women.2 While this study did not identify live births as a beneficial measurement of CVD risk prediction, there is still a need for more studies to evaluate the association of live births and CVD events in more socioeconomic inclusive cohorts of African American/ Black women.
The strength and potential limitations of our study warrant consideration. One strength is that this study included a large sample of middle-aged African American women in a well characterized cohort. This study also measured clinical, biologic, and anthropometric data, and collected information on reproductive history.
This study is also limited due to a lack of information on pregnancy complications, such as preterm births, gestational hypertension, or pre-eclampsia. All these factors have been linked to an increased risk of CVD events. There may be other measured or unmeasured confounders that were not included.
There was no relationship between live births and CVD events for this cohort of African American/Black women after controlling for CVD risk factors. Whether the lack of risk is due to a possible association between behavioral and lifestyle factors such as, stress and socioeconomic status is unclear, and requires further analysis. Further analysis is needed to ensure that marginalized African American/Black women can make informed sexual and reproductive health decisions to obtain optimum cardiovascular health. Through the implementation of further analysis, policies can be implemented that incorporate reproductive risk as an aspect of the assessment of cardiovascular risk in clinical guidelines.22 Therefore, these implementations will ensure that African American/ Black women have a thorough cardiovascular assessment and are given a candid view of their cardiovascular health status in relation to their reproductive health.
I would like to acknowledge the Jackson Heart Study and Jackson Heart Study Graduate Training Center for providing the opportunity perform cardiovascular research. I want to thank the JHS participants and the JHS community for their years of dedicated commitment and enthusiastic contributions. This research is supported by JHS Graduate Training and Education Center (GTEC) NHLBI-NIH Contract NHLBI-HHSN26818HV0009R and the National Institute on Minority Health and Health Disparities of the National Institutes of Health; “Research reported in this publication was supported by the National Institute on Minority Health and Health Disparities of the National Institutes of Health under Award Number U54MD015929. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.”
The authors have no relevant financial disclosures or conflicts of interest.
Elizabeth Jones is an epidemiologist II at the Mississippi State Department of Health and a human subject’s researcher at Jackson Heart Study GTEC program. Her research areas include maternal and child health and minority health. She received her formal training at Tougaloo College and Jackson State University.
Dr. Keith Norris is a clinician scientist and health policy leader. His research areas include nephrology and minority health. He received his formal training at Cornell University and Howard University.
Dr. Brenda Jenkins is the Director of Training and Education at the Jackson Heart Study GTEC Program. Her research areas include cardiovascular disease and minority health. She received her formal training at Mississippi State University.
Dr. Clifton Addison is a professor of epidemiology at Jackson State University and a senior research scientist for the Jackson Heart Study. He has received numerous honors, including JSU Staff Excellence Award for Outstanding Education Support/Initiative and Creativity, JSU Research & Sponsored Programs Excellence Awards, Marquis Who’s Who in Medicine and Healthcare, Sports Hall of Fame- University of Texas El Paso, and the Patricia Roberts Harris Doctoral Fellowship.
Marinelle Payton, MD, PhD, MS, MPH is the Principal Investigator of the Jackson Heart Study Graduate Training and Education Center. She is also the Assistant Dean for Research and Program Development, Professor of Epidemiology and Biostatistics, and Principal Investigator and Director of the Center of Excellence in Minority Health and Health Disparities/Institute of Epidemiology and Health Services Research, School of Public Health.
Dr. Payton is a physician with specialty training in Neurological Sciences, Surgery, Radiology, Epidemiology, Public Health, and Environmental and Occupational Medicine. She received the combined M.D.-Ph.D. degrees and the M.P.H. and M.S. degrees from Boston University School of Medicine and Harvard School of Public Health. Following clinical training at Yale University and Boston University hospitals, she held hospital and joint academic appointments at Harvard Medical School and Harvard School of Public Health in Boston, Massachusetts, prior to joining Jackson State University as the Founding Chair of Public Health (led the development of the M.P.H. Program and Dr.P.H. Program, the first and only program in the state of Mississippi).
Dr. Payton has over 25 years of experience in medicine, public health, community service, and research (awarded approximately 10 M over the last two years) and has made exemplary advances closing the disproportionate gap in health disparities—working with the National Institutes of Health to expand the traditional definition of research to encompass areas such as community based participatory research and health disparities.
Her longstanding research has focused on chronic disease and neurotoxins which resulted in peer reviewed publications including the first epidemiological study of blood and bone lead and cognitive function. She served on many boards including Children’s Health, commissioned by President William Jefferson Clinton, and the Science Advisory Board commissioned by Congress, and is the Chair of the Mississippi State Lead Poisoning Prevention Advisory Board. She also received many awards including the Alfred L. Frechette Award in Public Health, Clinical Environmental Medicine awards, and research achievement awards.
BCPHR.org was designed by ComputerAlly.com.
Visit BCPHR‘s publisher, the Boston Congress of Public Health (BCPH).
Email [email protected] for more information.
Click below to make a tax-deductible donation supporting the educational initiatives of the Boston Congress of Public Health, publisher of BCPHR.
© 2024 BCPHR: An Academic, Peer-Reviewed Journal