Drivers of Black-White Disparities in Hypertension Risk After Pregnancy: Findings from CARDIA

Abstract

BACKGROUND:

Hypertension is the leading risk factor for cardiovascular disease (CVD), with disparities observed by race. Few studies have examined how intersecting social and sex-based biological determinants influence hypertension incidence among parous individuals and across racial groups. The objective of this study was to examine whether female-specific CVD risk factors (e.g., adverse pregnancy outcomes), individual- (e.g., education), and structural-level (e.g., neighborhood) social drivers of health (SDoH), beyond traditional CVD risk factors (e.g., physical activity), explain the Black-White disparity in hypertension following pregnancy.

METHODS:

Parous individuals in CARDIA, a longitudinal cohort study beginning in 1985–1986, without hypertension prior to their first pregnancy were included. Race was self-reported at enrollment. Blood pressure was measured at each examination, with hypertension incidence was defined as the first elevated blood pressure (systolic ≥130 mm Hg or diastolic ≥80 mm Hg) or initiation of antihypertension mediation after baseline. Participant follow-up time began at the next exam following their first live delivery. Cox proportional hazards models with time-varying covariates calculated hazard ratios (HRs) and 95% confidence intervals (CIs). To quantify how much each set of covariates explained the Black–White difference in hypertension risk, the percentage reduction in the race coefficient was calculated following addition of female-specific CVD risk and SDoH in multivariable-adjusted regression models.

RESULTS:

Among 1653 participants (29.4±6.1 years), Black women (n=924) had higher hypertension incidence than White women (n=729): 41.4 vs. 18.5 per 1000 person-years; and an earlier median onset age (42 vs. 48 years). Following traditional risk factors, individual-level SDoH explained the largest disparity (7.52%), then female-specific factors (2.26%), and structural- level SDoH (1.50%). Black women remained at higher risk post-adjustment (HR 2.72, 95% CI 2.17 – 3.41).

CONCLUSIONS:

Parous Black women face substantially higher hypertension rates than White women, persisting after accounting for traditional risks, female-specific risk factors, and SDoH.

Introduction

Non-Hispanic Black men and women have higher rates of hypertension across the life course compared with other race and ethnicities.^1,2 Prior research from the Coronary Artery Risk Development in Young Adults (CARDIA) cohort found that over a 20-year-period, Black women had significantly greater rates of incident hypertension than both Black men and White women.³ These disparities likely reflect a complex interplay of structural, psychosocial, behavioral, and biological factors that combine to enhance risk beyond traditional cardiovascular risk profiles. For example, chronic stress, limited access to healthcare, neighborhood deprivation, and systemic racism collectively constrain health behaviors critical to cardiovascular health.^4–6 Moreover, female-specific cardiovascular risk factors (e.g., adverse pregnancy outcomes, high parity, and early menopause) disproportionately affect Black women, further driving disparities in hypertension.^7–9

Pregnancy functions as a “natural stress test”, with the ability to unmask an individual’s underlying susceptibility for cardiovascular disease (CVD).^10,11 Hypertension is estimated to affect up to 22% of individuals within the first year postpartum,^12,13 and its development is influenced by several factors including obesity, health behaviors (e.g. tobacco use, physical inactivity), and adverse pregnancy outcomes.¹ Since hypertension is the most common chronic condition among women and a potent risk factor for CVD, identifying drivers of hypertension in parous individuals is essential for mitigating long-term disease risk.^14,15

Despite these well-documented racial and sex-based disparities, few studies have examined how intersecting social and biological determinants jointly influence hypertension incidence after delivery across racial groups. Existing research often relies on short-term follow-up, limiting our understanding of the cumulative effects of these influences over time. To address these gaps, we leveraged the CARDIA cohort to assess the degree to which 1) known female-specific biological risk factors for CVD; and, 2) both individual- and structural-level social drivers of health (SDoH) explain the Black-White disparity in hypertension the years following pregnancy among parous individuals enrolled in the CARDIA study.

Methods

Study design and participants

We included data from the CARDIA cohort study and these data can be accessed via request from qualified researchers trained in human subject confidentiality protocols to CARDIA_Publications@uabmc.edu. The protocols of this cohort study have been cited elsewhere,¹⁶ but in brief, CARDIA launched in 1985–86 and recruited 5,115 Black and White individuals aged 18–30 years. Individuals were recruited from four locations including: Birmingham, Alabama; Chicago, Illinois; Minneapolis, Minnesota; and Oakland, California. Participants attended follow-up exams in 1987 – 1988 (Year 2), 1990–1991 (Year 5), 1992 – 1993 (Year 7), 1995 – 1996 (Year 10), 2000 – 2001 (Year 15), 2005 – 2006 (Year 20), 2010 – 2011 (Year 25), 2015 – 2016 (Year 30), and 2020 – 2022 (Year 35). At each study visit, factors associated with CVD were assessed including anthropometrics, blood pressure, cholesterol, glucose, lifestyle factors, substance use, behavioral and psychological factors, and medical history. Prior to the start of the study, each field center’s institutional review board approved of all protocols and participants gave written informed consent prior to enrollment.

To be included in the present set of analyses, participants must be assigned a female sex at birth and report one live birth throughout the study. Individuals diagnosed with hypertension prior to their first pregnancy were excluded (Figure 1).

Figure 1. Inclusion and exclusion flow chart from the CARDIA cohort for the present analysis.

An analytical sample was created from the CARDIA cohort which included parous females without hypertension prior to their first pregnancy. CARDIA: Coronary Artery Risk Development in Young Adults.

Race (Exposure)

The race variable was constructed with participant’s self-reported responses at the time of enrollment. In our analysis, race is used as a sociopolitical variable representing the effects of historical and contemporary inequities, avoiding the assumption of any biological distinctiveness between groups. Consistent with the 2025 National Academies of Science, Engineering, and Medicine Consensus Study Report which recommends against use of self-identified race as proxy for genetic ancestry, race in this study captures social positions rather than genetic heritage.¹⁷ Moreover, observed racial disparities in hypertension incidence should be interpreted through a sociopolitical lens and not a biological one.

Traditional Cardiovascular Disease Risk Factors (Covariates)

All variables were measured at the first examination following the delivery of the first child (baseline) and at each subsequent examination. Traditional CVD risk factors assessed included body mass index (BMI), smoking status, alcohol intake, physical activity, and depressive symptoms. BMI was calculated based on participant height and weight measured by clinical exam staff. Smoking status was self-reported and categorized as never, former, or current. Alcohol intake was self-reported and measured in milliliters consumed per week on average. An exercise score was generated from self-reported frequency and intensity of engagement in 13 activities. Depressive symptoms were measured with the validated Center for Epidemiological Studies - Depression questionnaire, where total scores ranged 0–60 (higher scores indicating greater depressive symptoms).¹⁸ Variables were treated as time-varying in models.

Female-specific Cardiovascular Disease Risk Factors (Covariates)

Parity, history of adverse pregnancy outcomes (APOs) across all pregnancies, and age at menopause were used to assess female-specific CVD risk enhancing factors. APOs were self-reported and defined as the occurrence of gestational diabetes, hypertensive disorders of pregnancy, and pre-term birth (gestational age at delivery of <37 weeks). Early age of menopause was defined as the self-reported age of menopause or double oophorectomy earlier than 40 years of age. Female-specific CVD risk enhancing factors were time-varying and assessed at or up until the Year 35 CARDIA exam.

Individual and Structural-Level Social Drivers of Health (Covariates)

SDoH was defined following recommendations by the World Health Organization and U.S. Centers for Disease Control and Prevention, and included the assessment of education access and quality, healthcare access and quality, neighborhood and built environment, social and community context, and economic stability.^19,20 Individual-level SDoH was assessed through self-reported measures, including highest level of education achieved (baseline through Year 35), ability to pay for necessities (baseline - Year 35), experiences of racial discrimination (Years 7, 15, 25, 30), and chronic burden (Years 15 – 30). Ability to pay for necessities was assessed via a single question “How hard is it for you (and your family) to pay for the very basics like food, medical care, and heating?” which participants ranked on a 5-point Likert scale from very difficult to not difficult (1–5). Participants were asked to report whether they experienced discrimination based on their race or color at school, work, getting healthcare, housing, or in public settings. Chronic burden, our measure of stress, was assessed by asking participants whether they experienced certain strains for longer than 6 months during the examination year including serious health problems, job stability, financial strain, and ongoing relationship difficulties. Both discrimination and chronic burden were included in models as categorical variables reflecting the distribution of scores.

Structural-level SDoH included the assessment of neighborhood cohesion (Years 15, 20, 30) and neighborhood physical activity environment (Years 20, 30). Neighborhood cohesion was assessed via survey questions addressing participant perceptions of whether they feel supported by neighbors, trusted their neighbors, and shared values with their community.²¹ Neighborhood physical activity environment was assessed via survey questions regarding access to an exercise facility, park, sidewalks, walking paths, and public transportation within a 10–15 mile radius from home.

Blood Pressure (Outcome)

Blood pressure was measured three times at 1-minute intervals by a trained technician using a random zero Sphygmomanometer (year 0 through year 20) or OMRON Sphygmomanometer (calibrated to the random zero device, year 20 through year 35). Calibration studies were conducted to eliminate any bias between measurement methods.^{21, 22} The average of the second and third measurements was used in analysis. Hypertension incidence was defined as the first occurrence after baseline of elevated blood pressure (systolic ≥130 mm Hg or diastolic ≥80 mm Hg) or initiation of antihypertensive medication, consistent with American Heart Association guidelines.²³

Statistical Analyses

Participant characteristics at baseline were summarized as means (standard deviations) for continuous variables and counts (percentages) for categorical variables. Incidence of hypertension after pregnancy per 1,000 person years was calculated for the total sample, and separately for Black and White women, by dividing the number of new hypertension cases by the total person-years of follow-up and multiplying by 1,000.

Cox proportional hazards models were used to evaluate the association between each covariate group and the incidence of hypertension. Hazard ratios (HRs) and 95% confidence intervals (CIs) were reported. Baseline was defined as the first CARDIA examination following the delivery of the first child. Time to event was calculated as the number of years from baseline to the first occurrence of hypertension. Participants were censored at withdrawal, loss to follow-up, or at the end of the study period if hypertension did not develop. The Breslow method was applied to handle tied event times, and the proportional hazards assumption was assessed using the Kolmogorov-type supremum test and by visually inspecting martingale residuals.

To quantify how much each group of covariates explained the Black–White difference in hypertension risk, we calculated the percent reduction in the race coefficient (β) where β₀ represents the coefficient for race in the unadjusted model and β_n represents the coefficient for race in each sequentially adjusted model using the formula:

$$\left(\right(\left[{\beta }_0\right]-\left[{\beta }_N\right])/{\beta }_0)\times 100$$

Subsequently, sequential model adjustment was also conducted to evaluate attenuation of racial disparities in hypertension incidence: Model 0 (base model): age and field center, Model 1: base model and traditional CVD risk factors, Model 2: model 1 and female-specific CVD risk factors, Model 3: model 2 and individual-level SDoH, Model 4: model 3 and structural-level SDoH (Table S1). Time-varying covariates were updated at each exam visit and assigned to the interval between that visit and the subsequent visit. For participants with intermittent missing measurements at a given exam, the most recently observed value was carried forward until the next available measurement. Missing covariate data were addressed using multiple imputation by chained equations. Five imputed datasets were generated, with all variables included in the analytic model (including outcome and covariate variables) into the imputation procedure. Parameter estimates were combined across datasets using Rubin’s rules. All analyses were conducted using R version 4.6.1 (R Core Team, Vienna, Austria) in RStudio (RStudio, PBC). All p-values were 2-sided and statistical significance was defined as a p-value of <0.05.

Results

Participant Characteristics

A total of 1653 parous participants were included in the present set of analyses, and their characteristics are described in Table 1. On average, participants were 29.4 (SD:6.1) years of age at the study visit following delivery of their first child, had a BMI of 25.7 (SD:6.1), and completed 15.1 (SD:2.5) years of education. Through Year 35, participants reported 1.9 (SD:0.9) live births with 32.7% (n = 540) experiencing at least one adverse pregnancy outcome across follow-up and only 2.4% (n = 39) experiencing menopause prior to the age of 40. On average, Black participants were younger at, had a higher BMI, and completed fewer years of education at baseline.

Table 1.

Participant characteristics at baseline (N= 1653)

	Overall (N=1653)	Black Females (n =924)	White Females (n =729)
Age (years), mean(SD)	29.4 (6.1)	27.6 (5.4)	31.7 (6.1)
Field center, n(%)
Birmingham, Alabama	289 (17.5%)	175 (18.9%)	114 (15.6%)
Chicago, Illinois	308 (18.6%)	168 (18.2%)	140 (19.2%)
Minneapolis, Minnesota	295 (17.8%)	126 (13.6%)	169 (23.2%)
Oakland, California	369 (22.3%)	207 (22.4%)	162 (22.2%)
Highest grade of education completed (years), mean(SD)	15.1 (2.5)	14.3 (2.2)	16.0 (2.7)
Difficulty paying for necessities, mean(SD)	3.4 (0.9)	3.4 (0.9)	3.5 (0.8)
Total physical activity, mean(SD)	259.0 (214.0)	244.0 (218.0)	277.0 (208.0)
Body mass index (kg/m 2 ), mean(SD)	25.7 (6.1)	26.8 (6.7)	24.4 (4.9)
Smoking status, n(%)
Never smoked	926 (56.0%)	529 (57.3%)	397 (54.5%)
Formerly smoked	294 (17.8%)	102 (11.0%)	192 (26.3%)
Currently smokes	427 (25.8%)	289 (31.3%)	138 (18.9%)
Alcohol intake (mL/week), mean(SD)	5.5 (12.0)	4.9 (12.3)	6.2 (11.6)
Depressive symptoms, mean(SD)	9.5 (7.8)	11.1 (8.8)	8.5(6.9)
Racial discrimination, median(range)	0 (0, 7.0)	1.0 (0, 7.0)	0 (0, 4.0)
Chronic burden, mean(SD)	1.6(0.5)	1.5(0.6)	1.6(0.5)
Neighborhood cohesion, mean(SD)	1.1(0.8)	0.9(0.8)	1.3(0.8)
Neighborhood physical activity environment, mean(SD)	1.7(1.2)	1.8(1.1)	1.5(1.2)

Baseline is defined as the first CARDIA exam after delivery of the first child. Difficulty paying for necessities self-reported on a scale from very difficult (1) to not difficult (5) with scores 1–3 defined as difficulty paying for necessities. Total physical activity in exercise units. Body mass index in kilograms per meters squared. Alcohol intake measured in milliliters per week. Depressive symptoms measured with the validated Center for Epidemiological Studies - Depression questionnaire (score range: 0–60, with higher scored indicating greater depressive symptoms). Racial discrimination was self-reported experiences of discrimination at school, work, getting healthcare, housing, or in public settings. Chronic burden was assessed by experiences of certain strains for longer than 6 months including a serious health problem, job instability, financial strain, and ongoing relationship difficulties. Neighborhood cohesion was self-reported perception of whether participants felt supported by their neighbors, trusted their neighbors, and shared values with their community. Neighborhood physical activity environment was assessed via survey questions regarding access to places to exercise, sidewalks, walking paths, and public transportation within a 10–15-mile radius from home. CARDIA: Coronary Artery Risk Development in Young Adults

New-Onset Hypertension Incidence by Race

A total of 966 participants (58.4%) developed hypertension overall across an average of 19.5 (SD: 11.0) years of follow-up, with a burden disproportionately higher among Black women than White women (72.6% vs. 40.5%). This pattern was reflected in survival free years from hypertension, with Black women experiencing significantly fewer years hypertension free over follow-up(41.4 per 1,000 person-years) compared with White women (18.5 per 1,000 person-years). Figure 2 presents Kaplan-Meier survival curves of survival free from hypertension, showing earlier divergence and higher cumulative incidence among Black women, showing an earlier and more pronounced decline among Black women. By the log-rank test (p < 0.001), hypertension-free survival differed significantly between groups. The median age at loss of hypertension-free survival, corresponding to the age at hypertension onset, was 48 years (IQR 40–55) for White individuals and 42 years (IQR 35–50) for Black individuals.

Figure 2. Probability Remaining Hypertension Free After Delivery of First Child in CARDIA by Race (N = 1653).

Baseline was defined as the first CARDIA examination following the delivery of the first child. Time to event was calculated as the number of years from baseline to the first occurrence of hypertension. Coronary Artery Risk Development in Young Adults.

Drivers of Racial Differences in New-Onset Hypertension Incidence

Adjustment for traditional CVD risk factors produced the greatest single attenuation (20.30% reduction), while adjustment for individual-level SDoH, structural-level SDoH, and female-specific CVD risk factors only reduced the association by 7.52%, 1.50%, and 2.26%, respectively (Table 2). When female-specific CVD risk factors, individual SDoH, and structural-level SDoH were included simultaneously, a 9.77% reduction in the racial disparity was observed.

Table 2.

Percent explain analysis assessing the association of race and hypertension occurrence after delivery of first child in CARDIA (N = 1653).

	Total (n = 1653)	White (n=729)	Black (n=924)
Frequency of events, n (%)	966 (58.4)	295(40.5)	671(72.6)
Incidence of hypertension per 1000 person-years (95% CI)	30.0 (28.2, 32.0)	18.5 (16.4, 20.7)	41.4 (38.3, 44.7)
	Beta coefficient (95% CI)	% Reduction*	Hazard Ratio (95% CI)
Base Model: Age + Field center	1.33 (1.15, 1.51)	Reference	3.78 (3.16, 4.52)
Base model + Traditional CVD risk factors *	1.06 (0.86, 1.26)	20.30	2.89 (2.37, 3.52)
Base model + Female CVD risk factors †	1.30 (1.12, 1.48)	2.26	3.67 (3.06, 4.39)
Base model + Individual SDoH ‡	1.23 (1.03, 1.44)	7.52	3.43 (2.79, 4.23)
Base model + Structural SDoH ||	1.31 (1.12, 1.49)	1.50	3.69 (3.06, 4.46)
Base model + Female CVD risk factor + Individual SDoH + Structural SDoH	1.20 (0.99, 1.42)	9.77	3.33 (2.68, 4.12)
	Beta coefficient (95% CI)	% Reduction*	Hazard Ratio (95% CI)
Base Model: Age + Field center	1.33 (1.15, 1.51)	Reference	3.78 (3.16, 4.52)
Model 1: Base model + Traditional CVD Risk Factors *	1.06 (0.86, 1.26)	20.30	2.89 (2.37, 3.52)
Model 2: Model 1 + Female CVD Risk Factors †	1.05 (0.85, 1.24)	21.05	2.84 (2.34, 3.46)
Model 3: Model 2 + Individual SDoH ‡	1.01 (0.79, 1.23)	24.06	2.74 (2.20, 3.42)
Model 4: Model 3 + Structural SDoH ||	1.00 (0.77, 1.23)	24.81	2.72 (2.17, 3.41)

Percent reduction in the beta coefficient representing race (black vs. white) is calculated as (([β₀]-[β_N])/ β₀) × 100 which is (Base Model – Model N) ×100 then divided by Base Model (N = 1,2,3,4,5, 6,7,8). Base model adjusts for age at assessment and field center. All the model adjustments, other than field center, are time varying covariates including age, body mass index, smoking status, alcohol intake, total physical activity, CES-D score, number of live births, and difficulty paying for necessities, racial discrimination, average burden, neighborhood cohesion, neighborhood physical activity environment. Missing covariate data were accounted for with multiple imputation by chained equations.

^*Traditional CVD Risk Factors: body mass index, smoking status, alcohol intake, total physical activity, and CES-D score;

^†Female CVD risk factor: history of adverse pregnancy outcomes, early menopause, and number of live births;

^‡Individual SDoH: highest education, difficulty paying for necessities, racial discrimination, average burden;

^||Structural SDoH: neighborhood cohesion, neighborhood physical activity environment

Abbreviations: β - beta coefficient; CVD - cardiovascular disease; SDoH - social drivers of health; CI - confidence interval

The sequential models demonstrated similar patterns, with the race–hypertension association incrementally attenuated as additional covariate groups were included. Compared with the base model (HR: 3.78, 95% CI: 3.16 – 4.52), the full model including adjustment of all covariate groups (Model 4) showed an 24.80% reduction in the beta coefficient and an HR of 2.72 (95% CI: 2.17 – 3.41).

Discussion

In this prospective cohort study, Black women had a significantly higher risk of hypertension after the delivery of their first child, as compared to white women. In addition, Black women developed hypertension approximately 6-years of age earlier than White women. These racial disparities persisted across multivariable-adjusted models that adjusted for traditional cardiovascular risk factors, female-specific factors, and SDoH. Outside of traditional CVD risk factors, individual-level SDoH accounted for the largest proportion of attenuation, followed by female-specific cardiovascular factors, and structural-level SDoH. In a fully-adjusted model, simultaneously adjusted for all covariates, Black women remained nearly 2.7 times more likely to develop hypertension postpartum, with traditional CVD risk factors, female-specific CVD risk factors, individual-, and structural-level SDoH explaining ~25% of the disparity.

The present study builds on current literature through the investigation of how both female-specific CVD risk factors and SDoH, beyond traditional CVD risk factors, attribute to racial disparities in hypertension among parous women. While racial disparities in hypertension incidence have been well documented, prior studies have largely focused on traditional CVD risk factors and individual-level mediators such as education, income, diet quality, and BMI.²⁴ The present study expands on this work by incorporating female-specific risk factors and broader social determinants of health to more comprehensively explain these disparities. Although CARDIA was not designed as a women’s health study, findings from this cohort are consistent with previous studies that underscore the significant influence female-specific factors, such as adverse pregnancy outcomes and reproductive transitions (e.g. menopause), have on hypertension development.^25–26 Likewise, robust evidence from the CARDIA cohort and other studies highlight the role SDoH play in shaping racial disparities in hypertension among women. Our results align with national data from the National Health and Nutrition Examination Surveys (2001 – 2018), in which even after adjustment for health factors, behaviors, and SDoH, Black women maintained a higher odds of uncontrolled blood pressure than White women (OR: 2.31; 95% CI: 1.08 – 4.92).²⁷

Despite the novelty of our study accounting for both female-specific CVD risk factors and SDoH simultaneously, the association of race and hypertension incidence remained. The persistence of the association between race and hypertension suggests that unmeasured structural or contextual influences continue to shape CVD risk. Given early-life exposures such as age at menarche and adverse childhood experiences are linked to CVD risk, future research should incorporate life course perspectives to better elucidate upstream determinants of hypertension.^28,29 Importantly, models including both SDoH and female-specific factors explained a greater proportion of racial disparities than either domain alone in our study, highlighting the need for intersectional approaches to understand how race, gender, and social context jointly shape CVD risk across the lifespan. Finally, it was observed that Black parous women developed hypertension at a median age of 44 years compared with 50 years among White parous women, indicating substantially earlier exposure to elevated blood pressure. The onset of hypertension among Black women in our study precedes or overlaps the average age of menopause, a period marked by declining cardioprotective effects of estrogen.³⁰ Consequently, early and sustained CVD prevention during this window may be critical to reduce the burden of disease among Black parous women.

Strengths and Limitations

There are numerous strengths of the present study including the use of a large population-based cohort study with 35 years of follow-up, with rigorous and repeated assessment of relevant covariates. We applied the most recent blood pressure guidelines published in 2017 by the American College of Cardiology and American Heart Association and followed WHO and US CDC recommendations for SDoH definitions. Our analysis focused on Black and White adults, which exhibit the largest racial/ethnic gap in the prevalence of hypertension according to national reports.³¹

Our findings must also be considered in the context of some limitations. Non-clinical measures were assessed by self-report and may be subject to measurement error. Hypertensive disorders of pregnancy, in particular, has been documented to have lower sensitivity in the CARDIA cohort relative to other APOs assessed in this analysis, which may increase the risk of misclassification and bias findings towards the null. Although we cannot entirely rule out the possibility of informative censoring due to death, hypertension status was ascertained at discrete study visits, meaning that death between visits does not introduce bias into the timing or ascertainment of hypertension events themselves. Additionally, censoring due to death was rare (1.4% of the analytic sample; n = 23) and occurred at similar rates among White (1.5%) and Black (1.1%) participants, suggesting censoring due to death was not differentially informative with respect to our primary exposure of race. Data on SDoH factors were not available at all years, with most assessed at the Year 20 exam. Given that SDoH factors and their influence may change across the life course, future studies are needed to identify key timepoints for intervention to equitably prevent the development of hypertension. Diet was assessed using the CARDIA Dietary History at only a few exam years, and the method shows variable validity across sociodemographic groups; as a result, diet could not be included as a time-updated covariate, and this limitation may contribute to residual confounding. Finally, given the descriptive nature of the analysis, results should not be interpreted causally.

Conclusions

Parous Black women experienced substantially higher rates of hypertension than White women after pregnancy, with this disparity persisting even after accounting for individual- and female-specific CVD risk factors, and SDoH measures. Given the average age of hypertension incidence occurred sooner among parous Black women relative to White women and the average age of menopause, midlife may be a critical period for intervention to reduce race-based disparities in CVD risk.

Supplementary Material

Supplemental Methods

Table S1: Variables included in each cox proportional hazard model and incident hypertension by race in the CARDIA cohort study.

What is Known

• Hypertension is the leading risk factor for cardiovascular disease (CVD), with disparities observed by race.

• Black women have significantly greater rates of incident hypertension than both Black men and White women across the lifespan

What the Study Adds

• Black parous individuals face substantially higher hypertension rates with incidence occurring at a younger age than White women, even after accounting for traditional CVD risk factors, reproductive history, and social drivers of health (SDoH).

• Beyond traditional CVD risk factors, individual-level SDoH (e.g., educational attainment, ability to pay for necessities, experiences of racial discrimination, and chronic burden) explained the largest proportion of the Black-White disparity in hypertension incidence among parous individuals.

Non-standard Abbreviations and Acronyms:

CARDIA

Coronary Artery Risk Development in Young Adults

APO

Adverse Pregnancy Outcomes

CES-D

Center for Epidemiological Studies Depression Scale

SDoH

Social Drivers of Health