Hypertensive disorders in pregnancy: differences by Hispanic ethnicity and Black race

Abstract

Objective:

Black individuals bear the greatest burden of maternal mortality, with hypertensive disorders in pregnancy emerging as a significant driving force of this disparity. However, research on maternal health disparities predominantly groups Hispanic Black individuals with all other individuals of Hispanic ethnicity. We hypothesized that this aggregation may obscure risk patterns of hypertensive disorders in pregnancy for Hispanic-Black and non-Hispanic Black individuals.

Methods:

We analyzed a California statewide dataset of vital records linked to hospitalization discharge data for births during 2007 to 2018. Using multivariable logistic regression models adjusted for age, pre-pregnancy BMI, parity, smoking status, diabetes, and chronic renal disease, we compared the odds of hypertensive disorders in pregnancy between Hispanic Black, non-Hispanic Black, and non-Black Hispanic racial-ethnic groups. Hypertensive disorders were grouped as: (1) any, and (2) chronic hypertension alone, non-severe hypertensive disorders and severe hypertensive disorders in pregnancy.

Results:

Non-Hispanic Black people had 75% increased odds of developing a hypertensive disorder in pregnancy (adjusted odds ratio 95% confidence interval (CI): 1.74-1.78) and Hispanic-Black people had 31% increased odds (95% CI 1.24-1.38) as compared with non-Black Hispanic people. When considering hypertensive disorders separately, the race-associated differences were largest for chronic hypertension alone (aOR: 2.35; 95% CI: 2.32, 2.38) for non-Hispanic Black people, and (aOR: 1.80; 95% CI: 1.66, 1.95) for Hispanic-Black people.

Conclusion:

Compared with non-Black Hispanic individuals, the prevalence of hypertensive disorders in pregnancy was higher in Black Hispanic individuals and highest in non-Hispanic Black individuals. Racial/ethnic differences were larger for chronic hypertension alone than for preeclampsia.

Introduction

Despite significant technological advancements in health care, alarming racial disparities persist in both maternal and infant health outcomes [11]. While various factors contribute to these stark differences, this public health crisis stems from the complex interactions of a multitude of economic and social factors. Legalized discrimination systematically disadvantages racial and ethnic minorities, subsequently, perpetuating health inequities across generations [38]. Race is a social construct and racism, based on perceived physical differences, predominantly skin color, intended to reinforce social hierarchies. In medicine, structural racism, ingrained stereotypes, lack of cultural competence and insufficient diversity have created major barriers to marginalized racial groups obtaining quality healthcare [38]. As a result, preventable conditions are often left untreated, further exacerbating existing disparities across the life-course, including pregnancy and childbirth [8,17]. In the most recent available data, the Centers for Disease Control and Prevention reported the maternal mortality rate in 2022 in the U.S. for non-Hispanic Black women to be 49.5 per 100,000 live births, as compared to 16.9 for Hispanic women and 19.0 for non-Hispanic white women [16].

One of the major drivers of the disparity in maternal mortality is hypertensive disorders. Hypertensive disorders in pregnancy encompass not only chronic hypertension, but also preeclampsia and eclampsia—high blood pressure conditions that develop during pregnancy and put pregnant people and infants at higher risk of severe complications and death [14]. Based on national data from 2017 to 2019, hypertensive disorders in pregnancy affect 20% of non-Hispanic Black women, compared with 15% of non-Hispanic white women and 13% of Hispanic women [14]. Relatedly, a report from maternal mortality review committees in 36 states found that hypertensive and other cardiovascular disorders are the leading cause (over 50%) of pregnancy-related death in non-Hispanic Black individuals—but not in other racial and ethnic groups [34]. To mitigate this crisis, we need to better understand who is most at risk and then develop interventions that fit the needs of diverse populations.

Awareness has grown in recent years of the negative consequences of health disparities research almost exclusively relying on a limited set of racial and ethnic categories. These racial and ethnic groups are social constructs that were created by the U.S. federal government, and fail to capture the true diversity that exists within the U.S. population [35]. This makes it difficult to perceive the full spectrum of health challenges faced by minorities within minority groups. In particular, there is increasing recognition that the aggregation of all Asian and Pacific Islander populations masks important health differences between these populations. Researchers found in a multi-state study that chronic hypertension in pregnancy affected Native Hawaiian and Other Pacific Islander individuals at twice the rate of Asian individuals—contributing to differences in adverse maternal and neonatal outcomes [22]. A California statewide study also demonstrated that hypertensive disorders in pregnancy vary considerably among Asian and Pacific Islander subpopulations [2]. The current federal standards for data on race and ethnicity now separate Asian from Native Hawaiian and Other Pacific Islander. However, the latest standards, released in March 2024, do not collect information on subpopulations and do not enable to ability to identify people who identify as Afro-Latino, or both Hispanic/Latino and Black [27, 28]. Yet, this group includes approximately six million adults in the U.S. [27].

For Hispanic and non-Hispanic Black individuals, the failure to capture the heterogeneity of experiences hinders the development of targeted interventions and policies to improve conditions at the population-level [5,9]. Borrell et al. developed a framework (updated in 2024) to consider the effects of racial identity in the Hispanic population on health and well-being [5]. Black and non-Black Hispanic people share similarities, but also differ in ways that are impactful on their health, in particular their experience of racial discrimination [5]. In an analytical review, Cuevas et al. found there is consistent, although limited, evidence that Hispanic Black individuals have worse health than Hispanic White individuals in the U.S., including a higher prevalence of self-reported hypertension [9]. These differences highlight the diversity of experience and the unique needs of the Hispanic Black community that are typically grouped together in health disparities research. In this study, our aim was to evaluate differences in hypertensive disorders in pregnancy across non-Black Hispanic, non-Hispanic Black, and Hispanic Black individuals. Identifying any such differences is critical to the development of tailored interventions to reduce disparities in maternal morbidity and mortality.

Methods

This population-based cohort study utilized a California statewide dataset of vital records linked to hospitalization discharge data for the years 2007 to 2018 (Figure 1). Linkage was performed previously with 95% success for live births and 85% for stillbirths, using data provided by the California Department of Health Care Access and Information [10]. California implemented the U.S. Revised Certificate of Live Birth in 2007 and data were available through 2018 at the time of analysis. In the vital record data from live birth and fetal death certificates, race and ethnicity were self-reported by the person giving birth. We categorized people as Black or non-Black race, and Hispanic or non-Hispanic ethnicity, using the first reported race if multiple were given. In line with our study question, we restricted the study population to people who identified as Black and/or Hispanic. We further restricted analyses to records with complete information on variables of interest (race, ethnicity, and covariates). As our goal was to focus on Hispanic and Black populations, and specifically the Hispanic-Black population, our study did not include other racial and ethnic groups, including non-Hispanic White people, who are most frequently used as the reference group in health disparities research.

FIGURE 1. STUDY SELECTION DIAGRAM.

The outcome of interest was hypertensive disorders in pregnancy. This included: chronic hypertension, gestational hypertension, mild preeclampsia (no severe features), severe preeclampsia, HELLP Syndrome, and eclampsia. Using the International Classification of Diseases – Clinical Modification (ICD-CM), we identified all people who received the diagnosis codes associated with each disorder during their hospitalization for birth. (Diagnosis codes provided in Supplemental Table 1.) We assessed the disorders as a composite (any versus none) and combined them into three hierarchical categories based on their association and known progression of disease: (1) chronic hypertension alone, (2) mild preeclampsia or gestational hypertension, and (3) severe preeclampsia, HELLP Syndrome, or eclampsia (including preeclampsia or eclampsia superimposed on chronic hypertension). For example, gestational hypertension typically presents after twenty weeks of pregnancy in patients with documented normal blood pressure readings prior to that time. Gestational hypertension can progress to the multi-system disorder pre-eclampsia which is characterized by hypertension with one or more of the following pathologies: proteinuria, thrombocytopenia, renal insufficiency, or hepatic dysfunction [21]. Furthermore, we combined preeclampsia with severe features, HELLP Syndrome, and eclampsia into a single category based on shared characteristics involving high blood pressure and end-organ dysfunction whilst also considering the additional complication of seizure disorder in eclampsia.

We used prior evidence and causal diagrams to select confounding variables a priori,based on their known contribution to increasing the risk of hypertensive disorders and association with race and ethnicity. These variables included maternal age, nativity (born in the U.S. or not), expected method of payment for delivery (Medicaid or other government-sponsored insurance, commercial insurance, or self-pay/other), pre-pregnancy body mass index (BMI), parity (nulliparous or multiparous), cigarette smoking (if reported smoking in the 3 months prior to pregnancy), chronic renal disease, and pre-gestational diabetes mellitus [1, 14, 22, 30, 33].

We assessed the frequency of the covariates and hypertensive disorders in the full study population and across the three racial and ethnic groups. We then employed multivariable logistic regression models to estimate odds ratios (OR) with 95% confidence intervals (CI) for hypertensive disorders (any vs none) across the racial and ethnic groups, adjusting for the confounding variables. We then used multinomial logistic regression models to assess differences in the four-category outcome: no hypertensive disorder, chronic hypertension alone, mild preeclampsia or gestational hypertension, or severe preeclampsia or eclampsia. We used multinomial logistic regression because of the relatively rare outcomes and to accommodate the four-category outcome. All statistical analyses were performed using RStudio version 4.2.2, using the ‘nnet’ package for multinomial logistic regression modeling [37].

Results

This study included 3,048,558 individuals who gave birth in the state of California between 2007 to 2018 and identified as having one of three racial and ethnic identities: non-Black Hispanic (n = 2,729,585), non-Hispanic Black (n = 304,881) or Hispanic-Black (n = 14,092) (Figure 1). On average, individuals in the study gave birth at 38.4 (±2.9) weeks of gestation, were 27.5 (±6.2) years of age and had a pre-pregnancy body mass index (BMI) of 27.3 kg/m² (±6.3) (Table 1). Additionally, the majority were multiparous, non-smokers, and recipients of public insurance benefits for delivery. Non-Hispanic Black individuals had the highest rates of obesity class 2 or 3 (11.2% and 8.0%, respectively), compared non-Black Hispanic (7.0% and 4.0%) or Hispanic-Black (7.9% and 5.8%) individuals. Also of note, 5.6% of non-Hispanic Black individuals and 5.3% of Hispanic-Black individuals reported smoking prior to pregnancy as compared with 1.3% of non-Black Hispanic individuals. The percentage of individuals born in the U.S. was much lower (51.7%) in non-Black Hispanic individuals than in non-Hispanic Black or Hispanic-Black individuals (89.7% and 91.6%, respectively). The prevalence of chronic kidney disease was higher among both non-Black Hispanic (0.4%) and Hispanic-Black (0,3%) individuals as compared with non-Black Hispanic individuals (0.2%). The prevalence of type I or type II diabetes was 1.0-1.2% across the racial and ethnic groups.

TABLE 1.

CHARACTERISTICS OF STUDY PARTICIPANTS BY RACIAL AND ETHNIC GROUP

Individual-Level Variables	Total n = 3,048,558	Non-Black Hispanic n = 2,729,585	Non-Hispanic Black n = 304,881	Hispanic-Black n = 14,092
Gestational Age (wk), mean (SD)	38.4 (2.9)	38.5 (3.4)	38.0 (3.9)	38.4 (2.9)
Maternal Age (y) (SD)	27.5 (6.2)	27.5 (6.2)	27.4 (6.3)	25.8 (6.0)
U.S. Born	1,698,055 55.7%	1,411,803 51.7%	273,346 89.7%	12,906 91.6%
Pre-pregnancy BMI (kg/m2), mean (SD)	27.3 (6.3)	27.3 (6.1)	27.6 (7.2)	27.3 (7.0)
Underweight	81,565 2.7%	69,168 84.8%	11,773 3.9%	624 4.4%
Normal Weight	1,204,963 39.5%	1,079,133 2.5%	120,033 39.4%	5,797 41.1%
Overweight	909,579 29.8%	824,995 30.2%	80,950 26.6%	3,634 25.8%
Obesity Class 1	506,558 16.6%	456,952 16.7%	47,494 15.6%	2,112 15.0%
Obesity Class 2	215,987 7.1%	190,598 7.0%	24,282 11.2%	1,107 7.9%
Obesity Class 3	129,906 4.3%	108,739 4.0%	20,349 8.0%	818 5.8%
Type I or type II Diabetes	37,878 1.2%	33,951 1.2%	3,788 1.2%	139 1.0%
Chronic Kidney Disease	6,757 0.2%	5,636 0.2%	1,075 0.4%	46 0.3%
Nulliparous	1,053,212 34.5%	928,295 34.0%	118,662 38.9%	6,255 44.4%
Insurance Type
Commercial	979,328 32.1%	865,416 31.7%	108,791 35.7%	5,121 36.3%
Medicaid/Government	1,999,663 65.6%	1,801,002 66.0%	189,894 62.3%	8,767 62.2%
Self-pay/Other	69,567 2.3%	63,167 2.3%	6,196 2.0%	204 1.4%
Smoking Status1	52,334 1.7%	34,370 1.3%	17,224 5.6%	740 5.3%

¹Defined as Chronic Hypertension, Gestational Hypertension, Preeclampsia, or eclampsia.

²Defined has smoking ≥1 cigarette/day 3 months prior to pregnancy.

³p-values were calculated from t-test for all continuous variables (gestational age, age, BMI and birthweight) and chi-square for all categorical variables.

In the overall study population, the prevalence of any hypertensive disorder in pregnancy was 7.9%. When separated into categories, the prevalence of chronic hypertension alone was 0.9%, of mild preeclampsia or gestational hypertension was 4.7%, and of severe preeclampsia or eclampsia was 2.3%. When comparing racial and ethnic groups, the prevalence of any hypertensive disorder in pregnancy was highest in non-Hispanic Black people (12.3%), lower in Hispanic Black people (9.5%) and lowest in non-Black Hispanic people (7.4%) (Figure 2). This pattern of differences across the racial and ethnic groups was similar for each hypertensive disorder category.

FIGURE 2. INCIDENCE OF HYPERTENSIVE DISORDERS IN PREGNANCY BY RACIAL AND ETHNIC GROUP.

In the multivariable regression analysis, non-Hispanic Black people had the highest odds of having a hypertensive disorder in pregnancy (adjusted OR: 1.52; 95% CI: 1.50, 1.54), and Hispanic-Black people also exhibited higher odds (adjusted OR: 1.18; 95% CI: 1.11, 1.25) than non-Black Hispanic people (Table 2; Figure 3). Odds ratios were attenuated by adjustment for confounding variables. As an example interpretation, the odds of having a hypertensive disorder in pregnancy were 52% higher (95% CI: 50%, 54%) in non-Hispanic Black people than in non-Black Hispanic people, independent of differences in age, pre-pregnancy BMI, insurance status, smoking, parity, nativity, chronic renal disease, and diabetes.

FIGURE 3. ASSOCIATION BETWEEN RACIAL/ETHNIC GROUP AND HYPERTENSIVE DISORDER IN PREGNANCY.

OR, odds ratio; CI, confidence interval

Multivariable logistic regression models adjusted for age, chronic renal disease, diabetes, pre-pregnancy BMI, insurance status, smoking, parity, and US-born.

In the multinomial analysis of specific hypertensive disorders, we found that non-Hispanic Black people experienced the highest odds of all three disorders: chronic hypertension alone (adjusted OR: 2.35; 95% CI: 2.32, 2.38), preeclampsia without severe features or gestational hypertension (adjusted OR: 1.32; 95% CI: 1.30, 1.33), and preeclampsia with severe features or eclampsia (adjusted OR: 1.67; 95% CI: 1.64, 1.69), compared with non-Black Hispanic people (Table 3; Figure 4). Hispanic-Black people also experienced increased odds of each disorder compared with non-Black Hispanic people, but the differences were smaller in magnitude than for non-Hispanic Black people. Among the three hypertensive disorders examined, differences between the three racial/ethnic groups were larger for chronic hypertension alone than for non-severe preeclampsia/gestational hypertension or severe preeclampsia/eclampsia.

FIGURE 4. ASSOCIATION BETWEEN RACIAL/ETHNIC GROUP AND TYPES OF HYPERTENSIVE DISORDERS IN PREGNANCY.

OR, odds ratio; CI, confidence interval

Multivariable logistic regression models adjusted for age, chronic renal disease, diabetes, pre-pregnancy BMI, insurance status, smoking, parity, and US-born

Discussion

We found that hypertensive disorders in pregnancy differed between non-Hispanic Black, Hispanic-Black and non-Black Hispanic individuals in a California statewide study of births. For all hypertensive disorders examined, prevalence was highest in non-Hispanic Black individuals and lowest in non-Black Hispanic individuals. Notably, differences between the three racial/ethnic groups were considerably larger for chronic hypertension alone—a chronic health condition—than for preeclampsia or gestational hypertension—conditions with onset during pregnancy.

The primary drivers of these health disparities are systemic racism and social determinants of health, both of which are deeply rooted within the institutions of our country. The construct of race reinforces social hierarchies that confer privilege upon White-presenting individuals while marginalizing others [6]. This systemic oppression has long adversely impacted the overall health of the Black community, influencing a wide range of health outcomes, including hypertension [19]. This holds true for Afro-Latino or Hispanic-Black patients who face institutionalized racism in the U.S. and in their countries of origin [5, 7]. Rooted in a legacy of colonialism, Afro-Latinos encounter limited educational attainment, economic marginalization and limited healthcare access due to the color of their skin [26, 28].

In the United States, Hispanic-Black patients face the compounded burden of navigating anti-Hispanic bias as well as anti-Black discrimination both of which create significant barriers to equitable care [5]. In contrast, Hispanic individuals who self-identify or who are socially perceived as White report better health outcomes, illustrating how race and phenotypic presentation can provide specific advantages [9, 24] This contrast demonstrates how the intersectionality of race and ethnicity can intensify or mitigate health risks. These dynamics may further explain the intermediate risk of hypertensive disorders in pregnancy observed in Hispanic-Black individuals, who are more likely to encounter discrimination that compromises their socioeconomic status and access to quality health care [5, 9].

Analysis of these findings within the framework of the Healthy Immigrant Effect provides another lens through which we can examine why these differences exist. The Healthy Immigrant Effect suggests immigrants to the United States have superior health outcomes as compared to their native-born counterparts [4, 32]. However, these benefits attenuate over time as these groups assimilate and adopt the same health-related behaviors common in the U.S. [15]. The erosion of these advantages changes the long-term health trajectories of immigrants but may have a more drastic impact on certain racial and ethnic groups [13]. For Hispanic-Black patients, any initial advantages conveyed by the Health Immigrant Effect may rapidly deteriorate due to long term acculturation as well as increased exposure to persistent racial and ethnic discrimination [5].

We observed larger racial/ethnic disparities for chronic hypertension than for preeclampsia/gestational hypertension, which is consistent with prior studies (14, 20]. Both a recent study using a national sample and another using Georgia statewide data reported non-Hispanic Black versus non-Hispanic white pregnant individuals to have more than double the rate of chronic hypertension, but only an approximately 30% higher rate of preeclampsia [14, 20]. Multi-state data have also reported the rates of chronic hypertension and preeclampsia to be the same or lower in Hispanic individuals compared with non-Hispanic white individuals, which explains why we observed similar magnitude differences in our study that used non-Black Hispanic individuals as the reference group [14, 22, 30]. Research is needed to understand the consistent finding that disparities in pregnancy are more pronounced for chronic hypertension than preeclampsia. Chronic hypertension is a condition that begins before pregnancy, versus preeclampsia is specific to pregnancy. Therefore, causes that may differentially affect chronic hypertension could include insurance status prior to pregnancy (the vast majority of women in the U.S. qualify for Medicaid during pregnancy) and quality of and access to primary care (versus obstetric care, which tends to be much more intensive [36].

Upon separating Black-Hispanic people into their own distinct category, we observed their risk for hypertensive disorders in pregnancy to be intermediate, positioned between the risk profiles of non-Hispanic Black individuals and non-Black Hispanic individuals. While Hispanic-Black people may share sociocultural and linguistic similarities with the broader non-Black Hispanic population, their experiences as Black-presenting individuals creates a unique risk profile for both chronic and acute disease processes that separates them from the broader Latino/Hispanic population. Recognizing these distinctions can have far-reaching clinical implications and help reshape clinical practice [6].

In the future, physicians and public health officials must adopt both conventional as well as innovative approaches to address this issue. A first step would be improvement of data collection, by adding questions that effectively capture Afro-Latino or Hispanic-Black self-identification [5]. Additional changes should include the creation of culturally sensitive community-based education initiatives and improved screening protocols aimed at reducing the prevalence of hypertensive disorders of pregnancy within this population. Furthermore, national organizations can update current prenatal care guidelines to include frequent blood pressure monitoring for Black-Hispanic women while organizing public health campaigns that better reflect the lived experiences of Black-Hispanic birthing people.

Our study was subject to several limitations. We selected people into the study based on self-identification as a member of one of three groups: non-Black Hispanic, non-Hispanic Black and Hispanic-Black. Self-identification or self-reported race, while widely considered the gold standard in research, is influenced by not only cultural factors, but perceptions as well as societal pressures [3, 24]. Additionally, we recognize that our categorization of patients aggregates data in much the same way as current investigators despite our attempts to advance research surrounding this question. Within the Hispanic community, colorism shapes the lens through which Hispanic individuals may view themselves [26, 28]. This is further complicated by how Hispanic-Black individuals identify racially. A recent study found that three-in-ten Hispanic-Black respondents identified as White [27]. Twenty-five percent identified as Black, and twenty-three percent identified their race as other [27]. As a result, it is possible that individuals who are of African-descent from a Spanish-speaking country may be less likely to classify themselves as Black, subsequently, affecting the data collection [31]. The data used in this study are susceptible to various biases and limits the accuracy of empirical findings. Data on chronic conditions and pregnancy-related complications in hospitalization discharge data likely underreports the prevalence of these conditions [23]. We were also constrained by the use of individual-level characteristics which do not account for the role system-level factors play and may result in residual confounding.

In conclusion, we found that the risk of hypertensive disorders in pregnancy differed between non-Hispanic Black, Hispanic-Black and non-Black Hispanic people. These findings emphasize the importance of data collection that goes further than the use of broad racial categorizations. Like prior studies, our results demonstrate how current aggregation of race hinders scientific and public health innovation through its oversimplification of the interplay between culture, environment and the social determinants of health. By grouping all patients of Hispanic origin into a single group, the intermediate risk of hypertensive disorders in pregnancy faced by Black-Hispanic patients was virtually invisible. By overlooking these distinctions, we create further challenges for the populations most in need whether in the form of ineffective public health strategies, poor resource allocation or by perpetuating biases in the healthcare setting. Disaggregation allows for a nuanced examination of health disparities, moving beyond a superficial analysis of variation in the prevalence and severity of disease within and across ethnic groups. It is an important tool through which we can develop comprehensive strategies to build a more equitable healthcare system in our country.