Sexual Orientation-Related Nondiscrimination Laws and Maternal Hypertension Among Black and White U.S. Women

Abstract

Background:

Black women and sexual minority women are more likely to report adverse maternal health. Little research has investigated maternal health disparities at the intersection of race/ethnicity and sexual orientation or the mechanisms that contribute to these disparities.

Materials and Methods:

We analyzed data from the National Longitudinal Study of Adolescent to Adult Health. Our sample was restricted to Black and White women who had at least one live birth and were followed-up in Wave V of the data (n = 3,396). We used multivariable logistic regression to analyze the associations between race, sexual orientation identity, and a four-item state-level index of sexual orientation-related nondiscrimination laws.

Results:

We found that higher numbers of state-level sexual orientation-related nondiscrimination laws were associated with lower risk of maternal hypertension among U.S. women overall (odds ratio [OR] = 0.82, 95% confidence interval [CI] 0.73–0.93), and Black women had a higher risk of maternal hypertension relative to White women (OR = 1.32, 95% CI 1.00–1.79). Interactions between race, sexual orientation identity, and sexual orientation-related policies show that, regardless of sexual orientation identity, sexual orientation-related nondiscrimination laws were associated with a lower risk of maternal hypertension among White mothers (OR = 0.80, 95% CI 0.70–0.92). However, among Black women, these laws were associated with a lower risk of maternal hypertension among lesbian and bisexual women (OR = 0.18, 95% CI 0.05–0.68) only.

Conclusions:

Laws that prevent discrimination related to sexual orientation in various societal domains may play an important role in improving maternal health outcomes among White women in general and Black lesbian and bisexual women in particular.

Introduction

The United States lags significantly behind other high-income nations in infant and maternal health outcomes.¹ In fact, new data suggest that U.S. maternal mortality rates have actually increased in recent years.² In particular, research shows that, in the United States, maternal morbidities and mortality are disproportionately experienced by Black women.^3–5 A growing body of research has examined how discrimination at the interpersonal, institutional, and structural level negatively impacts Black women's reproductive health.^6–10 Less research has interrogated how race (as a social construct that is co-determined with racism) and racism intersect with other social positions and inequalities along which power, privilege, and resources are distributed, such as sexual orientation and heterosexism.

Among U.S. women, roughly 27.4% of households headed by a female–female couple have a child younger than 18 years living in the household.¹¹ While more difficult to calculate, current estimates suggest that 59% of self-identified bisexual women and 31% of self-identified lesbian women report having had a child.¹² And, Black sexual minority women (i.e., women who do not identify as exclusively heterosexual, report same-sex sexual attraction, and/or same-sex romantic or sexual relationships SMW) are more likely to be parents than White sexual minority women.¹¹

Only recently has research investigated obstetrical outcomes among sexual minority women, and only one study to date has examined how race and sexual orientation intersect to shape birth outcomes among U.S. women.¹³ This study found that Black bisexual and lesbian U.S. women were more likely to report low birth weight and preterm infants compared with both Black heterosexual women and White sexual minority U.S. women. Understanding the mechanisms that lead to such different profiles in obstetrical outcomes at the intersection of both race and racism and sexual orientation and heterosexism is therefore critical.

Discriminatory policies are one critical avenue through which health inequities are perpetuated in at least two ways. First, these structural factors may increase or decrease access to material resources via the legal recognition of same-sex marriage, protection against eviction due to race, or being fired from a job due to race or sexual orientation. Second, changes in social policies may serve as a signal to Black women and SMW that the State and the people it represents are more or less complicit in institutional and structural discrimination.^14–16 In particular, positive changes, such as the repeal of discriminatory policies or the passage of nondiscrimination state laws related to racism or heterosexism, may result in stress reduction among Black women and SMW, leading to better maternal health outcomes and potentially reshaping the intergenerational effects of discrimination on families.

Previous research has linked legal and policy changes related to structural and institutional racism, such as the repeal of Jim Crow laws and the passage of the Civil Rights Act of 1964, to positive maternal and child health outcomes among Black U.S. women.^6,9,17–19 Similarly, a new study found that lesbian women giving birth in states with higher numbers of sexual orientation-specific protective laws had a lower risk of preterm birth and higher birth weight infants than lesbian women living in states with fewer or no such laws.²⁰ However, how institutional and structural discrimination impacts birth outcomes among U.S. women at the intersection of race and sexual orientation remains unclear.

Sexual orientation, racism, and maternal health

Intersectionality—an analytic tool rooted in Black feminist thought and practice that centers the lived experiences of multiply marginalized women and the compounding and mutually constitutive power relations (e.g., racism, sexism, heterosexism, cissexism) and systems of oppression (e.g., white supremacy, patriarchy, settler colonialism) that shape them in social and historical context^21–24—is critical to understanding social and health inequities in U.S. society. While a small but growing literature has documented the ways in which institutional and structural racism have negatively impacted maternal morbidity and mortality among Black U.S. women,^7–9,25,26 less research has examined the impact of both institutional and structural racism and heterosexism on maternal health outcomes among U.S. women at diverse intersections of race and sexual orientation.

Nonetheless, existing research suggests that the impact of inequities related to having a minoritized sexual orientation identity on maternal and child health outcomes among U.S. women may function in opposite ways among White and Black SMW. For example, data on birth outcomes suggest that, on average, lesbian and bisexual women have better birth outcomes than heterosexual women among White women; however, this is not the case among Black women. Indeed, researchers found that—Black lesbian and bisexual women experienced worse outcomes than their heterosexual counterparts,¹³ even as Black SMW were less likely to report unplanned pregnancies than White SMW.²⁷ Moreover, qualitative research indicates that institutional and interpersonal racism and heterosexism have a compounding negative effect on patient–provider interactions in the context of reproductive health care among Black SMW.²⁸

Given the historically and contemporarily different experiences of Black and White people—both in broader U.S. society and within lesbian, gay, bisexual, transgender, queer, and other sexual and gender minority (LGBTQ+) communities^29,30—the impact of sexual orientation-related nondiscrimination laws on maternal health outcomes among U.S. women may vary in relation to not only sexual orientation but also race. Indeed, Black women in the United States face multiple levels of discrimination in various societal domains that both directly and indirectly negatively impact maternal health. In particular, Black women face barriers to achieving optimal maternal health, including a lack of access to high-quality health care, structural barriers to economic capital, and health care provider bias and discrimination^7,10,31,32—all of which may be compounded by interpersonal, institutional, and structural heterosexism in undermining the health of Black SMW relative to their White and heterosexual counterparts.¹³

Thus, the potential health benefits of laws and policies that protect and support SMW may therefore disproportionately benefit Black SMW who experience the brunt of the compounding negative effects of heterosexism and racism, which are mutually constituted and reinforcing, on their access to social, economic, political, and health care resources.²² In line with this, one study demonstrated that after the passage and enactment of civil union legislation in the state of Illinois, Black and Latina SMW experienced greater reductions in discrimination, stigma consciousness, and hazardous drinking behaviors than White SMW.³³

Alternatively, given the extensive structural and institutional racism-related barriers to achieving optimal maternal health among Black women, the positive impact of sexual orientation-related nondiscrimination laws on Black SMWs maternal health may be blunted or limited by the deleterious impact of racism, which would make it more difficult for Black SMW to benefit from any added resources that sexual orientation-related policy protections may provide compared with White SMW.

Thus, this study used data from the National Longitudinal Study of Adolescent to Adult Health to assess the relationship between four sexual orientation-related nondiscrimination U.S. state laws—including legal protections for same-sex marriage, allowances for same-sex adoption, sexual orientation-related employment nondiscrimination state law, and sexual orintation-related hate crime law—and maternal hypertension among Black and White U.S. women. Importantly, we assessed whether the impact of sexual orientation-related nondiscrimination state laws on maternal health varied at the intersection of sexual orientation and race, a historically contingent social construct that influences individuals' exposure to racism.

Materials and Methods

Data

Data come from the National Longitudinal Study of Adolescent to Adult Health (Add Health), a prospectively collected cohort study that began in fall 1994 and enrolled a nationally representative sample of U.S. adolescents, now followed over 15 years into young adulthood. The Add Health sample was drawn from 80 high schools and 52 middle schools throughout the United States, with unequal probabilities of selection.^34,35 In the first wave of the study, a large sample of students (N = 20,747) were asked to complete an in-depth home interview survey. High school seniors in Wave I were not selected for follow-up in Wave II but were reclaimed in Wave III (2001–2002).

Response rates were 79% for Wave I, 88% for Wave II, 77.4% for Wave III, 80.3% for Wave IV (2008–2009), and 69.3% in Wave V (2016–2018). We use data from Wave V to assess maternal health complications, as it was the first time these survey items were asked. We also use contextual data from Wave IV that links respondents' location at the time of interview to a series of four LGB-specific state policies. Wave V LGB policy variables are not currently available; however, Wave IV more likely represents the policy environments before birth. Our sample is restricted to Black or White women who reported at least one live birth at Wave V (n = 3,396). One-hundred thirty-four women were removed from the sample due to missing data on the dependent variable. These women were slightly at Wave V (37.8 vs. 38.2, p < 0.05) and Black (5.16 vs. 2.33, p < 0.001).

Measures

Dependent variable

At Wave V, women who had ever had a live birth were asked if they had ever experienced “high blood pressure or hypertension” during pregnancy. A dichotomous variable was created that captured whether a woman reported maternal hypertension (1 = yes) or not (0 = no).

Independent variables

Sexual identity was measured at Wave V using a variable that asked respondents to please choose the description that best fits how you think about yourself: exclusively heterosexual, mostly heterosexual, bisexual, or gay/lesbian. This variable was recoded into a dichotomous variable that captures whether respondents identify as exclusively or mostly heterosexual (referent) or bisexual/lesbian. Race was measured with a dichotomous variable that captures whether respondents identify as non-Hispanic White or non-Hispanic Black.

State-level sexual orientation-related nondiscrimination laws were derived from a supplement contextual file added in 2019 to the Add Health data set.³⁶ It captures whether at Wave IV of the survey respondents lived in a state that: (1) allowed same-sex marriage; (2) allowed same-sex adoption; (3) had an LGB nondiscrimination employment policy; and (4) had an LGB-specific hate crime statute. We created a summed scale that ranges from 0 (no policies) to 4 (all the protective policies.)

Control variables

We additionally adjusted for several other covariates related to maternal health including age at Wave V of the survey; age at first birth, parity, education at Wave V; income at Wave V; whether at Wave I their parent reported that they lived under the income to poverty ratio threshold (i.e., adolescent poverty); and relationship status at Wave V.

Methods

Initially, we present descriptive statistics for the total sample. We then conduct a series of logistic regression analyses that assess the impact of sexual orientation-related nondiscrimination laws on maternal health. We test interactions between sexual orientation-related nondiscrimination laws and sexual orientation identity and between these laws and race to see if sexual orientation identity or race moderates the association between sexual orientation-related nondiscrimination laws and maternal health among U.S. women. We then conducted a three-way interaction between race, sexual orientation identity, and sexual orientation-related nondiscrimination laws. We estimated predicted probabilities of reporting maternal hypertension and present the results of these three-way interaction in Figure 1 for clarity. We additionally conducted analyses stratified by race as a robustness check for our results. All models account for Add Health's complex survey design.

FIG. 1.

Predicted probability of maternal hypertension.

Results

Descriptive statistics

Among U.S. women who had given birth, 19% reported hypertension during pregnancy (see Table 1). At Wave IV, respondents lived in states that had on average almost two LGBT protective policies. Our sample is majority exclusively heterosexual or mostly heterosexual (97%) and 80% non-Hispanic White. The average age at first birth was 25 years. Eleven percent of the sample fell under the poverty threshold at Wave I, and at Wave V, 47% of the sample had at least a college degree. The majority of the sample was married at Wave V (66%); however, 12% were divorced, 3% were separated, and 19% had never been married.

Table 1.

Descriptive Statistics (n = 3,396)

	% (n)/M
Hypertension during pregnancy (Wave V)	19.08 (648)
LGBT policies (Wave IV)	1.71
Bisexual/lesbian (Wave V)	2.56 (87)
Race/ethnicity (Wave I)
White	80.4
Black	19.6
Age at first birth (multiple waves)
<18	6.57 (223)
≥18 to <24	32.24 (1,095)
≥24 to <30	29.39 (998)
≥30 to <35	17.26 (586)
≥35	5.77 (196)
Missing	8.78 (298)
Age (Wave V)	37.77
Parity (Wave V)	2.08
Adolescent poverty (Wave I)	11.04 (375)
Education (Wave V)
High school or less	13.02 (442)
Some college	40.40 (1,372)
College	21.94 (745)
Graduate training	24.65 (837)
Income (Wave V)
<$20,000	24.97 (848)
≥$20,000 to <$50,000	38.07 (1,293)
≥$50,000 to <$100,000	27.18 (923)
≥$100,000	9.78 (332)
Marital status (Wave V)
Married	65.61 (2,228)
Widowed	0.53 (18)
Divorced	12.28 (417)
Separated	3.00 (102)
Never married	18.58 (631)

Source: National Longitudinal Study of Adolescent to Adult Health, Waves I–V.

Multivariable logistic regression results

Table 2 presents our multivariable logistic regression results. All models adjust for race, maternal age at first birth, annual household income, educational attainment, marital status, and adolescent poverty. Results show that, in the total sample, a higher number of sexual orientation-related nondiscrimination laws were associated with lower odds of maternal hypertension among U.S. women overall (odds ratio [OR] = 0.84, 95% confidence interval [CI] 0.74–0.95), and Black women had greater odds of reporting a history of maternal hypertension relative to White women (OR = 1.59, 95% CI 1.20–2.11; Model 1).

Table 2.

Results from Multivariate Logistic Regressions for Hypertension During Pregnancy (n = 3,396)

	Model 1			Model 2			Model 3			Model 4
	OR	95% CI	p	OR	95% CI	p	OR	95% CI	p	OR	95% CI	p
LGB policies	0.84	0.74–0.95	**	0.81	0.70–0.93	**	0.84	0.73–0.95	**	0.81	0.71–0.94	**
Black (ref White)	1.59	1.20–2.11	**	1.44	1.04–2.01	*	1.59	1.20–2.12	***	1.51	1.08–2.11	*
Bisexual/lesbian (ref heterosexual)	1.19	0.59–2.42		1.20	0.59–2.44		1.18	0.49–2.85		1.66	0.58–4.75
LGB policies × Black				1.25	0.92–1.69					1.25	0.92–1.69
LGB policies × bisexual/lesbian							1.02	0.60–1.71		0.94	0.52–1.70
Bisexual/lesbian × Black										0.29	0.06–1.39
LGB policies × Black × bisexual/lesbian										0.19	0.05–0.78	*

All models adjusted for age, age at first birth, education, income, adolescent poverty, relationship status, and parity.

Source: National Longitudinal Study of Adolescent to Adult Health.

CI, confidence interval; OR, odds ratio.

We did not detect a statistically significant difference in the odds of reporting maternal hypertension among U.S. women by sexual orientation identity. The interaction between sexual orientation-related nondiscrimination laws and race was not statically significant (Model 2), nor was the interaction between these laws and sexual orientation identity (Model 3). However, the three-way interaction between race, sexual orientation identity, and laws was statistically significant (Model 4).

We present the results of the three-way interaction in Figure 1 via predicted probabilities for reporting maternal hypertension. The figure is stratified by race; however, the predicted probabilities were derived from a single model. The results from the interactions show that a higher number of laws are associated with a lower risk of maternal hypertension among White women, regardless of sexual orientation identity. Among Black women, we observed a significantly lower risk of maternal hypertension among lesbian and bisexual women—but not heterosexual women—living in states with a higher number of sexual orientation-related nondiscrimination laws.

We conducted a series of supplementary analyses as robustness checks for our results. First, we restricted our sample to women who did not move during the survey period. Our results were robust to this restriction. Second, we adjusted for hypertension measured via biomarkers at Wave IV of the survey. While we cannot guarantee that these measures preceded the pregnancy during which a woman experienced hypertension, they provide some insights into the factors that may drive our findings.

Our three-way interaction continued to be statistically significant even after adjusting for Wave IV hypertension. Finally, we created a composite measure of exposure to sexual orientation-related nondiscrimination laws during Waves III and IV of the survey to create a proxy measure of duration of exposure to these laws. Our results were robust to the inclusion of Wave III legal variables in our scale and showed a protective effect of sexual orientation-related nondiscrimination laws on maternal hypertension among all White women as well as a specific beneficial effect of these laws among Black lesbian and bisexual women.

Discussion

The results of this study add to a growing body of research that demonstrates the role of structural discrimination on women's health and expands it by examining the impact of structural heterosexism, measured using sexual orientation-related nondiscrimination laws, on maternal health in relation to both sexual orientation and race among Black and White U.S. women. Overall, the results show a net benefit of these laws on maternal hypertension among U.S. women overall. However, there are important differences in terms of their impact across racial groups. Among White women, regardless of sexual orientation identity, a greater number of sexual orientation-related nondiscrimination laws were associated with a lower risk of maternal hypertension.

This finding suggests that the presence of state laws that prevent heterosexism and promote LGBTQ+ rights may provide additional resources or better sociopolitical climates for White women, regardless of sexual orientation. Feminist theorists have long argued that regulation of sexuality is inherently tied to systems of gender oppression.^37–39 The extent to which state laws pertaining to heterosexism are part of a broader system of heteropatriarchy, which may negatively impact all women's health, is likely the reason behind this general benefit of LGB policies for White women overall, SMW, and non-SMW alike.⁴⁰

Among Black women, however, the benefits of sexual orientation-related nondiscrimination laws were concentrated among bisexual and lesbian women. This finding is line with other work showing that the benefits of inclusive sexual orientation-related laws and policies are more pronounced among SMW of color compared with White SMW.³³ This may be because, given the historical and ongoing social, economic, and political marginalization of Black women in the United States as a result of anti-Black racism within and across the education, employment, housing, health care, and other social systems,⁴¹ access to social institutions that may improve access to social, economic, political, and health care resources, such as same-sex marriage and sexual orientation-related employment nondiscrimination laws, may disproportionately improve the social well-being and subsequently the maternal health of Black SMW compared with Black heterosexual women.

There are several limitations to consider in the interpretation of our study results. First, the measure of maternal hypertension is a measure of “ever” having had this condition. Thus, we cannot guarantee that the sociopolitical climate at Wave IV, when the sexual orientation-related nondiscrimination state laws were measured, matches that at the time the woman gave birth. We did, however, conduct supplementary analyses restricted to individuals who did not move between Waves I and V and our results are robust to this specification. Second, because the maternal health indicator is a measure of “ever,” we are unable to ensure that covariates included in the analysis (e.g., education, relationship status) reflect those indicators at the time of pregnancy.

Third, we were unable to include other racial/ethnic groups due to sample size and statistical power limitations. Finally, there may be other state-level laws and policies that may confound the associations of interest but were not available in the data set. At this point in time, however, Add Health is the only population-level data set that includes maternal outcomes, race/ethnicity, sexual orientation, and sexual orientation-related nondiscrimination laws. More data sets should include measures of sexual orientation and heterosexism at multiple levels to help elucidate the occurrence and mechanisms of maternal health outcomes across and within diverse sexual orientation and racial/ethnic subgroups of U.S. women.

Despite these limitations, our results suggest that sexual orientation-related nondiscrimination laws, which may result in greater access to resources and more supportive sociopolitical climates for SMW, may help improve maternal health among White women, regardless of sexual orientation. Subsequently, our results show that these protective sexual orientation-specific laws have a potentially disproportionate positive effect among Black bisexual and lesbian women in particular who bear the brunt of the compounding effects of heterosexism and racism, which are mutually constitutive.²² Increasingly, researchers have drawn attention to the role that structural and institutional barriers play in shaping maternal health, especially among Black women, whose maternal health and well-being are undermined by racism at all levels of society.

Identifying the modifiable mechanisms that can be leveraged to promote maternal health among diverse groups of Black women, including Black SMW who remain understudied and underserved, is critical to achieving maternal health equity in the United States. The results presented in this article suggest that state laws that prevent discrimination related to sexual orientation in various societal domains may play an important role in improving maternal health outcomes for Black lesbian and bisexual women in particular. However, more laws, policies, regulations, and institutional practices are needed at the national, state, county, and local levels to prevent structural heterosexism and other forms of structural discrimination across all social systems and institutions and instead promote equity and justice at all levels of society to advance maternal health for Black SMW and other multiply marginalized groups in the United States.

Ethics Approval

This study was approved by the institutional review board at the University of Utah.

Disclaimer

The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health under Award No. R01HD091405 and by the University of Colorado Population Center (Grant No. R24 HD066613) through administrative and computing support. M.A. is supported by grant No. K01CA234226-01 awarded by the National Cancer Institute of the National Institutes of Health.