Sexual Identity and Birth Outcomes: A Focus on the Moderating Role of Race-ethnicity

Abstract

Race-ethnic disparities in birth outcomes are well-established, and new research suggests that there may also be important sexual identity disparities in birth weight and preterm birth. This study uses the National Longitudinal Study of Adolescent to Adult Health and is the first to examine disparities in birth outcomes at the intersection of race-ethnicity and sexual identity. We use OLS and logistic regression models with live births (n = 10,318) as the unit of analysis clustered on mother-ID (n = 5,105), allowing us to adjust for preconception and pregnancy-specific perinatal risk factors as well as neighborhood characteristics. Results show a striking reversal in the effect of lesbian or bisexual identity on birth outcomes across race-ethnicities: for white women, a bisexual or lesbian identity is associated with better birth outcomes than their white heterosexual counterparts, but for Black and Latina women, it is associated with worse birth outcomes than their heterosexual peers.

The United States (US) lags significantly behind other developed nations in infant outcomes (MacDorman et al. 2014; March of Dimes 2012). New data suggests that the rate of preterm births in the US has increased in recent years (March of Dimes 2017; Martin and Osterman 2018). The risk of preterm and low birth weight births, however, are not evenly distributed across the population; instead, they are concentrated among marginalized groups who experience discrimination and limited access to health resources across the life course, particularly among women of color (Blumenshine et al. 2010; Lu and Halfon 2003; Martin and Osterman 2018; Willinger, Ko, and Reddy 2009).

Sexual minority women (SMW) (e.g., women who do not identify as heterosexual or report same-sex attraction or sexual relationships) represent a marginalized population, whose maternal, infant, and child health outcomes are understudied. This is an important area of research given current estimates that suggest that 59% of self-identified bisexual women and 31% of self-identified lesbians report having had a child (Goldberg, Gartrell, and Gates 2014). Additionally, among women, roughly 27% of households headed by a female-female couple have a child under the age of 18 living in the household (Gates et al. 2007).

Only one study has used nationally representative data to examine the obstetrical outcomes of SMW (Everett et al. 2019). This study found that bisexual and lesbian women were more likely to report pregnancies ending in miscarriage and stillbirth, as well as live births ending in preterm and low birth weight infants compared to heterosexual-identified women (Everett et al. 2019). This study, however, did not answer whether these outcomes varied by race and ethnicity. Racial disparities in birth outcomes are well-established (Acevedo-Garcia, Soobader, and Berkman 2007; Dominguez 2008; Rosenthal and Lobel 2011; Willinger et al. 2009). No research has investigated whether birth outcomes vary at the intersection of sexual identity and race-ethnicity. Addressing this gap in the literature is essential substantively because such disparities must be documented in order to inform public health interventions. It also has theoretical implications as we apply an intersectional framework (Agénor 2020; Bowleg 2008; Crenshaw 1990) that acknowledges the historically different experiences between white women and women of color, both within the LGBT community and in the context of birthing. To do so, we use the National Longitudinal Study of Adolescent to Adult Health (Add Health) to examine the relationship between sexual identity, race-ethnicity, and multiple birth outcomes (e.g., birth weight, low birth weight, and preterm birth).

BACKGROUND

Sexual Identity, Stress, and Infant Health

The biopsychosocial framework emphasizes the role of maternal stress in shaping infant and child outcomes (Dunkel Schetter 2011; Entringer, Buss, and Wadhwa 2015). This framework builds upon studies using both human and animal models that have linked maternal stress to low birth weight, preterm birth, and infant mortality (DiPietro 2012; Dole et al. 2003). Exposures to stressors are not uniform across social groups. Minority stress theory argues that in addition to everyday stressors experienced by everyone, individuals with stigmatized identities experience additional stressors that contribute to adverse health outcomes (Meyer 2003). SMW are more likely to experience multiple forms of minority stress at the individual, interpersonal, and structural level (Hatzenbuehler 2009) linked to adverse health behaviors and conditions (Blosnich, Lee, and Horn 2013; Hughes 2011; Katz-Wise and Hyde 2012).

Moreover, a large body of work has found that SMW experience excess stress and discrimination, specifically in medical settings (Ayhan et al. 2020). Few studies, however, have examined sexual orientation disparities in maternal and infant health. One recent study found that in the year prior to a live birth, SMW were more likely to report multiple health behaviors and conditions that are associated with adverse infant outcomes, including drug, alcohol, and tobacco use, as well as unmet medical needs and poor self-rated health (Limburg et al. 2020). Another study found that pregnant SMW were more likely to report unmet medical needs, chronic conditions, depression, and health risk behaviors than pregnant heterosexual women (Gonzales, Quinones, and Attanasio 2019).

Research on sexual orientation disparities in infant outcomes has mostly relied on samples of women in same-sex relationships using fertility treatments. This research suggests that the experience of stigma and discrimination within health care settings is common for these women (Dahl et al. 2013; Hayman et al. 2013; Ross et al. 2012), but that there were no differences in infant outcomes (Nordqvist et al. 2014). These studies are far from generalizable to the larger population of SMW, given that SMW report higher rates of both mistimed and unwanted pregnancies relative to heterosexual women (Charlton et al. 2019; Everett, McCabe, and Hughes 2017). Further, unplanned pregnancies, for example, with male romantic partners, are likely to be fundamentally different than planned pregnancies in the context of same-sex relationships with the use of expensive medical interventions.

The most comprehensive study to date on sexual orientation disparities in birth outcomes used the 2006–2015 National Survey of Family Growth (NSFG) (Everett et al. 2019). These data showed that compared with heterosexual women, pregnancies reported by lesbian women were more likely to be low birth weight and preterm, and those reported by bisexual women more likely to be very low birth weight compared to heterosexual women, even after adjusting for maternal age, parity, race-ethnicity, and several indicators of socioeconomic status. This study, however, did not investigate how sexual identity and race-ethnicity may overlap to shape birth outcomes. This gap is problematic given that a significant body of work has documented that Black women in the US have higher rates of preterm and low birth weight births than white women (Martin et al. 2018). New studies have also shown that Latina women are more likely to report preterm and low birth weight births than white women (Acevedo-Garcia et al. 2007; Gemmill et al. 2019). Applying an intersectional framework to this research area may reveal important insights into how power and privilege function across various identities to influence birth outcomes.

Race-ethnicity, Sexual Identity, and Birth Outcomes

Intersectionality argues that different social statuses connect individuals to different sources of power and legitimacy (Crenshaw 1990). As a legacy of Black feminism, it posits that the experience of health inequality must be understood in relation to a complex and interacting system of domination, including, but not limited to, patriarchy, racism, and heterosexism (Bowleg 2012). In the case of sexual orientation and race-ethnicity, these two identities may produce different outcomes given the historically different experiences of white SMW and SMW of color, particularly Black SMW (Cornwell 1983; Moore 2006). For example, applying minority stress theory in the absence of an intersectional lens, one may assume that the excess stress experienced by women with a sexual minority identity, regardless of race-ethnicity, would result in an increased risk of adverse birth outcomes for SMW. However, this straightforward application of minority stress theory elides both the contemporary and historical racialized abuses experienced by women of color, specifically in the area of Black women’s reproduction (Owens 2017; Roberts 1999). Moreover, previous research has revealed critical historical differences in the lived experiences of white SMW and SMW of color including, but not limited to, the marginalization of SMW of color within the LGBT rights movement and organizations (Cornwell 1983; Ghabrial 2016), differences in norms around parenting and motherhood (Mamo 2007; Moore 2011), and differential access to social resources which may serve as protective resources against sources of minority stress (Crisp 2014; McLemore et al. 2018).

With few exceptions (Agénor et al. 2014, 2020), much of the research on SMW’s sexual and reproductive health has not paid attention to how race and sexual identity work together to shape health experiences. However, limited research in this area has sometimes revealed a reversal of racial disparities observed among heterosexual women. For example, using the NSFG, Agénor and colleagues (2014) found that Black lesbian women were more likely to receive contraceptive services in the past year than white lesbian women. Using Add Health data, Everett and colleagues (2020) found that Black and Latina bisexual and lesbian women were less likely to describe their pregnancies as unwanted than white bisexual and lesbian women, again, a reversal of racial-ethnic disparities documented among heterosexual women. Thus, it is not simply an additive relationship between race and sexual identity that may produce reproductive health-related outcomes, but rather something unique about the experience of being a SMW of color and the specific combination of racism and heterosexism as interlocking systems of oppression (Bowleg 2008).

In the context of reproductive health, racism has a particularly pernicious effect on birth outcomes and healthcare experiences. Increasingly, research has documented how racism (Dominguez 2008; Rosenthal and Lobel 2011; Slaughter-Acey et al. 2016; Slaughter-Acey, Caldwell, and Misra 2013), discrimination (Collins et al. 2008; McLemore et al. 2018), and discriminatory policies (Almond, Chay, and Greenstone 2006; Gemmill et al. 2019; Krieger et al. 2013; Novak, Geronimus, and Martinez-Cardoso 2017) negatively impact birth outcomes among women of color. This is especially true for Black women (Owens and Fett 2019; Rosenthal and Lobel 2011; Slaughter-Acey et al. 2016). Moreover, for some SMW, their race-ethnicity, rather than their sexual identity, may trigger discrimination in health care settings. For example, in their qualitative study of Black SMW, Agénor and colleagues (2015) found that for some of their participants, sexual orientation was not an issue for receiving competent sexual and reproductive healthcare but being Black negatively impacted their healthcare experiences. These findings highlight the importance of using an intersectional lens that acknowledges historical oppression of and medical abuses of Black women and other women of color (Owens and Fett 2019; Ross and Solinger 2017).

Indeed, health resources are transmitted intergenerationally; thus, even while a white SMW may experience stigma and discrimination as a sexual minority, she may be able to tap into social networks and familial resources that buffer against such discrimination. Several studies that have examined health behaviors within same-sex couples have found that women in same-sex relationships are more likely to engage in cooperative health behavior work (Reczek and Umberson 2012). Same-sex couples also provide more support for their partners seeking health care services (Umberson, Donnelly, and Pollitt 2018) and they are less likely to report illness-related disagreement and stress (Reczek et al. 2020). While not specifically about white women, these studies are comprised of samples that range from 80% to 93% white couples, suggesting that perhaps in the context of pregnancy, white same-sex couples can provide better social support than opposite-sex couples. It is unclear whether such benefits translate to SMW of color. The application of an intersectional framework in the area of SMW’s reproductive health is therefore imperative to understand how white privilege may be wielded in the context of pregnancy to possibly produce as good or better outcomes for white SMW than their white heterosexual peers.

This study addresses this critical gap in the literature in several important ways. First, we use Add Health data—a nationally representative, longitudinal survey—to explore sexual orientation disparities in birth outcomes with a particular focus on the moderating role of race-ethnicity. Building upon an extensive body of work on racial disparities in birth outcomes, the emerging literature on sexual orientation disparities in birth outcomes, and intersectionality theory, we hypothesize that SMW of color will report poorer birth outcomes than both heterosexual Black and Latina women and white SMW (Hypothesis 1).

Second, this study expands upon previous research by using longitudinal data to account for preconception and perinatal risk factors (e.g., prenatal care, self-rated health, sedentary behavior, unmet medical need, smoking during pregnancy, BMI, unwanted pregnancy) (CDC 2014; Johnson et al. 2006), as well as neighborhood-level characteristics (Ncube et al. 2016). Preconception risk factors have been an important line of inquiry into the pathways that lead to both sexual orientation and racial-ethnic disparities in adverse birth outcomes (Atrash et al. 2006; Denny et al. 2012; Limburg et al. 2020). Previous research has relied on cross-sectional data, which has made it difficult to account for baseline disparities in health status prior to the pregnancy. This is a significant limitation of previous studies, given that a large body of research has demonstrated both sexual orientation and racial-ethnic disparities in clinically relevant health statuses and behavior. We hypothesize that any observed disparities may, in part, be explained by preconception health status and pregnancy contexts (Hypothesis 2).

Finally, the Add Health data provides a unique opportunity to examine the role of neighborhood characteristics and if they account for any observed disparities. A large body of research has documented that neighborhood context is associated with birth outcomes (Ncube et al. 2016). Given the high rates of segregation in the US, it may be that white SMW and SMW of color live in different spaces and that these differences in neighborhood contexts may contribute to any observed disparities in our results, particularly those between white SMW and SMW of color (Hypothesis 3).

DATA AND METHODS

Data

Data came from the Add Health survey Waves I through V. The initial Add Health sample was drawn from 80 high schools and 52 middle schools throughout the US with unequal probabilities of selection (Harris 2013). A subsample of students (n = 20,747) was asked to complete additional in-home interviews and contacted for follow-up interviews between 2001–2002 (Wave III) and 2007–2008 (Wave IV). Between 2016–2018 (Wave V), respondents were contacted again for follow-up interviews. Response rates for this survey were 77.4% for Wave III, 80.3% for Wave IV, and 69.3% for Wave V. By Wave V, a total of 12,945 live, singleton births were reported by women with valid sample weights. Measures of sexual identity were not assessed until Wave III of the survey; thus, we excluded births that occurred before Wave III to ensure that we had a measure of sexual identity prior to the birth (n = 2,351). We also excluded individuals who were missing data for key covariates (n = 276). Our final total sample included 10,318 births to 5,105 women for preterm birth and 10,146 births to 5,062 women for birth weight.

Measures

Dependent variables.

Two dependent variables were drawn from birth rosters reported by women in Wave IV and Wave V of the Add Health survey. Birth weight was measured in kilograms and ranges from .45 kg to 6.18. While birth weight is not a clinical indicator, it provides useful information on infant health and has precedence in health research (Dooley and Prause 2005; Earnshaw et al. 2013; Strutz et al. 2014). We also conducted an additional analysis using low birth weight as a clinical cutoff. This measure was coded as a dichotomous variable using the clinical cutoff for low birth weight for live births where the infant was born <2,500 grams (1=yes, 0=no). Preterm birth was coded as a dichotomous variable that captured whether the infant was born <37 weeks in the pregnancy (1=yes, 0=no).

Independent variables.

Sexual identity was assessed in Waves III, IV, and V of the Add Health survey. Respondents were asked, “Please choose the description that that best fits how you think about yourself: 100% heterosexual (straight); mostly heterosexual (straight), but somewhat attracted to people of your own sex; bisexual, that is, attracted to men and women equally; mostly homosexual (gay), but somewhat attracted to people of the opposite sex; 100% homosexual (gay); or not sexually attracted to either males or females.” Using the date on which the survey was conducted and the date of the birth, we assigned mothers the sexual identity that most closely preceded the birth. No differences in our outcomes were detected between mostly heterosexual and exclusively heterosexual women; therefore, we use a dichotomous variable to capture whether respondents identified either as heterosexual prior to the birth (referent) or bisexual, mostly lesbian/gay, or exclusively lesbian/gay.

Race-ethnicity (non-Hispanic white [referent], non-Hispanic Black, Latina, and non-Latina other) was measured using self-reported race and ethnicity.

Sociodemographic characteristics.

We adjust for several sociodemographic characteristics previous research has shown to be related to birth outcomes. Maternal age was coded as a continuous measure that ranges from 14 to 41. Education level prior to birth was measured using data from all waves of the Add Health sample. Respondents were asked about the highest degree they have received at all waves, and in Waves IV and V, the year they received that degree. From these data, we were able to calculate the age at which they finished their degree, capturing whether, before the birth, the respondent had less than a high school degree, a high school or equivalent degree, a terminal Associate’s or other terminal trade-school degree (some college), or a bachelor’s degree or more. Household income is a categorical variable that measured respondents reported total household income closest to the time of birth as being <$15,000 (referent); >=$15,000 & <$30,000; >=$30,000 & <$75,000; >=$75,000 & <$100,000; >=$100,000; or missing. Adolescent poverty was measured as a dichotomous variable, which captured whether at Wave I, respondents were in households <100% of the federal poverty level. Foreign born was measured as a dummy variable that captures whether the respondent was foreign born (1=yes) or born in the US (0=no).

Pregnancy-specific measures.

Parity was measured as a dichotomous indicator of whether this was a first birth (0=no) or not a first birth (1=yes). Previous low birth weight infant was measured by creating an indicator that determined if, prior to the index birth, a woman reported a previous birth that met the clinical cutoff for a low-birth-weight infant. Relationship status was derived from a question that asked for each pregnancy how respondents characterized their relationship to their pregnancy partner at the time of birth. We created a categorical variable that measures relationship to pregnancy partner as married (referent), cohabitating, dating, just friends/acquaintances, or missing.

Preconception and perinatal risk factors.

All preconception risk factors were created using survey information that most closely preceded the date of birth for each live birth and were therefore drawn from multiple survey data waves. Respondents were asked if they received prenatal care during their pregnancy and at what week they first accessed care. From these two survey measures of prenatal care, we created a dichotomous indicator that captured whether they reported receiving prenatal care in the first trimester (1=yes) or not (0=no). Self-rated health was derived from a subjective measure of health in which respondents were asked to rate their general health on a five-point scale that ranged from excellent (1) to poor (5).

Sedentary behavior is a dichotomous indicator of activity measured at all waves derived from a reporting of how many times a respondent engaged in various activities during the past week. A respondent was coded as sedentary if they did engage in any of the reported activities 0–3 times in the past week (1=yes). Respondents were coded as non-sedentary if they reported engaging in activities more than three times in a given week (0). At Wave I, these activities included leisure activities (e.g., bicycling or roller-skating), playing an active sport (e.g., baseball, soccer, swimming, etc.), or engaged in exercise (e.g., walking, jogging, dancing, etc.). At Wave III and Wave IV, respondents were asked more detailed specifications about their activities, but these still aligned relatively well with engagement in sports, leisure activities, and purposeful exercise.

Unmet medical need was derived from a variable that asked respondents, “Has there been any time over the past year when you thought you should get medical care, but you did not?” Respondents who answered “yes” were coded as 1, and those who responded “no” were coded as 0. Smoked during pregnancy was derived from a question in which respondents were asked, “during this pregnancy, how many cigarettes did you smoke?” Respondents who reported cigarette smoking were coded as having smoked during pregnancy (1=yes), while those who reported no cigarette smoking were coded as none (0=no). BMI was measured using height and weight measurements completed by the survey interviewer and coded as a categorical variable that measures whether respondents were underweight (BMI<18.5), healthy weight (BMI >=18.5 and <25 [referent]), overweight (BMI>=25 and <30) or obese (BMI>=30). Unwanted pregnancy was measured with a dichotomous variable that captured whether, at the time of pregnancy, participants wanted to have a child (0) or not (1).

Neighborhood characteristics.

Multiple studies have found that neighborhoods influence birth outcomes (Ncube et al. 2016). Given the highly segregated nature of the United States (Wright et al. 2014), we adjust for five neighborhood-specific measures at the census tract-level capturing the percent of respondents at the census tract living in poverty, the percent that had college degrees, percent Black, percent Latina, and percent foreign born. These measures were coded as deciles that range from one to ten.

Analytic Approach

First, we present descriptive statistics for both the complete analytic sample and the sample stratified by race-ethnicity. We use chi-square to test for differences across race-ethnicity for all categorical variables and t-tests to compare mean scores between white and each race/ethnic minority group. Second, because we use births as our unit of analysis, we use OLS and logistic regressions clustered on mother-ID to account for non-independence among multiple births reported by a single woman. All models include an interaction between lesbian/bisexual identity and race-ethnicity. A series of supplementary analyses found that the interaction between a mostly heterosexual identity and race-ethnicity was not significant for any outcomes, and there were no significant differences between mostly heterosexual and exclusively heterosexual-identified women; thus, we have combined these two groups.

Further sensitivity analysis revealed no significant differences between bisexual-identified and lesbian-identified women. Thus, these variables were combined to increase statistical power. We used a model-building strategy that first assessed the relationship between sexual identity, race-ethnicity, and our interactions between the two (Model 1). Model 2 additionally adjusted for additional sociodemographic characteristics (e.g., maternal age, education, adolescent poverty, household income, parity, relationship status, foreign born). Model 3 included preconception and perinatal risk factors (e.g., prenatal care, self-rated health, smoking during pregnancy, BMI, unmet medical need, sedentary behavior, unwanted pregnancy), and Model 4 included all five neighborhood characteristics. We then used the “margins” command in Stata to generate predicted probabilities for all our birth outcomes by race-ethnicity and sexual identity. All other covariates for these plots were held at their means.

RESULTS

Descriptive Statistics

Table 1 presents the descriptive statistics for the total population and stratified by race-ethnicity. The majority of births reported were to women who identified as exclusively or mostly heterosexual, with 4.2% of births reported by women who reported a bisexual or lesbian identity. The mean birth weight was 3.31 kg, 7.9% of births were reported as meeting the clinical cutoff for low birth weight, and 12.4% were reported preterm. The mean maternal age at birth was 26.2 years.

Table 1.

Descriptive Statistics for Total Population and by Race/Ethnicity

	Total Population		White		Black		Latina		Other
	N=10,318		N=5,900		N=2,209		N=1,558		N=651
	M/%	SE	M/%	SE	M/%	SE	M/%	SE	M/%	SE
Sexual Identity ***
Heterosexual	95.78		95.43		96.67		95.86		97.98
Lesbian/Bisexual	4.22		4.57		3.33		4.14		2.02
Maternal Age	26.22	.20	26.72	.23	24.3***	.29	25.77*	.37	26.92	.52
Education
LT High School	13.27		11.22		18.44		18.24		12.25
High School Graduate	47.16		44.5		55.79		53.13		38.61
Some College	15.39		15.45		14.36		14.5		21.73
College Graduate	24.18		28.82		11.4		14.13		27.41
HH Income ***
<$15,000	15.86		15.15		23.39		11.39		9.96
>=$15,000 & <$30,000	15.22		15.04		17.57		11.18		21.15
>=$30,000 & <$75,000	38.29		40.11		28.72		44.02		28.41
>=$75,000 & <$100,000	9.54		10.88		4.19		8.61		11.49
>=$100,000	8.53		9.64		3.17		6.78		17.6
Missing	12.57		9.18		22.96		18.03		11.4
Adolescent Poverty***	15.94		10.04		31.97		28.44		14.44
Foreign Born ***	4.76		1.33		1.06		20.67		34.2
Parity (M)	2.76	.03	2.69	.03	2.99***	.08	2.8	.11	2.61	.08
Previous Low Birthweight Infant***	4.75		4.17		7.41		4.04		6.03
Relationship Status ***
Married	55.15		63.95		21.43		51.56		53.6
Cohabitating	22.07		19.73		26.84		26.02		31.58
Dating	16.80		11.68		37.78		18.62		12.41
Not in romantic relationship	5.98		4.64		13.95		3.8		2.41
Missing	<1		.02		.27		.06		0
Prenatal Care, 1st Trimester***	93.15		93.15		85.11		88.38		90.91
Fair/Poor Self-Rated Health, Prior*	7.66		6.44		10.08		11.8		6.11
Sedentary, Prior*	17.22		16.14		22.68		16.48		17.22
Unmet Medical Need, Prior*	21.61		20.11		24.63		23.57		30.03
Smoked During Pregnancy ***	15.68		19.59		5.62		8.47		11.14
BMI, Prior ***
Underweight	5.54		5.76		4.03		4.66		11.19
Healthy Weight	47.43		50.54		36.13		42.12		57.31
Overweight	21.50		20.45		25.37		23.95		15.65
Obese	22.16		20.35		29.71		24.93		12.85
Missing	3.38		2.89		4.77		4.34		3
Unwanted Pregnancy ***	36.05		31.9		54.88		34.88		32.8
Neighborhood Characteristics, Prior (Tract, Deciles)
% Poverty	5.45	.17	4.81	.17	7.64***	.18	6.32***	.28	4.85	.41
% College Graduates	5.08	.17	5.34	.21	4.17***	.23	4.53*	.26	6.15*	.35
% Black	5.01	.20	4.11	.18	8.71***	.14	5.09***	.2	5.28**	.38
% Latina	4.56	.20	4.03	.19	3.77	.31	8.01***	.25	6.51***	.37
% Foreign Born	4.59	.18	4.1	.16	3.75	.26	7.78***	.3	7.1***	.38
Low Birthweight***	7.85		6.72		13.03		7.37		7.35
Birth weight	3.31	.01	3.34	.01	3.14***	.03	3.3	.03	3.27	.06
Preterm birth**	12.38		12.41		12.94		11.99		10.38

Source: National Longitudinal Study of Adolescent to Adult Health

Notes:

^*p < .05,

^**p < .01,

^***p < .001

M= mean; SE = standard error; LT = less than; HH = household

Several differences in our covariates of interest were detected across race-ethnic groups. First, there was a higher prevalence of births reported by bisexual and lesbian women among white (4.6%) and Latina (4.1%) women compared to Black (3.3%) women. Maternal age was lower among Black and Latina women than white women, and Black and Latina women were more likely to have grown up in poverty. Differences in several preconception and perinatal risk factors were also detected across race-ethnicity: Black and Latina women had a lower prevalence of receiving prenatal care in the first trimester. Black and Latina women had a higher prevalence of reporting fair or poor self-rated health and unmet medical needs but were less likely to report smoking during their pregnancy than white women. In line with other studies, the results show lower mean birth weights among Black women and a higher prevalence of low-birth-weight infants than white women.

Multivariate Models

Table 2 presents the results for preterm birth. Model 1 shows that white lesbian or bisexual identified women have a decreased risk of preterm birth (OR = .20, p < .01), however bisexual/lesbian Black (OR = 11.74, p < .001) and Latina (OR = 6.52, p < .05) women experienced an increased risk of preterm birth. This trend held even after the inclusion of sociodemographic characteristics, preconception and perinatal risk factors, and neighborhood characteristics. In fact, the disparity grew after the inclusion of all risk factors in Model 4 for both Black (OR = 12.04, p < .001) and Latina (OR = 7.43, p < .05) bisexual and gay women. Figure 1 presents the predicted probabilities (based on Model 4) and shows the interactive effect of race-ethnicity and sexual identity on preterm birth.

Table 2.

Multivariate Results from Logistic Regression for Preterm Birth

	Model 1		Model 2		Model 3		Model 4
	OR	(95% CI)	OR	(95% CI)	OR	(95% CI)	OR	(95% CI)
Lesbian/Bisexual Identity	.20**	(.07 – .57)	.19**	(.07 – .57)	.19**	(.06 – .56)	.20**	(.07 – .59)
Race-Ethnicity (Ref: White)
Black	1.09	(.82 – 1.46)	1.04	(.79 – 1.37)	1.09	(.83 – 1.43)	1.06	(.76 – 1.48)
Latina	.97	(.72 – 1.29)	.98	(.73 – 1.30)	.94	(.70 – 1.28)	1.00	(.72 – 1.39)
Other	.93	(.62 – 1.40)	1.01	(.64 – 1.59)	1.05	(.66 – 1.67)	1.12	(.70 – 1.80)
Sexual Identity * Race-Ethnicity
Lesbian/Bisexual * Black	11.74***	(2.86 – 48.10)	13.07***	(3.22 – 53.05)	12.07***	(3.06 – 47.66)	12.04***	(3.01 – 48.18)
Lesbian/Bisexual * Latina	6.52*	(1.10 – 38.46)	8.25*	(1.29 – 52.74)	7.71*	(1.16 – 51.08)	7.43*	(1.12 – 49.12)
Lesbian/Bisexual * Other
			.98*	(.95 – 1.00)	.97**	(.94 – .99)	.97*	(.95 – 1.00)
Maternal Age
Education			1.05	(.74 – 1.48)	1.16	(.82 – 1.63)	1.18	(.83 – 1.67)
LT High School			1.04	(.71 – 1.53)	1.22	(.83 – 1.80)	1.24	(.84 – 1.84)
Some College			.78	(.52 – 1.18)	.97	(.64 – 1.47)	1.00	(.65 – 1.53)
College Graduate
HH Income (Ref: <$15,000)
>=$15,000 & <$30,000			.98	(.70 – 1.37)	1.01	(.72 – 1.41)	.97	(.69 – 1.37)
>=$30,000 & <$75,000			1.04	(.78 – 1.38)	1.10	(.83 – 1.45)	1.10	(.82 – 1.47)
>=$75,000 & <$100,000			1.01	(.69 – 1.47)	1.08	(.74 – 1.58)	1.10	(.75 – 1.62)
>=$100,000			1.00	(.67 – 1.50)	1.05	(.69 – 1.58)	1.09	(.72 – 1.66)
Missing			1.15	(.80 – 1.64)	1.16	(.82 – 1.66)	1.16	(.81 – 1.66)
Adolescent Poverty			.97	(.74 – 1.26)	.94	(.72 – 1.23)	.94	(.72 – 1.22)
Foreign Born			.71	(.44 – 1.14)	.74	(.46 – 1.19)	.77	(.48 – 1.25)
Parity			1.01	(.94 – 1.08)	1.01	(.94 – 1.08)	1.00	(.93 – 1.07)
Previous Low Birthweight Infant			3.54***	(2.62 – 4.79)	3.52***	(2.60 – 4.78)	3.59***	(2.64 – 4.88)
Relationship Status (Prior) (Ref: Married)
Cohabitating			.99	(.80 – 1.23)	.99	(.80 – 1.24)	1.03	(.82 – 1.29)
Dating			.59***	(.45 – .78)	.61***	(.46 – .82)	.63**	(.47 – .84)
Not in romantic relationship			.74	(.50 – 1.09)	.73	(.48 – 1.10)	.73	(.48 – 1.12)
Missing			2.05	(.23– 18.19)	1.77	(.21 –15.19)	1.86	(.22 –15.48)
Prenatal Care, 1st Trimester					.74*	(.56 – .99)	.72*	(.54 – .96)
Fair/Poor Self–Rated Health (Prior)					1.64***	(1.24 – 2.17)	1.64***	(1.24 – 2.16)
Sedentary (Prior)					.82	(.66 – 1.03)	.83	(.66 – 1.03)
Unmet Medical Need (Prior)					1.18	(.97 – 1.44)	1.18	(.96 – 1.44)
Smoked During Pregnancy					1.31*	(1.01 – 1.71)	1.29	(.99 – 1.68)
BMI (Prior) (Ref: Healthy Weight)
Underweight					.99	(.68 – 1.44)	.97	(.66 – 1.41)
Overweight					.96	(.77 – 1.20)	.96	(.76 – 1.20)
Obese					1.28*	(1.02 – 1.60)	1.22	(.97 – 1.54)
Missing					1.23	(.64 – 2.38)	1.22	(.62 – 2.42)
Unwanted Pregnancy					.77**	(.63 – .94)	.77*	(.63 – .94)
Neighborhood Characteristics, Tracts and Deciles (Prior)
% Poverty							1.01	(.97 – 1.05)
% College Graduates							1.00	(.96 – 1.04)
% Black							1.00	(.96 – 1.04)
% Hispanic							1.02	(.98 – 1.07)
% Foreign Born							.96	(.91 – 1.01)
Constant	.14***	(.12 – .15)	.26**	(.12 – .58)	.34*	(.14 – .82)	.33*	(.13 – .86)

Source: National Longitudinal Study of Adolescent to Adult Health

Notes:

^*p < .05,

^**p < .01,

^***p <.001; n = 10,318

LT = less than; HH = household

Figure 1. Predicted Probability of Reporting a Preterm Birth by Race-ethnicity.

Source: National Longitudinal Study of Adolescent to Adult Health

Notes: n = 10,318

Table 3 presents the results for birth weight in Models 1–4 and an additional model using the clinical cutoff for a low-birth-weight infant adjusting for all risk factors (Model 5). In Model 1, white bisexual and lesbian women had higher birth weights (B = 0.18, p < .01), and while there were lower birth weights among Black (B = –0.23, p < .001) and Latina (B = –0.07, p < .05) women, the interactions between race-ethnicity and sexual identity were not significant. This trend persisted in Model 2; however, in Model 3, adjusting for preconception and perinatal risk factors functioned as a suppressor effect, such that Black SMW reported significantly lower birth weights (B = –0.33, p < .05). This finding held after adjusting for neighborhood characteristics in Model 4. The predicted birth weights in Model 4 are presented in Figure 2 and show that the mean predicted birth weight for white SMW is 3.53 kg compared to 3.03 kg among Black women, which translates into roughly a one-pound average difference in birth weights.

Table 3.

Multivariate Results from OLS and Logistic Regression for Birth Weight

	Model 1		Model 2		Model 3		Model 4		Model 5 (Low Birthweight Cutoff)
	B	(95% CI)	B	(95% CI)	B	(95% CI)	B	(95% CI)	OR	(95% CI)
Lesbian/Bisexual Identity	0.18**	(0.07 – 0.30)	0.21***	(0.10 – 0.32)	0.23***	(0.11 – 0.34)	0.20***	(0.09 – 0.32)	0.09**	(0.02 – 0.52)
Race/Ethnicity (Ref: White)
Black	−0.23***	(−0.29 − −0.17)	−0.18***	(−0.23 − −0.13)	−0.24***	(−0.29 − −0.18)	−0.19***	(−0.25 − −0.13)	1.91***	(1.38 – 2.66)
Latina	−0.07*	(−0.13 − −0.01)	−0.06	(−0.12 – 0.00)	−0.09**	(−0.15 − −0.04)	−0.11***	(−0.18 − −0.05)	1.13	(−0.18 − −0.05)
Other	−0.10*	(−0.19 − −0.01)	−0.10*	(−0.18 − −0.01)	−0.10*	(−0.18 − −0.01)	−0.11*	(−0.20 − −0.02)	1.06	(0.67 – 1.68)
Sexual Identity * Race/Ethnicity
Lesbian/Bisexual * Black	−0.22	(−0.50 – 0.07)	−0.26	(−0.54 – 0.02)	−0.33*	(−0.60 − −0.07)	−0.33*	(−0.59 − −0.07)	13.33*	(1.63 – 109.24)
Lesbian/Bisexual * Latina	−0.11	(−0.50 – 0.28)	−0.16	(−0.56 – 0.23)	−0.18	(−0.58 – 0.23)	−0.15	(−0.55 – 0.25)	32.78**	(3.53 – 304.01)
Lesbian/Bisexual * Other	−0.15	(−0.52 – 0.22)	−0.13	(−0.47 – 0.20)	−0.11	(−0.50 – 0.27)	−0.09	(−0.49 – 0.32)	---	---
Maternal Age			0.00	(−0.00 – 0.01)	0.00	(−0.00 – 0.01)	0.00	(−0.00 – 0.00)	0.99	(0.96 – 1.02)
Education
LT High School			0.04	(−0.02 – 0.10)	0.02	(−0.04 – 0.07)	0.01	(−0.04 – 0.07)	1.13	(0.83 – 1.54)
Some College			0.06	(−0.01 – 0.13)	0.02	(−0.05 – 0.09)	0.02	(−0.05 – 0.09)	1.20	(0.79 – 1.81)
College Graduate			0.09*	(0.01 – 0.16)	0.05	(−0.03 – 0.12)	0.05	(−0.03 – 0.12)	1.04	(0.65 – 1.66)
HH Income (Ref: <$15,000)
>=$15,000 & <$30,000			−0.01	(−0.08 – 0.06)	−0.01	(−0.08 – 0.05)	−0.01	(−0.08 – 0.05)	0.85	(0.65 – 1.66)
>=$30,000 & <$75,000			0.03	(−0.02 – 0.08)	0.02	(−0.03 – 0.07)	0.02	(−0.03 – 0.07)	0.71*	(0.52 – 0.96)
>=$75,000 & <$100,000			0.04	(−0.03 – 0.12)	0.04	(−0.03 – 0.11)	0.03	(−0.04 – 0.11)	0.70	(0.46 – 1.06)
>=$100,000			−0.02	(−0.08 – 0.05)	−0.02	(−0.08 – 0.05)	−0.02	(−0.09 – 0.05)	0.66	(0.42 – 1.03)
Missing			0.01	(−0.05 – 0.08)	0.02	(−0.04 – 0.08)	0.02	(−0.04 – 0.08)	0.79
Adolescent Poverty			−0.02	(−0.07 – 0.04)	−0.01	(−0.06 – 0.04)	−0.01	(−0.06 – 0.05)	0.96	(0.72 – 1.28)
Parity			0.03***	(0.01 – 0.04)	0.03***	(0.01 – 0.04)	0.03***	(0.01 – 0.04)	0.83***	(0.75 – 0.92)
Previous Low Birthweight Infant			−0.47***	(−0.56 − −0.39)	−0.45***	(−0.54 − −0.36)	−0.45***	(−0.53 − −0.36)	5.97***	(4.37 – 8.15)
Relationship Status (Prior) (Ref: Married)
Cohabitating			−0.06**	(−0.11 − −0.02)	−0.05*	(−0.10 − −0.00)	−0.06*	(−0.10 − −0.01)	1.29*	(1.00 – 1.66)
Dating			−0.02	(−0.07 – 0.03)	−0.01	(−0.06 – 0.04)	−0.01	(−0.06 – 0.04)	0.87	(0.62 – 1.21)
Not in romantic relationship			−0.03	(−0.10 – 0.03)	−0.02	(−0.09 – 0.05)	−0.02	(−0.09 – 0.05)	0.93	(0.61 – 1.42)
Missing			−0.01	(−0.20 – 0.19)	−0.03	(−0.19 – 0.14)	−0.05	(−0.25 – 0.16)
Foreign Born			0.05	(−0.04 – 0.15)	0.04	(−0.06 – 0.14)	0.04	(−0.06 – 0.14)	0.99	(0.60 – 1.63)
Prenatal Care, 1st Trimester					0.00	(−0.06 – 0.06)	0.00	(−0.06 – 0.07)	0.70*	(0.52 – 0.94)
Fair/Poor Self-Rated Health (Prior)					0.02	(−0.05 – 0.08)	0.01	(−0.05 – 0.08)	1.26	(0.90 – 1.76)
Sedentary (Prior)					0.01	(−0.03 – 0.05)	0.02	(−0.03 – 0.06)	0.80	(0.62 – 1.03)
Unmet Medical Need (Prior)					0.01	(−0.03 – 0.05)	0.01	(−0.03 – 0.05)	0.95	(0.75 – 1.21)
Smoked During Pregnancy					−0.21***	(−0.26 − −0.16)	−0.21***	(−0.26 − −0.16)	1.55**	(1.19 – 2.03)
BMI (Prior) (Ref: Normal Weight)
Underweight					−0.09**	(−0.15 − −0.03)	−0.09**	(−0.15 − −0.03)	1.11	(0.74 – 1.65)
Overweight					0.08***	(0.04 – 0.13)	0.08***	(0.04 – 0.12)	0.81	(0.62 – 1.06)
Obese					0.07**	(0.02 – 0.11)	0.07**	(0.03 – 0.12)	1.05	(0.79 – 1.38)
Missing					0.02	(−0.10 – 0.14)	0.02	(−0.10 – 0.14)	1.32	(0.79 – 2.19)
Pregnancy Unintended					0.05*	(0.01 – 0.09)	0.04*	(0.01 – 0.08)	0.91	(0.72 – 1.14)
Neighborhood Characteristics, Tracts and Deciles (Prior)
% Poverty							0.00	(−0.00 – 0.01)	0.96	(0.91 – 1.01)
% College Graduates							0.00	(−0.00 – 0.01)	0.97	(0.92 – 1.02)
% Black							−0.01**	(−0.02 − −0.00)	1.03	(0.99 – 1.08)
% Hispanic							0.01*	(0.00 – 0.02)	0.98	(0.93 – 1.03)
% Foreign Born							0.00	(−0.01 – 0.01)	1.03	(0.97 – 1.08)
Constant	3.36***	(3.34 – 3.38)	3.21***	(3.08 – 3.34)	3.26***	(3.11 – 3.40)	3.25***	(3.09 – 3.41)	0.27**	(0.10 – 0.70)

Source: National Longitudinal Study of Adolescent to Adult Health

Notes:

^*p<.05,

^**p<.01,

^***p<.001

Figure 2. Predicted Birth Weight of an Infant by Race-ethnicity.

Source: National Longitudinal Study of Adolescent to Adult Health

Notes: n = 10,146

We included an additional analysis using the clinical cutoff for low birth weight (Model 5). The results show a similar trend; white SMW have a significantly lower risk of reporting a low-birth-weight infant (OR = 0.09, p < .01), while Black (OR = 13.33, p < .05) and Latina (OR = 32.78, p < .01) SMW have a greater risk of a low-birth-weight infant. Due to the small sample sizes in these categories, the confidence intervals are quite large and should be interpreted cautiously.

We conducted a series of additional analyses to test the robustness of our results. Because sexual identities have been shown to shift over time (Ott et al. 2011), we analyzed our results using both sexual identity reported closest to the birth and an indicator of whether a respondent had ever identified as bisexual or lesbian. Our results yield similar results across all measurement strategies: white SMW report better birth outcomes than their heterosexual peers, and SMW of color report worse birth outcomes than their heterosexual peers.

DISCUSSION

Our results present the first estimates of birth outcomes at the intersection of race-ethnicity and sexual identity. They show a reversal in the impact of a bisexual or lesbian identity on birth outcomes, depending on the woman’s race-ethnicity. For white women, a lesbian or bisexual identity largely serves as a birth benefit, insofar as it is associated with an increase in birth weight and a decrease in the risk of preterm birth. However, for Black and Latina women, a lesbian or bisexual identity is associated with an additional risk for adverse birth outcomes. While we hypothesized that we would observe the greatest risk for adverse birth outcomes among SMW of color, we did not expect to see such a striking benefit for white SMW. The disparities we observed were not explained by preconception or perinatal risk factors (Hypothesis 2) or neighborhood characteristics (Hypothesis 3). These results highlight the need for an intersectional lens in health disparities research and the importance of incorporating sexual identity in conversations about maternal and infant health and broader conversations around reproductive justice.

The general paucity of research on SMW’s maternal and infant health is a significant barrier to understanding these results; however, the growing body of work on differences between white SMW and SMW of color’s health and exposure to minority stress provides some insights. First, only two studies have examined obstetrical outcomes among SMW. The first was drawn from clinical samples of women using reproductive services and found no difference in the birth outcomes of same-sex and other-sex couples (Nordqvist et al. 2014) and the second did not examine the moderating impact of race-ethnicity (Everett et al. 2019). The population of women with same-sex partners using assisted reproductive technologies is likely to be qualitatively different from the general population of SMW who have male partners or use means to achieve pregnancy outside of often cost-prohibitive fertility clinics. Moreover, these pregnancies are likely to be planned and intentional. Several studies have been published recently showing that for many SMW pregnancy is unplanned and with a male partner, even for women who may currently identify as a lesbian (Charlton et al. 2019; Everett et al. 2017). Thus, it is not surprising that we found somewhat different results in our study.

While the US has made some progress in terms of LGB rights, it appears that whatever benefits have accrued to LGB women in the area of pregnancy, these benefits do not appear to extend to women of color. Instead, the results suggest a form of double jeopardy for bisexual and lesbian Black and Latina women that likely reflect how systemic racism and homophobia interact to impact pregnant SMW. The results here are not simply an additive negative effect since, for white women, bisexual/lesbian identity was associated with a benefit. This finding may be demonstrative of how racial privilege operates in the medical landscape, even for women who may experience stigma in other ways. Even though research has found that women in same-sex relationships experience homophobia while pregnant (Hayman et al. 2013; Ross 2005), given the diversity of pregnancy contexts in our sample, it may be that many of the SMW in our sample were not perceived as such by providers limiting their experiences of stigma while pregnant. Additionally, white SMW may activate other axes of privilege in the context of their pregnancies while receiving care.

Prior research has revealed various ways the LGTBQ movement has historically centered whiteness in research (Ghabrial and Ross 2018) and marginalized sexual minority people of color from accessing relevant resources (Cornwell 1983; Ghabrial 2016). Regardless of sexual orientation, white women are able to access white privilege, not just in heterosexual spaces and communities but also within the LGBTQ community. In some ways, this may give white SMW access to a variety of resources during pregnancy that buffer against stigma and sources of minority stress. Studies that feature majority-white samples have found that women in same-sex couples provide more health-related support for each other than in heterosexual couples (Reczek et al. 2020; Reczek and Umberson 2012; Umberson et al. 2018). It is unclear whether similar patterns are present in the relationships of SMW of color, and even so, whether this social support would be enough to combat the adverse effects of systemic racism on birth outcomes.

It’s important to note that these disparities do not appear to be due to socioeconomic differences between white and non-white SMW. Supplementary analyses of the socioeconomic indicators included in this study showed smaller racial-ethnic gaps in education and income among SMW compared to the gaps observed among heterosexual women. Some research has found that higher socioeconomic status is associated with more enacted and anticipated stigma among sexual minorities of color but less among white sexual minorities (Shangani et al. 2020). This finding further suggests that racism and discrimination function differently at the intersection of race-ethnicity and sexual orientation to produce different birth outcomes among SMW.

In the case of pregnancy, the crisis of Black maternal health has garnered national attention (Rabin 2019). Studies have repeatedly found that racial discrimination and its associated excess stress has a cumulative effect that negatively affects birth outcomes (Wynn 2019). Our results support this finding but also suggest that a bisexual or lesbian identity adds to this cumulative stress exposure. Much of the recent work in the area of Black women’s reproductive health has used a Reproductive Justice framework that demands researchers pay attention to the historical and systemic trauma experienced by Black women and other women of color within the medical community (Dominguez 2008; Ross and Solinger 2017; Valdez and Deomampo 2019; Wynn 2019). This work highlights how distrust of the medical community has been fostered by decades of exploitation and abuse by medical providers, particularly regarding women of color’s reproductive lives. It should be noted that LGBTQ populations also have an extensive history of medical mistrust and abuse by providers. However, the explicit focus on women of color’s reproduction in the medical community may mean that SMW of color experience unique forms of discrimination from providers and distrust the medical community.

The fact that our results were robust to using an indicator of ever having identified as bisexual or lesbian suggest that women of color whose sexual minority status identification may have occurred after the birth of a child face additional challenges to healthy birth outcomes. Sexual identity shifts are associated with increased depressive symptoms (Everett 2015; Everett et al. 2016) and substance use behaviors (Fish and Pasley 2015), both of which may contribute to poorer birth outcomes. It may be that women who do not identify as bisexual or lesbian before the time of birth still experience forms of minority stress—such as internalized homophobia, stigma consciousness, or fear of rejection—that, in particular, for women of color may negatively impact their birth outcomes.

This study has several limitations that should be taken into consideration. First, we have a relatively small sample of lesbian and bisexual women who report pregnancies in the study. Few other nationally representative data sets that collect data on multiple indicators of maternal, infant, or child health also collect data on sexual identity, and none of these data sets are longitudinal. Thus, even though our sample size is relatively small, it is the best available data source for this study. More data sets, including the Pregnancy Risk Assessment Monitoring System (PRAMS), should incorporate sexual identity measures into their data set. Second, we could not determine the effect of partner gender on our outcomes because of the small sample sizes. Only 40 women were matched to female partners through the relationship rosters available in Add Health. Of these, the majority identify as heterosexual, suggesting that there may be a lack of clarity in how the survey item was presented to the participants.

We also rely on self-reported birth outcome data, which may result in some inaccuracies in our data. It is important to note, though, that other research has found women accurately report their infants’ birth weights and gestational ages (Dietz et al. 2014). We do not have reason to believe that these differences would systematically vary by sexual identity. Another indicator of socioeconomic status and the intendedness of the pregnancy would be the use of assisted reproductive technology services, but this question was not asked. To qualify for our sample, respondents had to have responded to the Wave III survey. While our sample did not differ from the original survey of Add Health respondents in terms of Wave I socioeconomic status, our sample had a higher proportion of white respondents compared to the Wave I sample and a lower proportion of Black respondents. No differences were detected in the proportion of Latina respondents. Loss to follow-up among women is often driven by socioeconomic factors and other stress-related factors (Powers et al. 2015). This means that our results for Black SMW’s birth outcomes are likely a conservative estimate. Finally, Add Health does not include measures of gender identity. Like SMW, trans and non-binary individuals, especially those of color, may also be at greater risk of adverse health experiences related to pregnancy. Future research should aim to investigate these potential disparities.

These results provide the first estimates of preterm and low birth weight risk among US women at the intersection of race and sexual identity despite these limitations. Our results add to a growing call for national efforts to address maternal health, and in particular, to growing conversations around reproductive justice and women of color’s ability to have and raise healthy children. Our results demonstrate that SMW of color face additional risks for adverse birth outcomes—it is likely that experiences of systemic racism are exacerbated by homophobia, particularly in clinical settings. It is imperative that efforts are made to include issues of sexual identity into discussions of reproductive justice and that any discussions of sexual orientation disparities in reproductive health incorporate issues of racism.

AUTHOR BIOS

Bethany Everett is associate professor of sociology and adjunct associate professor in the Department of Obstetrics and Gynecology at the University of Utah. Dr. Everett’s research focuses on the social determinants of health disparities with a specific focus on sexual orientation disparities in women’s health.

Aubrey Limburg is a Ph.D. Candidate in sociology and an affiliate of both the Health and Society Program and the CU Population Center at the Institute of Behavioral Science at the University of Colorado Boulder. Her research focuses on various aspects of health inequality and population health with special emphasis on gender and sexuality, reproductive health, aging, and mortality.

Brittany M. Charlton, ScD, is an assistant professor at Boston Children’s Hospital, Harvard Medical School, Brigham and Women’s Hospital, and the Harvard T.H. Chan School of Public Health. She is also Co-Director of the Harvard Sexual Orientation Gender Identity and Expression (SOGIE) Health Equity Research Collaborative. Dr. Charlton’s epidemiological research primarily focuses on sexual orientation-related disparities in reproductive health.

Jae Downing is an assistant professor in the School of Public Health at Oregon Health and Science University – Portland State University. Their research focuses on the health of transgender populations and parenthood among gender and sexual minorities.

Phoenix A. Matthews is a professor and associate dean for Equity and Inclusion in the College of Nursing at the University of Illinois at Chicago. Dr. Matthews’ research focuses on cancer prevention and control in underserved populations with a particular focus on smoking as a modifiable risk factor for cancer inequalities.