Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Mar 1.
Published in final edited form as: Womens Health Issues. 2021 Nov 10;32(2):147–155. doi: 10.1016/j.whi.2021.10.002

Experiences of Discrimination and Depression Trajectories over Pregnancy

Amanda Noroña-Zhou a,b,c,*, Özlü Aran c,**, Sarah E Garcia c, Dustin Haraden d, Sarah ED Perzow c, Catherine H Demers b,c, Ella-Marie P Hennessey c, Stephanie Melgar Donis e, Melanie Kurtz c, Benjamin L Hankin d, Elysia Poggi Davis c,f
PMCID: PMC9701536  NIHMSID: NIHMS1848790  PMID: 34774402

Abstract

Introduction:

Research on risk factors for prenatal depression is critical to improve the understanding, prevention, and treatment of women’s psychopathology. The current study examines the relation between experiences of racial discrimination and trajectories of depression symptoms over the course of pregnancy.

Method:

Participants completed standardized measures regarding symptoms of depression at four timepoints during pregnancy and reported on experiences of racial discrimination at one timepoint. Latent growth curve modeling was used to examine the relation between discrimination and initial levels (intercept) and trajectories (slope) of depression symptoms over pregnancy.

Results:

Participants were 129 pregnant individuals recruited from obstetric clinics and oversampled for elevated depression symptoms. Thirty-six percent of the participants were living at or below 200% of the federal poverty line. Fifty-four percent of the sample identified as non-Latinx White, 26% as Latinx, and 13% as non-Latinx Black. An unconditional latent growth curve modeling revealed a negative quadratic trajectory of depression symptoms during pregnancy. When women’s report of discrimination was added as a predictor of depression trajectories, discrimination predicted the initial value (intercept) of depression symptoms, but not change over the course of pregnancy (slope). Specifically, higher levels of experiences of discrimination were associated with higher levels of depression symptoms. When sociodemographic and contextual covariates were included in the model, a low family income-to-needs ratio was also related to higher levels of depression symptoms.

Conclusions:

These findings provide evidence that women’s experiences of racial discrimination and family financial strain are risk factors for prenatal depression, with implications for screening, treatment, and policy.

Depression is one of the most common prenatal complications, with up to 25% of pregnant women reporting elevated symptoms of depression (Hobfoll, Ritter, Lavin, Hulsizer, & Cameron, 1995; Marcus, Flynn, Blow, & Barry, 2003; Meaney, 2018) and 8.5%–11.0% meeting diagnostic criteria for depression (Gaynes et al., 2005) during pregnancy. In addition to causing significant concurrent impairment and distress and increasing risk for postpartum depression (Kim, Hur, Kim, Oh, & Shin, 2008), depression during pregnancy also constitutes a risk for child development (Capron et al., 2015; Davis et al., 2007; Plant, Pariante, Sharp, & Pawlby, 2015; Sandman, Buss, Head, & Davis, 2015). Given the prevalence of prenatal depression and its health sequelae for both birthing parents and their children, prenatal depression is a major public health issue. The prevalence of prenatal depression is exacerbated in the presence of adverse contextual factors, such as low socioeconomic status (Marcus et al., 2003) and experiences of discrimination (EOD) (Giurgescu et al., 2020), and these factors are hypothesized to contribute to existing disparities in prenatal health. For example, non-Latinx Black and American Indian/Alaska Native individuals are two to three times more likely to suffer pregnancy-related death compared with non-Latinx White individuals (Centers for Disease and Control and Prevention, 2019). Similarly, the rates of preterm birth are two to three times higher among Black individuals compared with White individuals (Goldenberg, Culhane, Iams, & Romero, 2008). These disparities have persisted for decades and not declined (Centers for Disease and Control and Prevention, 2020). Recent findings highlight that experiences of racial discrimination “get under the skin” and affect biological pathways associated with maternal mood and child health outcomes, such as inflammation (Sluiter at al., 2020). Therefore, studying discrimination at this developmental stage is especially important to understand the complex interactions between sociocultural environment and prenatal health. The current study examines the relation between discrimination and trajectories of depression symptoms over the course of pregnancy.

Not all women are equally likely to experience depression symptoms during pregnancy. In particular, cross-sectional studies show that Black and Latinx individuals report higher symptoms of depression during pregnancy than their non-Latinx White counterparts (Rich-Edwards et al., 2006; Mukherjee, Trepka, Pierre-Victor, Bahelah, & Avent, 2016). EOD likely play a role in racial and ethnic disparities in prenatal mental health (Canady, Bullen, Holzman, Broman, & Tian, 2008; Rosenthal et al., 2015). Pregnant women of color who have experienced higher levels of discrimination are more likely to experience prenatal depression than individuals who report no EOD (Bécares & Atatoa-Carr, 2016). Furthermore, risk for prenatal depression symptoms increases with the number of settings in which a woman has experienced discrimination (e.g., in the health care system, at work, getting housing; Bécares & Atatoa-Carr, 2016; Ertel et al., 2012) in addition to the frequency of experienced discrimination (Walker, Ruiz, Chinn, Marti, & Ricks, 2012).

A limitation of prior studies is the reliance on cross-sectional assessment of prenatal depression symptoms (e.g., Bécares & Atatoa-Carr, 2016; Canady et al., 2008; Walker et al, 2012), which does not allow for examination of whether discrimination affects change in depression symptom levels over the prenatal period. Maternal mental health is dynamic over the course of pregnancy (Glynn, Schetter, Hobel, & Sandman, 2008; Kane, Dunkel Schetter, Glynn, Hobel, & Sandman, 2014), and changes in prenatal depression might be due to differential social and environmental factors (Meijer et al., 2014); thus, there is a clear need for longitudinal investigation. One longitudinal study found that EOD predicted increases in depression symptoms at ensuing time points in a sample of low-income pregnant adolescents of color (Rosenthal et al., 2015). The current study, therefore, aims to examine how EOD relates to depression symptoms during pregnancy in a socioeconomically, racially, and ethnically diverse sample of adult women.

Aim of the Current Study

There has been a call for research on risk factors for prenatal depression, such as discrimination, within diverse populations in order to understand, prevent, and treat maternal psychopathology more effectively and efficaciously (Kuhlman, Urizar, Robles, Yim, & Dunkel Schetter, 2019). The current study aims to examine associations between maternal EOD and depression symptoms across pregnancy. Specifically, using repeated measurement of depression symptoms over the course of pregnancy in a diverse sample of women, we investigate whether discrimination experiences are associated with levels of depression symptoms when first assessed in the study or their trajectories over time across the prenatal period.

Methods

Participants

Participants included 129 women from the non-treatment arm of an ongoing randomized controlled trial investigating the impact of reducing prenatal maternal depression on infant developmental outcomes (Care Project, R01 MH109662, Davis et al., 2018). Participants were oversampled for recruitment of women with elevated depression symptoms. Participants identified1 as 53.5% non–Latinx-White, 25.6% Latinx, 13.2% non–Latinx-Black, 4.7% non–Latinx-Asian, 1.6% non–Latinx-American Indian, and 1.6% other race/ethnicity. Mean household income was $77,282 (SD = $58,644; range = $0–$315,000), with 36% of women living at or below 200% of the federal poverty line. Almost all participants had at least a high school degree, and 59% had completed a bachelor’s degree or higher. Women were on average 30 years old (SD = 5.3; range = 20–41 years) and 82% reported cohabiting with a spouse or partner. In terms of parity, 35% of the sample was primiparous; 36% of the participants reported one previous delivery, 11% reported two, and 12% reported three or more (Table 1).

Table 1.

Demographic Characteristics of the Sample

Mean (SD) or Percentage
Age, years 30.4 (5.3)
Income $77,282.3 (58,643.8)
Race and ethnicity
 Latinx 25.6%
 Non-Latinx–American Indian 1.6%
 Non-Latinx–Asian 4.7%
 Non-Latinx–Black 13.2%
 Non-Latinx–White 53.5%
 Other 1.6%
Education
 Less than high school 2.3%
 High school and higher 38.7%
 College degree 35.7%
 Graduate degree 23.3%
Cohabitating 82.2%
Married 65.9%
Parity
 Primiparous 35%
 One 36%
 Two 11%
 Three or more 12%
Open in a new tab

Procedures

Participants were recruited from obstetric clinics in the Denver metropolitan area. Initial eligibility criteria were maternal age between 18 and 45 years, singleton pregnancy, English speaking, and gestational age of less than 25 weeks. Exclusion criteria were current illicit drug use, current methadone use, major health conditions involving invasive treatments, past or current psychosis or mania, and/or current participation in cognitive behavioral therapy or interpersonal therapy.

Depression symptoms were measured at four time points during pregnancy (time 1: M = 16.50 (SD = 4.54); time 2: M = 21.89 (SD = 6.53); time 3: M = 25.74 (SD = 6.22); and time 4: M = 28.80 (SD = 4.23) weeks of gestation), and EOD were assessed once at the last prenatal assessment. All study procedures were approved by the Colorado Multiple Institutional Review Board and the University of Denver Institutional Review Board for Protection of Human Subjects. All participants provided written informed consent. The data collection period was from August 2017 to January 2020.

Measures

Sociodemographic characteristics

Participants self-reported sociodemographic characteristics, including age, household income, marital status, race, and ethnicity at the first study visit. An income-to-needs ratio was calculated by dividing the total reported household income by the poverty threshold corresponding to the number of persons living in the household at the time of assessment, specified by the U.S. Census Bureau (2020).

EOD

Participants reported experiences of racial/ethnic discrimination using the EOD Scale (Krieger & Sidney, 1996). The current study used the count score of the EOD, which sums the number of settings (e.g., work, school, health care) where participants have experienced racial/ethnic discrimination during their lifetime. Final sum scores range from 0 to 9, with higher scores indicating more settings at which racial discrimination is experienced. The EOD has shown adequate test–retest reliability (r = 0.69) and good internal consistency (α > 0.75) in previous studies (Krieger, Chen, Waterman, Kiang, & Feldman, 2005). In terms of construct validity, scores on the EOD correlate with scores on other established discrimination measures (e.g., Everyday Discrimination Scale; Williams, Yu, Jackson, & Anderson, 1997) and with measures of general psychological distress (Krieger et al., 2005). Internal consistency for the EOD was good (α = 0.80) in this sample.

Edinburgh Postnatal Depression Scale

Maternal depression symptoms were assessed using the 10-item Edinburgh Postnatal Depression Scale (EPDS) (Cox, Holden, & Sagovsky, 1987). The EPDS is a reliable and valid measure of depression symptoms in pregnancy and is widely used in the screening of maternal depression across the peri-partum period (Bergink et al., 2011). Participants provided ratings on a 4-point scale regarding depression symptoms (e.g., I have felt sad or miserable) over the past week. Final sum scores range from 0 to 30, with higher scores indicating greater levels of depression. Previous studies demonstrated both good internal consistency (α > 0.80) and validity of the EPDS (Cox et al., 1987). The internal consistency in this sample was good to excellent for all timepoints (M α = 0.91, SD = 0.02).

Data Analytic Plan

The data analytic plan of the present study was pre-registered on the Open Science Framework (https://osf.io/y2dhj). Analyses were conducted using the lavaan package for Structural Equation Modeling in R (Rosseel, 2012; R Core Team, 2013). Additional packages included tidyverse (Wickham et al., 2019), haven (Wickham & Miller, 2020), psych (Revelle, 2020), sjPlot (Lüdecke, 2020), broom (Robinson, Hayes, & Couch, 2020), and data table (Dowle & Srinivasan, 2020).

All data were missing completely at random or missing at random across the four timepoints. Full information maximum likelihood was used to estimate missing data.

Covariates

Key sociodemographic factors including family income-to-needs ratio, race, ethnicity, age, and marital status were considered as covariates and included in the adjusted latent growth curve model.

Growth curve models

To identify the longitudinal trajectory of self-reported symptoms of depression during the prenatal period, we fit a series of latent growth curve models to the EPDS data in several steps. To properly identify the best-fitting growth model, we examined convergence across multiple fit indices including the comparative fit index (CFI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR), following recommendations from Hu and Bentler (1999). Specifically, we considered good fit as RMSEA of less than 0.06, SRMR of 0.08 or less, and a CFI of 0.95 or more, and acceptable fit as a RMSEA of 0.08 or less, an SRMR of 0.10 or less, and a CFI of 0.90 or more. We also considered fit indices using the Akaike information criteria and Bayesian information criteria, with lower values indicating better fit. When examining model fit statistics, we prioritized convergence across fit indices as opposed to relying on a single fit measure (Barrett, 2007; Kenny, Kaniskan, & McCoach, 2015).

To determine the shape of depression symptom trajectories over pregnancy, we first fit an unconditional means (no-growth) model to the data; next we included a linear slope, and then a quadratic slope, using fit statistics to determine the best fit shape (intercept only, linear slope, and/or quadratic slope). Next, to evaluate whether women’s EOD were associated with the initial levels or trajectories in symptoms of depression, our primary analyses regressed the intercept, linear slope, and quadratic slope onto women’s self-reported EOD. In our secondary analysis, we computed an adjusted model that accounts for the aforementioned sociodemographic covariates in the relation between discrimination and depression trajectories.

Results

Descriptive Statistics

Means, standard deviations, and intercorrelations of the EPDS, EOD, and demographic variables are reported in Table 2. EOD were positively correlated with depression symptoms at 17 (r = 0.17; p < .05) and 29 (r = 0.18; p < .05) weeks of gestation. Women who were lower in income-to-needs ratio reported higher scores on the EOD (r = −0.28; p < .01) and more depression symptoms at all time points (all p < .05). Non-Latinx Black participants reported the highest scores on the EOD (M = 4.19, SD = 2.83), followed by Latinx individuals (M = 2.36, SD = 2.39) and non-Latinx White participants (M = 0.53, SD = 1.13).

Table 2.

Descriptive Statistics and Correlations among Study Variables

Mean (SD)/%,
Range		 1 2 3 4 5 6 7
1. EPDS at 17 weeks of gestation 7.16 (5.84), 0–26
2. EPDS at 22 weeks of gestation 6.12 (5.94), 0–25 0.82***
3. EPDS at 26 weeks of gestation 5.82 (5.77), 0–25 0.76*** 0.85***
4. EPDS at 29 weeks of gestation 5.75 (5.58), 0–23 0.78*** 0 77*** 0.83***
5. EOD 1.66 (2.26), 0–9 0.17* 0.10 0.12 0.18*
6. Age 30.42 (5.28), 20–41 0.03 0.12 0.03 0.03 −0.04
7. INR 412.56 (366.38), 25–2509 −0.30** −0.25** −0.27** −0.23* −28** 0.26**
8. Marital statusa 66% −0.21* −0.24** −0.17 −0.18* −0.26** 0.17 0.47***
Open in a new tab

Abbreviations: EPDS, Edinburgh Postnatal Depression Scale; EOD, experiences of discrimination; INR, income-to-needs ratio.

a

1 = married, 0 = not married.

*

p < .05

**

p < .01

***

p < .001.

Change in Depression Symptoms over the Prenatal Period

First, an unconditional means model, which assumes no change in depression symptoms during pregnancy, fit the data poorly (See Table 3 for fit indices of all compared models.). Next, the introduction of a linear slope, which tests change in symptoms along a linear trajectory, resulted in an improved and acceptable model fit to the data. Because the introduction of a quadratic slope to the model resulted in a Heywood case (negative variance in a single manifest variable), the variance was fixed to zero. This quadratic model resulted in excellent fit and included estimates for an intercept (B = 7.16; p < .001), a linear slope (B = −0.91; p = .02), and a quadratic slope (B = 0.16; p = .20) (Table 4).

Table 3.

Fit Statistics for Unconditional Growth Models

No Growth Linear Growth Quadratic Growth
CFI 0.94 0.97 1
SRMR 0.08 0.05 0.01
RMSEA 0.17 0.15 <0.01
AIC 2698.73 2687.1 2673.97
BIC 2715.89 2712.84 2708.28
χ2 37.16 19.54 0.4
df 8 5 2
p Value <.001 <.001 .82
Open in a new tab

Abbreviations: AIC, Akaike information criterion; BIC, Bayesian information criterion; CFI, comparative fit index; df, degrees of freedom; RMSEA, root mean square error of approximation; SRMR, standardized root mean squared residual.

Table 4.

Quadratic Growth Estimates for the Final Unconditional Model

Estimate Std. Error Std. Est. p Value Variance p Value
Intercept 7.163 0.513 1.23 <.001 33.888 <.001
Linear slope −0.914 0.392 −0.246 .02 13.756 <.001
Quadratic slope 0.159 0.123 0.15 .197 1.127 <.001
Open in a new tab

Because the quadratic model yielded the best fit indices, we applied this model for all subsequent analyses described below. These initial results suggest that, without accounting for any predictors or covariates, depression symptoms decreased throughout pregnancy in a quadratic (curvilinear) fashion (Figure 1). In addition, the intercept as well as linear and quadratic slopes in the quadratic model had levels of variance significantly different from zero, suggesting individual differences among the initial levels and trajectories of change of depression symptoms during pregnancy (Table 4).

Figure 1.

Figure 1.

Open in a new tab

Average depression trajectory over pregnancy.

Discrimination and Depression Symptoms during Pregnancy

Fit statistics

The primary model testing whether women’s EOD were related to individual differences in the intercept, linear slope, and quadratic slope had excellent fit (CFI = 1.00, SRMR = 0.01, RMSEA = 0.00).

Regression results

EOD were related to the intercept (B = .45; p = .05) but not linear or quadratic trajectories of prenatal symptoms of depression (Table 5). Thus, women who reported more EOD had higher depression symptoms at 17 gestational weeks, and these elevated levels persisted throughout pregnancy (Figure 2). EOD explained 5% of the variance for the intercept of depression symptoms.

Table 5.

Discrimination and Intercept and Slope of Depression over Pregnancy

Intercept Linear Slope Quadratic
Slope
B SE B SE B SE
Intercept 6.286** 0.646 −0.873* 0.457 0.142 0.145
Regression coefficient
Racial discrimination 0.446* 0.231 −0.199 0.168 0.062 0.053
Open in a new tab
*

p < .05

**

p < .01.

Figure 2.

Figure 2.

Open in a new tab

Experiences of racial discrimination (EOD) predicting depression trajectories over pregnancy. Experiences of discrimination was analyzed as a continuous variable and here we present estimated growth trajectories for depression symptoms at the mean ± 1 standard deviation of experiences of discrimination scores within the current sample.

Discrimination, Sociodemographic Factors, and Depression Symptoms during Pregnancy

Fit statistics

In the secondary model, the following covariates were added to the initial model: race (Black or not), ethnicity (Latinx or not), age, marital status (married or not) and income-to-needs ratio. This model revealed excellent fit (CFI = 1.00, SRMR = 0.01, RMSEA = 0.02).

Regression results

Income-to-needs ratio (B = −0.004; p = .01) and age (B = 0.21; p = .04) predicted the intercept of depression symptoms and these elevated levels persisted throughout pregnancy. Discrimination did not remain a significant predictor of the intercept with these other sociodemographic factors in the model (B = 0.22; p = .39). None of the demographic variables predicted linear or quadratic depression trajectories. As such, neither Black race nor Latinx ethnicity had a significant relation with intercept, linear slope, or quadratic slope. Hence, older pregnant women and those with lower income-to-needs ratio reported higher depression symptoms at 17 weeks of gestation, and these higher symptoms were maintained across pregnancy (Table 6). Income-to-needs ratio and age explained 10.67% of the variance for the intercept of depression symptoms.

Table 6.

Discrimination, Sociodemographic Factors, and Depression Trajectories over Pregnancy

Intercept Linear Slope Quadratic
Slope
B SE B SE B SE
Intercept 3.251 3.305 −02.127 2.498 0.458 0.792
Regression coefficients
Racial discrimination 0.224 0.261 0.010 0.198 0.021 0.063
Family INR −00.004* 0.002 −00.001 0.001 0.001 0.000
Blacka −01.331 1.808 −02.591 1.353 0.609 0.429
Latinxb 1.132 1.391 −01.553 1.050 0.523 0.331
Age 0.213* 0.105 0.060 0.079 −00.020 0.025
Marital statusc −01.770 1.295 0.432 0.966 −00.147 0.304
Open in a new tab

Abbreviation: INR, = income-to-needs ratio.

a

1 = Black, 0 = non-Black

b

1 = Latinx, 0 = non-Latinx

c

1 = married, 0 = not married.

*

p < .05

**

p < .01.

Discussion

Racial discrimination contributes to vast disparities in physical and mental health (Lewis, Cogburn, & Williams, 2015; Moody et al., 2019; Torres & Vallejo, 2015; Yip, Gee, & Takeuchi, 2008), but few studies have explored the importance of experiences of racial discrimination for depression symptoms over the course of pregnancy. Leveraging a longitudinal design with repeated measurement of prenatal depression, our primary analyses showed that more EOD were associated with higher depression symptoms that persisted throughout pregnancy. Experiences of racial discrimination did not predict trajectories of change in depression symptoms (i.e., getting better or worse over time) over the course of pregnancy. In addition, after including sociodemographic and contextual factors in the model, family income-to-needs ratio emerged as the strongest predictor of depression symptoms. Given the inter-relation between income and EOD in the current sample as well as previously published samples (Agency for Healthcare Research and Quality, 2009; Lott, 2002; Williams et al., 1997), these findings highlight that contextual risk factors, including EOD and income-to-needs ratio, are related to prenatal maternal depression. The interdependence of these risk factors, however, makes their independent contributions to maternal depression difficult to dissociate.

Our findings showed that depression symptoms decreased over time. This finding may be due to the oversampling of women with elevated depression symptoms early in pregnancy in the current study, though it is consistent with evidence of declining depression symptoms during pregnancy in Latinx populations (Lara-Cinisomo, Fujimoto, Oksas, Jian, & Gharheeb, 2019) and evidence that depression symptoms are generally likely to decrease in longitudinal assessments (Long, Haraden, Young, & Hankin, 2020). Notably, however, women who experienced more discrimination maintained relatively higher levels of depression symptoms over the entire course of pregnancy compared with those women who experienced less discrimination. Prior research has shown that contextual risk factors such as financial hardship, housing difficulties, and low education are correlated with elevated depression symptoms across pregnancy (Denckla et al., 2018; Korja et al., 2018). Our results suggest that discrimination is another salient risk factor for high depression symptoms among pregnant women.

When sociodemographic and contextual factors were included in analyses, household income-to-needs ratio emerged as a predictor of prenatal depression, and the relation between racial discrimination and depression attenuated. This finding may suggest that financial strain may be a stronger predictor of depression symptoms during pregnancy than EOD. However, these two risk factors are related, because women reporting lower household income also reported a greater number of EOD (Table 2). Racial discrimination is a major problem in the United States, with a system of racial injustice embedded in the society and its official institutions, including health care systems (Bailey et al., 2017). Thus, when women navigate the prenatal health care system, being a woman of color and having low family financial resources may elevate one’s risk for prenatal depression in tandem with exposure to racial discrimination.

Few prior studies of prenatal depression and discrimination have considered the role of household income or socioeconomic status. In one such study, Canady et al. (2008) compared and contrasted different forms of discrimination (race, gender, and socioeconomic) and their relations with prenatal depression. Each form of discrimination was associated with prenatal depression symptoms, but when all were examined in one model, only gender and socioeconomic discrimination emerged as significant predictors of prenatal depression. Given the high financial burden of pregnancy and childrearing in the United States, socioeconomic stress may be highly impactful on maternal mood. Further, race and socioeconomic security are related in the United States (Kaiser Family Foundation, 2019), and perceptions of race/ethnicity and socioeconomic security overlap (e.g., Kunstman, Plant, & Deska, 2016).

These results should be interpreted in the context of several strengths and limitations. First, although the current sample is diverse in terms of race, ethnicity, and household income, we lacked adequate statistical power to examine effects of discrimination separately for participants of different racial and ethnic backgrounds. Additional research should investigate associations of discrimination and depression during pregnancy within specific ethnic and racial groups. A notable strength of the current approach was its prospective, repeated assessment of depression symptoms during pregnancy. This repeated measurement is critical considering the dynamic nature of depression over this period. Although the current study used a well-established, widely used, self-reported measure of experiences of racial discrimination over the life course, women completed this measure only one time at the last assessment during pregnancy. Prior work shows that repeated measures of EOD demonstrate stability over time (Krieger et al., 2005), which suggests that assessing EOD once during pregnancy may represent a reasonable approach to ascertain individual differences in discrimination that do not change substantially over time. Next, the EOD questionnaire specifically assesses experiences of racial and ethnic discrimination. There are many other forms of discrimination, and a multidimensional approach assessing different facets of discrimination (e.g., gender or sexual identity) may be relevant to depression symptoms. The last and perhaps most notable limitation of the current article is the focus on interpersonal experiences of racial discrimination. Indeed, racial discrimination persists on multiple, co-occurring levels, including systemic, institutional, interpersonal, and internalized (Williams, Lawrence, & Davis, 2019). These manifestations of discrimination are significant and harmful, and undoubtedly affect health outcomes. There is an urgent need to study the cumulative and unique effects of these various levels of discrimination on health outcomes. Newer metrics have been developed for quantifying systemic racism using research participants’ address histories (see Alson, Robinson, Pittman, & Doll [2021] for a review of these measures). Addresses are linked to census tracts, which can be assessed for practices reflecting systemic racial discrimination such as higher rates of denials of home ownership (redlining), denial of home loans, and exposure to police violence (Krieger et al., 2015).

Implications for Policy and Practice

Sociocultural and contextual risk factors including EOD and household financial strain have important implications for prenatal health care and policies. Current findings demonstrate that experiences of racial discrimination and lower income-to-needs ratio may be two indicators that can help clinicians identify women’s risk for elevated levels of depression symptoms across pregnancy. Pregnancy also represents a sensitive period for families, particularly for families with strained socioeconomic resources (Braveman et al., 2010). Such sociocultural adversities impact mental health during the perinatal period and especially in response to stressors (Perzow & Hennessey et al., 2021). Thus, enhancing maternal wellness for under-resourced families remains an important public health priority. Our findings support the implementation of universal/community-based interventions aiming to address systemic risk factors for poorer health outcomes, such as supplemental income for pregnant individuals and families (Stepanikova & Oates, 2017). In summary, the current study provides evidence that a woman’s exposure to racial discrimination and current financial strain are important contributors to depression symptoms during pregnancy. Given well established links between parental prenatal well-being and child outcomes (Davis et al., 2007; Duan, Hare, Staring, & Deligiannidis, 2019), efforts to decrease discrimination, financial strain, and, consequently, depression will likely have downstream protective effects across generations.

Acknowledgments

Funded by the National Institute of Mental Health (R01 MH109662). A.N.Z. was partially supported by a training fellowship through the National Institute of Mental Health (T32MH015442). The authors thank the families who participated in this project. The assistance of the Care Project research team, who made this work possible, is gratefully acknowledged.

Biography

Amanda Noroña-Zhou, PhD, is Assistant Director, Developmental Medicine, University of California, San Francisco, Department of Psychiatry, Center for Health and Community, Weill Neurosciences Institute; University of Colorado Anschutz Medical Campus, Department of Psychiatry. Research interests include prenatal programming/early childhood development/maternal-child health/health equity.

Özlü Aran, MS, is a graduate student at the University of Denver, Department of Psychology. Özlü Aran’s research interests include intergenerational transmission of emotion regulation, fetal programming, early life stress, infant emotional development, mother-child interactions, cultural context, and disparities.

Sarah E. Garcia, PhD, is a postdoctoral fellow at the University of Denver, Department of Psychology. Dr. Garcia’s research interests include developmental psychopathology, intergenerational transmission of anxiety and depression, emotion processing and regulation.

Dustin Haraden, MS, is a graduate student at the University of Illinois at Urbana-Champaign, Department of Psychology. Dustin Haraden’s research interests include developmental psychopathology, circadian rhythms, depression and anxiety.

Sarah E. D. Perzow, PhD, is a Visiting Clinical Assistant Professor in the Department of Psychology at the University of Denver. Dr. Perzow’s interests include developmental psychopathology, particularly perinatal prevention, youth and family intervention, poverty-related stress, coping, and person-centered methods.

Catherine H. Demers, PhD, is a postdoctoral fellow at the University of Denver, Department of Psychology; University of Colorado Anschutz Medical Campus, Department of Psychiatry. Dr. Demers’s research interests include fetal programming, early life stress and fetal brain development.

Ella-Marie P. Hennessey, MA, is a graduate student at the University of Denver, Department of Psychology. Ella Hennessey’s research interests include early life stress, stress physiology, fetal programming, early child development.

Stephanie Melgar Donis, BS, is a research assistant affiliated with the University of Denver, Knoebel Institute for Healthy Aging. Stephanie Melgar Donis’ research interests include health disparities, preterm labor, fetal brain development.

Melanie Kurtz, BS, is a research assistant affiliated with the University of Denver, Department of Psychology. Melanie Kurtz’s research interests include perinatal mental health, antibiotic stewardship, healthy equity, and health care use.

Benjamin L. Hankin, PhD, is the Fred and Ruby Kanfer Endowed Professor at the University of Illinois at Urbana-Champaign, Department of Psychology. Dr. Hankin’s research interests include developmental psychopathology, depression and anxiety, intervention and prevention, risk factors and mechanisms.

Elysia Poggi Davis, PhD, is a Professor at the University of Denver, Department of Psychology; University of California, Irvine, Department of Psychiatry and Human Behavior. Dr. Davis’ research interests include: Pregnancy, fetal programming, health disparities, brain development, stress physiology, unpredictability.

Footnotes

1

The authors acknowledge that race is a political and social construct that often serves as a proxy for the impact of racist practices and structural inequality, not a biological variable, and we designed our analysis with this premise in mind. In addition, categorizations of race and ethnicity are not universally defined within the United States and, therefore, definitions are provided to denote participants’ self-reported race/ethnicity. In the current study, White is used to refer to people of European ancestry, Black to refer people of African ancestry, Asian to refer to people of Asian descent, Latinx to refer to participants from Spanish-speaking countries and/or Latin American descent, and American Indian to refer to participants of indigenous North American descent (American Psychological Association, 2019; Noe-Bustamante, Mora, and Lopez, 2020; U of SC Aiken, n.d.).

References

  1. Agency for Healthcare Research and Quality. (2009). National healthcare disparities report (ARHQ Publication Number 10-0004). Rockville, MD: Agency for Healthcare Research and Quality, 2010. Available: http://www.ahrq.gov/qual/nhdr09/nhdr09.pdf. Accessed: August 30, 2020. [Google Scholar]
  2. Alson JG, Robinson WR, Pittman L, & Doll KM (2021). Incorporating measures of structural racism into population studies of reproductive health in the United States: A narrative review. Health Equity, 5(1), 49–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. American Psychological Association. (2019).Bias-freelanguage.Available: https://apastyle.apa.org/style-grammar-guidelines/bias-free-language. Accessed: May 5, 2021.
  4. Bailey ZD, Krieger N, Agénor M, Graves J, Linos N, & Bassett MT (2017). Structural racism and health inequities in the USA: Evidence and interventions. Lancet, 389(10077), 1453–1463. [DOI] [PubMed] [Google Scholar]
  5. Barrett P (2007). Structural equation modelling: Adjudging model fit. Personality and Individual Differences, 42(5), 815–824. [Google Scholar]
  6. Be′cares L, & Atatoa-Carr P (2016). The association between maternal and partner experienced racial discrimination and prenatal perceived stress, prenatal and postnatal depression: Findings from the growing up in New Zealand cohort study. International Journal for Equity in Health, 15(1), 155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bergink V, Kooistra L, Lambregtse-van den Berg MP, Wijnen H, Bunevicius R, van Baar A, & Pop V (2011). Validation of the Edinburgh Depression Scale during pregnancy. Journal of Psychosomatic Research, 70(4), 385–389. [DOI] [PubMed] [Google Scholar]
  8. Braveman P, Marchi K, Egerter S, Kim S, Metzler M, Stancil T, & Libet M (2010). Poverty, near-poverty, and hardship around the time of pregnancy. Maternal and Child Health Journal, 14(1), 20–35. [DOI] [PubMed] [Google Scholar]
  9. Canady RB, Bullen BL, Holzman C, Broman C, & Tian Y (2008). Discrimination and symptoms of depression in pregnancy among African American and White women. Women’s Health Issues, 18(4), 292–300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Capron LE, Glover V, Pearson RM, Evans J, O’Connor TG, Stein A,., & Ramchandani PG (2015). Associations of maternal and paternal antenatal mood with offspring anxiety disorder at age 18 years. Journal of Affective Disorders, 187, 20–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Centers for Disease and Control and Prevention. (2019). Racial and ethnic disparities continue in pregnancy-related deaths. Atlanta: Author. Available: https://www.cdc.gov/media/releases/2019/p0905-racial-ethnic-disparities-pregnancy-deaths.html. Accessed: May 10, 2021. [Google Scholar]
  12. Centers for Disease and Control and Prevention. (2020). Infographic: Racial/ethnic disparities in pregnancy-related deaths — United States, 2007–2016. Atlanta: Author. Available: www.cdc.gov/reproductivehealth/maternal-mortality/disparities-pregnancy-related-deaths/infographic.html. Accessed: May 10, 2021. [Google Scholar]
  13. Cox JL, Holden JM, & Sagovsky R (1987). Detection of postnatal depression: Development of the 10-item Edinburgh Postnatal Depression Scale. The British Journal of Psychiatry, 150(6), 782–786. [DOI] [PubMed] [Google Scholar]
  14. Davis EP, Glynn LM, Schetter CD, Hobel C, Chicz-Demet A, & Sandman CA (2007). Prenatal exposure to maternal depression and cortisol influences infant temperament. Journal of the American Academy of Child and Adolescent Psychiatry, 46(6), 737–746. [DOI] [PubMed] [Google Scholar]
  15. Davis EP, Hankin BL, Swales DA, & Hoffman MC (2018). An experimental test of the fetal programming hypothesis: Can we reduce child ontogenetic vulnerability to psychopathology by decreasing maternal depression? Development and Psychopathology, 30(3), 787–806. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Denckla CA, Mancini AD, Consedine NS, Milanovic SM, Basu A, Seedat S, … Koenen KC (2018). Distinguishing postpartum and antepartum depressive trajectories in a large population-based cohort: The impact of exposure to adversity and offspring gender. Psychological Medicine, 48(7), 1139–1147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Dowle M, & Srinivasan A (2020). data.table: Extension of `data.frame`. R package version 1.13.2. Available: https://CRAN.R-project.org/package=data.table. Accessed: August 7, 2020. [Google Scholar]
  18. Duan C, Hare M, Staring M, & Deligiannidis KM (2019). Examining the relationship between perinatal depression and neurodevelopment in infants and children through structural and functional neuroimaging research. International Review of Psychiatry, 31(3), 264–279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Ertel KA, James-Todd T, Kleinman K, Krieger N, Gillman M, Wright R, & Rich-Edwards J (2012). Racial discrimination, response to unfair treatment, and depressive symptoms among pregnant black and African American women in the United States. Annals of Epidemiology, 22(12), 840–846. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gaynes BN, Gavin N, Meltzer-Brody S, Lohr KN, Swinson T, Gartlehner G, … Miller WC (2005). Perinatal Depression: Prevalence, screening accuracy, and screening outcomes. Agency for Healthcare Research and Quality. Available: https://www.ncbi.nlm.nih.gov/books/NBK11838/. Accessed: November 14, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Giurgescu C, Zhang L, Price MA, Dailey RK, Frey HA, Walker DS, … Misra DP (2020). Prenatal cigarette smoking as a mediator between racism and depressive symptoms: The Biosocial Impact on Black Births Study. Public Health Nursing, 37(5), 740–749. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Glynn LM, Schetter CD, Hobel CJ, & Sandman CA (2008). Pattern of perceived stress and anxiety in pregnancy predicts preterm birth. Health Psychology, 27(1), 43–51. [DOI] [PubMed] [Google Scholar]
  23. Goldenberg RL, Culhane JF, Iams JD, & Romero R (2008). Epidemiology and causes of preterm birth. Lancet, 371(9606), 75–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Hobfoll SE, Ritter C, Lavin J, Hulsizer MR, & Cameron RP (1995). Depression prevalence and incidence among inner-city pregnant and postpartum women. Journal of Consulting and Clinical Psychology, 63(3), 445–453. [DOI] [PubMed] [Google Scholar]
  25. Hu L, & Bentler PM (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. [Google Scholar]
  26. Kaiser Family Foundation. (2019). Poverty rate by race/ethnicity. Available from. https://www.kff.org. Accessed: August 25, 2020.
  27. Kane HS, Dunkel Schetter C, Glynn LM, Hobel CJ, & Sandman CA (2014). Pregnancy anxiety and prenatal cortisol trajectories. Biological Psychology, 100, 13–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kenny DA, Kaniskan B, & McCoach DB (2015). The performance of RMSEA in models with small degrees of freedom. Sociological Methods & Research, 44(3), 486–507. [Google Scholar]
  29. Kim Y-K, Hur J-W, Kim K-H, Oh K-S, & Shin Y-C (2008). Clinical application of Korean version of Edinburgh Postnatal Depression Scale. Journal of Korean Neuropsychiatric Association, 47(1), 36–44. [Google Scholar]
  30. Korja R, Nolvi S, Kataja E-L, Scheinin N, Junttila N, Lahtinen H, … Karlsson H (2018). The courses of maternal and paternal depressive and anxiety symptoms during the prenatal period in the FinnBrain Birth Cohort study. PLoS One, 13(12), e0207856. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Krieger N, Chen JT, Waterman PD, Kiang MV, & Feldman J (2015). Police killings and police deaths are public health data and can be counted. PLoS Med, 12(12), e1001915. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Krieger N, & Sidney S (1996). Racial discrimination and blood pressure: The CARDIA study of young Black and White adults. American Journal of Public Health, 86(10), 1370–1378. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Krieger N, Smith K, Naishadham D, Hartman C, & Barbeau EM (2005). Experiences of discrimination: Validity and reliability of a self-report measure for population health research on racism and health. Social Science & Medicine, 61(7), 1576–1596. [DOI] [PubMed] [Google Scholar]
  34. Kuhlman KR, Urizar GG, Robles TF, Yim IS, & Dunkel Schetter C (2019). Testing plausible biopsychosocial models in diverse community samples: Common pitfalls and strategies. Psychoneuroendocrinology, 107, 191–200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Kunstman JW, Plant EA, & Deska JC (2016). White ≠ poor: Whites distance, derogate, and deny low-status ingroup members. Personality and Social Psychology Bulletin, 42(2), 230–243. [DOI] [PubMed] [Google Scholar]
  36. Lara-Cinisomo S, Fujimoto EM, Oksas C, Jian Y, & Gharheeb A (2019). Pilot study exploring migration experiences and perinatal depressive and anxiety symptoms in immigrant Latinas. Maternal and Child Health Journal, 23(12), 1627–1647. [DOI] [PubMed] [Google Scholar]
  37. Lewis TT, Cogburn CD, & Williams DR (2015). Self-reported experiences of discrimination and health: Scientific advances, ongoing controversies, and emerging issues. Annual Review of Clinical Psychology, 11(1), 407–440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Long EE, Haraden DA, Young JF, & Hankin BL (2020). Longitudinal patterning of depression repeatedly assessed across time among youth: Different trajectories in self-report questionnaires and diagnostic interviews. Psychological Assessment, 32(9), 872–882. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Lott B (2002). Cognitive and behavioral distancing from the poor. American Psychologist, 57(2), 100–110. [DOI] [PubMed] [Google Scholar]
  40. Lüdecke D (2020). sjPlot: Data visualization for statistics in social science. R package version 2.8.5. Available: https://CRAN.R-project.org/package=sjPlot. Accessed: August 7, 2020. [Google Scholar]
  41. Marcus SM, Flynn HA, Blow FC, & Barry KL (2003). Depressive symptoms among pregnant women screened in obstetrics settings. Journal of Women’s Health, 12(4), 373–380. [DOI] [PubMed] [Google Scholar]
  42. Meaney MJ (2018). Perinatal maternal depressive symptoms as an issue for population health. American Journal of Psychiatry, 175(11), 1084–1093. [DOI] [PubMed] [Google Scholar]
  43. Meijer JL, Bockting CLH, Stolk RP, Kotov R, Ormel J, & Burger H (2014). Associations of life events during pregnancy with longitudinal change in symptoms of antenatal anxiety and depression. Midwifery, 30(5), 526–531. [DOI] [PubMed] [Google Scholar]
  44. Moody DLB, Chang Y-F, Pantesco EJ, Darden TM, Lewis TT, Brown C, … Matthews KA (2019). Everyday discrimination prospectively predicts blood pressure across 10 years in racially/ethnically diverse midlife women: Study of women’s health across the nation. Annals of Behavioral Medicine, 53(7), 608–620. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Mukherjee S, Trepka MJ, Pierre-Victor D, Bahelah R, & Avent T (2016). Racial/ethnic disparities in antenatal depression in the United States: A systematic review. Maternal and Child Health Journal, 20(9), 1780–1797. [DOI] [PubMed] [Google Scholar]
  46. Noe-Bustamante L, Mora L, & Lopez MH (2020). About one-in-four U.S. Hispanics Have heard of Latinx, but just 3% use it. Available: www.pewresearch.org/hispanic/2020/08/11/about-one-in-four-u-s-hispanics-have-heard-of-latinx-but-just-3-use-it/. Accessed: May 5, 2021. [Google Scholar]
  47. Perzow SED, Hennessey E-MP, Hoffman MC, Grote NK, Davis EP, & Hankin BL (2021). Mental health of pregnant and postpartum women in response to the COVID-19 pandemic. Journal of Affective Disorders Reports, 4, 100123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Plant DT, Pariante CM, Sharp D, & Pawlby S (2015). Maternal depression during pregnancy and offspring depression in adulthood: Role of child maltreatment. British Journal of Psychiatry, 207(3), 213–220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. R Core Team. (2013). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Available: http://www.R-project.org/. Accessed: August 7, 2020. [Google Scholar]
  50. Revelle W (2020). psych: Procedures for personality and psychological research. Evanston, Illinois: Northwestern University. Available: https://CRAN.R-project.org/package=psych Version=2.0.9. Accessed: August 7, 2020. [Google Scholar]
  51. Rich-Edwards JW, Kleinman K, Abrams A, Harlow BL, McLaughlin TJ, Joffe H, & Gillman MW (2006). Sociodemographic predictors of antenatal and postpartum depressive symptoms among women in a medical group practice. Journal of Epidemiology & Community Health, 60(3), 221–227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Robinson D, Hayes A, & Couch S (2020). broom: Convert statistical objects into tidy tibbles. R package version 0.7.2. Available: https://CRAN.R-project.org/package=broom. Accessed: August 7, 2020. [Google Scholar]
  53. Rosenthal L, Earnshaw VA, Lewis TT, Reid AE, Lewis JB, Stasko EC, … Ickovics JR (2015). Changes in experiences with discrimination across pregnancy and postpartum: Age differences and consequences for mental health. American Journal of Public Health, 105(4), 686–693. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Rosseel Y (2012). lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48(1), 1–36. [Google Scholar]
  55. Sandman CA, Buss C, Head K, & Davis EP (2015). Fetal exposure to maternal depressive symptoms is associated with cortical thickness in late childhood. Biological Psychiatry, 77(4), 324–334. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Sluiter F, Incollingo Rodriguez AC, Nephew BC, Cali R, Murgatroyd C, & Santos HP (2020). Pregnancy associated epigenetic markers of inflammation predict depression and anxiety symptoms in response to discrimination. Neurobiology of Stress, 13, 100273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Stepanikova I, & Oates GR (2017). Perceived discrimination and privilege in health care: The role of socioeconomic status and race. American Journal of Preventive Medicine, 52(1S1), S86–S94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Torres L, & Vallejo LG (2015). Ethnic discrimination and Latino depression: The mediating role of traumatic stress symptoms and alcohol use. Cultural Diversity and Ethnic Minority Psychology, 21(4), 517–526. [DOI] [PubMed] [Google Scholar]
  59. U of SC Aiken. (n.d. Inclusive Language Guide. Available: www.usca.edu/diversity-initiatives/training-resources/guide-to-inclusive-language. Accessed: May 5, 2021. [Google Scholar]
  60. Census Bureau US. (2020). How the Census Bureau measures poverty. Available: www.census.gov/topics/income-poverty/poverty/guidance/poverty-measures.html. Accessed: August 2020. [Google Scholar]
  61. Walker JL, Ruiz RJ, Chinn JJ, Marti N, & Ricks TN (2012). Discrimination, acculturation and other predictors of depression among pregnant Hispanic women. Ethnicity & Disease, 22(4), 497–503. [PMC free article] [PubMed] [Google Scholar]
  62. Wickham H, Averick M, Bryan J, D’Agostino McGowan L, Francois R, & Yutani H (2019). Welcome to the tidyverse. Journal of Open Source Software, 4(43), 1686. [Google Scholar]
  63. Wickham H, & Miller E (2020). Import and export. haven: SPSS. ’Stata’ and ’SAS’ Files. R package version 2.3.1. Available: https://CRAN.R-project.org/package=haven. Accessed: August 7, 2020. [Google Scholar]
  64. Williams DR, Lawrence JA, & Davis BA (2019). Racism and health: Evidence and needed research. Annual Review of Public Health, 40(1), 105–125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Williams DR, Yu Y, Jackson JS, & Anderson NB (1997). Racial differences in physical and mental health: Socio-economic status, stress and discrimination. Journal of Health Psychology, 2(3), 335–351. [DOI] [PubMed] [Google Scholar]
  66. Yip T, Gee GC, & Takeuchi DT (2008). Racial discrimination and psychological distress: The impact of ethnic identity and age among immigrant and United States-born Asian adults. Developmental Psychology, 44(3), 787–800. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES