Abstract
Objective:
Our goal was to examine the role of race, ethnicity, and immigration status on prenatal perceived stress and depressive symptomatology trends across pregnancy in a low-income, minority population of pregnant women in urban Los Angeles.
Methods:
Longitudinal data on 442 Hispanic and Black women were analyzed using mixed effects models with a random intercept for participant were employed to evaluate predictors of higher perceived stress levels and probable depression across pregnancy.
Results:
Compared to U.S. born Hispanic women, Black women had higher perceived stress (β 2.24; 95% CI: 0.41, 4.07) and higher odds of probable depression (OR 2.38; 95% CI: 0.95, 5.95) while Hispanic immigrants with greater than 20 years of residency in the U.S. had higher odds of probable depression (OR 2.93; 95% CI: 1.10, 7.79).
Conclusion:
Maternal mental health among Hispanic immigrants with longer U.S. residency and Black race warrants increased attention from public health and clinical agencies.
Keywords: Hispanic, Black, Immigrant, Pregnancy, Stress, Depression
Introduction
Socio-environmental inequities and racial/ethnic identity have been associated with maternal and infant health, where more marginalized groups tend to have poorer physical and mental health outcomes (1). Prenatal depression, for example, is highest in non-Hispanic Black and/or Hispanic females compared with Non-Hispanic White females. (2–6) In urban cities, such as Los Angeles, these rates can be as high as 20% or greater among low-income, minority women. (7) Similarly, according to the Center for Disease Control, non-Hispanic Black and Hispanic females rank highest in the proportion of stressful life events in the year prior to infant birth. (8) Despite these high exposures to stressors, few studies that address stress and depression risk factors have been conducted in exclusively racial/ethnic minority sample populations and few studies have looked at stressors longitudinally across pregnancy.
Among populations of U.S. Black and Hispanic women, pre-natal depression has been shown to be higher compared to post-partum depression (9–11), however studies that track prenatal stress and depression trajectories are uncommon. Instead, predictors of prenatal depression have been examined cross-sectionally during different stages of pregnancy or post-partum. Risk factors for stress and depression identified in these studies include young age (12), alcohol or smoking use (12), while predictors associated with sociodemographic factors include having a higher education, higher financial stress, and single or cohabitation status. (6, 12, 13) Important predictors for stress and depression in Hispanic populations include immigrant status and/or living in disadvantaged neighborhoods. (14) In this context, non-U.S. nativity and low socioeconomic status may serve as strong proxies for these risk factors for stress and depression. However, the literature is mixed, and some studies show that being U.S. born, rather than foreign-born, may be a risk factor for depression. (10, 15) Speaking Spanish, which may reflect a stronger tie to non-U.S. culture, has also been associated with higher stress (12). Other observed cultural stressors or social constructs include acculturative stress, (16) intense isolation, (17) and lack of support (10), all of which may be reflected in part by nativity. Collectively, the current body of literature reflects a suite of risk factors in Black and Hispanic mothers, including race and immigration status.
The public health importance of identifying pregnant women at greatest risk for high stress or depression cannot be understated, for only then can we potentially target prevention strategies in a timely manner. In this study, we will utilize our unique, longitudinal study with predominantly Black and Hispanic women to identify individual predictors or combinations of predictors of stress and depression during pregnancy. Therefore, our goals in this cohort study are to 1) describe trajectories of perceived stress and depressive symptoms across pregnancy, and 2) to examine if race, ethnicity, and immigration status, along with other risk factors, are associated with perceived stress and risk for depression at any time during pregnancy in a low-income population of minority pregnant women residing in the greater Los Angeles area.
Methods
Design Overview
The Maternal And Developmental Risks from Environmental and Social Stressors (MADRES) study is an ongoing prospective pregnancy cohort that was created to examine the social and environmental causes of maternal gestational weight gain, postpartum weight retention, and childhood obesity risk during the first years of life. Cohort participants are followed through their pregnancies, at birth, and through the infant’s first five years of life. Full details on study design has been published elsewhere. (18)
Enrollment in the MADRES cohort is initiated prior to the 30th week of gestation. Women are recruited from three prenatal clinic sites in Los Angeles. Eligible women are over 18 years of age and speak either English or Spanish. Exclusion criteria include: (1) HIV positive status, (2) physical, mental, or cognitive disability that prevents participation, (3) current incarceration, or (4) multiple pregnancy. For women enrolling at less than 20 weeks of gestation, the first study visit is scheduled within 2 weeks of recruitment to: (1) complete an interviewer-administered questionnaire in English or Spanish, (2) measure height and weight, and (3) collect biospecimens. During the period prior from 20–30 weeks of gestation, a set of follow-up questionnaires are conducted via phone (second set of observations). Lastly, a third set of observations are collected during a visit conducted in the gestational period between 30 and 34 weeks, during which a study interview is administered, height and weight are measured, and biospecimens are again collected. For women who enroll late (between 20 and 30 weeks of gestation), the first study visit occurs in person between 28 and 36 weeks to complete an interviewer-administered questionnaire, measure height and weight, and collect biospecimens.
For this analysis, because we were interested in risk factors related to stress and depression throughout pregnancy, we restricted analyses to pregnant women in MADRES who had completed their pregnancies and had data available for one, two, or three visits, completed by January 2, 2020, yielding a total of 467 unique participants. We further removed 25 individuals whose race/ethnicity was missing, or who indicated race/ethnicity other than Hispanic White or Black, given small numbers. This yielded a total of 1045 visits from 442 unique participants for this longitudinal analysis, where there were 85 participants had data from one visit, 111 participants had data from two visits, and 246 participants had data from three visits. Full study protocol was approved by the University of Southern California Institutional Review Board and informed consent was obtained prior to participation.
Measures
Migration/nativity Factors.
Of the existing variables collected for primary data collection, measures of cultural background and migration included the participants’ and baby’s biological father’s country of origin (U.S., Mexico, or Central America), length of U.S. residency, and preferred language. In the final models, we categorized participants based on length of U.S. residency as follows: U.S.-born, 1–10 years in U.S., 11–20 years in U.S., and >20 years in U.S. To evaluate both length of residency and country of nativity we further categorized participants as follows: 1) U.S.-born, 2) Born in Mexico and lived 1–10 years in U.S., 3) Born in Mexico and lived 11–20 years in U.S., 4) Born in Mexico and lived >20 years in U.S.; 5) Born in Central America and lived 1–10 years in U.S., 6) Born in Central America and lived >10 years in the U.S.
Socio-economic status variables.
Socio-demographic characteristics included mother’s employment status, marital/cohabitation status, education level, and annual household income. Those who were students or under temporary medical leave were categorized into one group and the employed category included part-time and full-time work.
Perceived stress.
Participants were administered Cohen’s 10-item Perceived Stress Scale (PSS-10) (19). The PSS-10 was used to measure the degree to which the participant perceives aspects of one’s life as uncontrollable, unpredictable, and over-loading during the past month. We slightly modified our text to read “in the past month (30 days)”. The form consists of 10 items on a 5-point Likert scale ranging from 0 (never) to 4 (very often), indicating how often they have felt or thought a certain way within the past month, and scores range from 0 to 40, with higher scores suggestive of greater perceived stress. The scale has good sensitivity and specificity and high internal consistency in English (Chronbach’s α = 0.78) and Mexican Spanish (Chronbach’s α = 0.82). (19, 20)
Depressive symptoms.
Depressive symptomatology among participants was measured using the Center for Epidemiologic Studies-Depression (CES-D) Scale which contains 20 items measuring six components of depression symptoms within the last week: depressed mood; feelings of guilt and worthlessness; feelings of helplessness and hopelessness; psychomotor retardation; loss of appetite; and sleep disturbance. The response alternatives were: rarely or none of the time (0), some or little of the time (1), occasionally or a moderate amount of time (2), and most of or all of the time. (3) The CES-D scores of the 20 items were summed, resulting in a range of total possible scores from 0 to 60, and scores of 16 or greater were used to identify individuals at risk for clinical depression (or probable depression). The scale has good sensitivity and specificity and high internal consistency (Chronbach’s α = 0.85) in English (21) and Spanish (Chronbach’s α = 0.88). (22)
Additional covariates.
A variety of maternal information was collected via questionnaire. Physical characteristics included maternal age, mother’s self-reported pre-pregnancy weight and height, estimated gestational age at questionnaire administration, and parity. Pre-pregnancy maternal body mass index (BMI in kg/m2) was calculated using self-reported weight and height according to CDC guidelines. Gestational age at questionnaire administration was derived from documented gestational age at birth by subtracting the dates of one from the other and back-calculating age. The best estimates of gestational age (GA) at birth were ascertained using a hierarchy of methods. A first trimester (<14 weeks GA) ultrasound measurement of crown-rump length was considered the most preferred and was used if available (n = 177). If unavailable, a second trimester (<28 weeks GA) ultrasound measurement of fetal biparietal diameter was used (n = 61). If measures from an early ultrasound were unavailable, GA at birth was determined based on the physician’s best clinical estimate, abstracted from the medical records (n = 48). If this information was missing, self-reported information obtained from the mothers was used as a proxy (n=96). Various behavioral factors collected during early pregnancy included self-reported pre-pregnancy physical activity (i.e., how many days physically active in one week for ≥30 minutes/day), multivitamins/prenatal vitamins intake before or during pregnancy, antibiotic use during pregnancy, ever having smoked tobacco, and ever having exposure to environmental tobacco smoke during pregnancy.
Statistical methods
Descriptive analyses were first conducted to examine the distribution of participant characteristics and outcomes, PSS-10 and CES-D, which were asked during each study visit across pregnancy.
We evaluated predictors of stress and depression symptomology at any time point during pregnancy by using mixed-effects linear regression models that accounted for repeated measurements, where an identify link function was fitted for PSS-10 and a log link function for CES-D. All models adjusted for recruitment site as a fixed effect as a design variable. Predictors included mother’s recruitment site, maternal age, gestational age at questionnaire administration, pre-pregnancy BMI, parity, categories of years mother having lived in the U.S., current employment status at the time of data collection, marital/cohabitation status, maternal education level, total household income in the last year, and mother’s smoking status and timepoint of data collection (i.e., 1st, 2nd or 3rd visit). Given collinearity between language and years of residency in the U.S., we subsequently removed language from the models. Hours working per week during pregnancy, multivitamins/prenatal vitamins intake before or during pregnancy, antibiotic use during pregnancy, ever exposed to environmental tobacco smoke, and pre-pregnancy physical activity were evaluated in the mixed models but were not associated with outcomes nor presented evidence of confounding the associations (determined by a change in the effect estimate of less than 10%) between other covariates and were therefore removed from final models. Participants with missing data for any of the covariates at a given timepoint were excluded from multivariable models. However, for the income variable, participants who responded as “don’t know” were included in the models as a distinct category because we felt there could be fundamental differences between those who did not know their household income versus those who abstained from answering.
All tests assumed a two-sided alternative hypothesis and a p-value<0.05 threshold of significance and were conducted using the R programming language, version 3.3.4.
Results
Physical, demographic, and behavioral characteristics of the 442 unique study participants in the sub-study are shown in Table 1. In general, PSS-10 scores were highest in the first visit, then declined to the lowest in the second visit, and slightly increased in the third visit. (Figure 1) These trajectories varied by race/ethnicity. Probable prenatal depression prevalence (CESD score >16) in this study was 28%, 20% and 18% in the first, second and third visits across pregnancy.
Table 1.
Physical and sociodemographic characteristics of 442 women during pregnancy
| Analytic dataset (N=442) | ||
|---|---|---|
| N | % or Mean ± SD | |
| Number of women who completed each prenatal visit | ||
| Visit 1 | 382 | |
| Visit 2 | 374 | |
| Visit 3 | 289 | |
| Maternal age (mean, SD), years | 442 | 28.1 ± 5.9 |
| Self-reported pre-pregnancy BMI (mean, SD), kg/m 2 | 373 | 29.7 ± 6.5 |
| Gestational age at questionnaire administration (weeks) | 382 | 12.9 ± 4.2 |
| Parity | ||
| 1 | 132 | 29.9% |
| 2 | 130 | 29.4% |
| 3 + | 153 | 34.6% |
| Years of mother having lived in USA | ||
| Born in USA | 217 | 49.1% |
| 1–10 years | 61 | 13.8% |
| 11–20 years | 83 | 18.8% |
| >20 years | 46 | 10.4% |
| Mother race and ethnicity | ||
| Hispanic white | 376 | 85.1% |
| Black | 53 | 12.0% |
| Country of origin | ||
| United States | 217 | 49.1% |
| Mexico | 123 | 27.8% |
| Other Central American or Caribbean | 67 | 15.2% |
| Asia/Europe/South America | 5 | 1.1% |
| Mother preferred language | ||
| Spanish | 165 | 37.3% |
| English | 277 | 62.7% |
| Current employment status in the 1st trimester | ||
| Employed | 184 | 48.2% |
| Homemaker | 107 | 28.0% |
| Student/temporary medical leave | 29 | 7.6% |
| Unemployed | 61 | 16.0% |
| Marital status | ||
| Married | 119 | 26.9% |
| Living together | 171 | 38.7% |
| Never married, single | 98 | 22.2% |
| Decline to answer/divorced/separated | 22 | 5.0% |
| Mother’s education level | ||
| < grade 12 (High School) | 122 | 27.6% |
| Completed grade 12 | 140 | 31.7% |
| Some college/tech school or completed college | 171 | 38.7% |
| Total household income in the last year | ||
| Less than $29,999 | 211 | 47.7% |
| $30,000 or more | 78 | 17.6% |
| Don’t know | 144 | 32.6% |
| Mother ever tobacco smoke | ||
| Yes | 100 | 26.2% |
| No | 281 | 73.6% |
Values may not add up to 100% due to missingness
Figure 1.
Average trajectories of PSS across pregnancy by race/ethnicity/years in the U.S.
When evaluating the associations between race/ethnicity with PSS-10 and probable depression, we found that Black women had consistently higher PSS-10 scores and higher odds of probable depression (OR 2.24; 95% CI 0.41, 4.07 and OR 2.38; 95% CI: 0.95, 5.95, respectively) compared to U.S. born Hispanic women (Table 2). However, when evaluating the influence of nativity and length of residency within the U.S. among Hispanic White women, we found that compared to being born in the U.S., Hispanic immigrants with greater than 20 years of U.S. residency had higher odds of probable depression (OR 2.93; 95% CI: 1.10, 7.79, Table 2). Longer residency (11–20 years) was also positively associated with PSS-10 scores (β 1.97; 95% CI: 0.39, 3.54) in Hispanic immigrants compared to U.S. born. Upon further examination of mother’s country of birth in conjunction with length of residency (Table 3), a difference emerged specifically between mothers born in the U.S., Mexico, and Central America. Compared to U.S.-born mothers, Central American-born mothers with 11–20 years of U.S. residency had higher odds of probable depression (OR 3.30; 95% CI:1.13, 9.67).
Table 2.
Physical, socio-demographic and behavioral predictors of perceived stress and depressive symptomatology (CES-D scores ≥16)
| Perceived Stress |
Depressive Symptomatology |
|||||
|---|---|---|---|---|---|---|
| β | 95% CI | p | OR | 95% CI | p | |
| Maternal age | 0.10 | (−0.02, 0.21) | 0.11 | 1.03 | (0.97, 1.10) | 0.28 |
| Gestational age at questionnaire administration | 0.10 | (−0.01, 0.21) | 0.09 | 1.00 | (0.93, 1.07) | 0.94 |
| Pre-pregnancy BMI | −0.01 | (−0.10, 0.08) | 0.85 | 0.99 | (0.94, 1.03) | 0.56 |
| Mother race/ethnicity/years in U.S. | ||||||
| Hispanic white, U.S.-born | Ref | Ref | Ref | Ref | Ref | Ref |
| Hispanic white, foreign-born, 1–10 years in U.S. | 0.98 | (−0.78, 2.74) | 0.28 | 1.43 | (0.57, 3.61) | 0.44 |
| Hispanic white, foreign-born, 11–20 years in U.S. | 1.97 | (0.39, 3.54) | 0.01 | 1.79 | (0.80, 4.01) | 0.16 |
| Hispanic white, foreign-born, >20 years in U.S. | 0.44 | (−1.45, 2.34) | 0.65 | 2.93 | (1.10, 7.79) | 0.03 |
| Black | 2.24 | (0.41, 4.07) | 0.02 | 2.38 | (0.95, 5.95) | 0.06 |
| Employment status during pregnancy | ||||||
| Employed | Ref | Ref | Ref | Ref | Ref | Ref |
| Homemaker | −0.47 | (−1.54, 0.61) | 0.39 | 0.73 | (0.38, 1.41) | 0.35 |
| Student/temporary medical leave | −0.93 | (−2.36, 0.49) | 0.20 | 0.97 | (0.41, 2.29) | 0.95 |
| Unemployed | 1.05 | (−0.22, 2.31) | 0.10 | 2.09 | (1.00, 4.37) | 0.05 |
| Marital/cohabitation status | ||||||
| Married | Ref | Ref | Ref | Ref | Ref | Ref |
| Living together | 2.09 | (0.72, 3.46) | 0.003 | 2.43 | (1.17, 5.05) | 0.02 |
| Never married, single | 1.22 | (−0.46, 2.90) | 0.15 | 2.25 | (0.94, 5.43) | 0.07 |
| Decline to answer/divorced/separated | 4.08 | (1.48, 6.68) | 0.002 | 7.99 | (2.25, 28.41) | 0.001 |
| Mother’s education level | ||||||
| < grade 12 | Ref | Ref | Ref | Ref | Ref | Ref |
| Completed grade 12 | −2.07 | (−3.56, −0.56) | 0.01 | 0.52 | (0.24, 1.14) | 0.10 |
| Some college/tech school or completed college | −1.66 | (−3.20, −0.12) | 0.03 | 0.77 | (0.35, 1.69) | 0.52 |
| Total household income in the last year | ||||||
| Less than $29,999 | Ref | Ref | Ref | Ref | Ref | Ref |
| $30,000 or more | 0.01 | (−1.48, 1.51) | 0.99 | 0.91 | (0.42, 1.97) | 0.81 |
| Don’t know | −0.12 | (−1.45, 1.21) | 0.86 | 0.79 | (0.40, 1.57) | 0.51 |
| Mother ever tobacco smoke during pregnancy | ||||||
| No | Ref | Ref | Ref | Ref | Ref | Ref |
| Yes | 0.76 | (−0.27, 1.78) | 0.15 | 2.01 | (1.11, 3.63) | 0.02 |
| Parity | ||||||
| 1 | Ref | Ref | Ref | Ref | Ref | Ref |
| 2 | −0.29 | (−1.75, 1.16) | 0.69 | 0.78 | (0.37, 1.67) | 0.53 |
| 3 + | −0.75 | (−2.49, 0.98) | 0.39 | 0.70 | (0.29, 1.73) | 0.45 |
| Timepoint | ||||||
| Trimester 1 | Ref | Ref | Ref | Ref | Ref | Ref |
| Trimester 2 | −2.21 | (−3.35, −1.07) | <0.001 | 0.61 | (0.30, 1.28) | 0.19 |
| Trimester 3 | −2.36 | (−4.66, −0.06) | 0.04 | 0.58 | (0.14, 2.45) | 0.46 |
Models were adjusted for site of recruitment and included a random intercept for participant
Table 3.
The effect of country of birth and years of residency in the U.S. with perceived stress and depressive symptomatology (CES-D scores ≥16)
| Perceived Stress | Depressive Symptomatology | |||||
|---|---|---|---|---|---|---|
| β | 95% CI | p | OR | 95% CI | p | |
| Mother country of birth and years having lived in USA combined | ||||||
| Born in USA | Ref | Ref | Ref | Ref | Ref | Ref |
| Born in Mexico and lived 1–10 years in U.S. | −1.48 | (−4.05, 1.08) | 0.26 | 0.37 | (0.08, 1.68) | 0.20 |
| Born in Mexico and lived 11–20 years in U.S. | 1.27 | (−0.40, 2.95) | 0.14 | 1.05 | (0.45, 2.46) | 0.91 |
| Born in Mexico and lived >20 years in U.S. | −0.21 | (−2.25, 1.83) | 0.84 | 2.22 | (0.80, 6.15) | 0.12 |
| Born in other Central America and lived 1–10 years in U.S. | 1.31 | (−0.76, 3.39) | 0.21 | 2.59 | (0.95, 7.06) | 0.06 |
| Born in other Central America and lived >10 years in U.S. | 2.02 | (−0.21, 4.25) | 0.08 | 3.30 | (1.13, 9.67) | 0.03 |
Model was adjusted for site of recruitment, maternal age, gestational age at consent, pre-pregnancy BMI, employment status, maternal education, income, tobacco smoke exposure, and parity. Random intercept for participant was used.
Martial/cohabitation status, education, and smoking status were predictive of perceived stress and probable depression (Table 2). Compared to being married, cohabitation with a partner associated with higher PSS-10 scores (β 2.09; 95% CI: 0.72, 3.46) and higher odds of probable depression (OR 2.43; 95% CI:1.17, 5.05). Next, compared to education level lower than grade 12, education levels higher than grade 12 associated with lower stress (Table 2). Finally, women who reported having ever smoked during pregnancy had higher odds of probable depression compared to those who had never smoked during pregnancy (OR 2.01; 95% CI:1.11, 3.63).
Discussion
Longitudinal studies on the effects of socio-demographic predictors on psychosocial stress and depression during pregnancy are lacking. In this cohort study of predominantly low-income, minority women, our goal was to examine predictors of perceived stress and depression across three time points during pregnancy. Higher perceived stress during at any point in pregnancy was associated Black race, longer time in the U.S. among Hispanic White women, lower education level and living together with a partner. Predictors of probable depression at any point in pregnancy included Black race, longer time in the U.S. among Central American born women, living together with a partner, and history of maternal smoking.
Our longitudinal results coincide with cross-sectional studies that report that race/ethnicity, specifically Black race and Hispanic ethnicity, are strong predictors of prenatal depression. (2–6) Black race was one of the strongest predictors of prenatal depression in this cohort study. Studies in some pregnancy cohorts have shown that racial discrimination could be a contributing factor for this finding, (23, 24) although not all studies agree. (3, 9) Potential explanations for the contradictory findings in the literature may include differences in methodology. A varied use of depression screening tools and differential thresholds to assess for depression risk have resulted in varied prenatal depression prevalence estimates. (25) Different tools have been used including the Physical Health Questionnaire (PHQ), the Edinburg Postnatal Depression Scale (EDPS), the CES-D, and even non-disclosed methods of measuring depression. Additionally, the covariates used in analyses have differed. Despite these differences, cohorts with urban, minority populations similar to ours have reported similar findings. While we observed odds ratios for probable depression ranging from 2.48–2.93 for Black race or Hispanic ethnicity with >20 years in U.S., others have reported odds ratios between 1.68–2.98. (5, 25)
Longer length of residency in the U.S. and Central American nativity were associated with higher depressive symptoms in Hispanic women during pregnancy, suggesting that an underlying construct, most likely acculturation or stress associated with acculturation, may play a role in this relationship. Acculturation, which is the process by which immigrants acclimate to a new home country environment and culture, can have negative health impacts on a variety of Hispanic groups.(26, 27) Previous work has demonstrated that Hispanic women who are more acculturated, as measured by a higher generational status, have higher prenatal stress (28) and depression, prenatal anxiety (28, 29), and postnatal (27) depression. Using migration and nativity factors as a proxy for acculturation, our results similarly suggest that compared to U.S. born women, immigrants who have been in the U.S. more than ten years and specifically, those from Central America, had higher prenatal stress and depressive symptoms. Although we cannot directly compare measures of migration and nativity to the acculturative process, we hypothesize that those from more distant native countries appear to have higher stressors. This may be due to further physical distance from the native country and family/friends, which may in turn negatively impact the degree of social support that may be available otherwise. Additionally, this finding may also be in part to differences in socio-political context driving migratory events in the countries of origin of the populations evaluated. Therefore, our results suggest that pregnancy may be a risk period for experiencing mental health symptoms in a low-income urban Hispanic population and they may be amplified during periods of political turmoil that have included increased discrimination against Hispanic immigrants.
Compared to their married counterparts in this study, pregnant women who were not married but living with a partner showed higher prenatal stress and risk for probable depression. Similar studies comprised of multi-racial/ethnic and/or exclusively Hispanic populations show that being single or cohabitating with a partner is predictive of depression during pregnancy (6, 9, 13, 30). In line with our findings, one study showed that in Mexican women who immigrated to the U.S., cohabitating with a stable partner was predictive of higher depressive symptomatology compared to those who were married. (13) Similarly, another multi-racial/ethnic study showed that either being single or cohabitating with a partner was associated high depression compared to those who were married. (9) In contrast, other studies have shown that only being unpartnered (i.e. single status) was associated with higher depression. (6, 30) A possible explanatory theory may be lower relationship quality in cohabitors vs. married women; also defined as “relationship instability” or an individual’s perception of how likely the relationship may be to dissolve. This construct has been linked to higher depressive symptoms. (31) Re-partnering may also be an explanatory factor of these findings. A study specifically in Mexican adults, also showed that compared to married women, re-partnered women who cohabitated with the new partner had higher depressive symptoms. (32) In our study, we did not have specific data on either relationship stability or re-partnering, as it was beyond the scope the study and therefore, we could not test these theories.
Predictive factors of perceived stress and depression related to health behavior or other socioeconomic factors are also worth noting. Our results showed that maternal smoking during pregnancy was associated with probable depression, which is consistent with studies in this population. (10, 12) With respect to socioeconomic factors, systematic reviews report that lower education as a predictor of stress and depression in various populations. (2, 33) Our study results showed that higher education was a protective factor for perceived stress, which is in line with most of the literature in pregnancy cohorts. Other socioeconomic factors such as annual income in prior year and employment status were not associated with stress or depression.
Several limitations to the study are worth noting. Given a time period of stricter immigration policies and deportation laws during which these data were collected (2015–2020), our cohort of immigrant women may have been exposed higher level of stressors, relative to the pre-2015 era. Finally, no data was collected on whether the married couples were cohabitating, whether the husband or cohabiting partner was the infant’s biological father, and/or if there was re-partnering (mother was previously in a different couple union prior to pregnancy).
In conclusion, the most notable predictors of preceived stress and depressive symptoms across pregnancy in this study of minority women were Black race, longer length of U.S. residency among Hispanic women, Central American nativity, living with a partner, lower education and maternal smoking history. These findings have clinical and public health implications. Clinically, assessment beyond physical and behavioral factors, and inclusion of stressors related to nativity and marital status at prenatal visits may aid in identifying those at early risk for adverse maternal mental health outcomes. Moreover, these results suggest the need for increased cultural sensitivity and cultural competence training for clinicians. From a public health perspective, development and implementation of culturally based interventions may decrease maternal prenatal risk of stress and depression in minority communities. For example, previous studies have shown that either community-planned, culturally tailored intervention or cognitive behavioral therapies have been successful in reducing prenatal depressive symptoms in Black and Hispanic women.(34, 35) Collectively, these results support other literature that certain sociodemographic prenatal risk factors in this population have important implications for both clinicians and maternal healthcare interventionists.
References
- 1.Howell EA. Reducing Disparities in Severe Maternal Morbidity and Mortality. Clin Obstet Gynecol 2018;61(2):387–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Mukherjee S, Trepka MJ, Pierre-Victor D, Bahelah R, Avent T. Racial/Ethnic Disparities in Antenatal Depression in the United States: A Systematic Review. Matern Child Health J 2016;20(9):1780–97. [DOI] [PubMed] [Google Scholar]
- 3.Shen JJ, Lin F, Jackson T. Risk of prenatal depression: differences by race. Ethn Dis 2010;20(1):35–9. [PubMed] [Google Scholar]
- 4.Gavin AR, Melville JL, Rue T, Guo Y, Dina KT, Katon WJ. Racial differences in the prevalence of antenatal depression. Gen Hosp Psychiatry 2011;33(2):87–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Melville JL, Gavin A, Guo Y, Fan MY, Katon WJ. Depressive disorders during pregnancy: prevalence and risk factors in a large urban sample. Obstet Gynecol 2010;116(5):1064–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Benatar S, Cross-Barnet C, Johnston E, Hill I. Prenatal Depression: Assessment and Outcomes among Medicaid Participants. J Behav Health Serv Res 2020. [DOI] [PubMed] [Google Scholar]
- 7.Health LCDoP. L.A. County Women Infant Children Survey 2005.
- 8.(CDC) CfDCaP. Stressful Life Events Experienced by Women in the Year Before Their Infants’ Births - United States, 2000–2010. Morbidity and Mortality Weekly Report (MMWR) 2015. p. 247–51. [PMC free article] [PubMed]
- 9.Rich-Edwards JW, Kleinman K, Abrams A, Harlow BL, McLaughlin TJ, Joffe H, et al. Sociodemographic predictors of antenatal and postpartum depressive symptoms among women in a medical group practice. J Epidemiol Community Health 2006;60(3):221–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sidebottom AC, Hellerstedt WL, Harrison PA, Hennrikus D. An examination of prenatal and postpartum depressive symptoms among women served by urban community health centers. Arch Womens Ment Health 2014;17(1):27–40. [DOI] [PubMed] [Google Scholar]
- 11.Silva R, Jansen K, Souza L, Quevedo L, Barbosa L, Moraes I, et al. Sociodemographic risk factors of perinatal depression: a cohort study in the public health care system. Braz J Psychiatry 2012;34(2):143–8. [DOI] [PubMed] [Google Scholar]
- 12.Silveira ML, Pekow PS, Dole N, Markenson G, Chasan-Taber L. Correlates of high perceived stress among pregnant Hispanic women in Western Massachusetts. Matern Child Health J 2013;17(6):1138–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lara MA, Le HN, Letechipia G, Hochhausen L. Prenatal depression in Latinas in the U.S. and Mexico. Matern Child Health J 2009;13(4):567–76. [DOI] [PubMed] [Google Scholar]
- 14.Sam DL, Berry JW. Acculturation: When Individuals and Groups of Different Cultural Backgrounds Meet. Perspect Psychol Sci 2010;5(4):472–81. [DOI] [PubMed] [Google Scholar]
- 15.Fleuriet KJ, Sunil TS. Perceived social stress, pregnancy-related anxiety, depression and subjective social status among pregnant Mexican and Mexican American women in south Texas. J Health Care Poor Underserved 2014;25(2):546–61. [DOI] [PubMed] [Google Scholar]
- 16.D’Anna-Hernandez KL, Aleman B, Flores AM. Acculturative stress negatively impacts maternal depressive symptoms in Mexican-American women during pregnancy. J Affect Disord 2015;176:35–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Bloom T, Glass N, Curry MA, Hernandez R, Houck G. Maternal stress exposures, reactions, and priorities for stress reduction among low-income, urban women. J Midwifery Womens Health 2013;58(2):167–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bastain TM, Chavez T, Habre R, Girguis MS, Grubbs B, Toledo-Corral C, et al. Study Design, Protocol and Profile of the Maternal And Developmental Risks from Environmental and Social Stressors (MADRES) Pregnancy Cohort: a Prospective Cohort Study in Predominantly Low-Income Hispanic Women in Urban Los Angeles. BMC Pregnancy Childbirth 2019;19(1):189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Cohen S, & Williamson GM Perceived stress in a probability sample of the United States. Spacapan S & Oskamp S (Eds.). The social psychology of health (pp 31–67): Newbury Park, CA: Sage.; 1988. [Google Scholar]
- 20.Gonzalez-Ramirez MT, Rodriguez-Ayan MN, Hernandez RL. The perceived stress scale (PSS): normative data and factor structure for a large-scale sample in Mexico. Span J Psychol 2013;16:E47. [DOI] [PubMed] [Google Scholar]
- 21.Lewinsohn PM, Seeley JR, Roberts RE, Allen NB. Center for Epidemiologic Studies Depression Scale (CES-D) as a screening instrument for depression among community-residing older adults. Psychology and aging 1997;12(2):277–87. [DOI] [PubMed] [Google Scholar]
- 22.Roberts RE, Vernon SW, Rhoades HM. Effects of language and ethnic status on reliability and validity of the Center for Epidemiologic Studies-Depression Scale with psychiatric patients. J Nerv Ment Dis 1989;177(10):581–92. [DOI] [PubMed] [Google Scholar]
- 23.Ertel KA, James-Todd T, Kleinman K, Krieger N, Gillman M, Wright R, et al. Racial discrimination, response to unfair treatment, and depressive symptoms among pregnant black and African American women in the United States. Ann Epidemiol 2012;22(12):840–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Giurgescu C, Zenk SN, Engeland CG, Garfield L, Templin TN. Racial Discrimination and Psychological Wellbeing of Pregnant Women. MCN Am J Matern Child Nurs 2017;42(1):8–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Gavin NI, Gaynes BN, Lohr KN, Meltzer-Brody S, Gartlehner G, Swinson T. Perinatal depression: a systematic review of prevalence and incidence. Obstet Gynecol 2005;106(5 Pt 1):1071–83. [DOI] [PubMed] [Google Scholar]
- 26.Ortega AN, Rosenheck R, Alegria M, Desai RA. Acculturation and the lifetime risk of psychiatric and substance use disorders among Hispanics. J Nerv Ment Dis 2000;188(11):728–35. [DOI] [PubMed] [Google Scholar]
- 27.Sumner LA, Valentine J, Eisenman D, Ahmed S, Myers H, Wyatt G, et al. The influence of prenatal trauma, stress, social support, and years of residency in the US on postpartum maternal health status among low-income Latinas. Matern Child Health J 2011;15(7):1046–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Ruiz RJ, Stowe RP, Brown A, Wommack J. Acculturation and biobehavioral profiles in pregnant women of Hispanic origin: generational differences. ANS Adv Nurs Sci 2012;35(3):E1–E10. [DOI] [PubMed] [Google Scholar]
- 29.Preciado A, D’Anna-Hernandez K. Acculturative stress is associated with trajectory of anxiety symptoms during pregnancy in Mexican-American women. J Anxiety Disord 2017;48:28–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Herbell K, Zauszniewski JA, Williams E. Stress and Depressive Symptoms Among Demographically Diverse American Pregnant Women. Issues Ment Health Nurs 2020;41(1):73–82. [DOI] [PubMed] [Google Scholar]
- 31.Brown SL. The effect of union type on psychological well-being: depression among cohabitors versus marrieds. J Health Soc Behav 2000;41(3):241–55. [PubMed] [Google Scholar]
- 32.Schmeer KK, Kroeger RA. Union Type and Depressive Symptoms Among Mexican Adults. J Fam Issues 2011;32(12):1597–621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Lancaster CA, Gold KJ, Flynn HA, Yoo H, Marcus SM, Davis MM. Risk factors for depressive symptoms during pregnancy: a systematic review. Am J Obstet Gynecol 2010;202(1):5–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kieffer EC, Caldwell CH, Welmerink DB, Welch KB, Sinco BR, Guzman JR. Effect of the healthy MOMs lifestyle intervention on reducing depressive symptoms among pregnant Latinas. Am J Community Psychol 2013;51(1–2):76–89. [DOI] [PubMed] [Google Scholar]
- 35.Le HN, Perry DF, Stuart EA. Randomized controlled trial of a preventive intervention for perinatal depression in high-risk Latinas. J Consult Clin Psychol 2011;79(2):135–41. [DOI] [PubMed] [Google Scholar]