Effects of Stress and Nativity on Maternal Antenatal Substance Use and Postnatal Mental Disorders

Abstract

Background:

Maternal substance use and common mental disorders (CMDs) during or after pregnancy can lead to negative health outcomes among mothers and infants. We examined whether nativity (US-born versus foreign-born) and stress levels during pregnancy were associated with antenatal substance use and postnatal CMDs.

Methods:

We analyzed the Boston Birth Cohort, a racially diverse cohort recruited at birth with rolling enrollment since 1998. Information on antenatal substance use (tobacco and/or alcohol use) was obtained using an in-person postpartum questionnaire (n = 6,514). Information on postnatal CMDs (depression and/or anxiety) was obtained from medical records (n = 2,052). Nativity and stress during pregnancy were self-reported. We performed multivariate logistic regression to examine how nativity and stress levels were jointly associated with antenatal substance use and postnatal CMDs. We further investigated if blacks, Hispanics, and whites were differentially at risk.

Results:

We found that US-born mothers were at higher risk of substance use and CMDs than their foreign-born counterparts. In analyses combining nativity and stress, being US-born with high stress was associated with increased odds of antenatal substance use (adjusted odds ratio [aOR] = 14.91, 95% confidence interval [CI]: 12.09–18.39) and postnatal CMDs (aOR = 4.09, 95% CI: 2.72–6.15) compared with foreign-born mothers with low stress. The results of the subanalyses limited to black and Hispanic women separately were similar; high stress alone was associated with fourfold increased odds of CMDs among foreign-born Hispanic mothers (aOR = 4.27, 95% CI: 1.96–9.33).

Conclusions:

Findings suggest that identifying and alleviating high stress among pregnant women may reduce their risk of antenatal substance use and postnatal CMDs.

Introduction

National US population statistics continue to demonstrate a high prevalence of substance use among pregnant women. In 2018, tobacco and alcohol use were estimated to be 11.6% and 9.9%, respectively.¹ While the use of tobacco and alcohol leads to adverse newborn outcomes, such as preterm birth, low birthweight, birth defects, and sudden infant death syndrome,^2
–4 they also lead to well-known negative maternal health outcomes.^5,6 Postnatal common mental disorders (CMDs), such as depression and anxiety, frequently occur among substance-using women.⁷

A study in Finland showed that substance users had about a three times higher rate of postpartum depression.⁸ Another study conducted in the US indicated that smoking throughout pregnancy was associated with postpartum anxiety.⁹ Approximately 11–21% of mothers experience postnatal CMDs,^10,11 increasing the risk of adverse outcomes in infants and children, including poor maternal/infant attachment, delayed cognitive and language development, and infant temperament.^12
–14

Growing literature supports the “healthy immigrant effect,” referring to the phenomenon that immigrants coming to high-income countries often have superior health relative to comparable native-born populations, despite often being from low- and middle-income countries with poorer health outcomes.¹⁵ Because prior research has largely focused on the association between nativity (US-born versus foreign-born) and physical health,^16
–18 relatively less is known about the role of nativity in mental health, including substance use and CMDs. Moreover, the study of nativity on substance use and CMDs has been limited to specific ethnicities (e.g., Asian Americans,¹⁹ African Americans,²⁰ and Latinos²¹), or to specific subpopulations (e.g., youth²²).

While a small body of literature posits that being a US-born woman is associated with higher risk of substance use and CMDs during pregnancy compared with a foreign-born woman, these studies have had relatively small sample sizes.^20,21,23 In addition to nativity, stress is another factor associated with mental health during and after pregnancy. However, only a handful of studies suggest positive associations of stress during pregnancy in relation to substance use and CMDs.^9,24,25

To date, independent associations between factors influencing maternal substance use and CMDs have been studied, but their combined effects have been largely neglected. To our knowledge, no prior research has examined the joint role of nativity and stress in relation to antenatal substance use or postnatal CMDs. In addition, investigating these associations with attention to race/ethnicity could highlight health disparities related to stress because stress may differentially affect the mental health of racial/ethnic groups.²⁶

The primary aim of this study was to examine whether nativity and stress levels were jointly associated with antenatal substance use (tobacco and/or alcohol use) and postnatal CMDs (depression and/or anxiety) in an ethnically diverse urban cohort that includes both US-born and foreign-born mothers. Secondarily, we aimed to investigate if risk differed by ethnicity.

Methods

Study population

We analyzed the Boston Birth Cohort, a racially/ethnically diverse mother/infant cohort recruited at birth with rolling enrollment. Since 1998, participants have been enrolled at the Boston Medical Center, an urban Boston hospital with a catchment of predominantly minority, low-income, inner-city patients. The parent study included 8,565 mothers, excluding those with multiple gestation pregnancies (twins or triplets), pregnancies resulting from in vitro fertilization, and infants born with major birth defects. Details of the parent study are described elsewhere.⁴

Relevant clinical data were obtained from the electronic medical record (EMR) system. Mothers self-reported sociodemographic information as well as information on stress and substance use through a postpartum maternal questionnaire within 24–72 hours postdelivery at the bedside in the patient room. For mothers who gave birth multiple times in the study hospital, only information from the first birth was included. After excluding mothers with missing data, our study included 6,514 mothers (76.1% of the parent study population). Of these, postnatal CMDs were evaluated among 2,052 mothers (24.0% of the parent study population), a subset of the cohort that continued medical care at Boston Medical Center.

All participants provided written informed consent. The study was approved by the Institutional Review Board (IRB) of Boston University Medical Center and the Johns Hopkins Bloomberg School of Public Health.

Variable construction

Our primary outcomes of interest were antenatal substance use (smoking and/or drinking) and postnatal CMDs (depression and/or anxiety). Information on antenatal substance use was obtained from the maternal questionnaire. Tobacco use was defined as smoking throughout the pregnancy, and alcohol use was defined as drinking anytime during the index pregnancy. Information on postnatal CMDs was retrieved from the EMR. We obtained physician diagnoses of depression and anxiety based on the International Classification of Diseases, Ninth Revision (ICD-9) or Tenth Revision (ICD-10) that was utilized after 2015.

The Diagnostic and Statistical Manual of Mental Disorders (DSM-4 or DSM-5) was used for mental disorder diagnoses. Adopting a broad definition of onset of postpartum CMDs, mothers who were diagnosed with depression and/or anxiety anytime between 1 month before delivery and 1 year after delivery were classified as having a CMD.²⁷ Because of the small number of participants with the outcomes, new variables for antenatal substance use and postnatal CMDs were created by combining tobacco and alcohol use as well as depression and anxiety, respectively.

Nativity was determined with the question “Where were you born?” Participants born in the US were classified as US-born, and those born in a US territory (i.e., Puerto Rico) or in a foreign country were classified as foreign-born. A one-item question, “What was the amount of stress in your life during the index pregnancy?” was used to assess stress levels during the index pregnancy. Response categories were recorded on a three-point Likert scale: “not stressful,” “average,” and “very stressful.” The answer option “very stressful” was classified as high stress, whereas the other answer options represented low-intermediate levels of stress.

Maternal age was recorded from the EMR and used as a continuous variable. Information on marital status, education level, and income was retrieved from the maternal questionnaire. Marital status was dichotomized into married and not married (single, divorced, widowed). Maternal education was categorized into middle school and lower, high school graduate, and some college and higher. Annual household income was dichotomized into ≥$30,000 versus <$30,000. Information on parity (nulliparous or multiparous) was obtained from the EMR. For the subanalyses, we retrieved race/ethnicity from the maternal questionnaire and categorized it into black, Hispanic, white, and other.

Statistical analyses

Maternal characteristics and the outcomes of interest were examined using cross-tabulations and were calculated for the total sample, foreign-born mothers, and US-born mothers. T-tests and chi-square tests were conducted to compare the differences in maternal characteristics between foreign-born and US-born mothers. Considering the large sample size, we also reported effect size estimates taking sample size into account. Cohen's d was used to measure the degree of difference for continuous variables (i.e., maternal age), defining 0.2, 0.5, and 0.8 as small, medium, and large effect sizes, respectively.²⁸

Effect sizes for categorical variables were estimated using Cramér's V. Ranging between 0 and 1, Cramér's V < 0.1 indicates a negligible association, whereas >0.1, >0.2, and >0.4 represent weak, moderate, and relatively strong associations, respectively.^29,30 Due to the aforementioned differences in sample sizes for each outcome of interest, we used separate sample sizes (n = 6,514 for substance use and n = 2,052 for CMDs). Logistic regression analyses were performed to compare tobacco and alcohol use, as well as depression and anxiety between foreign-born and US-born mothers, controlling for annual household income, education level, marital status, age, and parity.

After creating an interaction term between nativity and stress level, we performed logistic regression analyses to study whether nativity and stress level were jointly associated with these outcomes. Models were adjusted for maternal age, education level, marital status, annual household income, and parity. We followed Knol and VanderWeele's approach for presenting interaction analyses, and examined interaction effects both on the additive (using relative excess risk due to interaction [RERI]) and multiplicative scales.³¹ To assess these associations within specific race/ethnicities, we conducted three separate subanalyses including black, Hispanic, and white mothers separately. All analyses were conducted using STATA version 15 (StataCorp, College Station, TX).

Results

Table 1 displays maternal characteristics, suggesting that more US-born mothers reported high stress during pregnancy (28.8%) than foreign-born mothers (13.0%, p < 0.001), with a moderate association between nativity and stress levels (Cramér's V = 0.20). The majority of foreign-born mothers were black (46.5%) or Hispanic (39.6%), whereas US-born mothers were mostly black (54.5%) or white (26.6%, p < 0.001, Cramér's V = 0.41). US-born mothers were more likely to be younger (p < 0.001, Cohen's d = 0.55) and unmarried (83.6%, p < 0.001, Cramér's V = 0.30) compared with foreign-born mothers. The difference in age and marital status was significant and showed moderate effect sizes.

Table 1.

Maternal Characteristics of the Study Population, Stratified by Nativity (n = 6,514)

Variable	Total, n (%)^a	Foreign-born, n (%)^a	US-born, n (%)^a	p-Value^b	Effect size^c
Total	6,514 (100.00)	3,976 (61.04)	2,538 (38.96)	—	—
Stress during pregnancy				<0.001	0.20
High	1,250 (19.19)	518 (13.03)	732 (28.84)
Low-intermediate	5,264 (80.81)	3,458 (86.97)	1,806 (71.16)
Race/ethnicity				<0.001	0.41
Black	3,232 (49.62)	1,849 (46.50)	1,383 (54.49)
Hispanic	1,937 (29.74)	1,576 (39.64)	361 (14.22)
White	812 (12.47)	136 (3.42)	676 (26.64)
Others	533 (8.18)	415 (10.44)	118 (4.65)
Maternal age at delivery (year), mean (SD)	28.21 (6.48)	29.55 (6.26)	26.12 (6.26)	<0.001	0.55
Marital status				<0.001	0.30
Married	2,251 (34.56)	1,834 (46.13)	417 (16.43)
Single/divorced/widowed	4,263 (65.44)	2,142 (53.87)	2,121 (83.57)
Education				<0.001	0.06
Middle school and lower	2,007 (30.81)	1,299 (32.67)	708 (27.90)
High school	2,207 (33.88)	1,269 (31.92)	938 (36.96)
College and higher	2,300 (35.31)	1,408 (35.41)	892 (35.15)
Annual household income				0.008	0.03
<$30,000	2,949 (45.27)	1,748 (43.96)	1,201 (47.32)
≥$30,000	3,565 (54.73)	2,228 (56.04)	1,337 (52.68)
Parity				<0.001	0.06
Nulliparous	2,928 (44.95)	1,691 (42.53)	1,237 (48.74)
Multiparous	3,586 (55.05)	2,285 (57.47)	1,301 (51.26)

Column percentages are presented.

t-Test and chi-square tests were conducted to assess differences between foreign-born and US-born mothers.

Cohen's d and Cramér's V were used to determine strength of the association for the continuous variable of maternal age and for categorical variables, respectively.

About one-third of foreign-born and US-born mothers had some college or higher education level, and US-born mothers were more likely to be high school graduates (p < 0.001). US-born mothers were more likely to have a lower household income (p = 0.008) and be nulliparous (p < 0.001). Although p-values for education, annual household income, and parity were significant, their Cramér's Vs were <0.1, indicating that the differences may be negligible after taking into the sample size. Maternal characteristics for mothers with data on CMDs (n = 2,052) were similar to those in the larger sample (Supplementary Table S1).

US-born mothers showed a higher prevalence of antenatal substance use (37.8%) and postnatal CMDs (18.6%) compared with foreign-born mothers (7.6% and 8.4%; Table 2). Adjusting for covariates, US-born mothers showed 20.9 times higher odds of tobacco use compared with foreign-born mothers (adjusted odds ratio [aOR] = 20.89, 95% confidence interval [CI]: 16.39–26.61). US-born mothers showed 2–3 times the odds of alcohol use, depression, and anxiety compared with foreign-born mothers, after adjusting for covariates.

Table 2.

Distribution of Outcomes of Interest (Substance Use Before or During Pregnancy and Postpartum Common Mental Disorders), Stratified by Nativity, and Crude and Adjusted Associations Between Nativity and Outcomes of Interest

Outcomes	Total, n (%)^a	Foreign-born, n (%)^a	US-born, n (%)^a	p-Value^b	Cramér's V^c	Crude OR (95% CI)^d	Adjusted OR (95% CI)^e
(1) Substance use
Total, n (%)	6,514 (100.00)	3,976 (61.04)	2,538 (38.96)	—	—
Any	1,262 (19.37)	302 (7.60)	960 (37.83)	<0.001	0.37	7.40 (6.42–8.53)	7.73 (6.62–9.03)
Tobacco	825 (12.67)	94 (2.36)	731 (28.80)	<0.001	0.39	16.71 (13.38–20.86)	20.89 (16.39–26.61)
Alcohol	605 (9.29)	229 (5.76)	376 (14.81)	<0.001	0.15	2.85 (2.39–3.38)	2.64 (2.93–2.92)
(2) CMDs
Total	2,052 (100.00)	1,244 (60.62)	808 (39.38)	—	—
Any	254 (12.38)	104 (8.36)	150 (18.56)	<0.001	0.15	2.50 (1.91–3.27)	2.25 (1.67–3.01)
Depression	167 (8.14)	73 (5.87)	94 (11.63)	<0.001	0.10	2.11 (1.53–2.91)	1.76 (1.24–2.49)
Anxiety	129 (6.29)	46 (3.70)	83 (10.27)	<0.001	0.13	2.98 (2.06–4.32)	2.89 (1.92–4.35)

The proportion of the positive outcomes for the total, foreign-born, or US-born mothers.

Chi-square tests were conducted to assess differences between foreign-born and US-born mothers.

Cramér's V was used to determine the strength of the association between outcomes of interest and nativity.

Univariate logistic regression was performed to assess the associations between nativity (foreign-born mothers were used as the reference group) and the outcomes of interest.

Multivariate logistic regression was performed to assess the associations between nativity (foreign-born mothers were used as the reference group) and the outcomes of interest, adjusting for annual household income, education level, marital status, age, and parity.

CI, confidence interval; CMDs, common mental disorders.

More than half (369/(369 + 363)*100 = 50.4%) and a quarter (55/(55 + 157)*100 = 25.9%) of US-born mothers with high stress were using substances and were diagnosed with CMDs (Table 3). When nativity and stress levels were analyzed jointly, results indicated that US-born mothers with high stress had 14.9 times the odds of substance use compared with being a foreign-born mother with low stress (aOR = 14.91, 95% CI: 12.09–18.39). Among mothers with low stress, being a US-born mother was associated with sevenfold higher odds of substance use compared with a foreign-born mother (aOR = 7.48, 95% CI: 6.25–8.97).

Table 3.

Interaction Between Nativity and Stress Level on the Risk of Substance Use and Common Mental Disorders

	Maternal nativity
	Foreign-born		US-born
	No. of cases/controls	OR (95% CI)	No. of cases/controls	OR (95% CI)	OR (95% CI) for nativity within strata of stress
(1) Substance use (n = 6,514)^a
Stress
Low	222/3,236	1.0	591/1,215	7.48 (6.25–8.97)	7.48 (6.25–8.97)
High	80/438	2.58 (1.95–3.40)	369/363	14.91 (12.09–18.39)	5.79 (4.34–7.01)
OR (95% CI) for stress within strata of nativity		2.58 (1.95–3.40)		1.99 (1.67–2.38)
(2) CMDs (n = 2,052)^b
Stress
Low	75/1,007	1.0	95/501	2.31 (1.64–3.27)	2.31 (1.64–3.27)
High	29/133	2.69 (1.68–4.30)	55/157	4.09 (2.72–6.15)	1.52 (0.91–2.56)
OR (95% CI) for stress within strata of nativity		2.69 (1.68–4.30)		1.77 (1.21–2.59)

Relative excess risk due to interaction (RERI) represents the joint effect of nativity and stress levels estimated on the additive scale.

ORs are adjusted for annual household income, education level, marital status, age, and parity.

Measure of interaction on additive scale: RERI (95% CI) = 5.85 (3.33–8.37); p < 0.001. Measure of interaction on multiplicative scale: ratio of ORs (95% CI) = 0.77 (0.56–1.08); p = 0.126.

Measure of interaction on additive scale: RERI (95% CI) = 0.90 (−1.67 to 1.85); p = 0.920. Measure of interaction on multiplicative scale: ratio of ORs (95% CI) = 0.66 (0.36–1.21); p = 0.176.

Among mothers with high stress, being a US-born mother was related to six times the odds of substance use compared with being a foreign-born mother (aOR = 5.79, 95% CI: 4.34–7.01). Among foreign-born mothers, having high stress increased the odds of substance use by three times, compared with having low stress (aOR = 2.58, 95% CI: 1.95–3.40). Among US-born mothers, those with high stress were associated with twice the likelihood of substance use, compared with mothers with low stress (aOR = 1.99, 95% CI: 1.67–2.38). The RERI was 5.85 (95% CI: 3.33–8.37), indicating that the estimated joint effect on the additive scale of nativity and stress was greater than the sum of the estimated effects of nativity alone and stress level alone. However, the interaction on a multiplicative scale was not significant (ratio of ORs = 0.77, 95% CI: 0.56–1.46).

Among 2,052 mothers with information on postpartum CMDs, US-born mothers with high stress had 4.1 times higher likelihood of CMDs compared with foreign-born mothers with low stress (aOR = 4.09, 95% CI: 2.72–6.15). Stress alone was a significant factor that increased the likelihood of CMDs by two to three times (aOR = 2.69, 95% CI: 1.68–4.30 foreign-born; aOR = 1.77, 95% CI: 1.21–2.59 US-born). Among mothers with low stress, being a US-born mother was associated with 2.3 times the odds of CMDs compared with foreign-born mothers with low stress (aOR = 2.31, 95% CI: 1.64–3.27), but nativity alone did not significantly increase the odds of CMDs among mothers with high stress (aOR = 1.52, 95% CI: 0.91–2.56 for nativity within strata of stress). For CMDs, the estimated joint effect was not significant either on the additive or multiplicative scale.

These patterns were largely consistent in the subanalyses that included blacks, Hispanics, and whites separately (Table 4). Compared with foreign-born mothers with low stress, the likelihood of substance use increased in order from foreign-born mothers with high stress (lowest), US-born mothers with low stress, to US-born mothers with high stress (highest). Although the odds of substance use were not significantly different across the racial groups, US-born black, Hispanic, and white mothers with high stress showed 10, 11, and 6 times higher likelihood of substance use, respectively, compared with foreign-born black, Hispanic, and white mothers with low stress (aOR = 10.41, 95% CI: 7.53–14.38 black; aOR = 11.33, 95% CI: 7.09–18.11 Hispanic; aOR = 6.24, 95% CI: 3.42–11.35 white).

Table 4.

Subanalyses: Nativity and Stress Level in Association with Outcomes of Interest (Substance Use During Pregnancy and Postpartum Common Mental Disorders), Stratified by Ethnicity

	Substance use (n = 6,514)			CMDs (n = 2,052)
	Foreign-born	US-born		Foreign-born	US-born
	OR (95% CI)^a	OR (95% CI)^a	OR (95% CI) for nativity within strata of stress	OR (95% CI)^a	OR (95% CI)^a	OR (95% CI) for nativity within strata of stress
Black only	n = 3,232			n = 1,241
Stress
Low	1.0	5.78 (4.34–7.70)	5.78 (4.34–7.70)	1.0	1.74 (1.07–2.83)	1.74 (1.07–2.83)
High	2.63 (1.74–3.99)	10.41 (7.53–14.38)	3.95 (2.60–6.01)	2.55 (1.35–4.81)	2.98 (1.68–5.27)	1.16 (0.58–2.36)
OR (95% CI) for stress within strata of nativity	2.63 (1.74–3.99)	1.80 (1.40–2.33)		2.55 (1.35–4.81)	1.71 (1.02–2.89)
Hispanic only	n = 1,937			n = 484
Stress
Low	1.0	4.10 (2.78–6.06)	4.10 (2.78–6.06)	1.0	3.90 (1.87–8.12)	3.90 (1.87–8.12)
High	2.00 (1.20–3.33)	11.33 (7.09–18.11)	5.67 (3.04–10.55)	4.27 (1.96–9.33)	5.54 (2.04–15.08)	1.30 (0.42–4.01)
OR (95% CI) for stress within strata of nativity	2.00 (1.20–3.33)	2.76 (1.66–4.59)		4.27 (1.96–9.33)	1.42 (0.49–4.11)
White only	n = 812			n = 170
Stress
Low	1.0	3.31 (1.92–5.70)	3.31 (1.92–5.70)	1.0	1.02 (0.22–4.66)	1.02 (0.22–4.66)
High	2.31 (0.88–6.06)	6.24 (3.42–11.35)	2.69 (1.11–6.53)	1.97 (0.10–39.75)	1.42 (0.29–7.03)	0.72 (0.05–11.15)
OR (95% CI) for stress within strata of nativity	2.31 (0.88–6.06)	1.88 (1.30–2.72)		1.97 (0.10–39.75)	1.39 (0.61–3.17)

Adjusted for annual household income, education level, marital status, age, and parity.

Stress level was strongly related to postpartum CMDs. Compared with foreign-born black mothers with low stress, both foreign-born mothers and US-born mothers with high stress showed three times greater likelihood of CMDs (aOR = 2.55, 95% CI: 1.35–4.81 foreign-born; aOR = 2.98, 95% CI: 1.68–5.27 US-born).

Similarly, Hispanic mothers with high stress showed four to six times the odds of CMDs, compared with foreign-born Hispanic mothers with low stress (aOR = 4.27, 95% CI: 1.96–9.33 foreign-born; aOR = 5.54, 95% CI: 2.04–15.08 US-born). Notably, stress alone was associated with a fourfold increased odds of CMDs among foreign-born Hispanic mothers (aOR = 4.27, 95% CI: 1.96–9.33), but the effect of stress alone on CMDs was not significant among US-born Hispanic mothers (aOR = 1.42, 95% CI: 0.49–4.11). Nativity and stress levels were not clearly linked to CMDs among white mothers.

Discussion

We found that US-born mothers were more likely to report high levels of stress during pregnancy compared with foreign-born mothers, and US-born mothers with high stress were most vulnerable to substance use and CMDs (among the four groups of participants). Being a US-born mother and having high levels of stress independently increased the odds of substance use. High stress levels significantly increased the odds of CMDs among mothers, but nativity was associated with increased risk of CMDs only among mothers with low stress. The interaction effect between nativity and stress was only significant on the additive scale for the substance use outcome. Similar findings were observed in subgroup analyses, including black and Hispanic mothers separately, and the effect of stress alone on the risk of CMDs was the greatest among foreign-born Hispanic mothers.

Consistent with the majority of studies suggesting that immigrants exhibit similar or more favorable health outcomes than their US-born counterparts, our findings showed that US-born mothers were more likely to use substances during pregnancy than foreign-born mothers.^32,33 Among the few studies with contrasting results, Perreira and Cortes reported that US-born and foreign-born mothers (mostly Hispanic) were at an equal risk for alcohol consumption during pregnancy, while US-born mothers were at increased risk of smoking during pregnancy.

Different methods of measuring or including outcomes of interest and stress-related variables could explain these inconsistencies. Perreira and Cortes measured alcohol and tobacco use as separate outcomes, and included maternal stress during pregnancy as a covariate, which assessed the burden of unwanted pregnancy and perceived neighborhood insecurity.³³

In contrast, due to the small number of substance users, we combined alcohol and tobacco use when analyzing the joint effect of nativity and stress. Stress during pregnancy was included as one of the exposure variables to be examined in relation to antenatal substance use. Furthermore, half of our sample was black, and one-third was Hispanic. Given these differences in the measurement and composition of immigrants studied, further prospective studies are needed to confirm the association between nativity and substance use related to stress levels.

In our study, unmarried status was frequent, suggesting that these women may have less social support available from partners. While spouses play critical roles in provision of social support,³⁴ support from relatives or friends could be alternative resources that reduce stress and improve maternal mood.^35,36

Although only a few studies have examined the effects of social support on mental health of single mothers, one study showed that single mothers' participation in a 10-week group program offering social support clearly improved the mothers' mood and self-esteem compared with that of mothers in the control group.³⁷ Another study demonstrated that single mothers (mostly low-income black and white) who participated in a 13-week parenting and self-care group program showed reduced depression, posttraumatic stress disorder, and caregiving helplessness.³⁸ Similar interventions may be helpful in decreasing stress levels for unmarried mothers in our sample, thereby reducing the risk of CMDs.

Some researchers posit substance use as a coping strategy to deal with life stress,^39,40 and in our study, more than half of US-born mothers with high stress engaged in substance use. However, it is considered a dysfunctional coping strategy compared with other stress-reduction strategies given that sustained substance use is detrimental to physical and mental health.^39,40 One study reported that women who smoked to cope with emotional problems had double the odds of continuing to smoke during pregnancy.⁴¹ The tendency for highly stressed mothers to smoke during pregnancy may be one explanation for smoking among US-born participants.

The development of appropriate care strategies for stress management may be one direction for future interventions aiming to reduce substance use among pregnant women. Likewise, the development of social support interventions may be another strategy for improving postnatal mental health. While social support may be protective for mental health problems across diverse race/ethnicities,⁴² greater social support from family and friends among immigrants has been associated with higher life satisfaction, resulting in fewer mental health problems.⁴³ For example, one study demonstrated that familialism was stronger among pregnant Latinas (both foreign-born and US-born) than among pregnant non-Hispanic whites. High levels of familialism were associated with high social support as well as low stress, and associations were stronger for Latinas.⁴⁴

In the subanalyses, our study indicated that foreign-born mothers were more vulnerable to high levels of stress than US-born mothers, although differences in the effects of stress were not significantly different between the two groups. One plausible explanation may be that foreign-born mothers with less social support may perceive higher levels of stress, causing them to be at higher risk of CMDs compared with their US-born counterparts. Therefore, enhancing social support among foreign-born mothers may be necessary for reducing their stress levels.

Our study has several limitations. First, although stress levels and substance use during pregnancy were reported within 1–3 days of delivery, they could be subject to recall bias. Although the one item used to measure stress during pregnancy in our study has good face validity, further research is needed using more robust validated measures.

Second, only tobacco and/or alcohol use was included in our definition of substance use, limiting our ability to understand the effects of other substances used during pregnancy. Data on tobacco and alcohol use were derived from maternal self-report, and normative reporting on substance use may vary by country of origin. Therefore, social desirability bias could differentially affect the likelihood of substance use being reported between foreign-born and US-born mothers. In addition, the assessment of alcohol use was defined as consuming any alcohol during pregnancy, which may have captured a wide range of alcohol consumption. However, the prevalence of antenatal tobacco and alcohol use among our US-born sample was slightly greater than the national average.¹

Third, we conducted a complete case analysis, which could bias our estimates if missingness was not completely at random. For instance, if participants with high stress decided not to report stress levels in the maternal questionnaire, we may have a decreased prevalence of high stress in our sample. For the analysis of CMDs, we only included a subset of the cohort who continued medical care after delivery. These participants may have more physical or mental health problems than participants who discontinued utilizing the medical care.

Fourth, this study was cross-sectional; our stress during pregnancy exposure and antenatal substance use outcome may be concurrent. Therefore, the findings cannot be interpreted as causal. In addition, although we adjusted for factors related to maternal substance use and CMDs, residual confounding by unmeasured factors (such as a history of CMDs before pregnancy) remains a possibility. Finally, our participants were limited to foreign-born and US-born mothers enrolled in an inner-city hospital; thus, the results are likely to be only generalizable to similar contexts.

Despite these limitations, a considerable strength of our study was the use of a uniquely large birth cohort of predominantly low-income minority mothers. Having a large sample size, we were able to investigate less robust associations. Furthermore, examining the effect of perceived stress during pregnancy, a potentially modifiable risk factor, allowed us to provide evidence for future intervention strategies aimed at reducing the prevalence of antenatal substance use and postnatal CMDs.

Conclusions

Our findings suggest that in addition to nativity, stress is another factor that increases the risk of antenatal substance use and postnatal CMDs among US-born women, who were more likely to be young, unmarried, less educated, and low-income, compared with their foreign-born counterparts. We observed similar findings in subanalyses of black and Hispanic mothers. We may be able to reduce substance use and CMDs by developing appropriate care strategies for stress management among pregnant women. Furthermore, additional studies with a more robust measure of stress (e.g., with higher variance in stress levels and sources of stress) may provide new insights into preventing antenatal substance use and postnatal CMDs.

Footnotes

Author Contributions

All authors meet the journal's authorship requirements and approved the final article. S.P., Y.J., X.W., and P.J.S. developed the concept for this analysis. S.P. and P.J.S. drafted the article. S.P. conducted the data analysis. X.H. maintained and managed the database. B.Z. supervised the field data collection. X.W. was responsible for the initiation, overall development, and oversight of the study and its measures. All authors contributed to the analysis plan, edited the article, and interpreted data.

Author Disclosure Statement

The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

Funding Information

This work was supported by the Health Resources and Services Administration (HRSA) of the U.S. Department of Health and Human Services (HHS) under grant numbers R40MC27443 and UJ2MC31074. The Boston Birth Cohort (the parent study) is supported, in part, by the March of Dimes PERI grants (20-FY02–56 and 21-FY07–605); and the National Institutes of Health (NIH) grants (R21HD085556, R21ES011666, R21HD066471, R21AI079872, 2R01HD041702, R01HD086013, R01HD098232, and R01ES031272).

The content and conclusions contained in this article are those of the authors and should not be construed as the official position or policy of, nor should any endorsements be inferred by, the HRSA, HHS, or the U.S. Government.

Supplementary Material

Supplementary Table S1

References

Substance Abuse and Mental Health Services Administration (SAMHSA). (2017). Key substance use and mental health indicators in the United States: Results from the 2016 National Survey on Drug Use and Health (HHS Publication No. SMA 17-5044, NSDUH Series H-52). Rockville, MD: Center for Behavioral Health Statistics and Quality, SAMHSA. Retrieved from https://www.samhsa.gov/data/

Hackshaw

, Rodeck

, Boniface

. Maternal smoking in pregnancy and birth defects: A systematic review based on 173 687 malformed cases and 11.7 million controls. Hum Reprod Update, 2011; 17:589–604.

Shiono

, Klebanoff

, Rhoads

. Smoking and drinking during pregnancy: Their effects on preterm birth. JAMA, 1986; 255:82–84.

Wang

, Zuckerman

, Pearson

, et al. Maternal cigarette smoking, metabolic gene polymorphism, and infant birth weight. JAMA, 2002; 287:195–202.

Bradley

, Badrinath

, Bush

, Boyd-Wickizer

, Anawalt

. Medical risks for women who drink alcohol. J Gen Intern Med, 1998; 13:627–639.

Samet

, Yoon

S-Y

. Gender, women, and the tobacco epidemic. Geneva, Switzerland: World Health Organization, 2010.

Ross

, Dennis

. The prevalence of postpartum depression among women with substance use, an abuse history, or chronic illness: A systematic review. J Womens Health, 2009; 18:475–486.

Pajulo

, Savonlahti

, Sourander

, Ahlqvist

, Helenius

, Piha

. An early report on the mother-baby interactive capacity of substance-abusing mothers. J Subst Abuse Treat, 2001; 20:143–151.

Farr

, Dietz

, O'Hara

, Burley

, Ko

. Postpartum anxiety and comorbid depression in a population-based sample of women. J Womens Health, 2014; 23:120–128.

10.

Fairbrother

, Young

, Janssen

, Antony

, Tucker

. Depression and anxiety during the perinatal period. BMC Psychiatry, 2015; 15:206.

11.

Reck

, Struben

, Backenstrass

, et al. Prevalence, onset and comorbidity of postpartum anxiety and depressive disorders. Acta Psychiatr Scand, 2008; 118:459–468.

12.

Netsi

, Pearson

, Murray

, Cooper

, Craske

, Stein

. Association of persistent and severe postnatal depression with child outcomes. JAMA Psychiatry, 2018; 75:247–253.

13.

Park

, Greene

, Melby

, Fujiwara

, Surkan

. Postpartum depressive symptoms as a mediator between intimate partner violence during pregnancy and maternal-infant bonding in Japan. J Interpers Violence, 2021; 36:NP10545–NP10571.

14.

Britton

. Infant temperament and maternal anxiety and depressed mood in the early postpartum period. Women Health, 2011; 51:55–71.

15.

Fennelly

. The “healthy migrant” effect. Minn Med, 2007; 90:51–53.

16.

Kandula

, Lauderdale

, Baker

. Differences in self-reported health among Asians, Latinos, and non-Hispanic Whites: The role of language and nativity. Ann Epidemiol, 2007; 17:191–198.

17.

Jatrana

, Pasupuleti

SSR

, Richardson

. Nativity, duration of residence and chronic health conditions in Australia: Do trends converge towards the native-born population?. Soc Sci Med, 2014; 119:53–63.

18.

Cho

, Frisbie

, Hummer

, Rogers

. Nativity, duration of residence, and the health of Hispanic adults in the United States. Int Migr Rev, 2004; 38:184–211.

19.

Takeuchi

, Zane

, Hong

, et al. Immigration-related factors and mental disorder among Asian Americans. Am J Public Health, 2007; 97:84–90.

20.

Cabral

, Fried

, Levenson

, Amaro

, Zuckerman

. Foreign-born and US-born Black women: Differences in health behaviors and birth outcomes. Am J Public Health, 1990; 80:70–72.

21.

Escobar

, Nervi

, Gara

. Immigration and mental health: Mexican Americans in the United States. Harv Rev Psychiatry, 2000; 8:64–72.

22.

Bhattacharya

. Drug abuse risks for acculturating immigrant adolescents: Case study of Asian Indians in the United States. Health Soc Work, 2002; 27:175–183.

23.

Zambrana

, Scrimshaw

SCM

. Maternal psychosocial factors associated with substance use in Mexican-origin and African American low-income pregnant women. Pediatr Nurs, 1997; 23:253–259.

24.

Lancaster

, Gold

, Flynn

, Yoo

, Marcus

, Davis

. Risk factors for depressive symptoms during pregnancy: A systematic review. Am J Obstet Gynecol, 2010; 202:5–14.

25.

Havens

, Simmons

, Shannon

, Hansen

. Factors associated with substance use during pregnancy: Results from a national sample. Drug Alcohol Depend, 2009; 99:89–95.

26.

Williams

. Stress and the mental health of populations of color: Advancing our understanding of race-related stressors. J Health Soc Behav, 2018; 59:466–485.

27.

Stowe

, Hostetter

, Newport

. The onset of postpartum depression: Implications for clinical screening in obstetrical and primary care. Am J Obstet Gynecol, 2005; 192:522–526.

28.

Cohen

. Statistical power analysis for the behavioral sciences, 2nd ed. New York, NY: Lawrence Erlbaum Associates, 1988.

29.

Rea

, Parker

. Designing and conducting survey research. San Francisco, CA: Jossey-Bass, 1992.

30.

Cramér

. The two-dimensional case. In: Cramér H, ed. Mathematical methods of statistics. Princeton, NJ: Princeton University Press, 1946:260–290.

31.

Knol

, VanderWeele

. Recommendations for presenting analyses of effect modification and interaction. Int J Epidemiol, 2012; 41:514–520.

32.

Elo

, Culhane

. Variations in health and health behaviors by nativity among pregnant black women in Philadelphia. Am J Public Health, 2010; 100):2185–2192.

33.

Perreira

, Cortes

. Race/ethnicity and nativity differences in alcohol and tobacco use during pregnancy. Am J Public Health, 2006; 96:1629–1636.

34.

Dehle

, Larsen

, Landers

. Social support in marriage. Am J Fam Ther, 2001; 29:307–324.

35.

Murry

VMB

, Bynum

, Brody

, Willert

, Stephens

. African American single mothers and children in context: A review of studies on risk and resilience. Clin Child Fam Psychol Rev, 2001; 4:133–155.

36.

Luthar

, Ciciolla

. Who mothers mommy? Factors that contribute to mothers' well-being. Dev Psychol, 2015; 51:1812–1823.

37.

Lipman

, Waymouth

, Gammon

, et al. Influence of group cohesion on maternal well-being among participants in a support/education group program for single mothers. Am J Orthopsychiatry, 2007; 77:543–549.

38.

Muzik

, Rosenblum

, Alfafara

, et al. Mom Power: Preliminary outcomes of a group intervention to improve mental health and parenting among high-risk mothers. Arch Womens Ment Health, 2015; 18:507–521.

39.

Wills

, Shiffman

. Coping and substance use: A conceptual framework. In: Shiffman

, Wills

, eds. Coping and substance use. San Diego, CA: Academic Press, 1985:3–24.

40.

Wills

, Hirky

. Coping and substance abuse: A theoretical model and review of the evidence. In: Zeichnec

, Eudler

, eds. Handbook of coping: Theory research, and applications. New York, NY: Wiley, 1996:279–302.

41.

Lopez

, Konrath

, Seng

. Abuse-related post-traumatic stress, coping, and tobacco use in pregnancy. J Obstet Gynecol Neonatal Nurs, 2011; 40:422–431.

42.

Surkan

, Peterson

, Hughes

, Gottlieb

. The role of social networks and support in postpartum women's depression: A multiethnic urban sample. Matern Child Health J, 2006; 10:375–383.

43.

Hombrados-Mendieta

, Millán-Franco

, Gómez-Jacinto

, Gonzalez-Castro

, Martos-Méndez

, García-Cid

. Positive influences of social support on sense of community, life satisfaction and the health of immigrants in Spain. Front Psychol, 2019; 10:Article 2555.

44.

Campos

, Schetter

, Abdou

, Hobel

, Glynn

, Sandman

. Familialism, social support, and stress: Positive implications for pregnant Latinas. Cult Divers Ethn Minor Psychol, 2008; 14:155–162.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB