Maternity Leave, Race, and Postpartum Depression: The Impact of Public Policy on Mothers Living in Poverty

Abstract

Background:

Some maternity leave practices are associated with postpartum depressive (PPD) symptoms which in turn are linked with increased risk of poor maternal and infant outcomes. Mothers of color disproportionately experience poor maternity leave policies and elevated PPD symptoms.

Methods:

This cross-sectional study employed data collected at baseline of the Baby’s First Year study. Participants were 153 American mothers living below the poverty line. They were interviewed (2018-2019) in the days following birth regarding their race, maternity leave, and PPD symptoms.

Results:

Findings indicated that alone race and maternity leave practices were not associated with PPD symptoms. However, the interaction between the two accounted for significant variability in PPD symptoms. Black/African American mothers who reported that they used unpaid maternity leave (n=41, M=8.146, SD=0.708) and mothers in the “Other” race category who reported not taking maternity leave (n=7, M=8.857, SD=1.715) had higher than expected rates of PPD.

Conclusions:

These findings show that on the whole, poor maternity leave practices have a disproportionate impact on mothers of color. When women of color are exposed to either no maternity leave or unpaid maternity leave, they are at elevated risk of developing PPD.

Postpartum depression (PPD) is a serious psychiatric condition and a widespread public health concern. PPD is marked by depressed mood, loss of interest or pleasure in activities, crying, irritability, or restlessness, and disruptions to sleep and appetite and effects nearly one in seven mothers [1]. The widespread nature of PPD is particularly concerning as PPD increases the risk of maternal and infant mortality rates [2] and along with other postpartum mood disorders cost the country $14.2 billion per year [3]. For these reasons, PPD has been the focus of substantial study. Researchers have investigated individual-level interventions to reduce PPD, including Cognitive Behavior Therapy [4], Dialectical Behavior Therapy [5], and psychopharmacological interventions [6]; however, effects of those interventions are modest in that even after a full course of treatment many mothers still report depressive symptoms [7] Furthermore, individual-level interventions are associated with high costs and poor rates of both treatment initiation and retention, whereas primary prevention at the macro-level (e.g., public policy) may be a more cost-effective approach to reach the greatest number of individuals [8].

Recently, politicians, researchers, and healthcare providers have begun to call for system-level changes to prevent PPD [9]. Indeed, such changes would hopefully ease burdens on new families. One system-level policy of interest is nationwide paid family leave. Such a policy could reduce the likelihood PPD symptoms among new mothers who return to work in the weeks following delivery of a child. Importantly the literature reflects a strong link between PPD symptoms and maternal leave practices. For example, a prospective cohort study of 465 women from the Wisconsin Maternity Leave and Health Project found that mothers who returned to work in the four weeks following the birth of a child were significantly more likely to develop depression than those who did not [9]. Likewise, in a nationally representative sample of 177 mothers who took maternity leave of twelve weeks or less, every additional week of leave was associated with a reduced odds of experiencing PPD symptoms [10]. Further, the Maternal Postpartum Health Study, a prospective study with a year-long follow-up period among mothers living in St. Paul, Minnesota and Minneapolis, Minnesota, found that in the first postpartum year, an increase in leave duration was associated with a decrease in depressive symptoms until six months postpartum [11]. Together these studies evidence a link between maternity leave and reduced odds of PPD that is not entirely surprising. Indeed, countries that legislate maternity leave fare much better with regards to infant and maternal outcomes than the US [12]. For example, New Zealand, Portugal, Spain, and Germany all legislate paid family leave and have lower infant and maternal mortality rates and higher rates of breastfeeding and infant vaccinations than the US [12]. Across many metrics paid family leave improves outcomes for moms and infants [13,14, 15]. It has been hypothesized that paid maternity leave allows for mothers and infants to rest and recover from labor and birthing processes, bond emotionally, establish strong breastfeeding relationships, and improve availability to attend pediatrician appointments in the months following birth [16] . Regardless of the specific processes at play, a link between nationwide paid maternity leave and improved maternal and infant outcomes has been established.

Unfortunately, both PPD [17] and lack of paid maternity leave policies [18] disproportionately affect mothers of color as well as mothers who are poor. Poor women and Black/African American women report that they are more likely to have PPD and less likely to have paid maternity leave than women who are White and living above the poverty line [19]. There is a dearth of literature that examines the relationship between maternity leave policies and PDD symptom prevalence, and most of the work that has looked at maternity leave and PPD has been in nationally representative samples, not focusing on the populations most at risk. Looking specifically at the vulnerable populations of impoverished mothers of color could elucidate whether poor maternity leave practices are especially burdensome on this group of mothers who already face high barriers to physical and mental healthcare.

It is imperative that we study the link between maternity leave and PPD symptoms in impoverished mothers, particularly given recent political events that have and will continue to impact maternal and infant outcomes. Specifically, through a leaked report published in Politico (2022) and the 2022 ruling (Dobbs v. Jackson’s Women’s Health Organization; argued December 2021, decided June 2022), the Supreme Court of the United States overturned Roe v. Wade, the landmark case protecting a woman’s right to privacy in procuring abortion services. Indeed, overturning Roe v. Wade relinquished this decision to states, many of which passed “trigger laws” that immediately outlawed abortion services upon the overturning of Roe v. Wade. The consequences associated with denial of abortion health care services are deleterious. Indeed, women denied abortions were at 3.8-fold higher risk of poverty 6-months after abortion denial and were more likely to remain in poverty 4 years after the denial of abortion compared to women with access to abortion services [20]. The coming years will likely witness the ushering in of countless impoverished mothers and infants, and we need to understand how national policies related to maternity leave relate to PPD. Indeed, the changes in abortion access thus will increase the number of infants among mothers of lower socioeconomic status, and PPD and maternal leave need to be studied among this population.

Our investigation is also timely in that there are concerted efforts to legislate nationwide maternity leave. Quantitative investigation into the link between maternity leave and PPD would likely influence these legislation efforts and may provide additional support to adopt these practices in the US. Furthermore, identifying the demographic groups that might benefit the most from maternity leave practices would allow a more targeted roll out (and a more immediate improvement in PPD symptoms) following the potential implementation of a nationwide family leave policy. These efforts are likewise needed in that while millions of Americans support some form of paid family leave, politicians have been reluctant to pass legislation related to paid family leave, pledging to block all efforts for nationwide family. As such, investigations into the link between maternity leave and PPD are needed to promote evidence-based public policy.

We examined the link between maternity leave in the six months following their child’s birth and PPD symptoms in this population of interest. Furthermore, we investigated the specific role that race plays in the relationship, investigating if the link between maternity leave (paid, unpaid, and no leave) and PPD symptoms varies by race. Based on previous findings [10,11], we hypothesized that women who did not take maternity leave will exhibit greater depressive symptoms. Likewise, stemming from previous findings [17], we also hypothesized that this will be more pronounced in women of color compared to White women. Indeed, mothers of color have been found to encounter significant hurdles in mothering, more so than White mothers, that may contribute to PPD symptoms. For example, children of color have been found to be more likely than White children to be expelled from schools (even daycare facilities) even while exhibiting no more disruptive behavior than White children. Likewise, when diagnosed with development delays, children of color have also been found to be significantly less likely than White children to be offered, engage with, and complete early intervention services [21]. The burden of mothering, including providing care for children expelled from school or daycare and parenting young children without the support of early intervention services, appears to vary by race. While both of these findings relate closely to children beyond the immediate newborn phase, they are indicative of the widespread and seemingly extensive difficulties encountered by mothers of color. Findings from the present project may inform legislative and public policy efforts, potentially guiding implementation changes that would benefit the most vulnerable members of the United States population.

Methods

Study Setting

This is a cross-sectional study that employed baseline data from the Baby’s First Years project. In the present study we seek to investigate the link between maternity leave, race, and PPD among mothers who are living below the poverty line. We study this within the first of Baby’s First Years data collection. These data were collected during infants’ first days of life among babies born between May 2018 and June 2019. The overall goal of the Baby’s First Years project was to provide understanding of the extent to which household income plays a causal role in affecting children’s cognitive, socio-emotional, and brain development early in life for children born into low-income families. This data set is unique in that it specifically studies a portion of the population that is at heightened risk for negative maternal and infant outcomes, namely, mothers and infants below the federal poverty line. To be eligible, mothers’ self-reported income in the prior calendar year had to fall below the federal national poverty threshold for their family size.

Participants and Procedures

Participants were selected from the Baby’s First Year dataset (N = 1,050; [22]), which studied mothers living below the federal poverty line in New York City, greater New Orleans, the Twin Cities, and the Omaha metropolitan areas. From the entire sample of 1,050 mothers, participants from the present study were selected on the basis of answering “yes” to an item “Have you worked outside of the home in the last year?,” questions related to maternity leave, and questions related to PPD. This sample that endorsed having worked outside of the home in the last year consisted of 153 mothers (the representativeness of these 153 of the larger sample of 1,050 is addressed below in the Results section). Participants completed informed consent as a part of data collection, and the Institutional Review Board (IRB) of Teachers College Columbia University served as the single IRB of record for most of the study sites. Standalone IRB reviews were conducted in 5 of the 12 recruitment hospitals.

Measures

Race.

Mothers’ race was assessed during the demographic portion of the Baby’s First Year questionnaire. Race was coded for the current study as White, Black, “Other” which included Asian or Pacific Islander, American Indian, Eskimo, Aleut.

Ethnicity.

Mothers’ ethnicity was also assessed in the demographic portion of the study and was coded as Hispanic or Latino or Not Hispanic or Latino. This item was separate from items assessing race.

Maternity leave.

For mothers who were returning to work in the year following childbirth, they were asked about their maternity leave status: paid maternity leave, unpaid maternity leave, or no maternity leave.

Center for Epidemiological Studies Depression Scale (CES-D; [23]).

The CES-D is a ten-item measure of depressive symptoms including hopelessness, difficulty concentrating, loneliness, feeling like a failure, and feeling fearful. The CES-D was developed for use in the general public and has been deployed among mothers during the postpartum period as well (Van Lieshout et al., 2011). Items are scored on a 0 to 3 (0 = Rarely or none of the time; 1 = Some or a little of the time, 2= Occasionally or a moderate amount of the time, 3 = Most or all of the time). The ten items are then summed to create the total CES-D score, with elevated scores corresponding to elevated depressive symptoms. Any score at or above ten is consistent with meeting criteria for depression [24]. Reliability among participants in the present sample was found to be acceptable (α = 0.68).

Inclusion criteria

Beyond living below the federal poverty line, additional study inclusion criteria were (1) the mother was of legal age for informed consent (age 18 or older in NY, MN, and LA; 19 or older in NE); (2) the infant was admitted to the newborn nursery (not an intensive care unit); (3) the mother was residing in the state of recruitment; (4) the mother indicated that she is not highly likely to move to a different state or country in the next 12 months; (5) the infant was discharged into the custody of the mother; and (6) the mother spoke English or Spanish. Baseline data were collected via interview in person at the hospital bedside of new mothers.

Data Analytic Plan

Data were analyzed in three steps. First, descriptive statistics were conducted to determine features of the sample. Then a two-way analysis of variance (ANOVA) was conducted to determine if PPD symptoms varied based on race, maternity leave, and race x maternity.

Results

Descriptive Statistics

Mothers were selected for analyses in the present project on the basis of completing questions related to maternity leave, race, and PPD. With regard to maternity leave, 24.2% (n = 37) reported that they took paid maternity leave, 61.4% (n = 94) reported they used unpaid maternity leave, and 14.4% (n = 22) reportedly did not take maternity leave. With regard to PPD symptoms, the average CES-D 10 score was 6.55 (SD = 4.63; Min = 0, Max = 22). The 153 participants within our study ranged in age from 18 to 45 years (M = 27.71, SD = 5.74). All mothers earned below the federal poverty line (combined household income, M = $25,221, SD = $23,480). The sample was racially (White, n = 46; Black/African American, n = 62; Asian or Pacific Islander, n = 2; American Indian, Eskimo, Aleut, n = 6; Other, n = 36) and ethnically diverse (Hispanic or Latino, n = 56; Not Hispanic or Latino, = 96). Of respondents, 74.5% (n = 114) reported that they were born in the United States and 25.5% (n = 30) reported that they were immigrants. Seventy-nine percent (n = 122) of mothers were not married to the biological father of their children. With regards to education, 69.8% (n = 78) reported earning a twelfth grade education or lower, and only fourteen mothers reported earning a Bachelor’s degree (n = 9), Associate’s degree (n = 4), or Master’s degree (n = 1).

Notably, the participants who were selected for use in this study (n = 153) did not vary from the participants in the Baby’s First Years sample who were not selected for use in this study (n = 997) based on PPD depressive symptoms (t = 0.898, p = 0.42) or race (t = 1.629, p = 0.44). However, the sample used in the present study reported higher socio-economic status as measured by total annual household income (M = $14,924, SD = $15,569) than those who were not used in the present study (M = $6,371, SD = $8,219; t = 9.019, p = 0.01).

Maternity Leave’s and Race’s Relationships with PDD Symptoms

To investigate if PPD symptoms varied based on maternity leave policy, a one-way ANOVA was completed with means and standard deviations displayed in Table 1. Results from the one-way ANOVA indicated that individuals with access to paid maternity leave (M = 6.24) had fewer, albeit not statistically significant, PDD symptoms than individuals with unpaid maternity leave (M = 6.57). Further, mothers who had no maternity leave exhibited the highest number of PPD symptoms (M = 7.00). However, none of these differences were statically significant (F[2,142] = 0.167, p = 0.85).

Table 1.

Postpartum Depressive Symptoms by Maternity Leave Policy

	n	M	SD
Yes - Paid Maternity Leave	37	6.24	4.051
Yes - Unpaid Maternity Leave	93	6.57	4.598
No	22	7	5.936
Total	152	6.55	4.663
White	46
Black/African American	62
Asian or Pacific Islander	2
American Indian, Eskimo, or Aleut	6
Other	36
Hispanic or Latino	56
Not Hispanic or Latino	96

Note: Postpartum depressive symptoms = the sum score from the Center for Epidemiological Studies Depression Scale.

Next, based on results from the one-way ANOVA, the link between race and PDD symptoms was investigated via a two-way ANOVA. Results indicated that race did not predict PDD symptoms (F[2,145] = 0.193, p = 0.83; Table 2). Women identifying as Black/African American reported the highest level of PDD symptoms, whereas white women reported the lowest level of PDD symptoms. However, as noted, these differences were not significant.

Table 2.

Postpartum Depressive Symptoms By Race x Maternity Leave

Race	Maternity Leave Policy	n	M	SD
White	Yes - Paid Maternity Leave	5	3.00	2.029
	Yes - Unpaid Maternity Leave	33	5.939	0.79
	No	8	7.625	1.604
	Total	46	5.91	4.273
Black or African American	Yes - Paid Maternity Leave	14	6.857	1.212
	Yes - Unpaid Maternity Leave	41	8.146	0.708
	No	7	4.429	1.715
	Total	62	7.44	4.576
Other	Yes - Paid Maternity Leave	17	6.765	1.1
	Yes - Unpaid Maternity Leave	19	4.263	1.04
	No	7	8.857	1.715
	Total	43	6	5.122
Entire Sample	Total	151	6.56	4.677

Note. Post-partum depressive symptoms = the sum score from the Center for Epidemiological Studies Depression Scale.

Moderation Analysis

To investigate if the link between maternity leave and PDD symptoms varied based on race, results from the race x maternity leave level of the two-way ANOVA were examined. Findings indicated that race moderated the relationship between maternity leave and PDD symptoms (F[4,142] = 3.267, p = .013; Table 2). White mothers who did not take maternity leave, Black mothers who took unpaid maternity leave, and women who reported “Other” for race who did not take maternity leave reported elevated PPD symptoms. Conversely, mothers who took paid maternity reported lower PPD symptoms among every racial demographic group.

Pairwise comparisons were analyzed via Tukey’s posthoc tests. These results are displayed in Table 3 and indicate that Black/African-American mothers who used unpaid maternity leave (n = 41, M = 8.146, SD = 0.708) and mothers in the “Other” race category who did not take maternity leave (n = 7, M = 8.857, SD = 1.715) had higher than expected rates of PPD. Notably, to ensure that effects were not better accounted for by variability in socio-economic status, the above analyses were conducted with and without the covariate of socio-economic status as measured by annual household income. Whether or not socio-economic status was controlled, the results exhibited the same pattern: (1) race by itself was not predictive of PPD symptoms, (2) maternity leave by itself was not predictive of PPD symptoms, (3) race x maternity leave was predictive of PPD symptoms. See Figure 1 for the plotted interaction effect

Table 3.

Pairwise Comparisons of Postpartum Depressive Symptoms by Maternity Leave Policy x Race

Mother race	Maternity Leave Policy		Mean Difference	SE	p-value	95% Confidence Interval for Difference
White	Yes - paid maternity leave	Yes - unpaid maternity leave	−2.939	2.177	0.179	−7.243	1.364
		No	−4.625	2.586	0.076	−9.737	0.487
	Yes - unpaid maternity leave	Yes - paid maternity leave	2.939	2.177	0.179	−1.364	7.243
		No	−1.686	1.788	0.347	−5.219	1.848
	No	Yes - paid maternity leave	4.625	2.586	0.076	−0.487	9.737
		Yes - unpaid maternity leave	1.686	1.788	0.347	−1.848	5.219
Black or African American	Yes - paid maternity leave	Yes - unpaid maternity leave	−1.289	1.404	0.36	−4.065	1.487
		No	2.429	2.1	0.249	−1.722	6.58
	Yes - unpaid maternity leave	Yes - paid maternity leave	1.289	1.404	0.36	−1.487	4.065
		No	3.718*	1.855	0.047	0.051	7.385
	No	Yes - paid maternity leave	−2.429	2.1	0.249	−6.58	1.722
		Yes - unpaid maternity leave	−3.718*	1.855	0.047	−7.385	−0.051
Other	Yes - paid maternity leave	Yes - unpaid maternity leave	2.502	1.514	0.101	−0.492	5.495
		No	−2.092	2.037	0.306	−6.119	1.935
	Yes - unpaid maternity leave	Yes - paid maternity leave	−2.502	1.514	0.101	−5.495	0.492
		No	−4.594*	2.006	0.023	−8.559	−0.629
	No	Yes - paid maternity leave	2.092	2.037	0.306	−1.935	6.119
		Yes - unpaid maternity leave	4.594*	2.006	0.023	0.629	8.559

Note. Postpartum depressive symptoms = Center for Epidemiological Studies Depression Scale (CES-D) total score. * = significant at p < .05.

Figure 1.

Interaction Effects Between Maternal Race x Maternity Leave and PPD

Exploratory Analyses

Given that more than a third of the sample identified as Hispanic or Latino, an additional two-way ANOVA was conducted to investigate the relationships between ethnicity, maternity leave, and ethnicity x maternity leave with PPD symptoms. Notably, ethnicity and race were scored independently. These additional analyses were not directly related to either a priori hypothesis but instead were conducted to better understand the role that maternity leave practices play among Hispanic or Latino mothers. Given that Hispanic or Latino mothers accounted for such a large portion of the sample, it was imperative that we conduct this additional work to understand their unique experience.

Results from that two-way ANOVA indicated that Hispanic or Latino mothers (M = 6.18) reported fewer PPD symptoms as compared to mothers who were not Hispanic or Latino (M = 6.76). However, this difference was not statistically significant (F[1, 145] = .281, p = .59). Further, within this model, the association between maternity leave and PPD symptoms was likewise not significant (F[2, 151] = .686, p = .51). Finally, the two-way interaction between ethnicity x maternity leave was also not significant (F[2, 151] = 2.280, p = .11). Together, these findings indicate that ethnicity, maternity leave, and ethnicity x maternity leave were not associated with PPD symptoms within our sample.

Discussion

This study investigated the link between race, maternity leave, and PPD symptoms in a sample of mothers living below the federal poverty line. In this sample, when studied in isolation, race and maternity leave policies were not directly associated with PPD symptoms. That is, race alone and maternity leave policy alone did not confer risk for PPD. However, the interaction between the two was significantly associated with variation in PPD symptoms in mothers living below the poverty line. Specifically, Black/African American mothers who reported that they used unpaid maternity leave and mothers in the “Other” race category who did not take maternity leave had higher than expected levels of PPD symptoms. Likewise Hispanic or Latino mothers who had no maternity leave also experienced elevated symptoms of PPD. These results demonstrate that poor maternity leave policies place a disproportionate burden on mothers of color (i.e., mothers who are Black/African American, “Other”, Hispanic, or Latino).

The hypotheses driving the present work were agnostic to the role that race alone would play in the variation of PPD symptoms. Instead, our hypothesis proposed that race would interact with maternity leave policy to account for significant variation in PPD symptoms. The finding that race was not associated with PPD symptoms was not necessarily surprising in that there is wide experience of PPD symptoms among postpartum mothers of each race (Liu & Tronic, 2013). However, race may serve as a proxy measure for related factors including chronic exposure to institutional racism, experience of microaggressions, and exposure to traumatic events (Gee et al., 2019). These potential proxy measures may be increased among mothers exposed to poverty and thus interacting with and amplifying the deleterious effects of maternity leave policies.

These findings have clinical and policy implications that should be considered for implementation. Our findings echo that particularly among Black/African American mothers and those who endorsed “Other” as their race as well as mothers who are Hispanic or Latino, national paid family leave policies may confer measurable benefits on maternal outcomes. In other words, the findings from this work indicated that the lack of nationwide paid family leave policies took the greatest toll on impoverished mothers of color. This builds upon previous literature indicating that lack of abortion access is greatest among states with highest rates of poverty (Solazzo, 2019). This is a salient message that must be emphasized not only to researchers but also policy makers and elected officials. These leaders are tasked with the care of millions of families, among those are mothers in poverty who are Black/African American, “Other”, Hispanic, or Latino. Presently, based on our work, these groups are at greatest risk of the deleterious effects of poverty on their mental health.

Indeed, elected officials, largely without formal education in medicine, statistics, or epidemiology, disseminate rhetoric regarding “pro-life” policies which in practice appear to relate more closely to simply reducing access to abortion care. These efforts to restrict abortion care will result in the introduction of countless mothers and families of color who are forced to parent below the poverty line. These mothers and their children who are exposed to poverty are at great risk. Our study shows that these poor mothers of color with the lowest access to generous maternity leave policies will also be most likely to shoulder the burden of PPD. This is particularly alarming in that PPD symptoms have been repeatedly linked with developmental deficits in offspring, failure to achieve breastfeeding standards, and even elevated rates of maternal and infant mortality [25, 30]. Stated directly, we urge policy makers to legislate paid family leave particularly for Black/African American mothers, other mothers of color, and Hispanic mothers who are at risk of living below the poverty line. Indeed, with reduced access to abortion services that occurred as a result of the 2022 Dobbs decision and the increased number of mothers in poverty who will result, legislation for paid maternity leave policies is needed imminently. Indeed, Black/African American mothers, other mothers of color, and Hispanic mothers in poverty are at high risk for PPD which is associated with increased morbidity and mortality among infants and mothers [2,26]. Our results show that paid maternity leave could ease the burden of PPD.

It is with this in mind that we implore legislators to work together to pass a bill earmarking federal dollars for a nationwide paid maternity leave policy especially for the poorest of mothers/families of color. Poor mothers of color are the least likely group to have access to paid maternity leave [18] and they have poor outcomes related to maternal and infant morbidity and mortality [27]. These are exactly the people who would benefit from a policy that promotes the mental and physical health and the dignity of life in parenthood. Policies that promote the welfare of individuals including nationwide access to paid family leave would ease the suffering and burden of poverty and PPD on the most vulnerable of our nation, parents and infants who have few economic resources.

Notably the type of maternity leave policies plays an important role in the experiences of mothers [10]. For mothers living in poverty, the lack of any maternity leave presents mothers with a nearly impossible choice. Mothers who have no maternity leave option are forced to choose between (1) returning to the workplace rapidly following delivery as they continue to heal and recover from delivery or (2) staying at home with their child for an extended period of time, thus relinquishing their employment altogether. Either decision is met with significant consequences in that a mother who returns to work quickly following delivery risks elevated rates of morbidity and mortality for her and her new infant and a mother who loses employment risks financial decline for her growing family [28].

Likewise, unpaid maternity leave presents challenges to new mothers [29]. Unpaid maternity leave grants the benefit of allowing a mother to keep her employment while she rests, recovers, and bonds with her infant at home; however, it burdens her with the loss of income during that period. Loss of financial resources during the postpartum phase has been linked with unmet health care needs and higher odds of health care unaffordability [29]. Both lack of maternity leave as well as unpaid maternity leave are burdensome for mothers and are reflective of inhumane leave policies. For low-income families, when faced with the option of no maternity leave or unpaid maternity leave, mothers in poverty choose between the lesser of two evils.

While our study boasts many strengths, including a timely study of the link between maternity leave policies, race, and PPD symptoms in impoverished mothers, there are significant limitations that should be considered in the interpretations and implementations of our findings. Namely the data used in the present work are cross-sectional, leaving the authors unable to investigate the temporal precedence of constructs at hand. Further, these data were collected among a relatively small sample size from mothers living in relatively large metropolitan cities, thus these findings may not be generalizable to mothers in suburban or rural parts of America. It is possible that mothers in suburban or rural parts of America may experience vastly different associations between maternity leave policies, race, and PPD symptoms. It is also possible that with a larger sample, other smaller effects may emerge that were not seen in this relatively small group of mothers. Finally, there was not an investigation into the buffering effects of protective factors against PPD symptoms. While investigation into and the understanding of link between social support and PPD symptoms would be interesting and helpful in reducing burden of PPD symptoms, those constructs and phenomena were not studied or measured in this data set.

Conclusion

PPD symptoms are associated with worse maternal and infant outcomes. In this study, Black/African American mothers who reported using unpaid maternity leave and mothers in the “Other” race category who did not take maternity leave at all had higher than expected rates of PPD. Likewise Hispanic mothers who had no maternity leave also demonstrated higher than expected PPD symptoms. We strongly recommend nationwide paid family leave policies as this could improve outcomes in mothers of color living below the federal poverty line.