Risk and Resilience Factors Underlying Postpartum Depression and Impaired Mother-Infant Bonding During COVID-19

Abstract

Acute stress during pregnancy can have adverse effects on maternal health and increase the risk for postpartum depression and impaired mother-infant bonding. The COVID-19 pandemic represents an acute environmental stressor during which it is possible to explore risk and resilience factors that contribute to postpartum outcomes. To investigate prenatal risk and resilience factors as predictors of postpartum depression and impaired mother-infant bonding, this study recruited a diverse cohort of 833 pregnant women from an urban medical center in Philadelphia, Pennsylvania, and assessed them once during pregnancy in the early phase of the COVID-19 pandemic (April–July 2020) and again at approximately twelve weeks postpartum. Adverse childhood experiences, prenatal depression and anxiety, and COVID-19-related distress predicted a greater likelihood of postpartum depression. Prenatal depression was the only unique predictor of impaired maternal-infant bonding after postpartum depression was controlled for. Women reporting greater emotion regulation, self-reliance, and nonhostile relationships had healthier postpartum outcomes. Policies to increase the number of nonspecialty providers providing perinatal mental health services as well as reimbursement for integrated care and access to mental health screening and care are needed to improve lifelong outcomes for women and their children.

INTRODUCTION

Postpartum depression is an exceedingly common condition associated with childbirth and is a major public health problem. The average prevalence of postpartum depression diagnosis in the US is 13 percent within four weeks of childbirth,¹ and rates increase as measurement time frames lengthen,² with as many as 19 percent of women affected at three months postpartum.³ Untreated postpartum depression may result in a long-term course of chronic depression in women,⁴ as well as other poor maternal outcomes, including alcohol and illicit drug use,⁵ problems with weight gain and obesity,^6,7 breastfeeding difficulties,⁸ and relationship problems.⁹ Mothers who experience postpartum depression are less responsive to infant cues and less likely to engage in positive interactions with their infants,¹⁰ disrupting their emotional bond.^11,12 Such impaired mother-infant bonding has been linked to risk for infant maltreatment;¹³ poor socioemotional, behavioral, and cognitive problems in preschool;^14,15 and psychopathology in adulthood.¹⁶ The negative child outcomes linked to maternal postpartum depression and impaired mother-infant bonding can last even after remission of postpartum depression.^14,17,18 Thus, timely perinatal identification, intervention, and prevention of postpartum depression are necessary.

Given the high prevalence rates of postpartum depression and the poor associated outcomes for both mother and child, research has attempted to identify risk factors. Prenatal depression, anxiety, and stressful life events in general are known to increase the risk of developing postpartum depression.¹ Recent work links adverse childhood experiences to risk for postpartum depression.^19,20 Research suggests that prevalence of postpartum depression is elevated among low-income populations,^19,21 whereas data on racial differences have largely been equivocal.^21,22 Limitations of this literature, however, include a reliance on retrospective self-report data in clinical studies and a paucity of prospective studies of community samples with consecutive presentations.²³ Understanding how risk factors interact in concert can help improve targets for intervention and prevention efforts.

Acute stress during pregnancy can have adverse effects on maternal psychopathology and increase risk for postpartum depression. Biological measures of stress^24,25 also suggest that the timing of a stressor during gestation may be an important consideration for this risk. Many studies suggest that the second trimester is an especially critical window during which maternal environmental stress exposure can compromise fetal development.²⁶ Although some prior research has investigated the children of women exposed to intense acute stressors during gestation, most studies^27–29 have focused only on follow-up of the child, with less attention being paid to the mother’s postpartum experience and the mother-child relationship.

The global COVID-19 pandemic represents a stressor that may have significant repercussions for postpartum depression risk and mother-child relationship development. Several studies have documented a significant increase in rates of postpartum depression during the pandemic compared with prepandemic rates.^30–33 Furthermore, the level of COVID-19-related worry and distress has been linked to heightened depression in the perinatal period.^32,34 The pandemic has also placed specific restrictions and strains on the availability of family and community support, which may be particularly harmful to women during the perinatal period. Social isolation, lack of access to child care, limited in-person medical care, financial stress due to job loss, and restricted access to material items and resources may further contribute to risk of developing postpartum depression.^32,35

Aside from the examination of risk factors, far fewer studies have investigated resilience factors, which are protective factors that may prevent the development of postpartum depression and impaired maternal bonding. Broadly, resilience factors protect against adversity or promote positive outcomes when it occurs.³⁶ Thus, identifying and intervening on resilience factors in perinatal women to help reduce postpartum depression risk and possibly improve maternal-child bonding is important, especially during a stressful period such as the COVID-19 pandemic and postpartum period. Resilience factors can have important public health implications. Prior work has shown that these factors, which include self-reliance, emotion regulation, supportive close relationships (for example, high endorsement of care and support), nonhostile close relationships (for example, low frequency of arguments and hostility), and perceived neighborhood safety, are linked to decreased depression and anxiety; better occupational, social, and psychological functioning in a community sample;³⁷ and protection against COVID-19-related distress.^34,38 However, it is unclear whether these resilience factors protect against adverse postpartum outcomes.

This study evaluated the mental health and worries of a diverse cohort of pregnant women during the early lockdown phase (April–July 2020) of the COVID-19 pandemic in the US and followed up in the early postpartum period, which stretched into 2021. The goal was to prospectively assess the link between prenatal risk and resilience factors and postpartum depression and maternal-infant bonding during an acute stressor—in this case, the early months of the COVID-19 pandemic.

Specifically, we examined a broad set of individual risk factors, including prenatal anxiety and depression, adverse childhood experiences, and pandemic-related stress (both general and pregnancy-specific COVID-19 worries). We also examined a diverse set of individual resilience factors, including self-reliance, emotion regulation, supportive close relationships, non-hostile close relationships, and perceived neighborhood safety. Our aim was to identify specific factors that most contribute to postpartum health. We argue that such information points to potential targets to inform policy change.

STUDY DATA AND METHODS

Participants And Data Collection

The study included 833 perinatal women who were patients at an urban medical center in Philadelphia, Pennsylvania (see online appendix exhibit A1 for sample characteristics).³⁹ Participants were identified through Penn Medicine medical records if they were receiving prenatal care at one of nine University of Pennsylvania Hospital System prenatal clinics (see appendix exhibit A2).³⁹ Participants were initially contacted by email with an explanation of the purpose of this longitudinal study—to examine the impact of the COVID-19 pandemic on pregnancy and the postpartum period—and an invitation for them to complete an online survey.^34,39 Online surveys designed to measure risk and resilience factors and postpartum outcomes were distributed, and data were collected using REDCap at two points: prenatal (time 1) and postpartum (time 2); delivery information was also collected but was not used in the current study. Survey completion took approximately twenty minutes at each point; not all survey sections were completed by all participants. Prenatal data (mean gestational age at time 1: 25.27 weeks) were collected in Philadelphia at time 1 during a statewide stay-at-home order and a period of high community spread of the COVID-19 virus (April 17–July 8, 2020).³⁴ The time 2 survey was completed at an average of 11.83 weeks postpartum (August 11, 2020–March 2, 2021). The University of Pennsylvania Institutional Review Board approved the study.

Measures

Prenatal Risk Factors:

Prenatal anxiety was assessed using the Generalized Anxiety Disorder seven-item scale (GAD-7),⁴⁰ and prenatal depression was assessed using the Patient Health Questionnaire two-item scale (PHQ-2).⁴¹ Acute pandemic stress was measured using a COVID-19 stressor list, with six items assessing general pandemic worries and four items assessing pregnancy-related worries.³⁸ “General pandemic worries” refers to worries about contracting or spreading COVID-19, losing loved ones to the virus, and experiencing financial hardship or job loss. Questions assessing pregnancy-specific worries were focused on participants’ worries about receiving adequate prenatal care; being able to deliver their baby safely; and being able to obtain baby supplies, such as diapers and formula. Adverse childhood experiences were assessed via a nine-item version of the Adverse Childhood Experiences (ACE) Questionnaire ⁴² that eliminated the “parental divorce/separation” item.⁴³

Prenatal Resilience Factors:

Participants completed the Brief Risk and Resilience Battery, a twenty-one-item resilience questionnaire that assesses five distinct components of resilience.³⁷ In the original form of the battery, some of these factors were reverse scored such that higher scores indicated increased risk. For this study, all factors were scored such that higher scores on the five resilience factors represent increased resilience.

Postpartum Outcomes:

Postpartum depression was assessed with a nine-item Edinburgh Postnatal Depression Scale (EPDS). We used a clinical cutoff of a score of 10 or higher out of a possible 27 to identify possible postpartum depression status.^2,44 Item 10, assessing self-harm, was excluded given the difficulty of monitoring self-harm endorsements through online assessments. The EPDS is a validated scale commonly used to assess postpartum depression in diverse populations.⁴⁴ To assess mother-infant bonding, we used the twelve-item “general impairment” scale of the Postpartum Bonding Questionnaire (PBQ). We used a clinical cutoff of a score of 12 or higher to identify possible impaired bonding status.^45,46 This scale has been shown to be valid and reliable for detecting bonding dysfunction.^45,47

Data Analysis

Logistic regression models were conducted to predict postpartum depression status and impaired mother-infant bonding status. Separate models were run for the five risk factors and the five resilience factors (entered in block 2 of the model). We compared models when the risk (or resilience) factors were entered separately in individual models and when all five risk (or resilience) factors were simultaneously entered in block 2 (that is, to establish unique effects). Covariates were entered in block 1 of the model: socioeconomic status (higher scores relate to lower socioeconomic status), multiple gestations, parity, preterm birth (less than thirty-seven weeks), relationship status, maternal age, gestational week at the time of the prenatal survey (time 1), and weeks postpartum at the time of the postpartum survey (time 2). Moderation analyses examined whether interactions between COVID-19 worries and gestational age at time 1 completion or the other risk factors increased the odds of postpartum depression. Differences by race and ethnicity comparing effects for non-Latinx Black and non-Latinx White women are reported in the appendix.³⁹ Our interpretation of results focuses on the risk and resilience predictors included in block 2. For the main analyses, after Bonferroni correction for multiple comparisons, we considered results to be statistically significant at p ≤ 0.01. Listwise deletion per regression model was used to handle missing data.

Limitations

Our results should be considered in light of several limitations. First, the measures were self-reports from survey data and not diagnostic screens. Although the measures we used have established validity, future work should include more in-depth screening and in-person assessments. Given our desire to limit the burden on participants, we used the PHQ-2 at the initial survey administration during pregnancy and followed up with the EPDS during the postpartum period. We did not administer the self-harm question in the EPDS because of our inability to review the responses in real time and provide adequate clinical oversight. Second, we conducted the study during a global pandemic. Although the pandemic provided an opportunity to understand the links between an acute stressor and postpartum outcomes, it is unclear how this stressor may generalize to other acute stressors. Prenatal data were collected during the first three months of the COVID-19 shelter-in-place restrictions in Philadelphia, with 84 percent collected during the first four weeks. Thus, we referred to the stressor as it presented at that time as “acute.” Given that we did not collect data on pandemic-related distress at a later point in pregnancy, we cannot ascertain the chronicity of the stress conferred by ongoing pandemic conditions. Third, our study lacked a prepandemic comparison group, and we were unable to compare survey responses with those of matched controls who were not experiencing a pandemic. Last, we did not assess other factors that may contribute to postpartum outcomes, such as prepregnancy mental health,⁴⁸ gender identity,⁴⁹ or structural racism,^51,51 which are known to have important implications for perinatal health and policy.

STUDY RESULTS

Exhibit 1 presents descriptive statistics, scale response ranges, and scale sum scores. Positive postpartum depression screening status was reported by 23 percent of participants, and impaired mother-infant bonding status was reported by 11 percent of participants (data not shown).

Exhibit 1.

Descriptive Statistics

	N	Range	Mean (SD)
Time 1 Prenatal Risk Factors
ACE	732	0-9	1.04 (1.56)
COVID-19 General Worries	833	6-30	16.41 (5.16)
COVID-19 Pregnancy Worries	833	4-20	12.16 (3.84)
GAD-7	804	0-20	4.81 (4.15)
PHQ-2	804	0-6	0.85 (1.25)
Time 1 Prenatal Resilience Factors
Self-Reliance	831	3-21	17.83 (2.86)
Emotion Regulation	828	5-25	18.44 (4.03)
Supportive Close Relationships	827	5-20	17.37 (2.57)
Non-Hostile Close Relationships	827	6-25	22.01 (2.57)
Neighborhood Safety	826	4-20	14.06 (3.73)
Time 2 Postpartum Outcomes
Depression; EPDS	830	0-23	5.07 (4.66)
Impaired Bonding	814	0-36	5.67 (5.15)

ACE=Adverse Childhood Experiences; GAD=Generalized Anxiety Disorder Questionnaire 7; PHQ=Patient Health Questionnaire 2; EPDS= Edinburgh Postnatal Depression Scale

Postpartum Depression

Risk Factors:

As shown in appendix exhibit A3, all five risk factors predicted increased likelihood of postpartum depression when entered separately.³⁹ When they were entered in the model simultaneously, as shown in exhibit 2, three risk factors were significant. A one-unit increase in prenatal anxiety, prenatal depression, or adverse childhood experiences put women at 12 percent, 31 percent, and 18 percent higher odds, respectively, of screening positive for postpartum depression. Although COVID-19 general worries were not unique predictors of postpartum depression,⁵² we found that the timing of COVID-19 pregnancy worries during pregnancy mattered (appendix exhibit A6):³⁹ Higher prenatal COVID-19 pregnancy worries conferred greater postpartum depression risk among women assessed in their third trimester (odds ratio: 1.19; 95% confidence interval: 1.08, 1.33) but not in the other two trimesters (first trimester OR: 0.83 [95% CI: 0.58, 1.14]; second trimester OR: 1.00 [95% CI: 0.91, 1.10]). No significant interactions between COVID-19 worries and other risk factors were detected (appendix exhibit A7).³⁹

Exhibit 2.

Logistic Regressions Predicting Postpartum Depression

A.	Risk Factors Predicting Postpartum Depression (n=717; all risk factors entered simultaneously)
		B	S.E.	Wald	df	Exp(B)	Sig.
Block 1	SES	−.04	.11	.14	1	.96	.71
	Single Gestation (1=Yes)	−.63	.64	.98	1	.53	.32
	Parity (1=first Pregnancy)	−.12	.18	.41	1	.89	.52
	Preterm Birth	.35	.38	.82	1	1.41	.37
	Marital Group	−.30	.25	1.49	1	.74	.22
	Maternal Age	.00	.02	.03	1	1.00	.86
	Pregnancy weeks at Survey 1	−.01	.01	1.54	1	.99	.21
	Postpartum Weeks at Survey 2	.04	.03	1.47	1	1.04	.23
Block 2	GAD7 Score Time 1	.12	.03	13.93	1	1.12	.00
	PHQ2 Score Time 1	.27	.09	8.27	1	1.31	.00
	COVID-19 General Worry Time 1	−.05	.02	4.19	1	.95	.04
	COVID-19 Pregnancy-Related Worry Time 1	.07	.03	4.14	1	1.07	.04
	ACE	.16	.06	8.36	1	1.18	.00
B.	Resilience Factors Predicting Postpartum Depression (n=779; all resilience factors entered simultaneously)
		B	S.E.	Wald	df	Exp(B)	Sig.
Block 1	SES	.00	.10	.00	1	1.00	.98
	Single Gestation (1=Yes)	−.64	.58	1.20	1	.53	.27
	Parity (1=first Pregnancy)	−.13	.18	.56	1	.88	.45
	Preterm Birth	.33	.36	.86	1	1.39	.35
	Marital Group	−.33	.24	1.94	1	.72	.16
	Maternal Age	.00	.02	.03	1	1.00	.85
	Pregnancy weeks at Survey 1	−.01	.01	.91	1	.99	.34
	Postpartum Weeks at Survey 2	.03	.03	.92	1	1.03	.34
Block 2	Self-Reliance	−.07	.03	4.19	1	.94	.04
	Emotion Regulation	−.12	.02	26.57	1	.89	.00
	Supportive Close Relationships	.00	.04	.00	1	1.00	.96
	Negative Close Relationships	−.09	.04	5.69	1	.92	.02
	Perceived Neighborhood Safety	−.05	.03	2.74	1	.96	.10

ACE=Adverse Childhood Experiences; GAD=Generalized Anxiety Disorder Questionnaire 7; PHQ=Patient Health Questionnaire 2

Resilience Factors:

All resilience factors except for supportive close relationships showed protective effects against postpartum depression when entered as separate predictors (appendix exhibit A3).³⁹ However, as shown in exhibit 2, when the factors were entered in the model simultaneously, only emotion regulation met the significance threshold: A one-unit increase in emotion regulation was associated with an 11 percent lower likelihood of screening positive for postpartum depression.

Impaired Mother-Infant Bonding

Risk Factors:

Appendix exhibit A4 shows that higher prenatal anxiety, prenatal depression, and adverse childhood experiences scores significantly predicted greater odds of bonding impairment when examined separately.³⁹ When risk factors were entered in the model simultaneously (exhibit 3), only prenatal depression was a significant unique predictor: An increase in prenatal depression was associated with a 43 percent higher likelihood of having postpartum bonding impairment. Prenatal depression remained a unique predictor even after concurrent postpartum depression was controlled for (appendix exhibit A5).³⁹

Exhibit 3.

Logistic Regressions Predicting Impaired Mother-Infant Bonding

A.	Risk Factors Predicting Impaired Postpartum Mother-Infant Bonding (n=708; all risk factors entered simultaneously)
		B	S.E.	Wald	df	Exp(B)	Sig.
Block 1	SES	.03	.14	.04	1	1.03	.85
	Single Gestation (1=Yes)	−1.82	.72	6.47	1	.16	.01
	Parity (1=First Pregnancy)	.60	.25	5.61	1	1.82	.02
	Preterm Birth	−.06	.51	.01	1	.94	.91
	Marital Group	−.19	.33	.34	1	.82	.56
	Maternal Age	.06	.03	4.60	1	1.06	.03
	Pregnancy weeks at Survey 1	−.02	.01	1.83	1	.98	.18
	Postpartum Weeks at Survey 2	−.04	.05	.87	1	.96	.35
Block 2	GAD7 Score Time 1	.08	.04	4.52	1	1.09	.03
	PHQ2 Score Time 1	.36	.11	10.31	1	1.44	.00
	COVID-19 General Worry Time 1	−.06	.03	3.60	1	.94	.06
	COVID-19 Pregnancy-Related Worry Time 1	−.01	.04	.12	1	.99	.73
	ACE	.11	.07	2.35	1	1.12	.13
B.	Resilience Factors Predicting Impaired Postpartum Mother-Infant Bonding (n=768; all resilience factors entered simultaneously)
		B	S.E.	Wald	df	Exp(B)	Sig.
Block 1	SES	.05	.14	.16	1	1.06	.69
	Single Gestation (1=Yes)	−1.41	.67	4.39	1	.24	.04
	Parity (1=First Pregnancy)	.49	.24	4.05	1	1.63	.04
	Preterm Birth	−.23	.50	.20	1	.80	.65
	Marital Group	−.15	.32	.23	1	.86	.63
	Maternal Age	.06	.03	4.66	1	1.06	.03
	Pregnancy weeks at Survey 1	−.02	.01	1.67	1	.98	.20
	Postpartum Weeks at Survey 2	−.04	.04	.93	1	.96	.34
Block 2	Self-Reliance	−.20	.04	24.09	1	.82	.00
	Emotion Regulation	−.04	.03	1.34	1	.96	.25
	Supportive Close Relationships	−.08	.05	2.66	1	.93	.10
	Negative Close Relationships	−.15	.05	9.64	1	.86	.00
	Perceived Neighborhood Safety	−.03	.04	.59	1	.97	.44

ACE=Adverse Childhood Experiences; GAD=Generalized Anxiety Disorder Questionnaire 7; PHQ=Patient Health Questionnaire 2

Resilience Factors:

All resilience factors except perceived neighborhood safety predicted decreased likelihood of impaired mother-infant bonding when examined separately (appendix exhibit A4).³⁹ As shown in exhibit 3, when they were entered simultaneously, only greater self-reliance and nonhostile close relationships uniquely predicted decreased likelihood of impaired mother-infant bonding and remained significant after concurrent postpartum depression was controlled for (appendix exhibit A5).³⁹ An increase in self-reliance was associated with 18 percent lower likelihood of having impaired mother-infant bonding, and nonhostile close relationships were associated with a 13 percent decreased likelihood (exhibit 3).

DISCUSSION

The stresses of an uncertain future and social isolation specific to COVID-19 have intensified the experiences of the postpartum period, which are fraught, even under ideal circumstances, and need attention from policy makers to prevent worsening maternal mental health and impaired mother-infant bonding. This study examined a prospective cohort of pregnant women assessed during an acute moment in the pandemic to examine prenatal risk and resilience factors that contribute to postpartum outcomes (that is, postpartum depression and impaired mother-infant bonding). The results suggest that prenatal depression stands out as an important risk factor that predicts postpartum depression and uniquely contributes to impaired mother-infant bonding after postpartum depression is accounted for. Adverse childhood experiences and prenatal depression and anxiety also uniquely contribute to postpartum depression risk. Self-reliance and nonhostile relationships are resilience factors that appear protective for both postpartum depression and impaired mother-infant bonding, with emotion regulation showing strong protection against postpartum depression. These findings can support policy makers in promoting increased access to treatment programs that not only target mental health but also prioritize resilience building during the perinatal period for individuals and families.

Greater attention to mental health risk factors throughout pregnancy could have significant population health benefits via prevention efforts. Although individual hospital policies and guidelines from the American College of Obstetricians and Gynecologists⁵³ recommend universal mental health screening during pregnancy and the postpartum period, more specific policies surrounding the assessment of mental health and trauma in at-risk populations in combination with resilience training could be supported by provider training and increased funding for integrated mental health in obstetric care settings.

Pandemic-related distress should also be considered an important contributor to postpartum depression. When COVID-19 pregnancy worries and general worries were examined as separate predictors, the prospective link between COVID-19 pregnancy worries and postpartum depression was stronger than the link between COVID-19 general worries and postpartum depression, which suggests that an acute environmental stressor such as the COVID-19 pandemic may be perceived as more stressful when one considers how it may affect pregnancy, childbirth, and parenting. A rich literature on pregnancy-specific anxiety and its association with preterm birth and other adverse birth outcomes exists,⁵⁴ but data on the relationship between pregnancy-specific anxiety and postpartum depression are limited. Some have suggested that this is a result of the relatively short-term effects of pregnancy-specific anxiety that resolve after childbirth, unlike general anxiety disorders.⁵⁵ In our findings, however, COVID-19 pregnancy worries were more strongly predictive of postpartum depression than COVID-19 general worries, which may be related to continued distress around newborn care and maternal support during the COVID-19 pandemic. Understanding the unique postpartum vulnerabilities during the pandemic suggests potential insights into how acute stressors may increase the risk for postpartum depression in general.

Only prenatal depression was a unique predictor of impaired mother-infant bonding after concurrent postpartum depression and other risk effects were accounted for, dovetailing with prior work.⁵⁶ When these findings are taken together, this suggests that programs targeting women reporting prenatal depression may be necessary for improving mother-infant relationships in addition to postpartum depression. Local and federal policies that support screening women for mental health concerns during pregnancy and provide training and funding for provision of care to address prenatal mood symptoms may have substantial benefits, given that poor mother-infant bonding is linked to long-term negative outcomes for the offspring.^14–16

Recent literature has also reconceptualized postpartum depression as a condition affecting the whole family; fathers are affected by postpartum depression either by experiencing depression directly or via supporting and coping with their partner’s postpartum depression.⁵⁷ Our results suggest that a lack of hostility in close relationships is particularly important for reducing negative postpartum outcomes. As evidence suggests that men commonly experience depression as feelings of anger or hostility, psychoeducation about the negative impact of hostile communication in relationships and opportunities for supporting fathers⁵⁸ may be especially useful when policy makers and clinicians are considering ways to reduce maternal postpartum depression risk and improve mother-infant bonding. Educational policies that encourage medical school and residency curricula to promote training in family dynamics, family therapy, and the needs of fathers in the postpartum period would aid in the prevention of maternal postpartum depression, as would incentivizing hospitals and third-party payers to provide these types of services and care.

Several modifiable intrapersonal factors also showed strong protective effects in this cohort. Women with better self-reliance and emotion regulation skills were buffered against the likelihood of developing poor postpartum outcomes. Self-reliance and emotion regulation are skills that are routinely taught and are considered modifiable targets of group treatment.^59,60 Providing incentives to obstetrics clinics to offer integrated mental health care in group settings could enhance these resilience factors, representing an opportunity to apply these findings to the real world.

Our results highlight several policy priorities and considerations, including promoting preventive mental health services for at-risk people; increasing funding for integrated perinatal mental health services that allow for coordination between obstetric, pediatric, and behavioral health providers; training nonspecialty providers to offer group mental health care; and incentivizing obstetrics practices to offer these services by increasing insurance reimbursements. These measures would have long-lasting, cost-saving effects, given that maternal mental health has a reverberating impact on child development.^14–16

CONCLUSION

The COVID-19 pandemic has been an almost universally challenging experience. However, the experience of being pregnant and giving birth in the context of the social isolation and uncertain future wrought by the pandemic may represent a unique stressor. Although rates of general depression, anxiety, and stress also appear elevated in nonpregnant people during this time,^61,62 pregnancy confers additional risk as mother and child are both affected, leading to the potential for intergenerational transmission of trauma.

This study revealed risk and resilience factors that were significant prospective predictors of maternal postpartum outcomes, and its findings highlight targets for policy intervention and prevention efforts. Recognizing the contributing risk factors to postpartum depression and impaired mother-infant bonding and creating policies that will enable providers to offer preventive mental health services are important but certainly challenging in an environment where access to high-quality mental health care is limited. The benefits of increasing access to perinatal mental health treatment through policy are twofold: If postpartum depression prevention and intervention efforts improve the well-being of mothers, adverse childhood experiences or other negative consequences in their offspring may be prevented, impeding the continuation of the cycle of maternal depression, poor bonding, and adverse childhood outcomes in the next generation.⁶³