The Roles of Resilience and Childhood Trauma History: Main and Moderating Effects on Postpartum Maternal Mental Health and Functioning

Abstract

Objective

Recently postpartum women participated to investigate main and moderating influences of resilience and childhood history of maltreatment on posttraumatic stress disorder (PTSD), major depressive disorder (MDD), parental sense of mastery, and family functioning.

Method

At 4-months postpartum, 214 mothers (145 with a history of childhood abuse or neglect) completed interviews assessing mental health symptoms, positive functioning, resilience and trauma history. Multiple and moderated linear regression with the Connor- Davidson Resilience Scale (CD-RISC) and Childhood Trauma Questionnaires (CTQ) were conducted to assess for main and moderating effects.

Results

Resilience, childhood trauma severity, and their interaction predicted postpartum PTSD and MDD. In mothers without childhood maltreatment, PTSD was absent irrespective of CD-RISC scores. However, for those with the highest quartile of CTQ severity, 8% of those with highest resilience in contrast with 58% of those with lowest CD-RISC scores met PTSD diagnostic criteria. Similar, in those with highest resilience, no mothers met criteria for postpartum MDD, irrespective of childhood trauma, while for those with lowest quartile of resilience, 25% with lowest CTQ severity and 68% of those with highest CTQ severity were depressed. The CD-RISC, but not the CTQ, was predictive of postpartum sense of competence. The CD-RISC and the CTQ were predictive of postpartum family functioning, though no moderating influence of resilience on childhood trauma was found.

Conclusions

Resilience is associated with reduced psychopathology and improved wellbeing in all mothers. It further serves as a buffer against psychiatric symptoms following childhood trauma. Such findings may assist in identification of those at greatest risk of adverse functioning postpartum, utilization of resilience-enhancing intervention may benefit perinatal wellness, and reduce intergenerational transmission of risk.

Perinatal posttraumatic stress disorder (PTSD) and major depressive disorder (MDD) have myriad adverse proximal and distal cognitive, behavioral, physiological, and emotional sequelae that impact mothers and children and set the stage for intergenerational transmission of risk (Chemtob et al., 2010; Dubber et al., 2014; Bosquet Enlow, 2011; Marcus, 2009; Muzik and Borovska, 2010). In the context of this grave public health concern, comprehending fixed and adjustable threats and buffers specific to maternal functioning is paramount to identify and respond to those most vulnerable.

Substantial research has reliably linked a history of childhood maltreatment (CM) with PTSD and MDD in adulthood (Edwards et al., 2003; Koenen and Widom, 2009; Wingo et al., 2010). This risk factor is particularly salient postpartum as mothers who have experienced childhood abuse and neglect evidence increased risk for not only PTSD and MDD (Collishaw et al. 2007; Grekin and O'Hara, 2014; Lev-Wiesel et al., 2009), but also increased parenting stress (Ethier et al., 1995), difficulties in maternal-child interactions (Levendosky and Graham-Bermann, 2001), and maternal and infant impairments in hypothalamic-pituitary-adrenal axis functioning (Brand et al., 2010).

Yet, many individuals with a history of CM subsequently evidence an absence of pathology displaying positive functioning. Moreover, pregnancy and childbearing may represent a critical motivational window during which some women are uniquely inspired to improve health behaviors (Jagodzinski and Fleming, 2007; Peen et al., 1991; Severson et al., 1995; Wen et al., 2014). Additional research has identified relationships between other markers of wellbeing (e.g., greater parenting mastery) and improved postpartum psychiatric functioning (Eshbaugh, 2010; Fowles, 1998) and mother-infant attachment (Mercer and Ferketich, 2006). Thus, for many of women, childbearing itself may promote psychological health. However, greater attention is warranted to elucidate the role of resilience and adaptability in the context of postpartum adjustment, particularly in cases of elevated risk secondary to CM.

Recently, research has increasingly focused on the impact of protective factors that co-exist in contexts of adversity and have potential to influence the onset of psychiatric or physiological disease processes, and more generally, adaptation to life stressors and the attainment of wellness (Carver, 2005; Charney, 2004; Sharpley et al., 2014; Souza et al., 2013; Wright et al., 2008; Yi et al., 2008). A primary focus of this work has been on resilience, conceptualized as the ability of individuals to be “tested by adversity and continue to demonstrate adaptive psychological and physiological stress responses (Feder et al., 2009, p. 446). Resilience can be modified through intervention (Davidson et al., 2005) and serves as a predictor of mental health treatment response (Davidson et al., 2012), and thus is especially relevant in cases where other features are fixed (i.e., histories of childhood adversity).

Resilience may be particularly relevant for long-term adjustment in the context of CM exposure, as it may mitigate the association between historical adversity and poor adult functioning. For instance, Wingo et al. (2010) reported resilience attenuates existing links between CM and depression in adulthood and additionally, resilience diminished depressive symptoms in those with and without a trauma history. Resilience has demonstrated similar protective effects in studies assessing PTSD (Wrenn et al., 2011), depression and suicidal behaviors (Roy et al., 2011; Schulz et al., 2014), and substance use (Wingo et al., 2014) in community or high-risk adult populations following CM.

In contrast, while Spier and Seedat's (2014) research in women also confirms a relationship between resilience and depressive symptoms, the authors did not identify a moderating influence of resilience in conjunction with CM history on MDD in adulthood. Similar, a separate investigation of female survivors of childhood sexual assault found that resilience moderated sexual risk factors but did not mitigate interpersonal problems (Lamoureux et al., 2012). Given such mixed results, we propose the need for further inquiry before concluding that resilience uniformly mitigates CM-related risk and that findings are generalizable to other populations, particularly women.

To our knowledge, no study has investigated interrelationships between resilience, CM, and health among postpartum mothers. Investigating resilience and postpartum adaptation among mothers following CM will help characterize those most able to adjust and even thrive despite historical stress exposure in the context of postpartum demands. The identification of protective factors could lead to intervention development and implementation aimed to bolster wellbeing for those most at risk for postpartum mental illness. Further, though extant research has marked potential for understanding resistance to mental health symptoms, it has been less attentive to the broader context of healthiness following trauma. Consistent with the notion that resilience is not merely the “absence of pathology” (Bonanno, 2004, p. 20), we find it essential to expand beyond illness and concurrently attend to markers of wellness.

The primary aims of this research are twofold. First, we researched the associations between resilience, CM severity, mental health symptoms (postpartum PTSD and MDD), and positive functioning (self-perceptions of childrearing mastery and global family functioning), hypothesizing significant main effects for resilience and CM severity on outcomes. Second, we predicted that resilience would moderate relationships between CM and maternal illness and health sequelae postpartum.

2. Methods

2.1 Procedures

Participants in this study (N=214) were recruited from the [blind for review] Study as part of a an ongoing longitudinal study in which the overall aim was to investigate the effects of childhood maltreatment on mothers' psychosocial adjustment and parenting, and whether these factors predict children's behavioral and physiological outcomes. Participants were recruited in one of two ways: as postpartum follow-up to a parent study on the prenatal effects of CM on childbearing (previously reported in [blind for review] 2009) or through community advertisement at 6–8 weeks postpartum. Women in the community were recruited from obstetric clinics, childbirth classes, and newspaper advertisements.

Participants were non-psychiatrically referred English-speaking women, 18 and older, and mothers of singleton births. Exclusion criteria included diagnoses of schizophrenia or bipolar disorder, substance use problems within the last three months, and mothers of infants with severe health/developmental problems or more than six weeks premature. Data collection for the longitudinal study spanned from 4 to 18 months postpartum. Assessments were conducted in the home, the University-based playroom, or by phone. The results presented in this paper are restricted to data collected four months postpartum, when the CD-RISC was utilized.

At four months postpartum, mothers completed an interview to assess CM, mental health, parenting sense of competence, and family wellbeing. Women received an honorarium of $10 for this phone interview and a maximum honorarium of $130 for their participation in the overall longitudinal research. The Human Subjects Committee of the [affiliation removed for review] approved the research protocol and study participants initiated the informed consent process including verbal and written consent.

2.2 Measures

2.2.1 Demographics and health questionnaire

Participants completed a 28-item assessment of cohabitation status, race/ethnicity, employment, income, and educational history, maternal and child health concerns, and perinatal medication use.

2.2.2 Resilience

Resilience was assessed using the Connor-Davidson Resilience Scale (Connor and Davidson, 2003), a 25-item, 5-point Likert scale assessment of “personal qualities that enable one to thrive in the face of adversity” (Connor and Davidson, p. 76.). Scores range from 0 (not true at all) to 4 (true nearly all the time). The CD-RISC has been psychometrically investigated for use with individuals experiencing PTSD (Connor & Davidson, 2003) and nonclinical female populations (Lamond et al., 2008; Sexton et al., 2010). Internal reliability for this study was .92.

2.2.3 Childhood trauma

Participant CM history was evaluated using the Childhood Trauma Questionnaire (CTQ; Bernstein and Fink 1998), a 28-item self-report Likert scale. Responses range from 1 (never true) to 5 (very true). Scoring yields five subscales: Emotional Abuse, Physical Abuse, Sexual Abuse, Physical Neglect, and Emotional Neglect. For the purposes of this research, participants with scores exceeding established cut scores for any type of maltreatment on the CTQ were categorized as having a history of CM. The internal reliability for this study was .92.

2.2.4 Postpartum Mental Health Diagnoses

Posttraumatic stress

The National Women's Study PTSD Module (NWSPTSD; Resnick et al., 1993) was utilized to assess for postpartum PTSD. The model is a version of the Diagnostic Interview Schedule (DIS), is intended for use by trained lay researchers, and evaluates trauma-related symptoms with modifications based on a large epidemiological study of PTSD with women conducted with the National Crime Victim Center. The dichotomous scoring algorithm was used to identify those at greatest risk for postpartum PTSD.

Depression

Depression was evaluated with the Postpartum Depression Screening Scale (PDSS; Beck and Gable, 2000), a 35-item self-report instrument of depressive symptoms. It is often preferred to general depression screening instruments with this population because of potential confounding symptoms that may be normative in a postpartum context (e.g., frequent nighttime awakenings). Items are rated from 1 (strongly disagree) to 5 (strongly agree). Cutoff scores above 80 are used to indicate likely MDD. In this research, the dichotomous scoring algorithm was used to identify cases of probable postpartum depression.

2.2.4 Postpartum Positive Functioning

Family functioning

The Family Adaptation, Partnership, Growth, Affection, and Resolve (FAPGAR) Scale (Smilkstein et al., 1982) was used to evaluate global family functioning. The 5-item Likert scale (0 = Never, 4 = always) assesses maternal views of interpersonal supports in the areas of adaptation, partnership, growth, affection, and resolve with higher scores representing greater satisfaction with supports. Total scores range from 0 to 20. For this study, the internal consistency reliability was good (.86).

Postpartum sense of competence

Maternal perception of parental mastery was assessed via an adapted version of the Parenting Sense of Competence Scale (PSCS; Gibaud-Wallston and Wandersman, 1978 as adapted by Mowbray et al., 2005), an 11- item, 5-point Likert scale assessing satisfaction with maternal competence. Scores range from 11 to 55 with higher scores representing greater satisfaction. The internal reliability in this research was good (.80)

2.3. Data Analysis

We calculated descriptive statistics with percentage counts and means as appropriate. Multiple moderated regression analyses involved completion of two linear regression analyses per outcome variable (PTSD, MDD, PSCS, and FAPGAR). First, CM severity and resilience scores were used to attain an initial analysis of main effects with multiple linear regression (MLR). CD-RISC and CTQ scores were then centered to compute a resilience-by-trauma interaction analysis. The two predictors and the interaction term were then entered into a multiple moderated linear regression (MMLR) model and interpreted. In cases that indicated a moderation effect was present, CD-RISC and CTQ responses were stratified into quartiles. The lower and upper quartiles of resilience and childhood trauma scores were used to illustrate the moderation effect. An a priori alpha of p<.05 was established for this study.

3. Results

3. 1 Participant Characteristics

At four months postpartum, 214 women completed the assessment instruments. Participants were generally young adults, (mean age = 28.2 years, SD = 5.7), as would be expected in a study of childbearing women. Sixty one percent of participants identified as Caucasian. Eighty-six percent had completed at least some college. Most mothers cohabitated with a romantic partner (75%). Economically, fifty-eight percent described annual household incomes below $50,000.

3.2 Trauma and Postpartum Mental Health Characteristics

About two-thirds of the sample endorsed experiencing child abuse or neglect (67.8%). Of the five types of abuse and neglect assessed, most women endorsed experiencing multiple types (mean = 1.96, SD = 1.8). Emotional abuse and neglect were most commonly acknowledged. The average CTQ score was 43.08 (SD = 18.4). CM characteristics and severity data are presented in Table 1. The mean CD-RISC score was 76.6 (SD = 13.6. Prevalence rates were 19.6% and 21.5% for postpartum PTSD and postpartum MDD, respectively. The average PSCS and FAPGAR scores were 50.3 (SD=4.2) and 15.8 (SD=4.1.4), respectively. See Table 2 for correlations between resilience, childhood trauma, and investigated outcomes.

Table 1. Childhood Trauma Characteristics (N = 214).

Trauma Type and Severity	N	% Endorse	CTQ Subscale Mean (SD)
Emotional Abuse			10.3 (5)
Endorse Any	113	52.8%
Endorse Severe	39	18.2%
Physical			7.9 (4)
Endorse Any	68	31.8%
Endorse Severe	30	14%
Sexual Abuse			7.8 (5)
Endorse Any	75	35.2%
Endorse Severe	39	18.3%
Emotional Neglect			9.9 (5)
Endorse Any	97	45.3%
Endorse Severe	19	8.9%
Physical Neglect			7.1 (3)
Endorse Any	66	30.8%
Endorse Severe	17	7.9%

Note. CTQ = Childhood Trauma Questionnaire.

Table 2. Partial correlations between resilience, childhood trauma history, maternal pathology and maternal wellbeing.

Variable	1	2	3	4	5	6
1. CD-RISC	-
2. CTQ	-.289	-
3. PTSD	-.335	.339	-
4. MDD	-.441	.288	.557	-
5. PSC	.448	-.177	-.214	-.280	-
6. FAPGAR	.394	-.394	-.342	-.351	.330	-

Notes. CD-RISC = Connor-Davidson Resilience Scale, CTQ = Childhood Trauma Questionnaire, PTSD = posttraumatic stress disorder diagnosis, MDD = major depressive disorder diagnosis, PSC = Parenting Sense of Competence scale, FAPGAR = Family Adaptation, Partnership, Growth, Affection, and Resolve scale. All correlations significant at the p<.01 level.

3.3 Postpartum Mental Health Symptoms

3.3.1 Effects of Resilience and CM on PTSD

An initial MLR was evaluated to ascertain associations between resilience and childhood trauma and maternal diagnoses of PTSD. Results indicated that lower resilience and greater maltreatment severity were both associated with increased rates of PTSD (See Table 3) and the model was significant. Cohen's f² this model was a medium effect size.

Table 3. Multiple linear regression (MLR) models with CD-RISC and CTQ predicting maternal diagnoses of PTSD and MDD and multiple moderated linear regression (MMLR) models incorporating the interaction effect of CD-RISC and CTQ on maternal pathology.

Outcome	Predictor	β	SE	p	R2	f2	p	R2 (R2Δ)	f2	Δp
PTSD MLR					.180	.22	<.001
	CD-RISC	-.257	.002	<.001
	CTQ	.273	.001	<.001
PTSD MMLR								.204 (.024)	.26	<.00 1
	CD-RISC	-.221	.002	.001
	CTQ	.266	.001	<.001
	CD-	-.160	.021	.013
	RISC*CT
	Q
MDD					.223	.29	<.001
MLR
	CD-RISC	-.387	.002	<.001
	CTQ	.183	.001	.004
MDD MMLR								.239 (.016)	.31	.039
	CD-RISC	-.357	.002	<.001
	CTQ	.177	.001	.005
	CD-	-.129	.021	.039
	RISC*CT
	Q

Notes: PTSD = posttraumatic stress disorder, MDD = major depressive disorder, CD-RISC = Connor-Davidson Resilience Scale, CTQ = Childhood Trauma Questionnaire.

CD-RISC and CTQ scores were centered to compute a resilience-by-trauma history interaction MMLR analysis. The CD-RISC, CTQ, and their interaction were each significant, suggesting that resilience moderates the relationship between CM and maternal PTSD (See Table 3). The R²Δ score was also significant indicating improved explanatory power with the addition of the moderation variable. Cohen's f² for this model was a medium-to-large effect size.

Participant data was subsequently stratified into high (>53, n = 53) and low (<29, n = 56) levels of childhood trauma and high (>88, n = 49) and low (<68, n = 58) levels of resilience using categorization by first and fourth quartiles and used to evaluate postpartum PTSD diagnostic outcomes. As depicted, the lowest level of CM was associated with an absence of PTSD postpartum (See Figure 1 for the interaction model). However, for those with higher levels of maltreatment, resilience had a marked moderating effect on diagnostic outcomes with only 8% of those with high childhood trauma severity and high resilience meeting PTSD criteria in contrast with 58% of those with high childhood levels of childhood trauma but low resilience.

Figure 1. Moderating Effect of CD-RISC on CTQ for PTSD Diagnoses.

3.3.2 Effects of Resilience and CM on MDD

Parallel to the above analyses, a MLR was performed to ascertain main effect influences of resilience and childhood trauma on postpartum MDD. Results indicated that lower resilience and greater childhood trauma severity were significantly related to increased rates of postpartum depression (See Table 3). The initial model was significant and Cohen's f² indicated a medium-to-large effect size (.29).

The subsequent MMLR found the main effects retained significance and that relationships between CM and MDD outcomes were significantly moderated by resilience. The R²Δ score demonstrated a significant improvement with the addition of the interaction term. The effect size for the interactive model was in the moderate-to- high range (f² = .31; See Table 3).

Participant data for postpartum MDD were consequently stratified by using the upper and lower quartiles for high and low levels of CM and resilience described above. See Figure 2 for a representation of the moderation model. As illustrated, high resilience was particularly salient in predicting postpartum MDD with no individuals in the highest quartile of resilience meeting criteria for depression irrespective of trauma history. For those with low resilience, 25% of those with low levels of trauma met criteria for depression. In contrast, 68% of individuals with low resilience and a high rate of childhood trauma were depressed.

Figure 2. Moderating Effect of CD-RISC on CTQ for MDD Diagnoses.

3.4 Postpartum Positive Functioning

3.4.1 Effects of Resilience and CM on Maternal Sense of Competence

When initially evaluating the main associations between resilience, trauma history, and parental sense of competence, resilience was significantly associated with PSCS scores while CM severity was not. The MLR analysis was statistically significant and the effect size was in the medium-to-large range (f² = .24; See Table 4).

Table 4. Multiple regression linear (MLR) models with CD-RISC and CTQ predicting maternal PSCS and FAPGAR scores and multiple moderated linear regression (MMLR) models incorporating the interaction effect of CD-RISC and CTQ on wellbeing.

Outcome	Predictor	β	SE	p	R2	f2	p	R2 (R2Δ)	f2	Δp
PSC MLR					.195	.24	<.001
	CD-RISC	.423	.020	<.001
	CTQ	-.055	.015	.393
PSC MMLR								.195 (<.001)	.24	.725
	CD-RISC	.418	.020	<.001
	CTQ	-.054	.015	.404
	CD-RISC*CTQ	.022	.222	.725
FAPGAR MLR					.247	.33	<.001
	CD-RISC	.314	.018	<.001
	CTQ	-.306	.014	<.001
FAPGAR MMLR								.252	.37	.206
								(.006)
	CD-RISC	.296	.019	<.001
	CTQ	-.302	.014	<.001
	CD-RISC*CT Q	.078	.206	.206

Notes: Notes: CD-RISC = Connor-Davidson Resilience Scale, CTQ = Childhood Trauma Questionnaire, PSCS = Parental Sense of Competence Scale, FAPGAR = Family Adaptation, Partnership, Growth, Affection, and Resolve scale.

MMLR with data centralization was conducted to evaluate the predictive potential interactive associations between resilience and childhood trauma on parental sense of competence (See Table 4). Results again indicated resilience was significantly positively associated with sense of competence. In contrast, severity of childhood maltreatment was not a significant predictor nor was there a moderating relationship between resilience and childhood trauma on PSCS scores. While the moderated model version was significant, the R² Δwas not indicating the interaction had nothing of further significance to offer beyond the main effect earlier identified for the CD-RISC irrespective of CTQ.

3.4.2 Effects of Resilience and CM on Family Functioning

An MLR analysis was conducted to evaluate direct relationships between resilience, CM, and postpartum family functioning (See Table 4). The CD-RISC and CTQ were both significant predictors of the FAPGAR outcome and the model effect size was large (f² = .33).

The MMLR analysis was then used to assess relationships between resilience, childhood maltreatment, and the centralized interaction of the CD-RISC and CTQ on postpartum family functioning (See Table 4). Main effects remained for resilience and CM as with the first model. However, no moderation effect was observed. The final model was significant, yet the R²Δ indicated that the increased predictive utility of the model was not statistically significant beyond that presented in the initial main effect analysis.

4. Discussion

Based on a postpartum sample with a high proportion of CM, we found childhood trauma had a significant impact on postpartum psychiatric diagnoses, while resilience generated a significant buffering effect confirming our initial research hypotheses. In mothers with the lowest levels of resilience coupled with the most severe childhood maltreatment histories, the overwhelming majority exceeded PTSD and MDD cutoffs. In contrast, similarly high CM histories in women with highest levels of resilience evidenced low rates of PTSD and an absence of postpartum depression. Regarding assessment of postpartum positive functioning, CM was associated with worse family functioning but not with impaired maternal sense of parenting competence. Research hypotheses regarding benefits of resilience were confirmed for maternal competence and family supports. Irrespective of CM history, high resilience exhibited reduced psychopathology and improved wellbeing in all mothers. However, resilience did not moderate the link between CM history and postpartum functioning.

Our study findings on the moderating role of resilience on postpartum PTSD and depression is consistent with prior findings among community and at-risk populations that also found resilience to attenuate PTSD and MDD subsequent to childhood abuse and neglect (Schulz et al., 2014; Wingo et al., 2010; Wrenn et al., 2011). We found that resilience, CM severity, and their interaction yielded moderate-to-large effect sizes in explaining postpartum depression and PTSD. However, others have failed to find such moderation of resilience on CM and psychopathology in adult women (Spies and Seedat, 2014). A potential reason for such variability may be in the unique biopsychosocial context of the perinatal period itself which may confer additional benefits and bolster resilience. We speculate that postpartum may trigger more optimal coping, possibly facilitated by mothers' increased emotional investment and motivation for positive bonding with her infant. Several neuroscience studies report on normative enhanced brain activity in emotion and reward circuits relevant to parenting and mother-infant bond (Swain, 2011). Thus, such heightened emotional salience of mother-baby relationships may be in the service of resilience against postpartum psychopathology. Alternatively, other, currently unaccounted for, individual or contextual differences between these two studies may amend outcomes or overwhelm the mitigating influence of resilience in those with histories of CM. We propose this based on our own data as the amount of variability in mental health outcomes accounted for by resilience and CM was only 20- 25%, indicating a marked degree of unexplained variability warranting further study.

Our findings suggest assessment of resilience and CM history previous to childbirth may help identify those at greatest risk for postpartum mental illness and respond to those most likely to benefit from resilience-enhancing intervention. Irrespective of CM history, but particularly in light of it, our results suggest such targeted assessment and related responding may yield benefits to both mother and child given the far-reaching negative sequelae of postpartum pathology.

Our study is the first to investigate resilience as a moderating factor for CM's impact on maternal positive postpartum functioning including satisfaction with family supports and sense of parenting competence. No prior investigations can be used as benchmark, yet the lack of moderation effects for resilience suggest that adaptation trajectories following CM may be more flexible in positive domains than those associated with pathology.

Several limitations of this research are worth noting. First, one aim of this study was to evaluate resilience and CM in the context of postpartum functioning. These results may not extend to trauma survivors outside of a childbearing context or with childhood adversities beyond abuse and neglect. Second, our findings specific to the associations between self-reported resilience and measures of positive and negative functioning do not provide insight into underlying reliance-related processes or potential cause-and-effect relationships. Further research with genetic or endocrine assays or longitudinal methodology with this population could better augment these findings.

Despite these weaknesses, this study has several unique contributions. To our knowledge, this is the first study evaluating the moderating influence of resilience postpartum functioning while attending to CM risk. Second, consistent with an expanded definition that wellbeing extends beyond a lack of symptoms, this research further focused on positive outcomes of family functioning and maternal sense of mastery.

This research raises additional questions about the resilience in the context of childbearing. For example, does resilience reduce intergenerational transmission of psychological risk? Do the relationships between resilience and outcomes vary by specific types of CM? Is resilience related to parent-child attachment or postpartum engagement in infant- and health-care behaviors? Is postpartum resilience associated with maternal and infant HPA-axis functioning? Does perinatal intervention improve resilience? Extending these finings through additional research may enhance the understanding and fostering of positive adaptation in the context of stress. Overall, our findings indicate resilience is a key predictive characteristic for mental health symptoms and wellness for all mothers, particularly those with histories of maltreatment, and warrants inclusion in future postpartum and intergenerational transmission of risk research.

Highlights.

• In postpartum women, childhood trauma is predictive of maternal PTSD and MDD, while resilience reduced the prevalence of psychiatric diagnoses in all mothers and buffered the salience of childhood maltreatment.

• Improved resilience, though not trauma history severity, were predictive of greater maternal sense of mastery.

• Lowered resilience and greater trauma history are associated with poorer family functioning and resilience does not appear to independently buffer trauma effects in this domain.