Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Oct 1.
Published in final edited form as: Compr Psychiatry. 2018 Aug 3;86:123–130. doi: 10.1016/j.comppsych.2018.08.001

Moderators and Mediators of a Maternal Depression Treatment Study: Impact of Maternal Trauma and Parenting on Child Outcomes

Holly A Swartz 1, Jill M Cyranowski 2, Yu Cheng 3, Marlissa Amole 4
PMCID: PMC6142810  NIHMSID: NIHMS1503125  PMID: 30118995

Abstract

Objective:

Treatment of maternal depression with psychotherapy has been shown to confer indirect benefits to school-age offspring with psychiatric disorders. The current study sought to understand mechanisms by which improvement in depressed mothers, with and without histories of trauma and treated with psychotherapy, produce changes in children who struggle with psychiatric illnesses themselves. We hypothesized that maternal history of childhood trauma would moderate the relationship between maternal and child outcomes and that increased positive and decreased negative parenting behaviors would mediate the relationship between maternal and child outcomes. We also examined whether maternal history of trauma would moderate the mediational effects of parenting behaviors.

Methods:

Participants were dyads (n=62) of mothers with major depressive disorder and their children, ages 7–18, with at least one internalizing disorder. Mothers were treated with nine sessions of psychotherapy and children were treated openly in the community. Dyads were evaluated every three months over one year.

Results:

Maternal improvement in depressive symptoms was associated, in a lagged fashion, with child improvement in functioning six months later. There was a significant interaction of time and change in maternal symptoms [F(1, 45) = 5.84, p = 0.02], where change in maternal depressive symptoms from baseline to six months was robustly associated with change in child functioning from baseline to 12 months (β = 0.49, p = 0.0002). Maternal history of childhood sexual abuse moderated the association between change in maternal and child depressive symptoms [F(1,87) = 5.8, p = 0.02], and maternal history of physical neglect moderated the relationship between improvement in maternal depression and improvement in child functioning [F(1,36) = 4.34, p = 0.04], where significant associations between maternal and child outcomes were only found in mothers without histories of sexual abuse or physical neglect. Increase in positive parenting strategies (acceptance) by mothers mediated 6-month lagged associations between maternal and child outcomes, but reduction in negative parenting strategies (psychological control) did not. Maternal history of childhood emotional neglect moderated the mediational model, such that improved positive parenting did not explain lagged improvement in child depression among the subset of mothers with childhood histories of emotional neglect.

Conclusions:

In dyads comprised of depressed mothers and school-age children with internalizing disorders, children improved when mothers improved, but not among those whose mothers who had histories of sexual abuse or physical neglect. Increased use of positive parenting strategies among mothers accounted for lagged relationships between improvement in maternal depressive symptoms and improvement in child functioning. This pattern was not, however, observed among mothers with childhood histories of emotional neglect. Interventions that directly enhance positive parenting and more rapidly change these behaviors may hasten improvement in offspring. Offspring of depressed mothers with histories of early trauma are at high risk for poor outcomes, even when their mothers receive depression treatment.

Keywords: Psychotherapy, mothers, children, parenting, depression, trauma

1.1. INTRODUCTION

Children of depressed mothers are at increased risk for internalizing disorders [1, 2]. Maternal depression is also associated with worse treatment outcomes for psychiatrically ill children, even when children receive evidence-based care [3, 4]. Multigeneration studies have shown that successful treatment of maternal depression has an indirect positive outcome on children [5–8]; however, the mechanism by which improvement in maternal depression helps children fare better is unclear [9]. Even less is known about how this process works in families at very high risk for poor outcomes, defined as having multiple generations within the same family who concurrently experience psychiatric illness.

Quality of parenting is implicated in inter-generational transmission of depression. When depressed, parents experience disruptions in mood, energy, and cognition [10], symptoms which in turn interfere with parenting [11] and are associated with higher levels of child psychopathology [12]. Historically, two broad domains have been identified as essential to successful parenting: “demandingness” and “responsiveness” [13]. Demandingness involves consistent monitoring, discipline, and expectations (behavioral or psychological control). Responsiveness refers to affective warmth, attunement, and unconditional acceptance (emotional connection). Lower levels of maternal psychological control and higher levels of maternal warmth and acceptance are parenting styles associated with greater resilience in children [14]. In contrast, depressed caregivers demonstrate parenting styles that are characterized by high demandingness and low responsiveness: higher levels of hostility, increased behavioral control, and decreased warmth [15, 16].

Maternal history of childhood trauma also predicts poorer parenting [17, 18]. Maternal history of childhood physical or sexual abuse predicts increased use of overly permissive parenting strategies [19]. Childhood emotional abuse and neglect are associated with parenting characterized by less acceptance and greater psychological control, even after controlling for maternal depression severity [20].

Maternal depression and trauma are independently associated with increased rates of insecure attachment in children [21–23], likely related, at least in part, to decreased maternal responsiveness and affective availability [24] which are important components of parenting related to the domain of responsiveness. Attachment styles develop over the course of childhood and persist into adulthood, affecting lifelong capacity for intimacy in relationships [25]. Insecure attachment places children at risk for developing behavioral problems, including depression and anxiety [26]. Secure emotional attachments in children develop when they are cared for by a warm, consistent, and attentive caregiver or attachment figure, typically their mother [27]. Insecure attachment patterns are likely to develop when the caregiver is inattentive, insensitive, or inconsistent [24], as has been observed in both depressed mothers and mothers with histories of maltreatment [21–23].

History of childhood trauma is a non-specific predictor of poor depression outcomes [28, 29], and individuals with histories of abuse respond less well to psychotherapy [30] and pharmacotherapy [31]. Maternal history of maltreatment is a risk factor for offspring psychiatric illness [32]. Thus, when both depression and history of childhood trauma co-occur, negative effects on outcomes, parenting, and the motherchild relationship may be additive [20].

Maternal depression treatment favorably impacts child outcomes. This effect is explained, at least in part, by changes in parenting practices. In a non-randomized study where depressed parents of school-age children received “standard, evidence-based” depression treatments including medication and psychotherapy, investigators found that change in parental depressive symptoms significantly predicted change in child depressive symptoms and that increased use of parenting strategies characterized by acceptance partially mediated the relationship between parental and child depressive symptoms [33]. This finding was significant even though offspring were not necessarily ill (mean scores on a child depression scale were in the non-clinical range). Weissman and colleagues examined outcomes in children of depressed mothers randomized to treatment with escitalopram, buproprion, or the combination for 12 weeks, followed by an open trial for an additional 24 weeks [6]. Although investigators found no differences in maternal depression outcomes by treatment, children of mothers who were assigned to escitalopram had greater improvement in depression and functioning than mothers assigned to bupropion. Differential outcomes in children were explained by differential child-reported increase in maternal care and affection in the escitalopram-treated mothers compared to those who received combination treatment. Again, most offspring in this trial did not meet criteria for a psychiatric disorder.

The current study extends the work of prior investigations by examining the role of maternal trauma as a potential moderator and parenting as a potential mediator of maternal depression outcomes in families at very high risk for poor outcomes (i.e., both mothers and children suffer from psychiatric disorders) where depressed mothers were randomly assigned to one of two manualized, evidence-based psychotherapies. We conducted secondary analyses on a subset of individuals who had participated in a larger clinical trial [34]. Based on our earlier work [8], we hypothesized a lagged relationship between maternal and child outcomes such that mothers would show improvement prior to their offspring. We hypothesized that 1) maternal childhood trauma would moderate the relationship between maternal and child outcomes and 2) increased positive and decreased negative parenting behaviors would mediate the relationship between maternal and child outcomes. We further examined, in an exploratory analysis, the moderating effects of maternal trauma on the mediational model.

2.1. MATERIALS AND METHODS

2.2.1. Participants

Participants for the current analyses were drawn from a larger trial [34]. Potential participants provided informed written consent or assent (if < age 18) after receiving a complete description of the study. All study procedures were approved by the Biomedical Institutional Review Board of the University of Pittsburgh. Participants enrolled from 2010–2013 and were mother-child dyads recruited from child specialty mental health clinics, pediatric primary care practices, and advertisements.

Child participants were outpatients aged 7–18 with at least one internalizing disorder and no lifetime history of autism, schizophrenia, externalizing diagnosis, current substance use disorders, or developmental delays that would preclude study participation. Adult participants were outpatients aged 18–65 with non-psychotic major depressive disorder and no lifetime history of bipolar disorder, schizophrenia, schizoaffective disorder, current substance use disorder, borderline personality disorder, or antisocial personality disorder. In addition, mothers scored ≥ 15 on the 25item Hamilton Rating Scale for Depression (HRSD-25) [35], were biological or adoptive mothers of a child who met inclusion criteria, and lived with and had custody of the child participant. Concurrent treatment with antidepressant medication for mothers was permitted if they were on a stable dose for at least four consecutive weeks prior to entry and agreed to stay on the entry dose for the duration of the acute phase of the study (months 0–3). Mothers were not, however, permitted to participate individual psychotherapy other than the study intervention (see section 2.3) during the acute phase (month 0–3) of the study.

2.2.2. Study Sample

The study sample was drawn from participants from a larger study (characteristics described in section 2.2.1). The modified intent-to-treat sample for the larger trial (n=168 dyads) included dyads where children had internalizing diagnoses only (n=90) as well as dyads where children had both internalizing and externalizing diagnoses (n=78). Quantitatively, these groups differed significantly from each other on key child demographic and clinical variables such as age (14.8 ± 2.4 v. 12.9 ± 3.0, t=−4.39, p <0.0001), gender (% female; 69% v. 47%; Χ2=7.9, p=0.005), level of symptoms (Strengths and Difficulties Questionnaire Total Score; SDQ [41]; 12.8 ± 6.1 v. 16.9 ± 6.3, t=−4.23, p <0.0001), and functioning (Columbia Impairment Scale; CIS [42]; 14.5 ± 9.1 v. 17.4 ± 9.2; t= −2.03, p=0.04). In order to evaluate a more homogeneous group, the current study restricted analyses to those dyads where children suffered from internalizing diagnoses only and for whom all follow-up time point assessments were available for both mother and child (n=62). Within the original group of those with internalizing diagnoses only (n=90), those who were (n=62) and were not (n= 28) included in the current analyses did not differ in maternal age, baseline HRSD-25, child age, gender, SDQ, CIS, and Child Depression Inventory (CDI) scores (p > 0.14).

2.3. Interventions

As described in the initial report [34], mothers were randomly assigned to either brief interpersonal psychotherapy (IPT) for mothers (IPT-MOMS) [36] or brief supportive psychotherapy (BSP) [37] administered over a three month period. IPT-MOMS, described elsewhere [36], includes an initial engagement session based on principles of motivational interviewing [38] and ethnographic interviewing [39] designed to explore and resolve potential barriers to treatment-seeking [40], followed by eight IPT sessions [41]. IPT is an evidence-based therapy for depression [42] that helps individuals to understand links between mood and relationships, mobilize social support, and address interpersonal difficulties. IPT-MOMS incorporates specific strategies to help mothers manage problematic interpersonal relationships with their psychiatrically ill children. BSP is a rigorous, manualized intervention utilizing a non-directive approach to facilitate exploration of affect, emphasize patient strengths, and engender a therapeutic alliance [37]. It is rooted in Rogers’ Client-Centered Therapy [43]. Within BSP, the patient sets the treatment agenda and therapists employ strategies such as reflective listening and open-ended questions to explore feelings and provide empathic support.

Mothers in both conditions were provided nine weekly 45-minute psychotherapy sessions over three months, allowing for cancellations and rescheduling, which were common. All sessions were completed prior to the three-month assessment time point. Child participants received open treatment in the community which included medications and psychotherapy [34].

2.4. Assessments

Mother and child participants completed assessments at baseline, 3-, 6-, 9-, and 12-month follow-ups. Assessments were conducted by raters blind to treatment assignment. Separate assessors evaluated children and mothers. Maternal psychiatric diagnoses were determined with the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I) [44] and Personality Disorders (SCID-II) [45]. Depressive symptom severity was assessed with the clinician-rated HRSD-25, with higher scores indicating more depressive symptoms [35, 46]. Inter-rater reliability for HRSD-25 was high (intraclass correlation=0.989). Maternal functioning and history of childhood trauma were assessed with the self-report measures, Work and Social Adjustment Scale (WSAS), with a possible range of 0 (no impairment) to 40 (severe impairment) [47], and Childhood Trauma Questionnaire (CTQ), respectively. The 28-item CTQ (each item is scored 1—never true to 5—very often) includes five 5 subscales: emotional abuse, physical abuse, sexual abuse, emotional neglect, and physical neglect [48].Using previously validated cutpoints [49, 50], threshold scores were used to create five dichotomous variables defining presence or absence of trauma: ≥ 8 for physical abuse, physical neglect and sexual abuse; ≥10 for emotional abuse; and ≥15 for emotional neglect.

Child diagnoses were determined using the Kiddie Schedule for Affective Disorders and Schizophrenia – Present and Lifetime Version (K-SADS-PL) [51]. Child symptoms were assessed with self-report instruments: 27-item Child Depression Inventory (CDI), with a possible range of 0 (less depressed) to 54 (more depressed) [52] and the 34-item Strengths and Difficulties Questionnaire (SDQ), with a possible range of 0 (no symptoms) to 40 (many symptoms) [53]. Psychosocial functioning was assessed with the 19-item self-report Columbia Impairment Scale (CIS), with a possible range of 0 (not impaired) to 52 (highly impaired) [54].

Children reported on their perceptions of their mother’s parenting behavior on a 56-item version of the Child Report of Parent Behavior Inventory (CRPBI) [55], scored to yield two subscales, Acceptance and Psychological Control, using methods described in an earlier report [20]. Examples of Acceptance subscale items include: “My mother smiles at me often” and “My mother enjoys talking things over with me.” Examples of Psychological Control subscale items include, “My mother allows me to go out as often as I please” and “My mother thinks I am not grateful when I don’t obey.”

All self-reports measures were completed by the child. If the child was < age 11, mother completed the SDQ about the child. For children of any age who had difficulty completing the CDI, CIS, or CRPBI, a research assistant provided assistance with reading items to the child and recording responses.

2.5. Data Analysis

To examine the relationship between change in maternal symptoms and child outcomes, maternal depression change scores were calculated by subtracting baseline HRSD-25 scores from scores at later assessment points, such that negative scores indicated a reduction in depression and positive scores indicated an increase in depression. Change scores were similarly computed for maternal functioning (WSAS) and child outcomes as measured by the CDI, CIS and SDQ. Means and SDs were reported for those change scores. Because prior studies [5, 8] and previous analyses [56] have shown a delayed impact of change in maternal symptoms on child outcomes, lag analyses were conducted to evaluate the temporal relationship between maternal and child outcomes using mixed effects models. Lag-6 (6 month lag) analyses examine the impact of change in maternal symptoms from baseline to month 3 on change in child symptoms from baseline to month 9 and on the change in maternal symptoms from baseline to month 6 on change in child symptoms from baseline to month 12. Fixed terms included maternal change scores (HRSD-25, WSAS), time as a two-level factor corresponding to changes at 3 months, 6 months, the interaction terms of time and the change score, as well as group (IPT-MOMS v. BSP), child age, child gender, and family income. The inclusion of the interaction terms allowed the strength of the association between the maternal and child outcome variables to vary across time. Subject was included as a random term to account for dependence in repeated measures. If there was a significant time-by-maternal change interaction, contrasts were set up to evaluate slopes at each time point. If the interaction was not significant, the average slope was reported.

The following variables were examined as potential moderators of the lagged relationship between maternal and child outcomes: sexual abuse (as a dichotomous variable ≥ 8), emotional abuse (≥ 13), emotional neglect (≥ 15), physical abuse (≥ 8) and physical neglect (≥ 8). We evaluated whether each of these indicators of maternal trauma moderated the relationship between change in maternal outcomes and change in child outcomes six months later.

To examine mediation, we ran a series of mixed effects models to examine whether the lag-6 relationship between maternal outcomes and child outcomes was mediated by change in the child ratings of parenting (maternal acceptance and psychological control as measured by the CRPBI). We first examined whether maternal outcomes predicted maternal parenting at the next time point. Next, we tested whether the change in maternal acceptance or psychological control predicted child improvement at the subsequent time point. If both models were significant, we would then create a final model which would include changes in maternal depression at 3 and (or) 6 months and the changes in maternal parenting 3 months later to predict the child improvement 6 months later.

To examine whether maternal trauma moderated the above mediational effects of change in parenting on the relationship between maternal and child outcomes, we evaluated whether trauma history (each subscale) moderated the effect of change in maternal depression on change in parenting styles. In accordance with Muller [57], variables that had been shown in previous analyses to moderate the relationship between maternal and child outcomes were excluded as potential moderators for the mediational effects. This exclusion is based on the assumption that if a variable is found to be an overall moderator of the effect of the independent variable (change in maternal outcomes) on the dependent variable (change in child outcomes), then for a certain subset of participants there may be no effect to mediate as part of mediational analyses, which is analytically problematic. For full discussion of this issue, see Muller et al. [57] and Petty et al. [58]. We evaluated whether trauma history moderated the mediational effect of change in parenting by adding change in maternal depression, change in parenting, trauma history, and the interactions of trauma history with maternal depression and parenting into the model predicting changes in child outcomes.

3.1. RESULTS

3.2. Characteristics of Depressed Mothers and their Children

A summary of baseline demographic and clinical characteristics of mothers and children are provided in Tables 1 and 2. Mothers were moderately depressed (mean HRSD-17 score = 16.8 ± 3.7) as were children (mean CDI score = 12.3 ± 9.5). Over 60% of mothers reported histories of co-morbid anxiety disorders (n= 42; 67.7%). Approximately one third of mothers reported histories of emotional abuse (n=18; 31.0%) and neglect (n=17; 28.8%). One quarter reported histories of physical abuse (n=13; 22.8%) and neglect (14; 23.7%). Twenty percent (n=12) reported histories of sexual abuse.

Table 1.

Baseline Demographic and Clinical Characteristics of Mothers (N=62)

Variable N (%) or Mean ± SD
Ethnicity (% Hispanic) 0 (0)
Race (% White) 46 (74)
Age (years) 45.8 (± 6.5)
Married 27 (43.6)
Total family income < $30,000 per year 17 (27.4)
On an antidepressant medication at baseline 7 (11.3)
Lifetime Diagnosis of Anxiety – DSM IV 42 (67.7)
Current Diagnosis of Anxiety – DSM 5 36 (58.1)
More than 3 Lifetime Major Depressive Episodes 30 (48.4)
Hamilton Rating Scale for Depression – 17 Item 16.8 (± 3.7)
Hamilton Rating Scale for Depression – 25 Item 20.7 (± 4.8)
Childhood Trauma Questionnaire
Emotional Abuse 18 (31)
Emotional Neglect 17 (28.8)
Physical Abuse 13 (22.8)
Physical Neglect 14 (23.7)
Sexual Abuse 12 (20.3)
Parenting Scores (Child Reported)
CRPBI--Acceptance 59.4 (± 10.4)
CRPBI—Psychological Control 27.1 (± 6.4)
Open in a new tab

Table 2.

Baseline Child Demographic and Clinical Characteristics (n=62)

Variable N (%) or Mean ± SD
Gender (% female) 43 (69.4)
Ethnicity (% Hispanic) 2 (3.2)
Race (% White) 47 (75.8)
Age (years) 14.5 (± 2.5)
Number of Current Internalizing Diagnoses 1.77 (± 1.1)
Child Depression Inventory 12.3 (± 9.5)
Columbia Impairment Scale 14.2 (± 9.3)
Strengths & Difficulties Questionnaire 12.9 (6.0)
Open in a new tab

3.3. Lag-6 Association between Maternal and Child Outcomes

Improvement in maternal depression and functioning and improvement in child outcomes over time are reported in Table 3. In the 6-month (lag-6) lagged analyses, improvement in maternal depression was significantly associated with improvement in child depression (CDI scores) (β = 0.2, p = 0.05) and significantly associated with improvement in child SDQ scores (β = 0.19, p = 0.01). When predicting child improvement in functioning (change in CIS scores), there was a significant interaction (F(1, 45) = 5.84, p = 0.02), where change in maternal depression from baseline to 6 months was significantly associated with child improvement in functioning at 12 months (β = 0.49, p = 0.0002), but the association at the earlier time point was not significant, indicating a delayed association. There were no significant associations between change in maternal functioning scores and change in child outcomes.

Table 3:

Changes in maternal and child outcomes over time (n=62); Std= standard deviation

Group
BSP
time		 IPT-MOMS
time
Baseline to Month 3 Baseline to Month 6 Baseline to Month 9 Baseline to Month 12 Baseline to Month 3 Baseline to Month 6 Baseline to Month 9 Baseline to Month 12
Hamilton Rating Scale for Depression- 25 Item N 34 33 30 30 28 27 27 27
Mean −10.85 −8.82 −11.43 −10.37 −7.86 −10.26 −7.81 −10.48
Std 8.15 8.07 7.15 8.37 7.76 7.49 9.02 7.30
Work and Social Adjustment Scale N 30 31 28 23 26 26 26 26
Mean −7.50 −6.84 −9.11 −8.70 −5.96 −7.73 −4.73 −6.88
Std 8.65 10.49 8.15 8.40 8.67 9.22 8.77 8.00
Child Depression Inventory N 31 32 30 25 26 27 28 21
Mean −0.74 −2.41 −0.80 −3.72 −0.88 −3.15 −4.39 −5.76
Std 5.98 9.56 8.58 8.54 7.60 6.93 7.64 7.18
Columbia Impairment Scale N 29 32 30 26 25 27 24 21
Mean −2.24 −2.63 −2.90 −4.58 −1.20 −3.52 −4.63 −4.33
Std 6.45 8.28 8.14 9.50 7.84 6.99 8.33 10.17
Strengths and Difficulties Questionnaire N 27 29 29 24 22 23 24 19
Mean −1.41 −1.48 −1.21 −2.17 −2.59 −2.26 −3.75 −4.11
Std 5.25 6.55 5.84 6.75 4.27 5.84 6.05 4.59
Open in a new tab

3.4. Moderators of the Relationship between Maternal and Child Outcomes

CTQ scores were available in a subset of participants (n=53). Among the five indictors of childhood trauma, maternal history of sexual abuse and physical neglect were found to moderate the relationship between maternal and child outcomes. Maternal history of sexual abuse moderated the lag-6 association between maternal depression and child depression [F(1,87) = 5.8, p = 0.02]. Maternal history of physical neglect moderated the lag-6 relationship between improvement in maternal depression and improvement in child functioning [F(1,36) = 4.34, p = 0.04]. Other forms of early childhood trauma (emotional neglect and abuse, physical abuse) did not moderate the lag-6 relationship between maternal and child outcomes.

3.5. Role of Parenting in Child Outcomes

Child-reported increase in positive parenting strategies (acceptance) by mothers mediated 6-month lagged associations between maternal and child depression scores, and reduction in negative parenting strategies (psychological control) did not. Specifically, maternal improvement in depression from baseline to 3 months predicted increased acceptance scores on the CRPBI from baseline to 6 months with borderline significance (β = −0.21, p = 0.057), but did not predict change in psychological control scores. Increase in acceptance scores from baseline to 6 months in turn predicted improvement in child depression (CDI) from baseline to 9 months (β = −0.27, p = 0.01). When change in acceptance scores was added to the model, maternal depression was no longer significant, suggesting that the effect of change in maternal depression on improvement in child depression six months later was mediated by the change in positive parenting between the two time points. See Figure 1. This mediational model was not significant for the child outcome measures of the SDQ or the CIS.

Figure 1.

Figure 1.

Open in a new tab

Change in positive parenting (Acceptance) mediates lagged relationship between change in maternal depression and change in child depression (n = 53 dyads).

3.6. Impact of Maternal Childhood Trauma

Because maternal history of sexual abuse as measured by the CTQ was found to moderate the lag-6 association between maternal depression and child depression, it was ruled out as a potential moderator for the mediational effect reported in section 3.5 [57]. Other forms of early childhood trauma (emotional neglect, emotional abuse, physical neglect and physical abuse) did not moderate the lag-6 relationship between maternal and child depression. Thus, these four were further tested for their potential moderating effect on the above mediational model. None of them moderated the effect of maternal depression on positive parenting. However, when we included change in maternal depression, change in acceptance, emotional neglect, and its interactions with maternal depression and acceptance, the interaction between emotional neglect and change in acceptance was found to be just at the level of significance [F(1,69) = 4.07, p = 0.05], suggesting that emotional neglect moderated the mediational effect of positive parenting on the relationship between improvement in maternal depression and improvement in child depression six months later. More specifically, the mediational effect of positive parenting only existed among those dyads whose mothers did not have history of emotional neglect (β = −0.35, p = 0.007). See Figure 2. Similar models were fit for emotional abuse, physical abuse and physical neglect, which did not moderate the mediational effect.

Figure 2.

Figure 2.

Open in a new tab

Maternal history of childhood emotional neglect moderates the mediational effect on positive parenting (acceptance) in the lagged relationship between change in maternal depression and change in child depression.

4.1. DISCUSSION

In a sample of mothers and children both suffering from internalizing disorders, improvement in offspring’s functioning lagged relative to maternal improvement in symptoms. This lagged association coupled with a significant interaction between change scores and time underscores the delay in the relationship between child and maternal outcomes. Consistent with previous results [8, 34], these findings show that children improve when mothers get better, but it takes time for the effect to be detected—as if mothers first make changes in their own lives and then have the emotional and psychological resources to positively influence their children. This suggests that hastening improvement in maternal depression treatment through more rapidly acting treatments may be especially impactful in this population.

In our sample, increases in positive parenting strategies (acceptance) mediated the lagged relationship between improvement in maternal depression and improvement in child depression. This finding is similar to that observed by Garber’s group [33], underscoring that increased positive parenting behaviors is likely an important contributing factor to the indirect effects of maternal depression treatment on their children. Interestingly, decreases in negative (psychologically controlling) parenting behaviors did not mediate these outcomes, suggesting that an increase in positive parenting is more important than a decrease in this type of negative parenting. This perhaps can be best understood in the context of other studies that have found that maternal warmth, in particular, supercedes other parenting variables as a mediator of child outcomes [59].

Neither IPT-MOMS nor BSP focused directly on parenting behaviors. Although mothers’ relationships with their psychiatrically ill children were invariably a part of therapy, the therapist did not focus specifically on provision of parenting skills [60]. However, both IPT-MOMS and BSP are affect-focused therapies which help mothers develop greater capacity to recognize, tolerate, and manage feelings. Thus, both IPTMOMS and BSP may induce downstream effects on children by facilitating increases in maternal warmth which, in turn, appears to have salutary effects on child outcomes [59]. One might conceptualize this process as acting on the quality of the parent-child relationship which, perhaps, leads to enhanced secure attachment behaviors in children. A similar effect on maternal parenting was observed in a study that enrolled depressed mothers treated with antidepressant pharmacotherapy [6], suggesting that this may also be an indirect effect of reductions in depressive symptoms. Studies that compare maternal depression treatment with pharmacotherapy versus affect-focused psychotherapy would be needed to further explore this question.

In this study, maternal childhood history of sexual abuse and physical neglect moderated the relationship between maternal and child outcomes: the positive effects of improved maternal depression on child outcomes were not apparent in mothers who reported histories of early sexual abuse and physical neglect. Childhood trauma has been shown previously to be a non-specific predictor of worse outcomes in depression treatment [61], although this is the first study to show that maternal childhood trauma history affects the relationship between maternal depression treatment outcomes and child outcomes. Given prior studies showing intergenerational effects of trauma [23, 62, 63], it is perhaps not surprising that maternal history of childhood maltreatment offsets the positive effects of maternal depression treatment on offspring. The results of the mediational analyses (section 3.6) in the current trial show that improvement in maternal use of positive parenting did not occur in those who had histories of emotional neglect. Children of mothers with histories of childhood emotional neglect were not able to benefit from their mothers’ depression treatment because their mothers’ parenting strategies did not change in a way that helped their own depression. This again speaks to the enduring negative effects of childhood adversity [64] and suggests that for women with histories of emotional neglect, improvement in depression may not be sufficient to cause a change in parenting behaviors. Neither IPT-MOMS nor BSP directly addresses early childhood trauma; this subgroup of mothers may benefit from interventions that target the psychological sequelae of childhood trauma and, perhaps, parenting training.

History of sexual abuse and physical neglect moderated the lag-6 relationship between maternal and child outcomes but emotional neglect moderated the model showing that increased positive parenting mediated the relationship between changes in maternal and child depression. Forms of maltreatment in childhood are often highly comorbid [65], therefore downstream sequelae are often overlapping. Larger samples are needed to distinguish the specific effects of these experiences. The small sample size likely limited the ability to detect the effects of specific types of maltreatment in each of these domains.

This study has limitations. History of trauma is based on mothers’ retrospective self-reports. Studies examining the accuracy of retrospective reports have found that respondents may have difficulty recalling certain severe childhood events, may choose not to disclose shameful, personal experiences, or may engage in mood-congruent recall when depressed [66, 67]. Thus, recall bias may result in either over- or underreporting of abuse and neglect and thereby impact data reported here. Parenting measures were derived from child reports. These assessments may be biased by child’s mood or other factors and therefore do not constitute objective evaluations of parenting behaviors. Fathers/co-parents were not assessed in the current study. Coparents play an important role in family dynamics and undoubtedly contribute to child outcomes. The current study focused on the impact of maternal treatment on child outcomes because the trial was designed as a maternal treatment study. However, parent-child interactions are bi-directional, and child characteristics likely also influence parenting. For instance, mothers of depressed children have been observed to be less engaged and less active compared to mothers of non-depressed children [20], suggesting that some parents may withdraw in the face of their child’s internalizing symptoms. We did not specifically model these effects. We did not evaluate presence of trauma history in the children. The mediational model included vectors that were of borderline statistical significance, suggesting that these findings must be considered preliminary and hypothesis-generating. Similarly, the number of subjects with histories of childhood emotional neglect were very small in the moderated mediation model (n=17), underscoring the exploratory nature of these findings. Findings were not adjusted for the effects of multiple comparisons. Larger studies should be conducted to replicate the results.

5.1. CONCLUSIONS

The clinical implications of this trial are important. Treatment of maternal depression with evidence-based psychotherapy in families where children also experience internalizing disorders is likely to result in rapid improvement of maternal symptoms, however, one can expect a 6-month delay in positive impact on offspring as mothers make changes in their parenting behaviors after their depression resolves. Increases in positive parenting mediated the relationship between changes in maternal depression and child depression, suggesting that more intensive interventions that directly target positive parenting earlier in course of treatment may help to further improve outcomes for these very high risk families. This study also adds to our understanding of the enduring negative effects of childhood trauma by demonstrating that it continues to affect the next generation, even when mothers received treatment for depression. Treatments that directly address trauma may be needed for families affected by both depression and early childhood adversity.

Highlights.

  • Offspring of depressed mothers with histories of early maltreatment are at high risk for poor outcomes

  • Maternal history of sexual abuse and physical neglect moderate impact of maternal depression treatment on children

  • Positive parenting mediates relationship between change in maternal and child outcomes

  • Depressed mothers with histories of emotional neglect may need training in positive parenting to help improve outcomes in children

ACKNOWLEDGEMENTS

The authors would like to thank the families who participated in this research project and the clinics that helped us to recruit interested and eligible families.

FUNDING

This study was funded by the National Institute of Mental Health [grant number R01 MH083647].

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

  • [1].Swartz HA, Shear MK, Wren FJ, Greeno CG, Sales E, Sullivan BK, et al. Depression and anxiety among mothers who bring their children to a pediatric mental health clinic. Psychiatric Services. 2005;56:1077–83. [DOI] [PubMed] [Google Scholar]
  • [2].Beardslee WR, Gladstone TR, O’Connor EE. Transmission and prevention of mood disorders among children of affectively ill parents: a review. Journal of the American Academy of Child & Adolescent Psychiatry. 2011;50:1098–109. [DOI] [PubMed] [Google Scholar]
  • [3].Brent DA, Kolko DJ, Birmaher B, Baugher M, Bridge J, Roth C, et al. Predictors of treatment efficacy in a clinical trial of three psychosocial treatments for adolescent depression. J Am Acad Child Adolesc Psychiatry. 1998;37:906–14. [DOI] [PubMed] [Google Scholar]
  • [4].Garber J, Clarke GN, Weersing VR, Beardslee WR, Brent DA, Gladstone TR, et al. Prevention of depression in at-risk adolescents: a randomized controlled trial. JAMA. 2009;301:2215–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Pilowsky DJ, Wickramaratne P, Talati A, Tang M, Hughes CW, Garber J, et al. Children of depressed mothers 1 year after the initiation of maternal treatment: findings from the STAR*D-Child Study. American Journal of Psychiatry. 2008;165:1136–47. [DOI] [PubMed] [Google Scholar]
  • [6].Weissman MM, Wickramaratne P, Pilowsky DJ, Poh E, Batten LA, Hernandez M, et al. Treatment of Maternal Depression in a Medication Clinical Trial and Its Effect on Children. The American Journal of Psychiatry. 2015;172:450–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [7].Coiro MJ, Riley A, Broitman M, Miranda J. Effects on children of treating their mothers’ depression: Results of a 12-month follow-up. Psychiatric Services. 2012;63:357–63. [DOI] [PubMed] [Google Scholar]
  • [8].Swartz HA, Frank E, Zuckoff A, Cyranowski JM, Houck PR, Cheng Y, et al. Brief interpersonal psychotherapy for depressed mothers whose children are receiving psychiatric treatment. American Journal of Psychiatry. 2008;165:1155–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [9].Gunlicks ML, Weissman MM. Change in child psychopathology with improvement in parental depression: a systematic review. Journal of the American Academy of Child & Adolescent Psychiatry. 2008;47:379–89. [DOI] [PubMed] [Google Scholar]
  • [10].American Psychiatric Association; Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition. Washington, D.C.: American Psychiatric Publishing, Inc; 2013. [Google Scholar]
  • [11].Lovejoy MC, Graczyk PA, O’Hare E, Neuman G. Maternal depression and parenting behavior: a meta-analytic review. Clinical Psychology Review. 2000;20:561–92. [DOI] [PubMed] [Google Scholar]
  • [12].Oyserman D, Bybee D, Mowbray C, Hart-Johnson T. When mothers have serious mental health problems: parenting as a proximal mediator. Journal of Adolescence. 2005;28:443–63. [DOI] [PubMed] [Google Scholar]
  • [13].Baumrind D Rearing competent children In: Damon W, editor. Child Development Today and Tomorrow. San Francisco, CA: Jossey-Bass; 1988. [Google Scholar]
  • [14].Brennan PA, Le Brocque R, Hammen C. Maternal depression, parent-child relationships, and resilient outcomes in adolescence. Journal of the American Academy of Child & Adolescent Psychiatry. 2003;42:1469–77. [DOI] [PubMed] [Google Scholar]
  • [15].Goodman SH, Gotlib IH. Risk for psychopathology in the children of depressed mothers: A developmental model for understanding mechanisms of transmission. Psychol Rev. 1999;106:458–90. [DOI] [PubMed] [Google Scholar]
  • [16].Nicholson J, Sweeney EM, Geller JL. Mothers with mental illness: I. The competing demands of parenting and living with mental illness. Psychiatric Services. 1998;49:635–42. [DOI] [PubMed] [Google Scholar]
  • [17].Newcomb MD, Locke TF. Intergenerational cycle of maltreatment: a popular concept obscured by methodological limitations. Child Abuse & Neglect. 2001;25:1219–40. [DOI] [PubMed] [Google Scholar]
  • [18].DiLillo D, Damashek A. Parenting characteristics of women reporting a history of childhood sexual abuse. Child Maltreatment. 2003;8:319–33. [DOI] [PubMed] [Google Scholar]
  • [19].Appleyard K, Berlin LJ, Rosanbalm KD, Dodge KA. Preventing Early Child Maltreatment: Implications from a Longitudinal Study of Maternal Abuse History, Substance Use Problems, and Offspring Victimization. Prev Sci. 2011;12:139–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Zalewski M, Cyranowski JM, Cheng Y, Swartz HA. Role of maternal childhood trauma on parenting among depressed mothers of psychiatrically ill children. Depress Anxiety. 2013;30:792–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Martins C, Gaffan EA. Effects of early maternal depression on patterns of infantmother attachment: a meta-analytic investigation. J Child Psychol Psychiatry. 2000;41:737–46. [PubMed] [Google Scholar]
  • [22].Lyons-Ruth K, Block D. The disturbed caregiving system: Relations among childhood trauma, maternal caregiving, and infant affect and attachment. Infant Mental Health Journal. 1996;.17:pp. [Google Scholar]
  • [23].Berthelot N, Ensink K, Bernazzani O, Normandin L, Luyten P, Fonagy P, et al. Intergenerational transmission of attachment in abused and neglected mothers: The role of trauma-specific reflective functioning. Infant Mental Health Journal. 2015;.36:pp. [DOI] [PubMed] [Google Scholar]
  • [24].De Wolff MS, van Ijzendoorn MH. Sensitivity and attachment: a meta-analysis on parental antecedents of infant attachment. Child Dev. 1997;68:571–91. [PubMed] [Google Scholar]
  • [25].Bartholomew K Adult attachment processes: individual and couple perspectives. British Journal of Medical Psychology. 1997;70:249–63; discussion 81–90. [DOI] [PubMed] [Google Scholar]
  • [26].Brumariu LE, Kerns KA. Parent-child attachment and internalizing symptoms in childhood and adolescence: a review of empirical findings and future directions. Dev Psychopathol. 2010;22:177–203. [DOI] [PubMed] [Google Scholar]
  • [27].Ainsworth MD. Object relations, dependency, and attachment: A theoretical review of the infant-mother relationship. Child Development. 1969;.40:pp. [PubMed] [Google Scholar]
  • [28].Zlotnick C, Ryan CE, Miller IW, Keitner GI. Childhood abuse and recovery from major depression. Child Abuse Negl. 1995;19:1513–6. [DOI] [PubMed] [Google Scholar]
  • [29].Nanni V, Uher R, Danese A, Id, Uher ROhoo. Childhood maltreatment predicts unfavorable course of illness and treatment outcome in depression: A meta-analysis. The American Journal of Psychiatry. 2012;.169:pp. [DOI] [PubMed] [Google Scholar]
  • [30].Grote NK, Spieker SJ, Lohr MJ, Geibel SL, Swartz HA, Frank E, et al. Impact of childhood trauma on the outcomes of a perinatal depression trial. Depression & Anxiety. 2012;29:563–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [31].Lanquillon S, Krieg JC, Bening-Abu-Shach U, Vedder H. Cytokine production and treatment response in major depressive disorder. Neuropsychopharmacology. 2000;22:370–9. [DOI] [PubMed] [Google Scholar]
  • [32].Plant DT, Jones FW, Pariante CM, Pawlby S. Association between maternal childhood trauma and offspring childhood psychopathology: Mediation analysis from the ALSPAC cohort. The British Journal of Psychiatry. 2017;.211:pp. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [33].Garber J, Ciesla JA, McCauley E, Diamond G, Schloredt KA. Remission of depression in parents: links to healthy functioning in their children. Child Dev. 2011;82:226–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [34].Swartz HA, Cyranowski JM, Cheng Y, Zuckoff A, Brent DA, Markowitz JC, et al. Brief psychotherapy for maternal depression: Impact on mothers and children. Journal of the American Academy of Child & Adolescent Psychiatry. 2016;.55:pp. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [35].Thase ME, Carpenter L, Kupfer DJ, Frank EF. Clinical significance of reversed vegetative subtypes of recurrent major depression. Psychopharmacology Bulletin. 1991;27:17–22. [PubMed] [Google Scholar]
  • [36].Swartz HA, Zuckoff A, Frank E, Spielvogle H, Shear MK, Fleming MAD, et al. An open-label trial of enhanced brief interpersonal psychotherapy in depressed mothers whose children are receiving psychiatric treatment. Depression and Anxiety. 2006;23:398–404. [DOI] [PubMed] [Google Scholar]
  • [37].Markowitz JC. What is supportive psychotherapy? FOCUS: The Journal of Lifelong Learning in Psychiatry. 2014;XII:285–9. [Google Scholar]
  • [38].Miller WR, Rollnick S. Motivational interviewing: Helping people change. 3rd ed. New York, NY: Guilford Press; 2013. [Google Scholar]
  • [39].Schensul SL, Schensul JJ, LeCompte MD. Essential ethnographic methods: observations, interviews, and questionnaires Ethnographer’s Toolkit 2. Walnut Creek, CA: Alta Mira Press; 1999. [Google Scholar]
  • [40].Swartz HA, Zuckoff A, Grote NK, Spielvogle H, Bledsoe SE, Shear MK, et al. Engaging depressed patients in psychotherapy: Integrating techniques from motivational interviewing and ethnographic interviewing to improve treatment participation. Professional Psychology: Research and Practice. 2007;38:430–9. [Google Scholar]
  • [41].Swartz HA, Grote NK, Graham P. Brief interpersonal psychotherapy (IPT-B): Overview and review of the evidence. American Journal of Psychotherapy. 2014;68:443–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [42].Weissman MM, Markowitz JC, Klerman GL. Comprehensive guide to interpersonal psychotherapy. New York, NY: Basic Books; 2000. [Google Scholar]
  • [43].Rogers CR. Client-centered Therapy. Boston: Houghton Mifflin; 1951. [Google Scholar]
  • [44].First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clincal Interview for DSM-IV Axis I Disorders (SCID). New York: New York State Psychiatric Institute, Biometrics Research; 1995. [Google Scholar]
  • [45].First MB, Spitzer RL, Gibbon M, Williams JBW. Structured clinical interview for DSM-IV personality disorders, (SCID-II). Washington, D.C.: American Psychiatric Press, Inc; 1997. [Google Scholar]
  • [46].Hamilton M A rating scale for depression. Journal of Neurology, Neurosurgery, and Psychiatry. 1960;25:56–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [47].Mundt JC, Marks IM, Shear MK, Greist JH. The work and social adjustment scale: a simple measure of impairment in functioning. British Journal of Psychiatry. 2002;189:461–4. [DOI] [PubMed] [Google Scholar]
  • [48].Bernstein DP, Stein JA, Newcomb MD, Walker E, Pogge D, Ahluvalia T, et al. Development and validation of a brief screening version of the Childhood Trauma Questionnaire. Child Abuse & Neglect. 2003;27:169–90. [DOI] [PubMed] [Google Scholar]
  • [49].Bevilacqua L, Carli V, Sarchiapone M, George DK, Goldman D, Roy A, et al. Interaction between FKBP5 and childhood trauma and risk of aggressive behavior. Archives of General Psychiatry. 2012;.69:pp. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [50].Walker EA, Unutzer J, Rutter C, Gelfand A, Saunders K, VonKorff M, et al. Costs of health care use by women HMO members with a history of childhood abuse and neglect. Archives of General Psychiatry. 1999;.56:pp. [DOI] [PubMed] [Google Scholar]
  • [51].Kaufman J, Birmaher B, Brent D, Rao U, Flynn C, Moreci P, et al. Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version (K-SADS-PL): initial reliability and validity data. J Am Acad Child Psy. 1997;36:980–8. [DOI] [PubMed] [Google Scholar]
  • [52].Kovacs M Children’s Depression Inventory Manual. North Towanda, NY: Multi-Health Systems; 1992. [Google Scholar]
  • [53].Goodman R Psychometric properties of the strengths and difficulties questionnaire. Journal of the American Academy of Child & Adolescent Psychiatry. 2001;40:1337–45. [DOI] [PubMed] [Google Scholar]
  • [54].Bird HR, Shaffer D, Fisher P, Gould MS, Staghezza B, Chen JY, et al. The Columbia Impairment Scale (CIS): Pilot findings on a measure of global impairment for children and adolescents. International Journal of Methods in Psychiatric Research. 1993;3:167–76. [Google Scholar]
  • [55].Margolies PJ, Weintraub S. The revised 56-item CRPBI as a research instrument: Reliability and factor structure. Journal of Clinical Psychology. 1977;33:472–6. [Google Scholar]
  • [56].Zweben A, Zuckoff A. Motivational interviewing and treatment adherence In: Miller WR, Rollnick S, editors. Motivational interviewing: Preparing people for change. 2nd ed. New York: Guilford Press; 2002. [Google Scholar]
  • [57].Muller D, Judd CM, Yzerbyt VY. When moderation is mediated and mediation is moderated. Journal of Personality and Social Psychology. 2005;.89:pp. [DOI] [PubMed] [Google Scholar]
  • [58].Petty RE, Schumann DW, Richman SA, Strathman AJ. Positive mood and persuasion: Different roles for affect under high- and low-elaboration conditions. Journal of Personality and Social Psychology. 1993;.64:pp. [Google Scholar]
  • [59].Jones DJ, Forehand R, Rakow A, Colletti CJM, McKee L, Zalot A. The specificity of maternal parenting behavior and child adjustment difficulties: A study of inner-city African American families. J Fam Psychol. 2008;22:181–92. [DOI] [PubMed] [Google Scholar]
  • [60].Amole MC, Cyranowski JM, Conklin LR, Markowitz JC, Martin SE, Swartz HA. Therapist use of specific and nonspecific strategies across two affect-focused psychotherapies for depression: Role of adherence monitoring. Journal of Psychotherapy Integration. 2017;27:381–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [61].Nanni V, Uher R, Danese A. Childhood maltreatment predicts unfavorable course of illness and treatment outcome in depression: a meta-analysis. Am J Psychiatry. 2012;169:141–51. [DOI] [PubMed] [Google Scholar]
  • [62].McDonnell CG, Valentino K. Intergenerational Effects of Childhood Trauma: Evaluating Pathways Among Maternal ACEs, Perinatal Depressive Symptoms, and Infant Outcomes. Child Maltreat. 2016. [DOI] [PubMed] [Google Scholar]
  • [63].Yehuda R, Halligan SL, Grossman R. Childhood trauma and risk for PTSD: relationship to intergenerational effects of trauma, parental PTSD, and cortisol excretion. Dev Psychopathol. 2001;13:733–53. [DOI] [PubMed] [Google Scholar]
  • [64].Anda RF, Felitti VJ, Bremner JD, Walker JD, Whitfield C, Perry BD, et al. The enduring effects of abuse and related adverse experiences in childhood: A convergence of evidence from neurobiology and epidemiology. Eur Arch Psychiatry Clin Neurosci. 2006;.256:pp. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [65].Edwards VJ, Holden GW, Felitti VJ, Anda RF. Relationship between multiple forms of childhood maltreatment and adult mental health in community respondents: results from the adverse childhood experiences study. Am J Psychiatry. 2003;160:1453–60. [DOI] [PubMed] [Google Scholar]
  • [66].Fergusson DM, Horwood LJ, Woodward LJ. The stability of child abuse reports: a longitudinal study of the reporting behaviour of young adults. Psychol Med. 2000;30:529–44. [DOI] [PubMed] [Google Scholar]
  • [67].Williams LM. Recall of childhood trauma: a prospective study of women’s memories of child sexual abuse. J Consult Clin Psychol. 1994;62:1167–76. [DOI] [PubMed] [Google Scholar]

RESOURCES