Socioemotional outcomes

Socioemotional competencies underlie early childhood mental health and encompass “a child’s developing capacity from birth through five years of age to form close and secure adult and peer relationships; experience, regulate, and express emotions in socially and culturally appropriate ways; and explore the environment and learn—all in the context of family, community, and culture.” [7] (p. 2). The relationship between maternal depression and disruptions in a young child’s socioemotional development and related behavioral manifestations has been well replicated throughout infancy and early childhood [1,8].

Our review focuses on adverse early social and emotional outcomes associated with maternal depression and greater risk for psychopathology later in life [9]: 1) temperamental traits (e.g., higher negative emotionality and lower self-regulatory capacity) in infants that increase vulnerability to stressful caregiving environments; 2) infant social withdrawal behavior; 3) insecure mother-child attachment [10]; and 4) internalizing (e.g., fearfulness, anxiety) and externalizing (e.g., aggression) behaviors in toddlers and preschoolers [9]. We focus on maternal rather than paternal depression due to the high prevalence of maternal depression during sensitive developmental periods [11] and fetal exposure to prenatal depression during pregnancy.

Maternal sensitivity

Prenatally, maternal depression may alter biochemical pathways associated with physiological and emotional regulation, thus, undermining the foundation on which children develop socioemotional competencies [12,13]. Adverse outcomes, however, may still be mitigated by nurturing mother-child interactions [14]. The early social and cultural caregiving environment continues to shape a young child’s socioemotional developmental trajectory and, thereby, promotes future resiliency or exacerbates risk for mental health disorders over the life course [15]. The influence of maternal depression on parenting behaviors (e.g. maternal sensitivity) may underlie the relationship between maternal depression and child socioemotional maladjustment [16,17]. Mothers who are depressed may be less able to provide sensitive interactions needed to develop and sustain early socioemotional competencies, especially in the context of economic stressors [18,19].

Given that cultural beliefs and values influence parenting and subsequent child development [20,21], the interaction between maternal sensitivity and maternal mental health on young children’s socioemotional outcomes may differ based on sociocultural context [2,22,23]. In fact, similar parenting behaviors often result in contradictory socioemotional child outcomes depending on the cultural meaning underlying the behavior [20]. The effects of maternal depression on maternal sensitivity and subsequent mother-child interactions are both a modifiable and necessary target for intervention [24]. Empirical evidence suggests that integrated interventions during the first five years of life to reduce maternal depression and enhance parenting skills may ultimately improve children’s socioemotional developmental trajectories [16], particularly for the most vulnerable mother-child dyads [25]. However, culturally relevant interventions necessitate further research into the intersection of these variables and related child outcomes within the broader sociocultural and economic context.

Economic and sociocultural stressors

The relationship between mother and child occurs within the context of poverty in >26% of Latino families with young children [26]. Poverty-related stressors, which have also been used as a proxy for poverty, encompass numerous risk factors for children: food insecurity, overcrowded housing, and community violence. In fact, meta-analyses within the general population have shown that poverty heightens the significant positive association between maternal depression and children’s socioemotional difficulties [1,27]. Ethnic minority status also magnifies these effects, but understudied social and cultural pathways underlying such disparities have not been disentangled [1].

Theoretical frameworks such as the Family Stress Model (FSM) hypothesize that economic stressors engender parental stress and depression, compromising parental quality which could otherwise serve as a child’s protective buffer from poverty; as such, maternal psychological functioning may serve as a mediating mechanism by which poverty-related stressors predict child maladjustment [28]. The few FSM studies examining Latino children have focused primarily on middle childhood and adolescence and provide evidence for the mediating role of maternal depression between poverty-related stressors and socioemotional child outcomes [29,30] as well as revealing culture-specific ways in which these variables may interact [31,32]. However, not only do we know little about the validity of these models for Latino children during early childhood [33] but culture-specific stressors have rarely been used as anchors for these studies–despite the integral role of cultural influences on developmental processes [34,35].

Culture-specific stressors (e.g., discrimination, stressful life events, and acculturative stress) have predicted parental depression [36,37] and provide key sociocultural contexts in which to examine maternal and child health [38]. Acculturative stress, stemming from psychosocial adjustment to a host culture, may affect both immigrants and later generations who have simultaneously assimilated to a new culture and maintained cultural heritage [39]. In immigrant women, acculturative stress is often accompanied by family separation, social and linguistic isolation, and discrimination [40]—salient contributors of perinatal depression among Mexican-American mothers [41]. The cumulative effects of these stressors on the relationship between maternal mental health and the child’s gestational and early childhood experiences may alter not only the child’s life course but the physiology of subsequent generations via biological (e.g., epigenetic) mechanisms [38,42,43] and interpersonal (e.g., parenting) processes [42]. The Latino paradox, wherein less acculturated Latinos experience better health outcomes than more acculturated Latinos, has been documented for maternal and child health: birth weights [44] and breastfeeding rates [45]. However, investigators have not, to our knowledge, explored the relationship between maternal biopsychosocial functioning (e.g., depression) and early child socioemotional outcomes in the context of the Latino paradox.

Current study

Although the Latino population comprises the largest ethnic minority group in the US [3], young Latino children remain an understudied population. For public policies to be effective, we need to know what harms and what protects children most profoundly amid the economic and sociocultural contexts in which their lives unfold. Research relevant to understanding Latina mothers and their young children may inform interventions that meaningfully support mothers struggling with depression, thereby potentially minimizing or even averting a developmental cascade that undermines critical socioemotional competencies formed early in life. Therefore, this systematic review addresses the following question: during the foundational years from the prenatal period to early childhood (0–5 years), how is maternal depression in Latina mothers related to their children’s socioemotional outcomes?

Study inclusion criteria

We screened studies for eligibility based on predetermined inclusion criteria per the study protocol. To be considered for inclusion, the study needed to assess the relationship between maternal depressive symptoms and children’s socioemotional outcomes within Latina mother-child dyads living in the US. We used the term “Latina” to refer to a female of Latin American origin or descent (e.g., Mexico, Cuba, Puerto Rico, South and Central America), regardless of race. Maternal depression could encompass depressive symptoms on a continuum, elevated scores on self-report measures, or clinical diagnosis of a depressive disorder. We did not exclude for co-morbid symptoms or diagnoses such as anxiety, stress, or another form of psychological distress. Self-report depression measures (e.g. CES-D) often used in population-based studies may more broadly capture psychological distress and diverse affective symptoms (e.g. anxiety) rather than symptoms specific to a depressive disorder [18,47,48]. We defined socioemotional development as a child’s emerging capacity for the emotional, behavioral, and social competencies underlying early childhood mental health [9]. Further inclusion criteria were a) young children exposed to maternal depression at any point prenatally up until age five, allowing us to focus on foundational sensitive early developmental periods; b) adult mothers ≥ 18 years old; and c) specified maternal economic status. Studies of subpopulations with confounding secondary diseases or conditions altering maternal depressive symptoms or socioemotional outcomes (e.g., cancer, natural disasters, or developmental delays) were excluded. There were no time restrictions applied to ensure comprehensive results; we completed the search in October 2018. Studies could be in English or Spanish.

Search strategy

We identified studies through manual searches and through PubMed, CINAHL, and PsycINFO. We applied a comprehensive search string to all three databases to capture studies examining the impact of maternal depression on young children’s socioemotional and cognitive outcomes in the Latino population: (Postpartum OR Depression OR Depressive Disorder OR Depress*) AND (Postpartum Period OR Postnatal Care OR Prepartum OR pre-partum OR prenatal OR pre-natal OR antenatal OR perinatal OR peripartum OR postpartum OR post-partum OR postnatal OR post-natal OR puerperium OR parturition OR pregnanc* OR pregnant or mother* OR newborn*) AND (Hispanic Americans OR Latina* OR Latino* OR Hispanic*) AND (child* OR infant OR toddler OR preschooler OR pediatric) AND (develop* OR behavior* OR cognitive OR emotion* OR conduct OR language OR temperament OR socio* OR regulation OR internalizing OR externalizing OR psych*). We manually searched reference lists from relevant literature for any additional eligible studies.

Study selection and data extraction

We selected studies using the four-phase process for systematic reviews recommended by the PRISMA Group [46] (Fig 2). From the electronic database search, we identified 914 studies after removing 466 duplicates. We screened titles and abstracts from identified studies for eligibility and excluded 858 studies not meeting inclusion criteria. We retrieved the full text of the 56 remaining to determine eligibility, and recorded reasons for exclusion for the 38 non-eligible studies. In the final phase, 17 studies from the database search and one study from the manual search met all eligibility criteria and were selected for data extraction. Data from included literature was extracted onto a template which included study purpose, first author, publication year, study design, sample characteristics, maternal depression measures and data collection points, socioemotional or cognitive outcome measures and data collection points, main analytical approach and covariates, and main results. When available, country of origin and acculturation data was extracted. A second reviewer evaluated the extracted data for thoroughness. During the extraction phase, two studies did not include maternal age ranges, and three studies included broad maternal age ranges that extended into adolescence. Reviewers agreed to include these studies; the broad age range would provide a representative sample of Latina mothers in the United States. We contacted authors for maternal age distribution if needed. Moreover, we included three dissertations meeting inclusion criteria.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. PRISMA flow chart diagram.

https://doi.org/10.1371/journal.pone.0230256.g002

We extracted data for cognitive outcomes because they are interwoven with developing socioemotional competencies early in life [9]. However, these were excluded from our final qualitative synthesis and placed in a supplemental table in order to focus our attention on early social and emotional outcomes (S1 Table).

Quality assessment

Study quality (Table 1) was assessed by two reviewers following the guidelines established in the National Heart, Lung, and Blood Institute (NHLBI) Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies [49].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Quality assessment tool for observational cohort and cross-sectional studies.

https://doi.org/10.1371/journal.pone.0230256.t001

Study characteristics

In this systematic review, we had a final sample of 15 studies examining Latina maternal depression in relation to child socioemotional development (Table 2). Study samples ranged from 26 to 1,600 mother-child participants per study (totaling 5,656 dyads). Participants in most studies (n = 14) came from low-income backgrounds, except one study using data from the Early Childhood Longitudinal Study–Birth Cohort (ECLS-B)—a nationally representative cohort of children born in 2001. Income levels for participants in the ECLS-B ranged from ≤ 20,0000 to ≥75,0000. Country of origin was reported in seven studies; six studies had mostly Mexican descent participants while one study was predominantly represented by participants of Puerto Rican descent [50]. Five studies employed samples from Head Start programs, which serves low-income families. Twelve of 15 studies were completed within the past ten years, and nine of those studies were published within the past five years.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Summary of studies included in the review (n = 15, sample size = 5,656).

https://doi.org/10.1371/journal.pone.0230256.t002

Study designs included five cross-sectional studies and 10 longitudinal studies. Most studies measured child socioemotional outcomes through mother-reported scales, such as a) Child Behavior Checklist for ages 1 ½ to five (CBCL, n = 8), which measures internalizing and externalizing behavioral functioning [51]; and b) Infant Behavior Questionnaire (IBQ-R, n = 2) which assesses infant’s regulatory capacity, negative effect, and extraversion [52]. Maternal depressive symptomology was measured by self-reported symptoms in all but one study, which used the DSM-IV diagnostic criteria for Major depressive disorder [53]. The most prevalent measure was the Center for Epidemiologic Studies-Depression Scale (CES-D, n = 9), which measures depression symptoms within the past week. Maternal acculturation was measured in nine studies, mostly through nativity (n = 6). Two studies measured maternal sensitivity: a mother’s ability to perceive, interpret, and appropriately respond to child signals [54]. Measures of maternal sensitivity included The Three Bag Task or Two Bag Task [55], in which the dyad is asked to play with three toys during a 15-minute video-recorded session; trained coders later assess parent behavior.

Data synthesis

We grouped key findings from selected studies into three categories based on individual, interpersonal, and contextual pathways linking maternal depression to child development (Fig 3): 1) child’s socioemotional outcomes in relation to chronological developmental windows and the course and timing of maternal depression (prenatal, infancy, toddlerhood, preschool, and infancy through early childhood); 2) associations between maternal depression and maternal sensitivity in relation to child socioemotional outcomes; and 3) contextual stressors culturally relevant to Latina mothers and their young children living in poverty (economic and sociocultural stressors).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Timeline of studies reviewed (n = 15) based on maternal depression and child outcome assessments.

https://doi.org/10.1371/journal.pone.0230256.g003

Socioemotional outcomes

Maternal sensitivity

Two studies examined the relationship between maternal depressive symptoms, maternal sensitivity, and acculturation during toddlerhood and preschool [2,64]. In the ECLS-B’s full sample representing the general population, maternal depressive symptoms measured at nine months (CES-D ≥ 16) and 24 months (Composite International Diagnostic Interview Short Form Instrument) were inversely associated with maternal sensitivity scores [2]. However, in analyses stratified by ethnicity, patterns of depression at nine months and 24 months (e.g., remittent, late onset, chronic) were not significantly associated with maternal sensitivity in Latina mothers [2]. In contrast, Palermo et al. found a significant inverse association between maternal sensitivity and maternal mental health problems–a variable combining maternal depressive symptoms (CES-D), parenting stress, and self-efficacy measures. In both studies, more acculturated mothers scored significantly higher on maternal sensitivity measures than less acculturated mothers. In fact, maternal acculturation (as measured by either primary language use, nativity, or a combination of generational status and language proficiency), not maternal depressive symptoms, was most consistently and significantly associated with maternal sensitivity scores [2,64].

Maternal sensitivity was also tested as a mediator between maternal depressive symptoms and child outcomes. Combining stress and depression measures into one variable again yielded the stronger association: maternal sensitivity significantly mediated the association between maternal mental health problems and child behavioral functioning (Problem Behavior Scale) in preschoolers [64]. Although Huang et al. found that maternal sensitivity mediated the associations between maternal depressive symptoms and child attachment in toddlers, the mediation was not significant in Latina mother-child dyads–a finding that differed from the ECLS-B’s full sample [2]. In summary, unless maternal depressive symptoms were combined with maternal stress, maternal sensitivity scores were not correlated with maternal depression and did not function as a significant mediator between maternal depressive symptoms and child outcomes. Both studies assessed and adjusted for demographic moderators (e.g., socioeconomic status, maternal education).

Economic and sociocultural stressors

Maternal sensitivity

Maternal sensitivity scores were more consistently correlated with maternal acculturation than maternal depression, whereby high levels of maternal acculturation were linked to high levels of maternal sensitivity scores—thus raising the question “Are maternal sensitivity measures more closely linked to cultural assimilation than maternal processes that protect children?” Although previous research has supported maternal sensitivity as a protective cross-cultural construct critical for healthy child developmental outcomes [68], some evidence asserts that measures used to operationalize this complex construct may be culture-specific [69]. In fact, the Nursing Child Assessment Teaching Scale, used to measure maternal sensitivity during infancy in the ECLS-B, may detect maternal knowledge of early child development more sensitively than either maternal depression or self-efficacy [70]. Although positive correlations between maternal sensitivity and preschooler’s behavioral functioning suggest that maternal sensitivity measures may at least partially apply to Latina mothers in relation to child well-being [64], measures are likely not capturing key maternal processes specific to less acculturated Latina mothers; these processes warrant further exploration. Moreover, considering that increased maternal sensitivity scores in more acculturated Latina mothers did not moderate the relationship between maternal depression and child outcomes in toddlers or preschoolers [2,64], unexplored interpersonal or contextual factors in relation to maternal depression may play a superseding role in determining child well-being.

Economic and sociocultural stressors

Depression has long been recognized as an interdependent phenomenon inextricably linked to interpersonal and contextual factors [71]. As such, economic stressors and lack of social support were significantly associated with maternal depression in our studies. We also found some evidence for the mediating [67] and moderating [56,63] role of maternal depression between poverty-related stressors and child socioemotional outcomes. Our review found no support for using the FSM during the preschool years, in contrast to FSM studies conducted during middle childhood and adolescence within the broader literature [31,32,72]. Perhaps the cumulative effects of chronic stressors are not measurable until later in childhood. The preschool years, however, were the least likely to yield significant associations between maternal depression and child development in our studies unless maternal depression was combined with a measure of stress. Elevated scores on both maternal depression and stress measures may reflect more chronic depressive symptoms [73], a pattern of depression associated with greater risk for child maladjustment in our longitudinal studies [2,62,63]. Moreover, some empirical evidence suggests that maternal stress (e.g., parenting stress, limited social support, marital conflict) may mediate the relationship between maternal depression and behavior problems in toddlers [74] and preschoolers [73]. In immigrant Latino families, economic stressors combined with parental immigration-related stress have predicted increased behavioral difficulties in preschoolers [75].

Although studies in our review did not address specific sociocultural stressors, a small but growing body of literature shows that culturally relevant precursors of maternal depression in Latina mothers (e.g., stressful life events, discrimination, acculturative stress) have been associated with socioemotional maladjustment during early developmental windows, suggesting that maternal stress in general may affect a child’s capacity for healthy socioemotional outcomes beginning as early as the prenatal period [38,76–78]. Moreover, bio-behavioral profiles indicative of chronic stress across succeeding generations have been found in Mexican-origin women, whereby third generation women exhibited blunted maternal cortisol during pregnancy [79].

These increasing socioemotional vulnerabilities echo our findings linking consistently poorer socioemotional outcomes to children of U.S.-born mothers compared to children of foreign-born mothers. Such outcomes occurred even within the context of higher family incomes, maternal education, and higher levels of maternal sensitivity scores frequently found in succeeding generations [2]. A previous study examining school readiness also noted healthier socioemotional adjustment in kindergarten children of foreign-born Latino parents compared to children of U.S.-born parents of other races and ethnicities, as rated by teachers [80]. Similarly, a study examining the relationship between cultural adaptation in parents and behavioral functioning in preschoolers found that high US identity in Latino parents was associated with more externalizing behaviors, whereas higher ethnic parental identity predicted lower levels of child externalizing and internalizing behaviors [81]. Increased parental acculturation, as measured by language usage, has also been linked to lower (i.e., less adaptive) cortisol responses in preschoolers living in poverty [75]. Although cross-generational research supports an immigrant paradox in behavioral outcomes during adolescence (e.g., substance abuse, teen pregnancy, delinquency), research examining socioemotional outcomes in children under five years of age has been scarce [82]. This is the first review, to our knowledge, to have found support for the Latino paradox in early childhood mental health. If an immigrant advantage exists throughout early developmental windows in the socioemotional domain, why and when does it erode?

The interaction between maternal nativity and maternal mental health in relation to child outcomes was rarely examined in our studies, preventing us from fully grasping the underlying mechanisms for this apparent decline. However, our findings clearly indicate that a) this pattern can be observed within sensitive early developmental windows; b) maternal mental health and early child development are intimately connected; and c) contextual stressors are salient precursors of maternal mental health. Maternal biopsychosocial stress across generations stemming from the chronic stressors inherent in unequitable societal structures may be a salient contributing factor; protective cultural factors such as collectivism that confer psychological resilience [83] may also be lost in subsequent generations. Combining models based on risk (e.g., developmental-risk framework) and resilience (e.g., ecocultural theory) may help elucidate underlying mechanisms for the Latino Paradox [84].

Clinical implications

Maternal depression is treatable, and its effects on child socioemotional outcomes can be minimized. A meta-analysis revealed that non-pharmaceutical treatment of depression during pregnancy not only improved maternal mood but improved child functioning–particularly infant regulatory capacity [85]. These outcomes provide causational support for the link between prenatal depression and infant socioemotional outcomes as well as subsequent evidence for the intergenerational transmission of mental health disparities. Furthermore, the World Health Organization recommends that interventions should treat maternal depression and support the mother-child relationship concurrently [24]. Well-designed, culturally responsive interventions that mitigate maternal stress and strengthen the mother-child relationship may also foster healthy child development. Although depression is treatable and support is critical during early developmental windows when interventions would be most preventative and cost-effective, treatment often remains inaccessible to Latina mothers even after controlling for income levels [6,86]. Given that maternal depression is highly correlated with adversity, lack of appropriate treatment for mothers suffering from depression and their young children may further compound disparities in children’s socioemotional development. Policies should address the need to support maternal mental health as early as the prenatal period.

Effective interventions for maternal depression in Latinas have focused on culturally sensitive psychoeducation, emotional support, and access to community resources [87]. Exploratory family-level interventions have also embraced potential sociocultural strengths such as kinship and familism (i.e., family interdependence), resulting in decreased maternal depressive symptoms and decreased internalizing and externalizing behaviors in children [88]. Moreover, a culturally-tailored intervention targeting postpartum depression and mother-child interactions in infants and toddlers lowered maternal depression scores and improved maternal perceptions of the child’s socioemotional competence [89]. Upstream approaches in immigrant women should address culturally relevant stressors and barriers that contribute to depression and stress, such as potential and poignant losses experienced during immigration: social support networks lost through family separation, a sense of belonging dislodged by discrimination, and access to community resources impeded by language barriers [41].

Limitations

Due to scarce research examining maternal depression in Latinas and child socioemotional outcomes, we had a limited number of studies–five of which were cross-sectional. Studies were too heterogeneous to conduct a meta-analysis. Furthermore, most studies during the preschool years used maternal-reported measures of child behavior such as the CBCL. Behavior ratings may be influenced by maternal cultural expectations [20] and maternal depression. Mothers with depression may be more likely to view their children negatively, resulting in potentially biased ratings [1]. However, three out of five studies using maternal reported child measures in longitudinal studies adjusted for concurrent depression.

We also only included studies conducted in the US. Although this focus prevented confounding factors that may have arisen from research in several host countries, our findings cannot be generalized outside of the US. Moreover, the Latino population is not homogeneous; subpopulations from Latin American countries have various socioeconomic and cultural backgrounds as well as reasons for coming to the United States [90]. Furthermore, acculturation may differ among Latino sub-groups [38]. Although we extracted data such as country of origin and acculturation, eight studies did not include country of origin, and seven did not measure acculturation.