Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Dec 1.
Published in final edited form as: Depress Anxiety. 2021 Aug 26;38(12):1279–1288. doi: 10.1002/da.23211

Within- and Between-Family Transactions of Maternal Depression and Child Engagement in the First Two Years of Life: Role of Prenatal Maternal Risk and Tobacco Use

Rachel A Level 1, Shannon M Shisler 2, Danielle M Seay 1, Miglena Y Ivanova 1, Madison R Kelm 1, Rina D Eiden 1, Pamela Schuetze 3
PMCID: PMC8665021  NIHMSID: NIHMS1731501  PMID: 34435727

Abstract

Background:

This study examined transactional associations among maternal depression, maternal sensitivity, and child engagement in the context of a low-income, diverse sample with maternal cigarette smoking during pregnancy (MSDP) as a moderator of these transactions.

Methods:

A random-intercept cross-lagged panel model was used to investigate within- and between-family variability from infancy to toddlerhood. The sample included 247 mother-child dyads (47% girls; 51% African-American; 178 MSDP, 69 non-MSDP). Assessments were conducted once during each trimester of pregnancy and at 2, 9, 16, and 24-months of child ages.

Results:

Between-family associations revealed that children exposed to higher levels of sensitive parenting across time had higher behavioral engagement from infancy to toddlerhood. At the within-family level, increased sensitive parenting at 9-months was predictive of increased child engagement at 16-months which in turn predicted increases in sensitive parenting at 24-months. Increased maternal depression was concurrently associated with lower maternal sensitivity at 2-months and lower child engagement at 16-months. Contrary to hypotheses, changes in maternal depression were not associated to changes in parenting or child engagement. These associations did not vary between prenatally smoking and non-smoking mothers. However, there was significantly higher stability in maternal depression across time among non-smoking mothers compared to those in the MSDP group. Additionally, increased maternal depression was related to lower-than-expected child engagement at 9 months only for the non-smoking group.

Conclusions:

Results highlight transactional processes at the within-family level and the importance of timing for parent and child effects on transactional processes.

Keywords: depression, maternal-child, addiction, child/adolescent, pregnancy and postpartum, smoking, maternal depression, prenatal tobacco exposure, mother-child interactions, parenting, bidirectional associations, random intercept cross-lagged panel model

Parent-child interactions, while demanding, can be very rewarding. Viewing an infant’s smiling face elicits a dopaminergic reward response (Strathearn et al., 2008), and the pleasurable nature of infant cues may function as a method of self-regulation that diminishes parents’ stress and promotes sensitive caregiving (Belskey et al., 1991). Maternal depression, characterized by fewer positive emotions and prolonged negative emotions (Joorman & Gotlib, 2010), may disrupt this process. Depression is associated with less positive and more negative parent-infant interactions, as well as diminished pleasure in response to infant cues (Campbell et al., 2004; Cohn et al., 1990; Graham et al., 2000). Thus, parent-infant interactions may be less rewarding to depressed mothers and lead to withdrawn or hostile caregiving. Over time, less sensitive caregiving may decrease children’s engagement with their mothers, further reducing the quality of these interactions.

Individual differences in child responsiveness and positivity may also influence parent-child interactions. Interacting with a difficult or less engaged infant may be more demanding and less rewarding for parents, which may decrease maternal sensitivity and lead to higher levels of depression over time (Choe et al., 2013; Putnam et al., 2002; Roben et al., 2015). In this way, parent- and child-driven processes may influence each other, creating bidirectional feedback loops (Masten & Cicchetti, 2010; Sameroff & Mackenzie, 2003). Evidence suggests that depression and sociodemographic risk during pregnancy may contribute to early differences in child behavior and emotion regulation (i.e., through fetal programming; Dunkel-Schetter & Tanner, 2015), but the scarcity of longitudinal investigations beginning in the prenatal period has limited our ability to investigate these risk factors. To delineate these important mechanisms, we examined bidirectional processes between maternal depression and child engagement via sensitive parenting in a sample of mother-child dyads recruited in the first trimester of pregnancy and assessed across the first two years of life.

Depression and Substance Use

Depression’s effect on parenting and child outcomes may be moderated by several factors, including maternal tobacco use (Goodman et al., 2011). Maternal smoking during pregnancy (MSDP) is associated with higher depression and stress, especially among low-income pregnant smokers (Bullock et al., 2001; Ludman et al., 2000; Munafò et al., 2008). Depression is also associated with continued MSDP (Eiden et al., 2013), suggesting that the combination of depression and MSDP prolongs the effect of nicotine exposure, potentially magnifying depression’s impact on parenting. In fact, per Rutherford and Mayes’ (2017) reward-stress dysregulation model, nicotine exposure damages and disrupts parents’ reward and regulation systems, lessening the saliency of infant signals and decreasing sensitive parenting. Therefore, mothers who smoke during pregnancy may be at increased risk of disrupted stress/reward regulation, and thus show more salient alterations in transactional processes during mother-child interactions compared to demographically similar non-smoking mothers.

MSDP is also linked to higher levels of irritability and emotion dysregulation in children (Cornelius & Day, 2009; Schuetze et al., 2017), suggesting that nicotine-exposed infants may be more difficult, less positive, and less responsive during mother-infant interactions than non-exposed infants. These infant behaviors may increase the likelihood of more demanding and less rewarding mother-infant interactions, potentially decreasing maternal sensitivity and increasing maternal depression over time. Therefore, the effect of child engagement on maternal depression and parenting may be most salient for these children.

Current Study

Despite substantial links among depression, MSDP, child engagement, and parenting, previous literature has not examined transactional mechanisms between these factors. Thus, three components guided our study. First, we included prenatal depression and sociodemographic risk into our model as exogenous variables. Second, we tested reciprocal associations among maternal depression, sensitive parenting, and child engagement during mother-child interactions from infancy to toddler age. Third, we tested MSDP as a moderator of transactional associations.

Our hypotheses were threefold: (a) Based on evidence that stress and depression during pregnancy are independent risk factors for postnatal depression and adverse child outcomes (Dunkel-Schetter & Tanner, 2015), we predicted that prenatal depression and sociodemographic risk would be related to higher levels of postnatal maternal depression and lower levels of sensitive parenting and child engagement; (b) Consistent with transactional models of parent-child interactions (e.g., Patterson & Reid, 1970), we predicted that there would be bidirectional associations among maternal depression, sensitive parenting, and child engagement across time, such that maternal depression would predict decreases in child engagement via decreases in sensitive parenting whereas decreases in child engagement would predict increases in maternal depression and decreases in maternal sensitivity over time; (c) Consistent with Rutherford and Mayes’ reward-stress dysregulation model (2017), we expected that MSDP would make all associations among maternal depression, sensitive parenting, and child engagement more pronounced.

Method

Participants and Procedure

Pregnant women who presented for care at an urban prenatal clinic completed eligibility screening at their first prenatal appointment (see author citation, 2018 for procedural details). The final sample consisted of 247 mother-infant dyads (178 MSDP group, 69 non-MSDP group, 47% female). Mothers ranged in age from 18 to 39 years (M = 24.09, SD = 5.00). Mothers were 51% African American, 31% White, and 6% other or mixed race with several mothers identifying more than one race, and 19% Hispanic ethnicity. Most mothers in this sample were high school graduates (60%), 65% were on Medicaid, 53% on Food Stamps. Forty six percent were married or cohabiting, 32% were in a relationship but not cohabiting, 21% were single, and 1% were divorced. Mothers in the MSDP group were individually matched on age and highest education level, but smokers were oversampled to allow for a full range of light to heavy smokers. The protocol was approved by the Institutional Review Board. Assessments were conducted once during each trimester and again at 2, 9, 16, and 24-months of child age.

Maternal Smoking During Pregnancy

MSDP was measured by maternal reports on the Timeline Follow-Back Interview (Sobell & Sobell, 1992), maternal saliva assayed at the end of each trimester, and infant meconium assayed after delivery (see author citation, 2018 & author citation, 2017 for further details). Families were assigned to the smoking group based on positive results on any of these criteria.

Cumulative Demographic Risk

A cumulative demographic risk (CDR; Rutter, 1981) was computed from 4 variables: maternal race, education, occupation, and partner status. If participants belonged to any racial minority group (69%), had below high school education (40%), or were single status (not married, cohabiting, or in a relationship; 21%), they were assigned a score of 1 for that risk category. Maternal occupation was coded using the Hollingshead scale (1975; M = 2.09, SD = 1.61, range = 1–8), divided by the maximum value of 9 to compute a proportion, and then re-coded so higher numbers indicated lower occupation status (greater risk). The final CDR variable was created by averaging the four risk variables and thus had a possible maximum score of 1 (M = 0.67, SD = 0.27, range = 0.07–0.96), with higher scores indicating greater risk. First proposed by Rutter (1979, 1981), the CDR has a long and rich history of use in the field of child development (see Evans, Li, & Whipple, 2013 for a critical review).

Maternal Depression

Maternal depression was assessed during the second and third trimester and postnatally using the Beck Depression Inventory (BDI-II; Beck et al., 1996). As 2nd and 3rd trimester BDI-II scores were highly correlated (r = 0.69, p < .001), they were averaged (M = 15.54, SD = 7.83, Range = 0–48.5) to indicate prenatal depression. The BDI-II had good internal consistency (Cronbach’s α ranged from 0.86 to 0.92; see Table 2).

Table 2.

Bivariate Correlations of Study Variables

Variable 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
Prenatal
1. Cumulative Demographic Risk .05 .00 −.08 −.10 .01 −.10 −.24** −.16* −.11 −.26** −.12 −.12 −.24**
2. Maternal Depression .07 .56** −.11 .02 .52** −.10 .05 .55** .03 −.04 .43** −.16*
3. Child Sex (female = 0, male = 1) .11 −.10 −.08 −.00 .05 −.24** .07 −.10 −.19** −.00 −.11
2 Months
4. Maternal Depression −.16* .01 .47** −.02 .11 .47** .12 .11 .41** 0.11
5. Sensitive Parenting −.01 −.00 .37** .00 .01 .38** .23** −.06 .37**
9 Months
6. Child Engagement .05 .27** .28** .01 .14 .28** .09 .07
7. Maternal Depression .00 .14 .55** .17* .00 .49** −.05
8. Sensitive Parenting .26** −.04 .58** .25** .03 .43**
16 Months
9. Child Engagement −.07 .40** .40** .09 .32**
10. Maternal Depression .04 .03 .54** −.11
11. Sensitive Parenting .23** .07 .52**
24 Months
12. Child Engagement .10 .52**
13. Maternal Depression .04
14. Sensitive Parenting
Open in a new tab
*

p < .05

**

p < .01.

Sensitive Parenting and Child Engagement

Sensitive parenting was assessed from observed mother-infant interactions at all postnatal assessments. At 2-months of infant age, mothers were asked to feed their infants, as they would at home. During the subsequent laboratory visits, mothers were asked to interact with their children as they would at home for 10 minutes in a room with age-appropriate toys. Child engagement was assessed during the free-play tasks at age 9, 16, and 24-months. Interactions were coded using the Parent-Child Early Relational Assessment (Clark et al., 1997) by two sets of coders trained to reliability and blind to smoking status. Sensitive parenting was the average of 11 items reflecting positive tone of voice, affect, verbalizations, responsiveness, sensitivity, involvement, and connectedness with the child rated on a scale of 1–5 and averaged (higher scores = more sensitivity). Child engagement scale was the average of 8 items that reflected child’s positive affect and responsiveness during the interaction rated on a scale of 1–5 and averaged (higher scores = more child engagement). Cronbach’s α ranged from α = 0.73 to α = 0.86. Inter-rater reliability was calculated on 10–13% of the tapes and ranged from intraclass correlation coefficients ranged from .90 to .95.

Missing data

Of the 247 enrolled dyads, five were missing at 9-month assessment, nine dyads were missing at 16-month assessment, and 15 dyads were missing at 24-months assessment. Missing data were handled using full-information maximum likelihood estimates. Dyads with complete versus missing data did not significantly differ on maternal age, education, or any of the study variables.

Analytic Plan

We used the random intercept cross-lagged panel model (RI-CLPM: Hamaker et al., 2015; Keijsers, 2016) to examine relations among maternal depression, sensitive parenting, and child engagement across the first two years of life. While traditional cross-lagged panel modeling (CLPM) is commonly used to test reciprocal influences, this approach fails to distinguish between-person estimates from within-person processes. Unlike traditional CLPM, RI-CLPM allows for the disentanglement of within- and between-person variability, and as such is an ideal method to model transactional associations over time within families. In this paper, within-person and between-person results are described as within-family and between-family, respectively.

Three latent random intercepts (RIs) were included in the model to capture stability in maternal and child variables across time. Correlations between the RIs reflect associations among stable between-family differences in maternal depression, sensitive parenting, and child engagement, while the concurrent and across time correlations capture the within-family associations at each time (e.g., if a mother scores higher on depression than expected, does her child also score lower on child engagement than expected?). The auto-regressive paths capture the extent to which within-person deviations at t + 1 can be predicted by deviations from their expected scores at t. Similarly, cross-lagged effects reflect the extent to which within-family deviations in child engagement at t + 1 can be predicted by deviations from a family’s expected level of maternal depression and sensitivity at t. To test the invariance of the associations across time, we compared a baseline model in which these effects were freely estimated to models in which the autoregressive and lagged effects were constrained to be equal across time.

We first analyzed RI-CLPMs with all structural paths constrained to be equal between the MSDP and non-MSDP group. Next, we used multiple-group path analysis with exposure group (Non-MSDP = 0; MSDP = 1) as the grouping variable. Using a stepwise process, we examined whether freeing the lagged effects (i.e., cross paths and stability paths for parenting and child engagement) and correlated change parameters in our baseline RI-CLPMs resulted in a significantly better model fit. All analyses were completed using Mplus v.7.2 (Muthén & Muthén, 1998–2014).

The degree of model fit was assessed using the comparative fit index (CFI), the Standardized Root Mean Square Residual (SRMR), and the root mean square error of approximation (RMSEA) with its 90% confidence interval, using conventional standards (see Hu & Bentler, 1999). Model selection was based on the Chi-square difference (Δχ2) test where a significant Δχ2 indicates the model with more freely estimated parameters fits the data better. If there was no significant fit improvement, the model with equality constraints was selected to preserve parsimony.

Results

Preliminary Analysis

Relations between study variables and covariates

Means and standard deviations of the study variables for both groups are presented in Table 1. We computed bivariate correlations to examine relations among the covariates and the study variables and to detect stability of variables across time (see Table 2).

Table 1.

Means and Standard Deviations of Study Variables

Variable Total Non-MSDP MSDP
M SD M SD M SD t-test
Prenatal Period
Cumulative Demographic Risk .49 .25 .51 .25 .48 .25 .78
Maternal Depressive Symptoms 15.27 7.73 13.45 7.55 15.98 7.70 −2.36*
2-Months
Maternal Depressive Symptoms 10.08 7.43 8.12 6.59 10.85 7.62 −2.63**
Sensitive Parenting 4.06 0.67 4.13 0.64 4.03 0.67 1.07
9-Months
Child Engagement 4.36 0.46 4.39 0.40 4.34 0.48 0.80
Maternal Depressive Symptoms 10.25 8.42 7.54 7.61 11.38 8.51 −3.15**
Sensitive Parenting 3.87 0.81 3.89 0.77 3.86 0.83 0.22
16-Months
Child Engagement 4.36 0.44 4.34 0.42 4.37 0.46 −0.40
Maternal Depressive Symptoms 9.18 8.38 7.63 8.08 9.81 8.44 −1.71
Sensitive Parenting 3.87 0.72 3.83 0.83 3.88 0.69 −0.51
24-Months
Child Engagement 4.03 0.56 4.02 0.59 4.04 0.55 −0.19
Maternal Depressive Symptoms 8.42 7.81 7.30 6.80 8.89 8.17 −1.36
Sensitive Parenting 3.39 0.79 3.42 0.76 3.37 0.81 0.39
Open in a new tab

Note.

*

p < .05

**

p < .01

Between- and within- family variance

To examine the proportion of variance explained at the between- and within-family levels, we computed intraclass correlations (ICCs). The ICC for maternal depression suggested that the variance explained within- and between-family was relatively equal (ICC = .502). Variations in child engagement were due more to within-family changes across time than to between-family differences (ICC = .312), as were variations in maternal sensitivity, though to a lesser extent (ICC = .443).

Model comparisons

To make the model as parsimonious as possible, we first compared baseline models with lagged (i.e., cross paths and stability paths) and correlated errors constrained to be equal across time to models in which these effects were freely estimated. We also compared the RI-CLPM to the traditional CLPM and found that extending the CLPM with RIs significantly improved the model fit Δχ2 (6) = 122.44, p < .001, suggesting that the disentanglement of within- and between person variance offers a better representation of the data. Specific constraints for the RI-CLPM with the most optimal fit are noted in Figure 1.

Figure 1.

Figure 1.

Open in a new tab

Multiple group RI-CLPM examining relations between maternal depression, sensitive parenting, and child engagement across the first two years of life, controlling for demographic risk and child sex. Standardized beta coefficients are presented. Only significant paths for the multigroup model are presented for ease of presentation. In the full group model, the following paths were constrained: autocorrelations for maternal depression and child engagement; correlated change between maternal depression and sensitivity; cross-lagged paths between engagement and depression, depression and sensitivity, and depression and engagement. In the multigroup model all paths were free to vary except autocorrelation for depression between groups, autocorrelation for depression in the non-MSDP group, autocorrelation between sensitivity at 2 and 9 months, and between-level associations. *p < .05, *** p < .01.

Primary Analysis

Child sex was related to sensitive parenting at 16-months (β = −0.15, p = .03) with mothers of boys showing less sensitivity. Child sex was also associated with child engagement at 16 (β = −0.28, p < .001) and 24 (β = −0.21, p = .007) months with boys being less engaged. Higher demographic risk was only associated with higher levels of prenatal depression (β = 0.18, p = .01).

RI-CLPM full sample

The final model showed good fit, χ2 (57) = 96.39, p = .0142, CFI = .95, RMSEA = .05 [.03, .07], SRMR = .07. Concurrent associations suggested that at 2-months, higher maternal depression was related to less sensitive parenting (see Table 3). At 9-months, higher-than-expected sensitive parenting was associated with higher child engagement. At 16-months, higher-than-expected levels of maternal depression were related to lower-than-expected levels of child engagement, while higher-than-expected sensitive parenting was associated with higher-than-expected child engagement. At 24-months higher-than-expected sensitive parenting was related to higher-than-expected child engagement.

Table 3.

Between and Within -Level Estimates from the RI-CLPM

Associations
Overall MSDP Non-MSDP
Between-level associations
Depression ←→ Sensitive parenting −.06 −.14 −.14
Depression ←→ Child engagement .14 .17 .17
Sensitive parenting ←→ Child engagement .26* .29* .26*
Within-level associations
2 Months
Depression ←→ Sensitive parenting −.26** −.28** −.28**
9 Months
Depression ←→ Sensitive parenting .04 .03 .03
Depression ←→ Child engagement .04 .19 −.32*
Sensitive parenting ←→ Child engagement .27** .29** .29**
16 Months
Depression ←→ Sensitive parenting .05 .04 .04
Depression ←→ Child engagement −.28** −.26** −.26**
Sensitive parenting ←→ Child engagement .43*** .40*** .40***
24 Months
Depression ←→ Sensitive parenting .05 .04 .04
Depression ←→ Child engagement .06 .05 .05
Sensitive parenting ←→ Child engagement .55*** .55*** .55***
Open in a new tab

Note.

*

p < .05

**

p < .01

***

p < .001.

Model fit indices were CFI = .95 for the full sample and CFI = .90 for the moderated model.

Higher-than-expected sensitive parenting at 9-months predicted higher-than-expected sensitive parenting at 16-months. There was also evidence of a transactional loop such that higher than expected sensitive parenting at 9-months was predictive of higher-than-expected child engagement at 16-months, which was then prospectively predictive of higher-than-expected sensitive parenting at 24-months (see Figure 1).

RI-CLPM multi-group by exposure group

Results from multi-group RI-CLMP are presented in Table 3. Maternal depression appeared stable from 2 to 24-months for the non-MSDP group, but not for the MSDP group. The only path that was different between groups was the concurrent association between depression and child engagement at 9-months. For the non-MSDP group, higher-than-expected maternal depression was related to lower–than-expected child engagement.

Discussion

Using a longitudinal design with a high-risk sample, this study extends research on transactional processes within families in multiple ways. First, our study adds to prior research by examining moderation of these processes by MSDP. Second, we utilized a novel statistical approach, the RI-CLPM, which allowed us to examine change and transactional processes at the within-family level (i.e., where processes between children and their environments occur). Finally, examining these processes during the first two years of life allowed us to investigate a critical period of development where maternal depression rates increase (Gelave et al., 2016) and children are most sensitive to depression’s effects on parenting (Field, 2010). Findings highlight transactional processes at the within-family level with higher-than-expected maternal sensitivity at 9-months prospectively predicting higher-than-expected child engagement at 16-months, which in turn was predictive of higher-than-expected maternal sensitivity at 24-months.

Prenatal Risk Factors

Bivariate associations revealed moderate stability for maternal depression from the prenatal period to toddler age. Additionally, prenatal depression was negatively associated with sensitive parenting at child age 24-months. CDR was negatively associated with sensitive parenting at child age 9, 16, and 24-months. Mothers’ CDR was related to increased prenatal, but not postnatal depression. Within our model, mothers’ prenatal depression and CDR were not related to future changes in parenting or child engagement. These findings suggest that a child’s postnatal environment accounts for unique variance in changes in maternal and child interactive behavior across the first two years of life.

Within- and Between- Family Effects

Sensitive Parenting and Child Engagement

At the between-family level, children exposed to more sensitive parenting across time had higher levels of engagement from infancy to toddlerhood. These findings are consistent with CLPM studies revealing more sensitive parents experience more child engagement in parent-child interactions compared to less sensitive parents (Eiden et al., 2011; Perry et al., 2014). Furthermore, within-family cross-lagged effects confirmed reciprocal associations between sensitive parenting and child engagement. Mothers’ increased sensitivity at 9-months predicted increased child engagement at 16-months, which in turn was predictive of increased sensitive parenting at 24-months. These findings suggest mother-driven effects on dyad interactions such that increased sensitive parenting led to more child engagement which led to future increases in sensitive parenting. Interestingly, these transactional processes were delayed until child ages 16 and 24-months. This suggests that child-driven effects may be initially smaller, but gain more salience as children develop their social, cognitive, language, and motor abilities.

Maternal Depression

There were no significant associations between maternal depression and mother-child interactions at the between-family level. Contrary to our hypotheses, maternal depression was not associated with changes in sensitive parenting within mother-child interactions over time. Indeed, increased depression was associated with lower levels of sensitive parenting only at 2-months of age. However, we found concurrent associations between maternal depression and child engagement at 16-months that were not related to changes in sensitive parenting. Additionally, there were no significant causal pathways from child engagement to maternal depression, suggesting that prospective associations between maternal depression and parent-child interactions may be primarily parent-driven. Given the lack of association between maternal depression and sensitive parenting, further investigation is necessary to identify other maternal factors that may contribute to this association. For example, mothers’ use of language and vocabulary has been found to contribute to their child’s engagement during mother-child interactions (Pearson et al., 2016). Previous studies have found that mothers with depression speak less and use more negative verbal expressions during their interactions with their children, potentially diminishing child’s engagement (Bornstein et al., 2010). Given our findings that transactional processes between parenting and child engagement were only apparent at critical periods of language development (16 and 24-months; McMurray, 2007), it is possible that differences in maternal language may contribute to the association between maternal depression and child engagement. However, as we found no association between maternal depression and child engagement at 24-months, it is possible that sensitive parenting may buffer the direct association between maternal depression and child engagement and may still represent an intervention target.

Group Differences

Overall, mothers in the MSDP group had higher levels of depression prenatally and at 2 and 9-months postpartum compared to non-smoking mothers. However, results from our RI-CLPM suggested that while maternal depression was stable from 2 to 24-months for non-smoking mothers, it was not stable for the MSDP group. Although we hypothesized that co-occurring depression and smoking would magnify disruption to parents’ stress/reward regulation, it is possible that the positive mood or relief parents experience when smoking cigarettes may contribute to greater instability in depressive symptoms. Indeed, there is evidence suggesting that cigarette smoking is associated with greater day-to-day variation in internalizing symptoms and mood (Bares et al., 2018).

No group differences were found at the between-family level. At the within-family level, increases in maternal depression at 9-months were only associated with less child engagement at 9-months for the non-smoking group. It is possible that the greater stability of depression for non-smoking mothers may lead their children to be more sensitive to changes in maternal depression. No other within-family differences were found. It seems that in our diverse, low-income sample, depression was associated with maternal behavior similarly across the groups. It is possible that in such a socio-demographically similar group, other risk factors (e.g., food instability, community violence) may influence the relationship between maternal depression and child engagement more strongly than MSDP. Furthermore, the greater day-to-day variation in maternal symptomatology associated with tobacco-use (Bares et al., 2018) may have limited detection of within-family change for the MSDP group. Future studies incorporating maternal smoking in the postnatal period may yield different results.

In our preliminary analyses, we found significant differences in parenting and child engagement for boys and girls. Mothers of boys were less sensitive at 16-months and, in turn, boys were less engaged at 16 and 24-months. Given previous findings that boys are more sensitive to maternal depression than girls (Park et al., 2018), there may be more pronounced effects in families of boys. Future studies may examine whether these transactional processes operate differently for boys and girls.

Limitations and Future Directions

The current study is among the first to examine transactional dynamics between mother-child dyads within the context of MSDP, but was not without limitations. First, our study did not include measures of child behavior beyond 24-months of age, limiting our ability to make predictions of child outcomes and to examine the intergenerational transmission of psychopathology. Second, our study did not include a measure of mothers’ postnatal smoking. Although strongly correlated with prenatal smoking (Shisler et al., 2016), change in postnatal smoking may influence the relationship between maternal depression and child engagement via sensitive parenting (i.e., increases in maternal irritability). Future research should examine how day-to-day variation in substance use may affect mother-child interactions. Given the significant sex differences found in our study, it may also be useful to investigate how MSDP and sensitive parenting affect boys and girls differently.

Despite these limitations, this study highlights the transactional nature of mother-child interactions and is one of the first to examine these processes differentiating within- and between-family effects in the context of MSDP, using a low-income and diverse sample. Additional strengths included the multi-method assessments of MSDP and observational assessments of maternal and child behavior. These results increase our understanding of transactional processes in the first two years of life and within time-direct associations between increases in maternal depression and decreases in child engagement.

Footnotes

We have no conflicts of interests to disclose.

References

  1. Bares CB, Dick DM, & Kendler KS (2018). Nicotine dependence, internalizing symptoms, mood variability and daily tobacco use among young adult smokers. Addictive Behaviors, 83, 87–94. 10.1016/j.addbeh.2017.09.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Beck AT, Steer RA, & Brown GK (1996). Manual for the Beck Depression Inventory-II. San Antonio, TX: Psychological Corporation. [Google Scholar]
  3. Belsky J, Youngblade L, Rovine M, & Voling B (1991). Patterns of marital change and parent-child interactions. Journal of Marriage and the Family, 52, 83–96. 10.2307/352914 [DOI] [Google Scholar]
  4. Bullock LF, Mears JL, Woodcock C, Record R (2001). Retrospective study of the association of stress and smoking during pregnancy in rural women. Addictive Behaviors, 26, 405–413. 10.1016/S0306-4603(00)00118-0 [DOI] [PubMed] [Google Scholar]
  5. Campbell SB, Cohn JF, & Meyers T (1995). Depression in first-time mothers: Mother-infant interaction and depression chronicity. Developmental Psychology, 31(3), 349–357. 10.1037/0012-1649.31.3.349 [DOI] [Google Scholar]
  6. Choe DE, Sameroff AJ, & McDonough SC (2013). Infant functional regulatory problems and gender moderate bidirectional effects between externalizing behavior and maternal depressive symptoms. Infant Behavior and Development, 36(3), 307–318. 10.1002/da.22688 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Choe DE, Shaw DS, Brennan LM, Dishion TJ, & Wilson MN (2014). Inhibitory control as a mediator of bidirectional effects between early oppositional behavior and maternal depression. Development and Psychopathology, 26(4 Pt 1), 1129–1147. 10.1017/s0954579414000613 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Civic D & Holt VL (2000). Maternal depressive symptoms and child behavior problems in a nationally representative normal birthweight sample. Maternal and Child Health Journal, 4, 215–221. 10.1023/A:1026667720478 [DOI] [PubMed] [Google Scholar]
  9. Cohn JF, Campbell SB, Matias R, & Hopkins J (1990). Face-to-face interactions of postpartum depressed and nondepressed mother-infant pairs at 2 months. Developmental Psychology, 26(1), 15–23. 10.1037/0012-1649.26.1.15 [DOI] [Google Scholar]
  10. Cornelius MD, & Day NL (2009). Developmental consequences of prenatal tobacco exposure. Current opinion in neurology, 22(2), 121–125. 10.1097/WCO.0b013e328326f6dc [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Eiden RD, Homish GG, Colder CR, Schuetze P, Gray T, & Huestis M (2013). Changes in smoking patterns during pregnancy. Substance use and misuse, 48, 513–522. 10.3109/10826084.2013.787091 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Eiden RD, Schuetze P, & Coles CD (2011). Maternal cocaine use and mother-infant interactions: Direct and moderated associations. Neurotoxicology and Teratology, 33(1), 120–128. 10.1016/j.ntt.2010.08.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Eiden RD, Zhao J, Casey M, Shisler S, Schuetze P, & Colder C (2018). Pre- and postnatal tobacco and cannabis exposure and child behavior problems: Bidirectional associations, joint effects, and sex differences. Drug and Alcohol Dependency, 185, 82–92. 10.1016/j.drugalcdep.2017.11.038 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Elgar FJ, Curtis LJ, McGrath PJ, Waschbusch DA, & Stewart SH (2003). Antecedent-consequence conditions in maternal mood and child adjustment: A four-year cross-lagged study. Journal of Clinical Child and Adolescent Psychology, 32(3), 362–374. 10.1207/s15374424jccp3203_05 [DOI] [PubMed] [Google Scholar]
  15. Evans GW, Li D, & Whipple SS (2013). Cumulative risk and child development. Psychological Bulletin, 139(6), 1342–1396. DOI: 10.1037/a0031808. [DOI] [PubMed] [Google Scholar]
  16. Field T (2010). Postpartum depression effects on early interactions, parenting, and safety practices: A review. Infant Behavior & Development, 33(1), 1–9. DOI: : 10.1016/j.infbeh.2009.10.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gelave B, Rondon M, Araya R, & Williams M (2016). Epidemiology of maternal depression, risk factors, and child outcomes in low-income and middle-income countries. Lancet Psychiatriy, 3(10), 973–982. DOI: 10.1016/S2215-0366(16)30284-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Goodman SH, Rouse MH, Connell AM, Robbins Broth M, Hall CM, & Heyward D (2011). Maternal depression and child psychopathology: A meta-analytic review. Clinical Child Family Psychological Review, 14(1), 1–27. 10.1007/s10567-010-0080-1 [DOI] [PubMed] [Google Scholar]
  19. Graham CA, & Easterbrooks MA (2000). School-aged children’s vulnerability to depressive symptomatology: The role of attachment security, maternal depressive symptomatology, and economic risk. Developmental Psychopathology, 12(2), 201–213. 10.1017/s0954579400002054 [DOI] [PubMed] [Google Scholar]
  20. Graham KA, Blissett J, Antoniou EE, Zeegers MP, & McCleery JP (2018). Effects of maternal depression in the still-face paradigm: A meta-analysis. Infant Behavior & Development, 50, 154–164. 10.1016/j.infbeh.2017.12.001 [DOI] [PubMed] [Google Scholar]
  21. Gruber J, Kogan A, Quoidbach J, & Maus IB (2013). Happiness is best kept stable: Positive emotion variability is associated with poorer psychological health. Emotion, 13(1), 1–6. 10.1037/a0030262 [DOI] [PubMed] [Google Scholar]
  22. Hahn-Holbrook J, Haselton MG, Schetter CD, & Glynn LM (2013). Does breastfeeding offer protection against maternal depressive symptomatology?. Archives of women’s mental health, 16(5), 411–422. 10.1007/s00737-013-0348-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hamaker EL, Kuiper RM, & Grasman RPPP (2015). A critique of the cross-lagged panel model. Psychological Methods, 20(1), 102–116. 10.1037/a0038889. [DOI] [PubMed] [Google Scholar]
  24. Hoffman C, Dunn DM, & Njoroge WFM, (2017). Impact of postpartum mental illness upon infant development. Current Psychiatry Report, 6, 100–111. 10.1007/s11920-017-0857-8 [DOI] [PubMed] [Google Scholar]
  25. Hollingshead AB (1975). Four factor index of social status. Unpublished manuscript. Yale University, New Haven, CT. [Google Scholar]
  26. Hu L, & Bentler PM (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. 10.1080/10705519909540118 [DOI] [Google Scholar]
  27. Hummel AC, Kiel EJ, & Zvirblyte S (2016). Bidirectional effects of positive affect, warmth, and interactions between mothers with and without symptoms of depression and their toddlers. Journal of child and family studies, 25(3), 781–789. 10.1007/s10826-015-0272-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Joormann J & Gotlib IH (2010). Emotion regulation in expression: Relation to cognitive inhibition. Cognition and Emotion, 24(2), 281–298. 10.1080/02699930903407948 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Keijsers L (2015). Parental monitoring and adolescent problem behaviors: How much do we really know? International Journal of Behavioral Development, 40(3), 10.1177/0165025415592515. [DOI] [Google Scholar]
  30. Kuckertz JM, Mitchell C, & Wiggins JL (2018). Parenting mediates the impact of maternal depression on child internalizing symptoms. Depression and anxiety, 35(1), 89–97. 10.1002/da.22688 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lohaus A, Rueth JE, & Vierhaus M (2020). Cross-informant discrepancies and their association with maternal depression, maternal parenting stress, and mother-child relationship. Journal of Child and Family Studies, 29(3), 867–879. 10.1007/s10826-019-01625-z [DOI] [Google Scholar]
  32. Ludman EJ, McBride CM, Nelson JC, Curry SJ, Grothaus LC, Lando HA, & Pirie PL (2000). Stress, depressive symptoms, and smoking cessation among pregnant women. Health Psychology, 13, 149–155. 10.1037//0278-6133.19.1.21 [DOI] [PubMed] [Google Scholar]
  33. Mackler JS, Kelleher RT, Shanahan L, Calkins SD, Keane SP, & O’Brien M (2015). Parenting Stress, Parental Reactions, and Externalizing Behavior from Ages 4 to 10. Journal of Marriage and Family, 77, 388–406. 10.1111/jomf.12163 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Masten AS, & Cicchetti D (2010). Developmental cascades. Development and Psychopathology, 22, 491–495. 10.1017/S0954579410000222 [DOI] [PubMed] [Google Scholar]
  35. McMurray B (2007). Defusing the childhood vocabulary explosion. Science, 317, 631–637. 10.1126/science.1144073 [DOI] [PubMed] [Google Scholar]
  36. Morelen D, Menke R, Rosenblum KL, Beeghly M, & Muzik M (2016). Understanding bidirectional mother-infant affective displays across contexts: Effects of maternal maltreatment history and postpartum depression and PTSD symptoms. Psychopathology, 49(4), 305–314. 10.1159/000448376 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Morozova M, Rabin RA, George TP (2015). Co-morbid tobacco use disorder and depression: A re-evaluation of smoking cessation therapy in depressed smokers. The American Journal on Addictions, 24(8), 687–694. 10.1111/ajad.12277 [DOI] [PubMed] [Google Scholar]
  38. Munafò MR, Heron Jon, & Araya R (2008). Smoking patterns during pregnancy and postnatal period and depressive symptoms. Nicotine and Tobacco Research, 10, 1620. 10.1080/14622200802412895 [DOI] [PubMed] [Google Scholar]
  39. Narayanan MK, & Nærde A (2016). Associations between maternal and paternal depressive symptoms and early child behavior problems: Testing a mutually adjusted prospective longitudinal model. Journal of affective disorders, 196, 181–189. 10.1016/j.jad.2016.02.020 [DOI] [PubMed] [Google Scholar]
  40. National Research Council (US) and Institute of Medicine (US) Committee on Depression, Parenting Practices, and the Healthy Development of Children; England MJ, Sim LJ, editors. Depression in Parents, Parenting, and Children: Opportunities to Improve Identification, Treatment, and Prevention. Washington (DC): National Academies Press (US); 2009. 4, Associations Between Depression in Parents and Parenting, Child Health, and Child Psychological Functioning. Available from: https://www.ncbi.nlm.nih.gov/books/NBK215128/ [PubMed] [Google Scholar]
  41. Neill EL, Weems CF, & Scheeringa MS (2018). CBT for child PTSD is associated with reductions in maternal depression: Evidence for bidirectional effects. Journal of Clinical Child & Adolescent Psychology, 47(3), 410–420. 10.1080/15374416.2016.1212359 [DOI] [PubMed] [Google Scholar]
  42. Park H, Sundaram R, Gilman SE, Bell G, Louis GMB, & Yeung EH (2018). Timing of maternal depression and sex-specific child growth, the upstate KIDS study. Obesity, 26(1), 160–166. 10.1002/oby.22039 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Patterson GR & Reid JB (1970). Reciprocity and coercion: Two facets of social systems. In: Neuringer C, Michael J, editors. Behavior modification in clinical psychology. New York: Appleton-Century-Crofts. [Google Scholar]
  44. Pearson RM, Bornstein MH, Cordero M, Scerif G, Mahedy L, Evans J,...Stein A (2016). Maternal perinatal mental health and offspring academic achievement at age 16: The mediating role of childhood executive function. Journal of Child Psychology and Psychiatry, 57(4), 491–501. 10.1111/jcpp.12483 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Perry NB, Mackler JS, Calkins SD, & Keane SP (2014). A transactional analysis of the relation between maternal sensitivity and child vagal regulation. Developmental Psychology, 50(3), 784–793. 10.1037/a0033819 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Putnam SP, Sanson AV, & Rothbart MK (2002). Child temperament and parenting. In Bornstein MH (Ed.), Handbook of parenting: Children and parenting (p. 255–277). Lawrence Eribaum Associates Publishers. [Google Scholar]
  47. Roben KP, Moore GA, Cole PM, Molenaar P, Leve LD, Shaw DS, Reiss D & Neiderhiser JM (2015). Transactional patterns of maternal depressive symptoms and mother-child mutual negativity in an adoption sample. Infant Child Dev., 24(3), 322–342. 10.1002/icd.1906 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Rutherford HJV, & Mayes LC (2017). Parenting and addiction: Neurobiological insights. Current opinion in psychology, 15, 55–60. 10.1016/j.copsyc.2017.02.014/ [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Sameroff AJ, & Mackenzie MJ (2003). Research strategies for capturing transactional models of development: The limits of the possible. Development and Psychopathology, 15, 613–640. 10.1017/s0954579403000312 [DOI] [PubMed] [Google Scholar]
  50. Savory K, Garay SM, Sumption LA, Kelleher JS, Daughters K, Janssen AB, Van Goozen S, & John RM (2020). Prenatal symptoms of anxiety and depression associated with sex differences in both maternal perceptions of one year old infant temperament and researcher observed infant characteristics. Journal of Affective Disorders, 264, 383–392. 10.1016/j.jad.2019.11.057 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Schuetze P, Eiden RD, Leornard KE, Huestis MA, & Colder CR (2018). Prenatal risk and infant regulation: Indirect pathways via fetal growth and maternal prenatal stress and anger. Child Development, 89(2), 123–137. 10.1111/cdev.12801 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Scott JK, Nelson JA, & Dix T (2018). Interdependence among mothers, fathers, and children from early to middle childhood: Parents’ sensitivity and children’s externalizing behavior. Developmental Psychology, 54(8), 1528–1541. 10.1037/dev0000525 [DOI] [PubMed] [Google Scholar]
  53. Shisler S, Eiden RD, Molnar DS, Schuetze P, Huestis M, & Homish G (2017). Smoking in pregnancy and fetal growth: The case for more intensive assessment. Nicotine & Tobacco Research, 19(5), 525–531. 10.1093/ntr/ntx018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Shisler S, Homish G, Molnar DS, & Eiden RD (2016). Predictors of changes in smoking from 3rd trimester to 9 months post-partum. Nicotine and Tobacco Research, 18, 1–4. 10.1093/ntr/ntv057 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Strathearn L, Li J, Fonagy P, & Montague PR (2008). What’s in a smile? Maternal brain responses to infant facial cues. Pediatrics, 122(1), https://doi.org/40-51.10.1542/peds.2007-1566 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Sweeney S, & MacBeth A (2016). The effects of paternal depression on child and adolescent outcomes: a systematic review. Journal of Affective Disorders, 205, 44–59. 10.1016/j.jad.2016.05.073 [DOI] [PubMed] [Google Scholar]
  57. Taraban L, Feldman JS, Wilson MN, Dishion TJ, & Shaw DS (2020). Sad dads and troubled tots: protective factors related to the stability of paternal depression and early childhood internalizing problems. Journal of abnormal child psychology, 1–15. 10.1007/s10802-020-00649-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Taraban L, & Shaw DS (2018). Parenting in context: Revisiting Belsky’s classic process of parenting model in early childhood. Developmental Review, 48, 55–81. 10.1016/j.dr.2018.03.006 [DOI] [Google Scholar]
  59. Wiggins JL, Mitchell C, Stringaris A, & Leibenluft E (2014). Developmental trajectories of irritability and bidirectional associations with maternal depression. Journal of the American Academy of Child & Adolescent Psychiatry, 53(11), 1191–1205. 10.1016/j.jaac.2014.08.005 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES