The Association Between Maternal Depressive Symptoms and Toddlers’ Developmental and Behavioral Problems: A Population-Based Study

Abstract

This study aims to compare the developmental-behavioral profiles of 2-year-olds of mothers who experienced postpartum and/or current depression with profiles of toddlers of mothers without depression at either time using population-based Rhode Island data. Weighted data from Rhode Island Department of Health’s Pregnancy Risk Assessment Monitoring System and Rhode Island’s follow-up Toddlers Wellness Overview Survey distributed to mothers giving birth between 2006 and 2008 were analyzed. Compared with non-depressed mothers, those with any depression following childbirth reported more concerns with their toddlers’ receptive language, social-emotional development, and their sleep and feeding behaviors. When adjusted for demographics, persistent depression remained associated with social-emotional (adjusted odds ratio [aOR] = 7.53, 2.78-20.34) and feeding concerns (aOR = 3.13, 1.36-7.22), and current depression was associated with social-emotional concerns (aOR = 2.52, 1.26-5.01). We conclude that pediatric providers should explore maternal mental health as a mediating and potentially modifiable factor beyond the postpartum period when toddlers present with developmental-behavioral challenges.

Keywords

maternal depression toddler development behavior population-based study

Key Findings

This population-based study shows mothers with depression following childbirth reported more concerns with their toddlers’ receptive language, social-emotional development, and their sleep and feeding behaviors. When adjusted, maternal depression is associated with more social-emotional and feeding concerns in 2-year-olds in Rhode Island’s diverse population. Persistently depressed mothers have the highest worries in these domains.

Introduction

Maternal depression has been linked to adverse outcomes in children’s social-emotional, behavioral, language, and cognitive development.^1-5 Recurrent symptoms of maternal depression have been associated with increased externalizing, internalizing, and total behavioral problems at kindergarten entry.² However, prior studies had small sample size, and population-based studies are limited.³,^6-8 Few studies have investigated the longitudinal relationship between maternal depressive symptoms and children’s behavior and development, including language, adaptive and regulatory skills, and potential impact of maternal depression on toddlers.^9-13

In an effort to mitigate these adverse outcomes, early identification of social-emotional, cognitive, and regulatory concerns in early childhood is of utmost importance. Using population-based survey data, the goal of this study is to investigate the developmental-behavioral (DB) profiles of 2-year-olds born to mothers with postpartum depression (PPD) and/or current depression (CD) as compared with children whose mothers did not endorse depression. We also aim to examine the association between persistent maternal depression and DB concerns among toddlers in the following domains: social-emotional, language, sleep, and feeding. We hypothesize that maternal depression is independently linked to toddler DB concerns.

Methods

Data Sources

The 2006-2008 Rhode Island (RI) Pregnancy Risk Assessment Monitoring System (PRAMS) and 2008-2010 Toddler Wellness Overview Survey (TWOS) linked data from RI Department of Health (RIDOH) were analyzed for this study.

Pregnancy Risk Assessment Monitoring System

The PRAMS is a surveillance project of the Centers for Disease Control and Prevention (CDC) and state health departments that collect state-specific, population-based data on maternal behaviors and experiences before, during, and shortly after pregnancy (https://www.cdc.gov/prams/). Each year, about 10 200 babies are born in RI, and approximately 1900 new mothers are randomly selected from the state’s birth file as the PRAMS sample.¹⁴ Total respondents for 2006-2008 PRAMS were 4029 with a weighted response rate between 70.4% and 72.5% per year. Rhode Island PRAMS selects all mothers delivering a low–birth weight baby (<2500 grams), and over-samples mothers who reside in cities with more than 15% of children living below the federal poverty line to ensure adequate data in these smaller but higher-risk populations.¹⁵ The survey is initiated 2 months postpartum and is available in both English and Spanish. The survey is mailed to mothers up to 3 times. Phone follow-ups are made for non-respondents. Data are weighted to reflect all women giving birth in RI during the collected year.

Toddler Wellness Overview Survey

Rhode Island TWOS was RI’s state-specific follow-up survey to PRAMS. The 2008-2010 TWOSs had a response rate of 39% representing 1226 respondents. Mothers who completed PRAMS surveys were asked if they would consent to be re-contacted at their child’s second birthday to participate in an additional survey. At 2 years of age, TWOS questionnaires were mailed along with a $5.00 incentive gift card to consenting mothers. If no response was received, a second questionnaire was mailed. Data were weighted to represent all women giving birth each cohort year. The TWOS queried mothers regarding many developmental milestones and behavioral concerns. The TWOS was discontinued by RIDOH, and data are only available through 2010 due to limited funding.

Measures/Operational Definitions

Depression categories

On PRAMS, self-reported PPD symptoms were assessed by 5 responses (“always,” “often,” “sometimes,” “rarely,” and “never”) to 2 questions: (1) “Since your new baby was born, how often have you felt down, depressed, or hopeless?” and (2) “Since your new baby was born, how often have you had little interest or little pleasure in doing things?” A response of “always” or “often” to either question was defined as positive for PPD (Figure 1).

In TWOS, maternal CD was assessed using 2 questions: (1) “Over the past 12 months, have you had 2 or more weeks in a row when you felt sad, blue, or depressed, or lost pleasure in things that you usually cared about or enjoyed?” and (2) “Over the past 12 months, has a doctor, nurse, or other health care worker diagnosed you with depression?” A response of “yes” to either question was defined as positive for CD.

Maternal depression was categorized into 4 groups for this analysis: (1) Never depressed (ND) mothers did not endorse depressive symptoms and chose the response options “never,” “rarely,” or “sometimes” in PRAMS and “no” in TWOS. (2) Postpartum depression only (PPD only) mothers only endorsed depressive symptoms in PRAMS but not in TWOS. (3) Current depression only (CD only) mothers endorsed depression in TWOS but not in PRAMS. (4) Mothers with depression at both time points (Both) endorsed depressive symptoms in both PRAMS and TWOS.

Developmental and behavioral profiles

Toddlers’ DB profiles were defined in 5 domains: Expressive Language, Receptive Language, Social-Emotional, Sleep, and Feeding. The TWOS asked mothers about various developmental milestones and behavioral concerns, with response options: “most of the time,” “sometimes,” and “rarely or never.” Mothers choosing one or more response options at either extreme (“rarely or never,” or “most of the time”) in the problematic direction for that DB category were defined as having concerns in the respective domain.

(1) Expressive language concern was present when toddlers “rarely/never” use 2-word phrases such as “Mommy come,” use at least 20 words, or people outside of the family understand at least half of what he or she says. (2) Receptive language issue was present when toddlers “rarely/never” point to familiar objects in picture books, point to at least 6 body parts, or follow 2-part directions such as “please pick up your toys and put them away.” (3) Social-Emotional difficulty was present when toddlers “rarely/never” like to be hugged or cuddled or settle himself or herself down after periods of exciting activity; or if they cry a lot or act aggressively when frustrated “most of the time.” (4) Sleep concern was present when toddlers “rarely/never” follow a bedtime routine or have a regular bedtime; or if they fall asleep with a bottle or cup, have trouble falling asleep, or wake up at night and need help to get back to sleep “most of the time.” (5) Feeding problem was present when toddlers “rarely/never” sit for meals at regular times, or if they drink from a bottle or refuse to eat “most of the time.” Each domain was coded as a binary variable (concern present/absent).

Covariates

Aligning with previous research, PRAMS demographic factors included in this analysis were maternal age, annual household income, maternal educational attainment, ethnicity, race, marital status at birth, child sex, and birth weight given their associations with PPD and/or DB concerns.^2,3,7,8,^16-19

Data analysis

Analyses were conducted in STATA (Version 11.2; Stata Corp, College Station, TX). Chi-square was used to examine associations between PRAMS demographic variables, TWOS toddlers’ DB profiles, and maternal depression status. Multivariable regression models for concern in each DB domain were generated and adjusted for demographic covariates. In all regression models, 3 minority groups (non-Hispanic black [NHB], Hispanic, and non-Hispanic Others) were collapsed into “Others” to account for small proportions in these categories. Maternal age and maternal education were dichotomized at 24 years and high school graduation for the same reason. Annual household income was dichotomized into <$10 000 and ≥$10 000 to capture the impact of extreme poverty.

Results

Demographics

Data were available from 1226 mothers who responded to both PRAMS (2006-2008) and TWOS (2008-2010). Mothers included in the analysis were more likely to be non-Hispanic white (NHW), between the age of 25 and 34 years, have some post–high school education, and be married. Of the respondents, 12.8% were mothers living in extreme poverty with an annual household income less than $10 000. An estimated 7.1% of infants were low birth weight, and half were male sex (Table 1).

Table 1.

Weighted Proportions of Mother and Infant Characteristics by Self-Reported Depression Status (RI PRAMS 2006-2008 and TWOS 2008-2010, n = 1226), Row Percentage.

	Never depressed (n = 854, 68.0%)	PPD only (n = 56, 5.8%)	CD only (n = 244, 19.2%)	Both (n = 72, 7.0%)
Maternal demographics (at birth)
Age, y
<20 (5.9%)	58.8%	8.2%	19.2%	13.8%
20-24 (18.9%)	58.3%	8.9%	21.6%	11.2%
25-34 (58.1%)	69.6%	5.1%	19.7%	5.6%
≥35 (17.1%)	76.1%	3.9%	15.4%	4.6%
P				NS
Race/ethnicity
NHW (72.1%)	70.7%	5.6%	18.7%	5.0%
NHB (5.8%)	60.5%	3.2%	33.0%	3.3%
H (16.3%)	63.5%	7.0%	18.0%	11.5%
NH Others (5.8%)	55.1%	8.2%	18.0%	18.7%
P				.024*
Education, y
<12 (12.2%)	58.4%	8.4%	24.8%	8.4%
12 (32.3%)	58.9%	8.0%	20.8%	12.3%
>12 (55.5%)	75.8%	4.1%	16.7%	3.4%
P				.001**
Marital status
Married (58.7%)	75.6%	3.9%	17.8%	2.7%
No (41.3%)	57.1%	8.4%	21.5%	13.0%
P				<.001***
Annual household income, $
<10 000 (12.8%)	46.2%	15.9%	20.7%	17.2%
10 000-24 999 (22.3%)	55.7%	9.0%	22.2%	13.1%
≥25 000 (64.9%)	75.0%	3.0%	18.8%	3.2%
P				<.001***
Infant demographics
Birth weight
<2500 (7.1%)	49.7%	6.4%	34.5%	9.4%
≥2500 (92.9%)	69.4%	5.7%	18.1%	6.8%
P				<.001***
Infant sex
Male (50.1%)	68.1%	5.1%	19.2%	7.6%
Female (49.9%)	67.9%	6.4%	19.4%	6.3%
P				NS

The χ2 analysis of mother and infant characteristics by maternal mental health status. Bolded % indicates the highest proportion of participants in each demographic category for each depression status group.

Abbreviations: RI, Rhode Island; PRAMS, Pregnancy Risk Assessment Monitoring System; TWOS, Toddler Wellness Overview Survey; PPD, postpartum depression; CD, current depression; NS, not significant; NHW, non-Hispanic white; NHB, non-Hispanic black; NH, non-Hispanic.

P ≤ .05. **P ≤ .01. ***P ≤ .001.

Depression groups

As illustrated in Figure 1, 12.8% of mothers reported PPD symptoms in PRAMS, and 26.2% noted depressive symptoms for more than 2 weeks or were diagnosed with depression in the year prior to their toddlers’ second birthday. Mothers were categorized into 4 depression groups as described previously: 68.0% were ND, 5.8% had PPD only, 19.2% had CD only, and 7.0% had depression at both time points (Both).

Bivariate analysis of weighted demographic data and maternal depression groups demonstrated that maternal age and infant sex were not linked to specific maternal depression categories (Table 1). Compared with their counterparts, mothers with depression at either time point were more likely to be self-identified as minority, have lower educational attainment, be unmarried, have an annual household income <$25 000, and to have delivered an infant <2500 grams. The NHB mothers were over-represented in the CD group (33.0%), while the non-Hispanic Others mothers were over-represented in PPD only and Both groups (8.2% and 18.7%, respectively).

Toddler DB profiles

In the TWOS, 41% of mothers expressed at least 1 concern about their toddlers’ development or behavior. Sleep difficulty was the highest reported concern with 18.9% of mothers reporting sleep issues. The DB domains in toddlers that most closely align with later academic success and cognition are receptive language and social-emotional development. Concern in these areas were endorsed among 6.4% and 11.6% of mothers, respectively (Table 2).

Table 2.

Weighted Proportions of Mother and Infant Characteristics by Self-Reported DB Domain (RI PRAMS 2006-2008 and TWOS 2008-2010, n = 1226).

	Expressive language (n = 182, 13.1%)	Receptive language (n = 78, 6.4%)	Social-emotional (n = 127, 11.6%)	Sleep (n = 220, 18.9%)	Feeding (n = 197, 16.4%)
Maternal demographics (at birth)
Age, y
<20 (5.9%)	19.0%	1.0%	32.2%	25.9%	18.8%
20-24 (18.9%)	14.5%	10.1%	16.3%	27.6%	18.8%
25-34 (58.1%)	13.1%	5.3%	10.3%	16.2%	13.4%
≥35 (17.1%)	10.9%	7.5%	4.2%	16.7%	21.3%
P	NS	NS	<.001***	.061	NS
Race/ethnicity
NHW (72.1%)	11.6%	3.2%	7.6%	17.0%	13.5%
NHB (5.8%)	9.2%	4.5%	23.0%	8.6%	6.4%
H (16.3%)	17.0%	15.0%	19.8%	28.7%	22.7%
NH Others (5.8%)	30.0%	21.3%	27.4%	26.8%	38.6%
P	.013*	<.001***	<.001***	.011*	<.001***
Education, y
<12 (12.2%)	18.1%	14.5%	28.9%	33.7%	18.1%
12 (32.3%)	15.7%	6.7%	14.4%	23.0%	18.5%
>12 (55.5%)	11.2%	4.4%	6.4%	14.0%	14.6%
P	NS	.013*	<.001***	.001***	NS
Marital status
Married (58.7%)	13.0%	4.6%	7.5%	14.7%	13.6%
No (41.3%)	13.9%	8.7%	17.7%	25.0%	19.6%
P	NS	.035*	<.001***	.002**	.047*
Annual household income, $
<10 000 (12.8%)	22.7%	17.7%	27.7%	40.0%	23.4%
10 000-24 999 (22.3%)	15.4%	7.2%	18.0%	20.1%	20.4%
≥25 000 (64.9%)	10.4%	3.0%	5.4%	13.8%	12.8%
P	.017*	<.001***	<.001***	<.001***	.033*
Infant demographics
Birth weight
<2500 (7.1%)	18.6%	10.9%	21.6%	25.4%	20.8%
≥2500 (92.9%)	13.0%	5.9%	10.9%	18.5%	15.7%
P	.029*	.014*	<.001***	.025*	NS
Infant sex
Male (50.1%)	18.5%	9.9%	14.4%	16.4%	16.7%
Female (49.9%)	8.2%	2.7%	9.0%	21.5%	15.5%
P	<.001***	<.001***	NS	NS	NS

The χ² analysis of mother and infant characteristics by each DB domain. Bolded % indicates the highest proportion of participants in each DB domain for each demographic variable.

Abbreviations: RI, Rhode Island; PRAMS, Pregnancy Risk Assessment Monitoring System; NS, not significant; NHW, non-Hispanic white; NHB, non-Hispanic black; NH, non-Hispanic; DB, developmental-behavioral.

P ≤ .05. **P ≤ .01. ***P ≤ .001.

In bivariate analysis, younger mothers reported the most concerns regarding their toddlers’ social-emotional development (Table 2). Maternal race/ethnicity and family income were significantly associated with DB concerns in all 5 categories. Mothers who were unmarried or had less than a high school education expressed more concerns regarding their toddlers’ receptive language, social-emotional behaviors, and sleep habits, as were mothers with income <$10 000 and those with infants weighing <2500 grams at birth. Unmarried mothers also endorsed more feeding problems. Male sex was linked to increased language delay.

Overall, mothers with depression expressed more concerns around their toddlers’ development and behavior compared with non-depressed mothers (Table 3). Within the 3 depression groups, mothers depressed at both time points disproportionately expressed more concern in 6 of these 17 toddler DB indicators/questions.

Table 3.

Weighted Proportions of Toddler DB Concerns by Maternal Depression Status (RI PRAMS 2006-2008 and TWOS 2008-2010, n = 1226).

	ND (68.0%)	PPD (5.8%)	CD (19.2%)	Both (7.0%)	Total (n = 1213)	P value
Expressive language concern
Rarely uses 2-word phrases	6.0%	12.6%	5.5%	11.4%	6.6%	NS
Rarely has speech intelligibility >50%	5.3%	2.1%	6.4%	18.1%	6.3%	.004**
Rarely uses at least 20 words	8.5%	11.1%	12.1%	16.7%	9.9%	NS
Any expressive language concern	11.0%	15.5%	16.3%	22.6%	13.1%	NS
Receptive language concern
Rarely points to pictures in books	1.5%	0.0%	0.4%	7.3%	1.7%	NS
Rarely points 6 body parts	3.8%	10.0%	6.5%	9.5%	5.1%	NS
Rarely follows 2-step directions	2.5%	5.5%	3.4%	10.9%	3.4%	NS
Any receptive language concern	4.8%	15.2%	6.7%	14.3%	6.4%	.037*
Social/emotional concern
Rarely likes to be cuddled or hugged	0.2%	0.0%	0.0%	0.0%	0.2%	NS
Rarely settles self down after period of exciting activity	1.4%	1.5%	3.0%	6.3%	2.1%	NS
Cries a lot most of the time	1.2%	3.2%	4.3%	12.2%	2.6%	<.001***
Acts aggressively when frustrated most of the time	4.8%	14.1%	14.9%	31.6%	9.2%	<.001***
Any social/emotional concern	7.2%	18.2%	15.7%	38.3%	11.6%	<.001***
Sleep concern
Falls asleep with a bottle or a cup most of the time	10.6%	28.6%	15.4%	16.0%	12.9%	.045*
Rarely or never follows a bedtime routine or has a regular bedtime	1.1%	1.3%	5.3%	9.3%	2.5%	.003**
Has trouble falling asleep most of the time	1.0%	6.6%	6.0%	9.9%	2.9%	.007**
Wakes up at night and needs help to get back to sleep most of the time	4.3%	8.9%	6.7%	12.8%	5.6%	NS
Any sleep concern	15.8%	35.5%	23.9%	22.3%	18.9%	.027*
Feeding concern
Drinks from a bottle most of the time	12.6%	22.8%	12.1%	25.4%	14.0%	NS
Rarely or never sits for meals at regular time	0.9%	0.2%	1.0%	4.6%	1.1%	NS
Refuses to eat most of the time	1.1%	6.1%	5.1%	15.1%	3.2%	<.001***
Any feeding concern	13.9%	22.8%	16.1%	36.1%	16.4%	.004**

The χ² analysis of toddler DB concern items by maternal depression status. Shaded rows are toddler DB concerns significantly related to maternal depression; bolded number is the highest % in maternal depression groups.

Abbreviations: DB, developmental-behavioral; RI, Rhode Island; PRAMS, Pregnancy Risk Assessment Monitoring System; ND, never depressed; PPD, postpartum depression; CD, current depression; NS, not significant.

P ≤ .05. **P ≤ .01. ***P ≤ .001.

In multivariable analyses that included maternal and infant demographic factors as well as maternal depression groups (Table 4), male sex was the strongest predictor of expressive language delay with male children having 2.6 times the odds of having difficulty with expressive language. In a similar regression analysis for receptive language, male children had 3.9 times the odds of having receptive language problems. Similarly, children in extreme poverty had 5 times the odds, and toddlers with minority mothers had 3.5 times the odds of having receptive language problems. Mothers’ depression status was not significantly associated with toddlers’ expressive or receptive language when models were adjusted for the socio-demographic variables.

Table 4.

Multivariable Regressions of Each Toddler DB Domain With Demographics and Maternal Depression Status (RI PRAMS 2006-2008 and TWOS 2008-2010, n = 1226).

	Expressive language (13.1%)	Receptive language (6.4%)	Social-emotional (11.6%)	Sleep (18.9%)	Feeding (16.4%)
	aOR (95% CI)	aOR (95% CI)	aOR (95% CI)	aOR (95% CI)	aOR (95% CI)
Maternal age <24 years	1.22 (0.65-2.28)	0.84 (0.35-1.99)	1.90 (1.04-3.47)*	1.54 (0.89-2.69)	1.09 (0.61-1.96)
Minority (NHB, Hispanic, Others)	1.20 (0.66-2.17)	3.52 (1.61-7.69)**	1.97 (0.99-3.93)	1.11 (0.64-1.94)	1.67 (0.97-2.86)
Maternal education less than high school	0.75 (0.27-2.09)	0.60 (0.20-1.84)	1.54 (0.67-3.55)	1.16 (0.50-2.72)	0.66 (0.25-1.73)
Annual household income < 10 000	1.92 (0.81-4.55)	5.05 (1.88-13.57)***	1.65 (0.76-3.55)	2.37 (1.13-4.97)*	1.55 (0.67-3.54)
Birth weight < 2500 grams	1.37 (0.87-2.17)	1.33 (0.67-2.65)	1.84 (1.09-3.11)*	1.24 (0.81-1.89)	1.08 (0.70-1.67)
Male sex	2.57 (1.55-4.27)***	3.94 (1.66-9.33)**	1.90 (1.04-3.48)*	0.69 (0.45-1.07)	1.00 (0.64-1.56)
Never depressed	ref	ref	ref	ref	ref
Postpartum depression only	1.36 (0.46-4.08)	3.14 (0.80-12.42)	2.59 (0.83-8.03)	2.03 (0.86-4.77)	1.55 (0.60-4.01)
Current depression only	1.62 (0.91-2.87)	1.40 (0.56-3.46)	2.52 (1.26-5.01)**	1.58 (0.94-2.66)	1.23 (0.70-2.17)
Depressed at both time points	2.07 (0.88-4.86)	2.19 (0.69-6.97)	7.53 (2.78-20.34)***	1.19 (0.46-3.10)	3.13 (1.36-7.22)**

Bolded/shaded aOR is significant.

Abbreviations: DB, developmental-behavioral; RI, Rhode Island; PRAMS, Pregnancy Risk Assessment Monitoring System; aOR, adjusted odds ratio; CI, confidence interval; NHB, non-Hispanic black; ref, reference group.

P ≤ .05. ** P ≤ .01. *** P ≤ .001.

In the regression model for social-emotional concerns, maternal depression was a strong predictor of social-emotional difficulty. In this model, mothers who were depressed at both time points had more than 7.5 times the odds of having concern for their toddlers’ social-emotional development. In addition, mothers currently depressed had 2.5 times the odds of having this concern. Male sex, low birth weight, and younger maternal age also had significant associations with social-emotional concerns.

In the similar regression model for sleep issues, only children in extreme poverty were significantly associated with sleep difficulties with 2.4 times the odds of having this issue. Maternal depression was not significantly associated with sleep problems in this adjusted model. In the adjusted model for feeding challenges, mothers depressed at both time points had more than 3 times the odds of having concern for their toddlers’ feeding habits. Demographic variables were not linked to feeding problems in this model.

Discussion

This is the first population-based study investigating the relationship between early maternal depression and concerns regarding toddlers’ language, social-emotional, sleep, and feeding in a single longitudinal cohort. Previously published population-based studies mostly focused on social-emotional behaviors among toddlers and preschoolers.^3,8 Junge et al reviewed maternal depression at 32 weeks during pregnancy, 8 weeks postpartum, and 2 years postpartum. Data from these 1235 Norwegian dyads showed that perinatal maternal depression was strongly associated with social-emotional problems in 2-year-olds, while CD at 2 years postpartum was not.⁸ Their findings differ from ours; however, we did see that mothers with persistent depression were 7.5 times more likely to report social-emotional difficulties in their toddlers. Their study showed that preterm birth, male sex, and poor child health were associated with more social-emotional concerns.⁸ Our analysis also revealed a similar association between social-emotional challenges and both male sex and low–birth weight infants. While their research used The Ages and Stages Questionnaire: Social-Emotional, which screened more broadly for social-emotional difficulties, our study looked beyond this domain and included assessments of toddlers’ language-related, sleep-related, and feeding-related concerns.

Another population-based study surveyed 5303 mothers of infants born in the late 1980s; however, it only assessed social-emotional behaviors in preschoolers.³ They showed that mothers with depressive symptoms at a mean of 17 and 36 months postpartum had significantly greater risk of reporting their preschoolers as having social-emotional problems. This association was greatest for mothers reporting depression at both time points. We noticed similar associations between both current and persistent maternal depression and toddler social-emotional concerns. However, our PPD only mothers did not acknowledge significantly more social-emotional challenges in their toddlers. Our mothers were surveyed during the toddlers’ first 2 to 6 months of life, and as such, our PPD group had symptoms slightly earlier than those in this larger cohort. Also, their study excluded children with low birth weight or congenital anomalies, which may have decreased its external validity.³ In contrast, our study oversampled infants born in centers of poverty and those of low birth weight, making our cohort more demographically representative. These factors may help explain the differences between these studies.

A more recent study examined the connection between maternal depressive symptoms at 4 antenatal time points with development delay in language, cognitive/adaptive, motor, and social-emotional domains at age 2.⁷ This work retrospectively surveyed 2679 Los Angeles County mothers and demonstrated a dose-response relationship between maternal depressive symptoms and presence of social-emotional delay or any developmental delay in toddlers. In our work, we sampled mothers prospectively at only 2 time points. Although we did not see a link between PPD only mothers and their toddlers’ social-emotional development, we did demonstrate associations between current and persistent maternal depression and more social-emotional challenges in their toddlers. These authors noted that maternal depression was not associated with higher risk for language delay, while Hispanic mothers were nearly twice as likely to report language delay. Similar to their findings, our analysis did not identify a link between maternal depression and toddlers’ expressive/receptive language outcomes, and minority mothers reported more receptive language issues compared with NHW mothers. While their study examined a wide range of developmental outcomes, they did not explore toddlers’ sleep and feeding patterns, which are common parental concerns. They used 7 questions from the CDC’s “Learn the Signs. Act Early” Milestones Checklist as proxy for each domain, whereas our study used 17 questions to capture toddlers’ DB profiles. In addition, the cohort in this Los Angeles study was more reflective of an urban population and may be less generalizable than our state-wide cohort.

Other investigations with somewhat larger cohorts have concentrated on externalizing and internalizing behaviors and school readiness in older children.^16,20 A larger study of 52 103 Canadian maternal-child dyads used the Early Development Instrument (EDI) completed by kindergarten teachers to measure school readiness.²⁰ Maternal depression at any time point in their children’s life was associated with more difficulties in Physical Health and Well-Being, Social Competence, and Emotional Maturity domains on the EDI. This is consistent with our findings associating current and persistent maternal depression with difficulties in social-emotional and feeding domains in toddlers, long before kindergarten entry.

Concern around toddler’s behaviors is more common in our cohort than is concern for development, specifically language development. The most common domain of DB concern is toddler’s sleep (Table 3). Within this domain, toddlers of depressed mothers in all 3 groups had significantly more trouble falling asleep, needing a bottle to fall asleep, rarely following a bedtime routine, and with any sleep problem. This aligns with a previous study which found an association between perinatal depression and more frequent nighttime awakenings among 2-year-olds.¹³ However, in our adjusted model of any sleep concern, only family low-income status was significantly associated with toddler sleep difficulty. This differs from another cross-sectional study of 174 racially diverse mother-child dyads living in low-income households in which maternal depression independently predicted toddler difficulties in initiating or maintaining sleep.²¹ Our definition for low-income status is lower than Bates et al’s cohort, which may explain these differences.

The second most common domain of concern was around toddler’s feeding behaviors with 1 in 6 mothers acknowledging problem feeding. Toddlers of depressed mothers in all 3 groups had significantly more feeding refusal and any feeding problems than those of ND mothers. In our adjusted model, toddlers of mothers with persistent depression had more than 3 times the odds of having feeding challenges, compared with toddlers of ND mothers. Our findings build on previous research, which investigated the relationship between maternal mental health issues and regulatory problems (defined as issues with eating, sleeping, or excessive crying) among a large sample of Norwegian infants from birth to 11 months of age.²² Their assessments were performed by community health nurses, and a validated infant mental health screening tool was used to measure the presence or absence of problems related to feeding, sleeping, and fussiness. Olsen et al found that mothers with early mental health problems had more than 3 times the risk of their children demonstrating regulatory problems in late infancy. Increased odds for infant feeding or crying difficulties were noted in mothers with mental health problems. No association between maternal mental health and infant sleep problems was found. We observed a similar connection between maternal depression and toddlers’ social-emotional and feeding concerns. Our study advances their findings by more specifically identifying maternal mental health challenges and examining children’s DB profiles beyond the first year of life.²²

Our analysis did not identify a link between maternal depression and toddlers’ expressive/receptive language development. This differs from a previous Japanese population-based study which found that exposure to PPD may lead to a persistent decline in the rate of expressive language development in offspring during infancy and early childhood.¹⁰ While some studies reported similar connections between PPD and delayed language development and reduced cognitive scores in toddlers, others reported no significant reduction in expressive language at 6 and 12 months of age.^11,12,23

As hypothesized, maternal depression was found to be independently linked to toddler DB concerns in our cohort. Adjusted regression models suggested a strong association between maternal depression and behavioral/regulatory domains (Social-Emotional, Feeding) but not with developmental outcomes (Expressive/Receptive Language) among these toddlers. In particular, a strong link was seen between persistent maternal depression and Social-Emotional challenges in toddlers primarily around aggression and irritability (Table 3). This is consistent with several previous studies in toddlers and somewhat older children.^1-3,^{7,8,18,20,24,25} Our finding of a strong connection between persistent maternal depression and toddler feeding challenges is novel and has not been formally examined in other population-based studies to our knowledge.

Although the primary focus of this analysis is on the relationship between maternal depression and toddler DB outcomes, it is important to note the high prevalence of self-reported maternal depressive symptoms during toddlers’ second year of life. While 1 in 8 mothers reported early PPD symptoms, more than a quarter experienced depressive symptoms for more than 2 weeks or were diagnosed with depression in the year prior to their toddlers’ second birthday. A meta-analysis which investigated prevalence estimates of PPD globally and at different postpartum periods suggested an overall prevalence of 17% with increasing prevalence beyond 6 months postpartum.²⁶ In this meta-analysis, maternal depression prevalence was highest at 1 year postpartum (25%), closely followed by 7 to 12 months postpartum (20%), 4 to 6 months postpartum (16%), then 0 to 3 months (14%).²⁶ Other studies also observed a high prevalence of maternal depression up to 5 years after delivery (19.1%-29.6%).^3,7,20 While our study notes slightly lower PPD prevalence (12.8%), the prevalence for maternal depression (26.2%) during the toddlers’ second year of life is similar to global data. Our PPD prevalence is also consistent with overall national PPD average of 11.5%.²⁷

The limitations of this study include the relatively small sample size in each depression group; however, there are more than 1200 dyads in this study making it one of the largest population-based investigations to include both developmental and behavioral concerns among toddlers currently in the literature. Another limitation is the low response rate for the TWOS follow-up survey. We attempted to control for this by weighting data to reflect that of the birth cohort. In addition, all the data in the study were collected through parent report and are therefore susceptible to reporting bias. However, as childhood DB concerns rely mostly on caregiver report, our results should represent clinically significant parent perceptions.

Another limitation is the relatively older data set compared with recent studies. While updated data would reflect more recent maternal report of their emotional status around the time of their infant’s birth and subsequent toddler DB profiles, we were not able to expand our data source despite PRAMS being updated each year because follow-up TWOS data were only available during the time period used in our analysis. The TWOS was designed by the RIDOH and was fully funded by RIDOH as a follow-up to the CDC-funded PRAMS survey. The RIDOH was especially interested in toddler developmental and behavioral outcomes as they might be related to perinatal risks. With limited funding available for printing, mailing, data entry, weighting of data, and analysis, the TWOS was discontinued by RIDOH, and data are only available through 2010. Our study uses both PRAMS and the corresponding TWOS providing a longitudinal, population-based assessment of perinatal maternal mental health as it is associaed with toddler behavioral challenges in multiple domains. Only a few other states have tried to create follow-up surveys to PRAMS, and no other state has ever funded a follow-up survey with our toddler behavior questions. This makes our study unique and important as these outcomes have not been reported or described elsewhere.

We identified not only a high rate of depression among toddlers’ mothers but also a correlation between problematic toddler behaviors and mother’s mental health status. This study highlights the important role that pediatric medical homes may serve by screening for maternal depression beyond the postpartum period, which has been recommended by previous studies.^17,18,^28-30 Our findings further support the importance of exploring maternal mental health as a mediating and potentially modifiable factor when toddlers present with developmental or behavioral concerns.

Author Contributions

P-CW: Conceptualized the project, reviewed literatures, analyzed data, and prepared the manuscript. MADB: Contributed to literature review and preparation of the manuscript. HK: Contributed to project conceptualization and preparation of the manuscript. SV-B: Contributed to project conceptualization and preparation of the manuscript. PH: Contributed to project conceptualization, data analysis, and manuscript preparation. All authors discussed the results and contributed to the final manuscript.

Footnotes

Authors’ Note

Hanna Kim, PhD, has retired from Rhode Island Department of Health and currently resides in Seoul, South Korea.

Availability of Data and Material

The RI PRAMS-TWOS linked data are not publicly available but can be requested from the Center for Health Data Analysis at the RIDOH.

Code Availability

The STATA custom codes for this study are available on request.

Consent to Participate

The TWOS was obtained only from PRAMS participants who opted in for follow-up.

Consent for Publication

No elements of this study have been previously published in peer-reviewed journals.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Disclaimer

The views expressed herein are those of the authors and do not necessarily reflect the views of The Warren Alpert Medical School of Brown University/Rhode Island Hospital/Hasbro Children’s Hospital, and Rhode Island Department of Health.

Ethics Approval

The RI PRAMS and TWOS were reviewed by the Rhode Island Department of Health (RIDOH) IRB. The data were de-identified and publicly available.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Maternal and Child Health Bureau, Project T77MC09797-06-00, Brown University Leadership in Education in Developmental-Behavioral Pediatrics.

ORCID iD

Pei-Chi Wu

References

Luoma

Tamminen

Kaukonen

, et al. Longitudinal study of maternal depressive symptoms and child well-being. J Am Acad Child Adolesc Psychiatry. 2001;40(12):1367-1374. doi:10.1097/00004583-200112000-00006.

Giles

Davies

Whitrow

Warin

Moore

. Maternal depressive symptoms and child care during toddlerhood relate to child behavior at age 5 years. Pediatrics. 2011;128(1):e78-e84. doi:10.1542/peds.2010-3119.

Civic

Holt

. Maternal depressive symptoms and child behavior problems in a nationally representative normal birthweight sample. Matern Child Health J. 2000;4(4):215-221. doi:10.1023/a:1026667720478.

Camerota

McGowan

Aschner

, et al. Prenatal and perinatal factors associated with neonatal neurobehavioral profiles in the ECHO Program. Pediatr Res. Feb 25 2023. doi:10.1038/s41390-023-02540-2.

Liu

Bann

Lester

, et al. Neonatal neurobehavior predicts medical and behavioral outcome. Pediatrics. 2010;125(1):e90-e98. doi:10.1542/peds.2009-0204.

Smith-Nielsen

Tharner

Krogh

Vaever

. Effects of maternal postpartum depression in a well-resourced sample: early concurrent and long-term effects on infant cognitive, language, and motor development. Scand J Psychol. 2016;57(6):571-583. doi:10.1111/sjop.12321.

Chorbadjian

Deavenport-Saman

Higgins

, et al. Maternal depressive symptoms and developmental delay at age 2: a diverse population-based longitudinal study. Matern Child Health J. 2020;24(10):1267-1277. doi:10.1007/s10995-020-02990-8.

Junge

Garthus-Niegel

Slinning

Polte

Simonsen

Eberhard-Gran

. The impact of perinatal depression on children’s social-emotional development: a longitudinal study. Matern Child Health J. 2017;21(3):607-615. doi:10.1007/s10995-016-2146-2.

Van der Waerden

Galéra

Larroque

, et al. Maternal depression trajectories and children’s behavior at age 5 years. J Pediatr. 2015;166(6):1440-1448. doi:10.1016/j.jpeds.2015.03.002.

10.

Aoyagi

Takei

Nishimura

Nomura

Tsuchiya

. Association of late-onset postpartum depression of mothers with expressive language development during infancy and early childhood: the HBC study. PeerJ. 2019;7:e6566. doi:10.7717/peerj.6566.

11.

Quevedo

Silva

Godoy

, et al. The impact of maternal post-partum depression on the language development of children at 12 months. Child Care Health Dev. 2012;38(3):420-424. doi:10.1111/j.1365-2214.2011.01251.x.

12.

Koutra

Chatzi

Bagkeris

Vassilaki

Bitsios

Kogevinas

. Antenatal and postnatal maternal mental health as determinants of infant neurodevelopment at 18 months of age in a mother-child cohort (Rhea Study) in Crete, Greece. Soc Psychiatry Psychiatr Epide-miol. 2013;48(8):1335-1345. doi:10.1007/s00127-012-0636-0.

13.

Kim

Bird

Peterson

Underwood

Morton

SMB

Grant

. Maternal antenatal depression and early childhood sleep: potential pathways through infant temperament. J Pediatr Psychol. 2020;45(2):203-217. doi:10.1093/jpepsy/jsaa001.

14.

Monteiro

Cooper

Viner-Brown

. Rhode Island Pregnancy Risk Assessment Monitoring System Data Book. Providence, RI: Rhode Island Department of Health; 2018.

15.

Kim

Yang

. Rhode Island Pregnancy Risk Assessment Monitoring System Data Book. Providence, RI: Rhode Island Department of Health; 2012.

16.

Kingston

Kehler

Austin

, et al. Trajectories of maternal depressive symptoms during pregnancy and the first 12 months postpartum and child externalizing and internalizing behavior at three years. PLoS ONE. 2018;13(4):e0195365. doi:10.1371/journal.pone.0195365.

17.

Putnick

. Trajectories of maternal postpartumdepressive symptoms. Pediatrics (Evanston). 2020;146(5):e20200857. doi:10.1542/peds.2020-0857.

18.

Park

Brain

Grunau

Diamond

Oberlander

. Maternal depression trajectories from pregnancy to 3 years postpartum are associated with children’s behavior and executive functions at 3 and 6 years. Arch Womens Ment Health. 2018;21(3):353-363. doi:10.1007/s00737-017-0803-0.

19.

Weinberg

Tronick

. Emotional characteristics of infants associated with maternal depression and anxiety. Pediatrics. 1998;102(5)(suppl E):1298-1304.

20.

Wall-Wieler

Roos

Gotlib

. Maternal depression in early childhood and developmental vulnerability at school entry. Pediatrics. 2020;146(3):e20200794. doi:10.1542/peds.2020-0794.

21.

Bates

Singletary

Dynia

Justice

. Maternal mental health symptoms and clusters predict toddler sleep in low-income homes. J Genet Psychol. 2021;182(4):252-268. doi:10.1080/00221325.2021.1910125.

22.

Olsen

Ammitzbøll

Olsen

Skovgaard

. Problems of feeding, sleeping and excessive crying in infancy: a general population study. Arch Dis Child. 2019;104(11):1034-1041. doi:10.1136/archdischild-2019-316851.

23.

Murray

. The impact of postnatal depression on infant development. J Child Psychol Psychiatry. 1992;33(3):543-561. doi:10.1111/j.1469-7610.1992.tb00890.x.

24.

Stein

Malmberg

Sylva

Barnes

Leach

; FCCC Team. The influence of maternal depression, caregiving, and socioeconomic status in the post-natal year on children’s language development. Child Care Health Dev. 2008;34(5):603-612. doi:10.1111/j.1365-2214.2008.00837.x.

25.

Netsi

. Association of persistent and severe postnatal depression with child outcomes. JAMA Psychiatry. 2018;75(3):247-253. doi:10.1001/jamapsychiatry.2017.4363.

26.

Shorey

Chee

CYI

Chan

Tam

WWS

Chong

. Prevalence and incidence of postpartum depression among healthy mothers: a systematic review and meta-analysis. J Psychiatr Res. 2018;104:235-248. doi:10.1016/j.jpsychires.2018.08.001.

27.

Rockhill

Tong

Morrow

Farr

. Trends in postpartum depressive symptoms—27 states, 2004, 2008, and 2012. MMWR Morb Mortal Wkly Rep. 2017;66(6):153-158. doi:10.15585/mmwr.mm6606a1.

28.

Chaudron

Szilagyi

Kitzman

Wadkins

Conwell

. Detection of postpartum depressive symptoms by screening at well-child visits. Pediatrics. 2004;113(3, pt 1):551-558. doi:10.1542/peds.113.3.551.

29.

Haight

. Postpartum depressive symptoms and screening opportunities at health care encounters. J Womens Health (Larchmt). 2021;30:731-738. doi:10.1089/jwh.2020.8373.

30.

Hunt

Uthirasamy

Porter

Jimenez

. Parental depression screening in pediatric health care settings: a scoping review. Pediatrics. 2022;150(1):e2021055804. doi:10.1542/peds.2021-055804.