The Associations Between Pre- and Postnatal Maternal Symptoms of Distress and Preschooler’s Symptoms of ADHD,Oppositional Defiant Disorder,Conduct Disorder,and Anxiety

Abstract

Objective: The objective of this article is to examine the associations between pre- and postnatal maternal distress and preschooler’s symptoms of ADHD, Oppositional Defiant Disorder (ODD), Conduct Disorder (CD), and anxiety, by timing and gender. Method: Children, aged 3.5 years (N = 1,195), recruited from the Norwegian Mother and Child Cohort Study, were assessed with a semistructured parental psychiatric interview. Perinatal maternal symptoms of distress were assessed by Symptom Checklist (SCL-5); Poisson regression was used to examine the associations. Results: Mid-gestational maternal distress significantly increased the average number of child symptoms, ranging between 3.8% for ADHD hyperactive–impulsive (ADHD-HI) and 8.7% for anxiety. The combination of high maternal scores of distress both pre- and postnatally were associated with increased risk of child symptoms of anxiety (relative risk [RR] = 2.10; 95% confidence interval [CI] = [1.43, 3.07]), CD (RR = 1.83; 95% CI = [1.33, 2.51]), and ODD (RR = 1.30; 95% CI = [1.03, 1.64]), with minor sex differences. Conclusion: Maternal distress during mid-gestation was associated with ADHD, behavioral, and emotional symptoms in preschool children. Continued exposure into the postnatal period may further increase these risk associations .

Keywords

perinatal distress ADHD behavioral problems anxiety preschoolers

Introduction

Maternal mental health across the pregnancy to the postnatal period represents a major public health concern because up to 25% of women show significant symptoms of depression or anxiety during this period (Heron, O’Connor, Evans, Golding, & Glover, 2004). Prenatal maternal anxiety, depression, and distress may have long-term impact on the offspring’s cognitive, emotional, and behavioral development, which through fetal programming mechanisms is hypothesized to alter the regulation of the hypothalamic–pituitary–adrenal axis in response to stress (Glover, 2011; Sandman, Davis, & Glynn, 2012). However, human studies of particularly sensitive time points during gestation have shown inconsistent results.

Prenatal maternal anxiety during the second trimester (Loomans et al., 2011; O’Connor, Heron, & Glover, 2002; O’Connor, Heron, Golding, & Glover, 2003), as well as first trimester maternal distress (Rodriguez & Bohlin, 2005), is associated with an increased risk of symptoms of ADHD in young children. Mid-gestational prenatal maternal anxiety, or distress (O’Connor et al., 2002; Robinson et al., 2011), early gestational anxiety (Loomans et al., 2011), and both anxiety and depression during mid- or late gestation (O’Donnell, Glover, Barker, & O’Connor, 2014; Van Batenburg-Eddes et al., 2013) are reported to predict behavior and/or emotional problems. Some studies report that the increased risk of child symptom development could be equally related to pre- and postnatal exposure to maternal emotional complaints (Carter, Garrity-Rokous, Chazan-Cohen, Little, & Briggs-Gowan, 2001; Giallo, Woolhouse, Gartland, Hiscock, & Brown, 2015; O’Donnell et al., 2014), whereas still others have found the risk associated with only postnatal maternal depression (Kim-Cohen, Moffitt, Taylor, Pawlby, & Caspi, 2005; Sciberras, Ukoumunne, & Efron, 2011). Impaired maternal sensitivity, insecure infant attachment, and less optimal mother–child interactions have been found associated with postpartum depression, which may negatively affect children’s further cognitive, socioemotional, and behavioral development (Carter et al., 2001; Shaw & Vondra, 1995; Stein et al., 2012).

Another issue that warrants attention is the potential moderating role of the child’s sex. Some preschool studies have provided support for an overall pattern of greater risk of ADHD and behavior problems in boys than girls from both pre- and postnatal maternal anxiety, distress, or depression (Carter et al., 2001; Loomans et al., 2011; O’Connor et al., 2002; O’Connor et al., 2003; Rodriguez & Bohlin, 2005; Shaw & Vondra, 1995). However, emotional problems have been found more prominent in girls (Robinson et al., 2011; Sandman, Glynn, & Davis, 2013).

Limitations in previous preschool studies include insufficient assessment of ADHD, behavioral, and emotional problems, where symptom checklists comprise the basis for phenotype definition in most preschool studies. For that reason, preschool studies on this topic have most often merged Oppositional Defiant Disorder (ODD) and Conduct Disorder (CD) into one behavior disorder construct, while findings from longitudinal studies have shown that the two constructs may be distinguished in terms of symptomatology, patterns of comorbidity, and developmental trajectories (Lahey et al., 2009; Nock, Kazdin, Hiripi, & Kessler, 2007; Stringaris & Goodman, 2009). Likewise, the ADHD subtypes are found to be associated with different comorbidities and neurocognitive correlates (Willcutt et al., 2012) and perinatal risk relationships may thus potentially differ for ADHD inattentive (ADHD-IA) compared with ADHD hyperactive–impulsive (ADHD-HI) symptoms. The present study extends most previous preschool research on this topic using a structured clinical interview for the assessment of child diagnostic symptoms and thus takes into consideration the distinctions between the types of disruptive behavior problems and the ADHD subtypes.

The overall aim of the study was to investigate the relationships between perinatal maternal symptoms of anxiety and depression and preschoolers’ symptoms of ADHD-HI, ADHD-IA, ODD, CD, and anxiety, by examining timing effects, whether there was a dose-dependent relationship, and to examine for potential sex-differences.

Method

Design and Participants

“The Preschool ADHD study” is a longitudinal study with participants recruited from the Norwegian Mother and Child Cohort Study (MoBa). MoBa is a prospective birth cohort study run by the Norwegian Institute of Public Health, following about 107,000 pregnancies before and after birth (Magnus et al., 2006). Mothers completed the MoBa questionnaires at Weeks 17 and 30 of gestation, and when children were 6, 18, and 36 months old.

The recruitment of participants to the Preschool ADHD Study has been described in more detail previously (Bendiksen et al., 2017). In brief, this study included 1,195 mother and child pairs. Of these, 1,048 were recruited based on scores above the 90th percentile on 11 questions about hyperactivity, impulsivity, and inattention, or parental report of hyperactivity as a health problem, in the MoBa questionnaire at 36 months. Six questions were from the Child Behavior Checklist (CBCL; Achenbach & Ruffle, 2000) and five from the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; American Psychiatric Association [APA], 2000) criteria. The comparison group of 147 children was randomly recruited from the MoBa study.

After parental consent, the children, aged 36 to 44 months, participated in a 1-day clinical assessment at Oslo University Hospital together with (one of) the parents. The questionnaires used in the MoBa and the preschool ADHD study were written in Norwegian, which implied that one of the parents had to speak Norwegian. The great majority of participants were therefore Caucasians. The exclusion criteria were severe medical conditions or high scores for autistic symptoms.

Ethics

The Regional Ethics Committee approved the study and the Norwegian Data Inspectorate granted a license in 2007. Assessments were carried out according to the principles of the Helsinki Declaration. Parents returned a written consent prior to the clinical assessment.

The enrollment is shown in Figure 1.

Figure 1.

Enrollment into the MoBa and the preschool ADHD study.

Measures

Perinatal maternal symptoms of anxiety and depression

In the MoBa questionnaire at Week 17 of gestation, the mothers were asked whether or not they had had any previous history of anxiety or mood disorder (yes: anxiety and/or depression prior to this pregnancy/no: no previous history of anxiety and/or depression before this pregnancy). If yes, they were rated as preconception anxiety/depression positive.

A widely used self-administered instrument of psychological distress, the Symptom Checklist (SCL-5), derived from the SCL-25 (Hesbacher, Rickels, Morris, Newman, & Rosenfeld, 1980), was included in the MoBa questionnaires at Gestational Week 17 (T₁; that is, early gestation), at Week 30 (T₂; that is, mid-gestation), and at 6 months postnatally (T₃). The SCL-5 includes two symptoms of depression (“feeling hopeless about the future,” “feeling depressed/sad/blue”) and three questions of anxiety (“constantly frightened/anxious/fearful,” “nervous, inner turmoil,” “frequently worried or uneasy”). Each question is rated on a 4-point scale (“not at all,” “a little,” “quite a bit,” “extremely,” rated 1 to 4). The SCL-5 scales showed satisfactory internal consistency with Cronbach’s alpha values of .82 (T₁), .79 (T₂), and .84 (T₃). The scores on SCL-5 have been shown to correlate strongly with the SCL-25, r = .91 (Strand, Dalgard, Tambs, & Rognerud, 2003).

We created a categorical composite variable by combining prenatal maternal anxiety and depression (T₁ and/or T₂) into Group 1, “prenatal maternal distress, only” (n = 105). Group 2 included women who only had reported anxiety and depression postpartum (T₃), “postnatal distress, only” (n = 61), whereas Group 3 included those who reported symptoms of anxiety and depression both during pregnancy and postnatally (T₁/T₂ and T₃), the “postconception distress” (n = 55). Group 4 consisted of women who had reported anxiety and/or depression previous to conception, but not during pregnancy or postnatally, “preconception anxiety/depression, only” (n = 58), whereas Group 5, “distress at all time points” (n = 36) included those who reported both anxiety and depression previous to conception, during pregnancy, and postpartum. The reference category (Group 0) included women with no reports of anxiety, depression, or distress at any of the assessed time points (n = 714). We excluded missing data (n = 138), women who reported symptoms of anxiety and depression both during pregnancy and previous to conception (n = 21), and those who reported symptoms both previous to conception and postpartum (n = 8).

The suggested cutoff of the SCL-5 score for mental disorder/distress has been set at or above 2 (Strand et al., 2003), but we decided to use a quasi-clinical cutoff above the 85th percentile (i.e., we dichotomized the SCL-5 scores into distress yes/no using a cutoff between 8 and 9) in the absence of an established cutoff of SCL-5 during the perinatal period. This cutoff is more in line with other studies using different scales for the assessment of perinatal maternal anxiety (eg., the Crown–Crisp index, “CCEI”) (Clavarino et al., 2010; O’Connor et al., 2002; O’Donnell et al., 2012).

Child symptoms of ADHD, ODD, CD, and anxiety

One of the parents, most often the mother, was interviewed with the “Preschool Age Psychiatric Assessment” (PAPA; Egger & Angold, 2004), a comprehensive psychiatric interview for the assessment of psychiatric symptoms and disorders in preschool children (Egger et al., 2006). The PAPA interviews were performed by trained psychology students and supervised by clinically trained psychologists or child psychiatrists. This semistructured interview provides information about the scale and frequency of psychiatric symptoms according to DSM-IV-TR (APA, 2000). To be recorded as “present,” symptoms had to have lasted for at least 3 months. Outcome measures were number of symptoms of ADHD-IA (nine symptoms), ADHD-HI (nine symptoms), ODD (eight symptoms), and CD (eight symptoms). We used a sum score of child symptoms of specific phobia, social phobia, separation anxiety, and generalized anxiety as a measure of anxiety symptoms in 3-year old children. Interrater reliability of the number of Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; APA, 1994) symptoms of ADHD, ODD, and CD was good with average intraclass correlations (ICCs) ranging from .91 to .99, whereas ICC for anxiety symptoms was .86 in this sample.

Covariates

We assessed a range of possible confounders, but we only included covariates that were associated with maternal symptoms of distress at one time point and/or number of ADHD, ODD, CD, or anxiety symptoms in the preliminary analyses. The included covariates were as follows: birth weight, child’s sex, maternal age, caesarean section (yes/no) from The Norwegian Medical Birth Registry (Magnus et al., 2006), and maternal educational level, marital status (i.e., cohabit/single parent), parity, social support, maternal smoking during pregnancy—assessed at Week 17 and Week 30 during gestation and postnatally at 6 months (i.e., the mean number of cigarettes per mother was calculated by summing the number of cigarettes at the three time points and divide by 3)—and breast-feeding duration (yes: 9 months or more/no: less than 9months) from the MoBa questionnaires . In addition, from the Medical Birth Registry we chose to include APGAR score ( i.e. score given for the the newborn’s physical condition including Apparance/ skin color, Pulse/ heart rate, Grimace response/ reflexes, Activity/ muscle tone and Respiration/ breathing rate and effort. For each item, the score is 0, 1, or 2, and the total score range from 0 to 10) .

The following potential covariates were not included, as they were not related to either exposure or outcome in our data: maternal alcohol use during pregnancy, antenatal maternal urogenital and bladder infection, preeclampsia, and neonatal complications. Furthermore, children’s intellectual functioning (IQ) at 3.5 years was not included because it was considered to be an outcome and potentially associated with the same risk and protective factors as the other outcome measures. In the unadjusted analyses, gestational age was not significantly associated with outcome measures, and we chose to exclude the variable from the multivariate analyses, because it was strongly correlated with birth weight (Pearson product–moment correlation: r = .72).

Data Analyses

We conducted Poisson regression analyses, unadjusted and adjusted, to examine the relationships between maternal symptoms of distress measured and the child’s number of ADHD-HI, ADHD-IA, ODD, CD, and anxiety symptoms at 3.5 years of age. Three variables of maternal symptoms of distress were studied: the SCL-5 sum score at Week 17 (T₁) and Week 30 (T₂) of gestation, and at 6 months postpartum (T₃). To test whether maternal distress at one time point remained significant when other time points were covaried, we examined the associations between child symptoms and SCL-5 scores at each time point and covaried the SCL-5 scores at different time points, and in addition, preconception maternal anxiety and/or depression was covaried. The sum scores of the three time points were all included in the analyses to examine at which time point maternal distress had the strongest impact on child symptoms. Furthermore, we conducted Poisson regression models to examine the associations between child symptoms and the composite variable of maternal distress (described in the “Method” section) to investigate whether there was a combined perinatal effect of maternal distress on child symptom load. Covariates, which in the preliminary analyses were found associated with predictors or outcomes, were included in the multiple Poisson regressions for adjustment. In the multivariate analyses, we also examined for possible interactions between maternal symptoms of distress and other covariates included in the model. We examined for possible interactions between SCL-5 scores and child sex, but in addition, we conducted stratified analyses for girls and boys.

Crude and adjusted relative risk (RR) with 95% confidence intervals (CIs) was given using robust sandwich estimator of variance. The two-sided p values were not corrected for multiple testing.

The analyses were performed with IBM SPSS Statistics for Windows, Version 21, and STATA/IC 13 for Windows, Version 13.1.

Results

Background and clinical characteristics of 1,195 children and mother and child pairs are presented in Table 1. As seen from the table, average scores for maternal symptoms of anxiety and depression were pretty stable over time. The individual SCL scores were moderately correlated over time. Spearman’s correlation coefficient between SCL-5 score at Week 17 of gestation (T₁) and Week 30 of gestation (T₂) was .58, between (T₁) and postnatal (T₃) was .47, and between T₂ and T₃ was .52.

Table 1.

Background and Perinatal Factors for 1,195 Mother and Child Pairs.

	Total N = 1,195	M	SD	Median	Interquartile range	Minimum	Maximum
Mother’s age (years)	1,194	30.62	4.29	31	28-34	19	43
Mother’s education (years)	1,146	15.24	2.36	15.5	13.6-18	9	18
^aMaternal smoking, continuous measure	1,195	0.39	1.75	0	0-0	0	16.7
Parity	1,195	0.46	0.70	0	0-1	0	4
Maternal distress (T₁)	1,134	6.61	2.18	6	5-7	5	20
Maternal distress (T₂)	1,128	6.53	2.02	6	5-6	5	17
Maternal distress (T₃)	1,123	6.63	2.33	6	5-6	5	20
Gestational age (week)	1,187	39.29	2.21	40	39-41	24	43
Birth weight (gram)	1,192	3,499.57	615.34	3,630	3,181-3,896	550	5,224
APGAR score, 5 min	1,191	9.37	0.90	10	9-10	0	10
Child’s age (months)	1,193	41.74	1.34	41.8	40.8-42.6	37.15	46.72
ADHD-IA symptoms	1,195	1.33	1.81	1	0-2	0	9
ADHD-HI symptoms	1,195	2.69	2.49	2	0-4	0	9
ODD symptoms	1,195	1.68	1.55	1	0-3	0	8
CD symptoms	1,195	0.66	0.98	0	0-1	0	6
Anxiety symptoms	1,195	0.77	1.24	0	0-1	0	8
					Total N	n	%
Child’s sex (girls)					1,195	569	47.3
Preeclampsia					1,194	49	4.1
^bMaternal smoking (T₁)					1,147	54	4.5
^bMaternal smoking (T₂)					1,061	48	4.0
^bMaternal smoking (T₃)					1,090	158	14.4
Caesarean					1,195	239	20
Breast-feeding ≥ 9 months					1,195	681	57
Marital status (single parent)					1,194	38	3.2
Partner satisfaction (good)					1,134	1,028	86
Social support (good)					1,161	1,130	94.6
Preconception anxiety/depression					1,195	135	11.3

Note. APGAR score: score given for the newborn’s physical condition (i.e. Apparance/ skin color. Pulse/ heart rate. Grimace response/ reflexes. Activity/ muscle tone. Respiration/ breathing rate and effort. Score for each of the five subscales: 0, 1, 2). ADHD-IA = ADHD Inattentive. ADHD-HI = ADHD hyperactive–impulsive; ODD = oppositional defiant disorder; CD = conduct disorder; T₁ = Gestational Week 1 to Week 17; T₂ = Gestational Week 18 to Week 30; T₃ = 0 to 6 months postpartum; Maternal distress = SCL-5 scores (i.e Symptom Checklist) of maternal symptoms of anxiety and depression; Breast-feeding ≥ 9 months = breast-feeding duration past 9 months; Preconception anxiety/depression = maternal anxiety, depression, or both previous to conception. ^aMaternal smoking, continuous variable = the average number of cigarettes per day (M score, T₁, T₂, T₃). ^bMaternal smoking = percentage of women who smoked cigarettes, T₁, T₂, T_3. Caesarean = elective or acute operative delivery by caesarean section.

The Associations Between Perinatal Maternal Distress on Child Symptoms of ADHD, ODD, CD, and Anxiety

Crude and adjusted RRs of maternal distress for the increased mean numbers of ADHD-IA, ADHD-HI, ODD, CD, and anxiety symptoms are shown in Table 2. Unadjusted, SCL-5 score at (T₁), (T₂), and (T₃) was significantly associated with child symptoms of ADHD-IA, ADHD-HI, and anxiety, whereas distress at (T₂) and (T₃) most significantly was associated with ODD and CD symptoms. The prenatal effects at T₁ and T₂ on child symptom loads were equally attenuated by postnatal distress, and compared with the crude RR (Table 2), the relative risks decreased on average by 2% after adjustment. Postnatal maternal distress was no longer associated with child symptoms after the adjustment of maternal distress at T₂, with the exception of the association with ADHD-IA symptoms (RR = 1.03, 95% CI = [1.00, 1.06], p = .03 [T₃] vs. RR = 1.06, 95% CI = [1.03, 1.10], p = .001 [T₂]). However, after the adjustment of early gestational distress (T₁), the postnatal effects on ADHD-IA turned out nonsignificant (RR = 1.03, 95% CI = [0.99, 1.06], p = .07 [T₃] vs. RR = 1.07, 95% CI = [1.03, 1.10], p = .001 [T₁]).

Table 2.

Crude and Adjusted Relative Risks for the Relationship Between Perinatal Maternal Symptoms of Distress and Children’s Symptoms of ADHD-IA, ADHD-HI, ODD, CD, and Anxiety.

N = 1,195	Scale	Maternal distress
N = 1,195	Scale	Crude RR	95% CI	p value	Adjusted RR	95% CI	p value
ADHD-IA symptoms	0-9
SCL-5 (T₁)	5-15	1.084	[1.05, 1.12]	<.001	1.021	[0.98, 1.07]	.33
SCL-5 (T₂)	5-12	1.082	[1.05, 1.12]	<.001	1.028	[0.99, 1.07]	.19
SCL-5 (T₃)	5-15	1.067	[1.04, 1.10]	<.001	1.003	[0.97, 1.03]	.86
Preconception anxiety/depression	Yes	1.493	[1.20, 1.86]	<.001	1.143	[0.88, 1.48]	.32
Sex	Girls	0.884	[0.76, 1.03]	.119	0.913	[0.77, 1.08]	.28
Maternal age (years)	19-43	0.960	[0.94, 0.98]	<.001	0.972	[0.95, 0.99]	.007
Maternal education (years)	9-18	0.932	[0.90, 0.96]	<.001	0.970	[0.93, 1.01]	.11
Maternal smoking (cigarette/day)	0-17	1.036	[1.00, 1.07]	.041	1.021	[0.98, 1.06]	.27
Caesarean section	Yes	1.179	[0.98, 1.42]	.085	1.278	[1.04, 1.57]	.02
Breast-feeding ≥ 9 months	Yes	0.729	[0.63, 0.85]	<.001	0.772	[0.65, 0.91]	.002
ADHD-HI symptoms	0-9
SCL-5 (T₁)	5-15	1.040	[1.02, 1.06]	<.001	0.982	[0.95, 1.01]	.25
SCL-5 (T₂)	5-12	1.058	[1.04, 1.08]	<.001	1.038	[1.01, 1.07]	.013
SCL-5 (T₃)	5-15	1.050	[1.03, 1.07]	<.001	1.014	[0.99, 1.04]	.25
Preconception anxiety/depression	Yes	1.236	[1.06, 1.44]	.008	1.056	[0.88, 1.27]	.56
Sex	Girls	0.856	[0.77, 0.95]	.004	0.853	[0.76, 0.95]	.006
Maternal age (years)	19-43	0.973	[0.96, 0.99]	<.001	0.976	[0.96, 0.99]	.001
Maternal education (years)	9-18	0.938	[0.92, 0.96]	<.001	0.968	[0.94, 0.99]	.013
Parity, total	0-4	1.003	[0.93, 1.08]	.94	1.094	[1.01, 1.19]	.035
SCL-5 (T₁) × Parity	Interaction term				1.028	[1.00, 1.05]	.030
Maternal smoking (cigarette/day)	0-17	1.041	[1.02, 1.06]	<.001	1.021	[0.99, 1.05]	.09
SCL-5 (T₁) × Maternal Smoking	Interaction term				1.019	[1.01, 1.03]	.009
Breast-feeding ≥ 9 months	Yes	0.801	[0.72, 0.89]	<.001	0.866	[0.77, 0.97]	.013
ODD symptoms	0-8
SCL-5 (T₁)	5-15	1.034	[1.01, 1.06]	.002	0.988	[0.96, 1.02]	.45
SCL-5 (T₂)	5-12	1.077	[1.03, 1.12]	<.001	1.039	[1.01, 1.07]	.019
SCL-5 (T₃)	5-15	1.040	[1.02, 1.06]	<.001	1.015	[0.99, 1.04]	.25
Preconception anxiety/depression	Yes	1.177	[1.01, 1.37]	.036	1.085	[0.91, 1.30]	.37
Sex	Girls	1.069	[0.96, 1.10]	.21	1.018	[0.91, 1.14]	.76
Mother’s age (years)	19-43	0.978	[0.97, 0.99]	<.001	0.981	[0.97, 0.99]	.007
Mother’s education (years)	9-18	0.968	[0.95, 0.99]	.004	0.976	[0.95, 1.00]	.058
Birth weight	pr.100 g	0.990	[0.982, 0.998]	.018	0.990	[0.98, 1.00]	.058
Breast-feeding ≥ 9 months	Yes	0.946	[0.85, 1.05]	.30	0.985	[0.88, 1.10]	.080
CD symptoms	0-6
SCL-5 (T₁)	5-15	1.048	[1.01, 1.09]	.02	0.963	[0.91, 1.02]	.16
SCL-5 (T₂)	5-12	1.077	[1.03, 1.12]	<.001	1.065	[1.00, 1.13]	.047
SCL-5 (T₃)	5-15	1.059	[1.03, 1.09]	.001	1.027	[0.99, 1.07]	.17
Preconception anxiety/depression	Yes	1.556	[1.25, 1.94]	<.001	1.316	[1.00, 1.73]	.048
Sex	Girls	0.895	[0.75, 1.06]	.21	0.890	[0.74, 1.07]	.21
SCL-5 (T₁) × Sex	Interaction term				0.872	[0.79, 0.97]	.014
Maternal smoking (cigarette/day)	0-17	1.057	[1.02, 1.10]	.004	1.031	[0.99, 1.07]	.14
SCL-5 (T₁) × Maternal Smoking	Interaction term				1.013	[1.00, 1.03]	.037
SCL-5 (T₃) × Maternal Smoking	Interaction term				1.019	[1.01, 1.03]	.007
Parity, total	0-4	1.208	[1.09, 1.34]	<.001	1.296	[1.14, 1.47]	<.001
Anxiety symptoms	0-8
SCL-5 (T₁)	5-15	1.101	[1.07, 1.14]	<.001	1.044	[0.99, 1.10]	.085
SCL-5 (T₂)	5-12	1.114	[1.08, 1.15]	<.001	1.087	[1.03, 1.14]	.001
SCL-5 (T₃)	5-15	1.066	[1.03, 1.10]	<.001	1.004	[0.96, 1.05]	.85
Preconception anxiety/depression	yes	1.227	[0.96, 1.56]	.097	0.934	[0.69, 1.26]	.65
Sex	Girls	1.006	[0.84, 1.21]	.95	1.005	[0.83, 1.22]	.96
Mother’s education (years)	9-18	0.943	[0.91, 0.97]	<.001	0.947	[0.91, 0.99]	.013
SCL-5 (T₃) × Mother’s Education	Interaction term				1.023	[1.01, 1.04]	.027
Marital status (single parent)	Yes	1.064	[0.67, 1.69]	.79	0.696	[0.43, 1.14]	.15
SCL-5 (T₁) × Marital Status	Interaction term				1.461	[1.17, 1.83]	.001
SCL-5 (T₃) × Marital Status	Interaction term				1.529	[1.22, 1.91]	<.001
Preeclampsia	Yes	1.234	[0.83, 1.85]	.31	1.219	[0.81, 0.86]	.35
SCL-5 (T₁) × Preeclampsia	Interaction term				1.26	[1.03, 1.54]	.023

Note. Poisson regression model. Nonsignificant associations of covariates are not reported. The interaction term was not included in the model when estimating the adjusted relative risks reported in the table . ADHD-IA = ADHD Inattentive. ADHD-HI = ADHD hyperactive–impulsive; ODD = oppositional defiant disorder; CD = conduct disorder; Crude RR = unadjusted relative risk; CI = confidence interval; Adjusted RR = adjusted relative risk; SCL-5 = maternal depressive and anxiety symptoms at three different time points: during gestation at Week 17 (T₁) or Week 30 (T₂), and postnatal at 6 months (T₃); SCL-5 = Symptom Checklist; Preconception anxiety/depression = maternal anxiety, depression, or both previous to conception.

In the adjusted multivariate model, one additional unit increase in maternal scores of prenatal distress (T₂) increased the mean number of ADHD-HI symptoms by 3.8%, ODD by 3.9%, CD by 6.5%, and child anxiety symptoms by 8.7%, whereas the effects of early gestational- and postnatal maternal distress (T₁/ T₃) turned out nonsignificant for all outcomes. The effect of maternal distress on ADHD-IA decreased about 6.3% after the adjustments, whereas respective figures for ADHD-HI, ODD, CD, and anxiety symptoms were 2.5%, 2.8%, 1.8%, and 4.9%. The attenuation of the postnatal SCL-5 scores was more evident than during gestation for all the outcome variables.

Confounding and Moderating Effects of Covariates

Positively directed interactions between maternal distress (T₁) and maternal smoking were shown for the effect on ADHD-HI (p = .009) and CD symptoms (p = .037). Maternal smoking increased the risk by moderating the effect of postnatal maternal distress (T₃) on child CD symptoms (p = .007).

For child anxiety symptoms, there were positively directed interactions between SCL-5 scores (T₃) and maternal educational level (p = .027), and between SCL-5 scores (T₁ and T₃) and marital status (p = .001) and (p < .001), respectively. Parity (i.e., multiparity) moderated the effect of SCL-5 (T₁), which increased risk of ADHD-HI symptoms (p = .030). Maternal educational level, maternal smoking, breast-feeding duration, maternal age, social support and maternal mood previous to conception attenuated the effects of SCL-5 scores (T₁ and T₃) on ADHD-IA symptoms, whereas the association between maternal distress (T₂) and symptoms of ADHD-HI was attenuated by maternal smoking. There was minor confounding on SCL-5 scores by obstetric factors, but a positively directed interaction was shown between preeclampsia and SCL-5 score (T₁) on child anxiety symptoms (p = .023).

The Dose-Dependent Effects of Pre- and Postnatal Maternal Distress

The adjusted analyses for the composite variable of maternal distress showed that when mothers experienced distress at all time points, average number of ADHD-IA and anxiety symptoms significantly increased compared with absence of any maternal distress (Table 3). The most significant associations were found between maternal postconception distress (i.e., both pre- and postnatal distress) and the numbers of ODD, CD, and anxiety symptoms. Child anxiety symptoms were also significantly associated with prenatal maternal distress, whereas a small association was found between preconception maternal anxiety and/or depression and child CD symptoms.

Table 3.

The Association Between Maternal Distress and ADHD, ODD, CD, and Anxiety Among 1,195 Children, Stratified on Timing of Exposure.

N = 1,030	Scale	Maternal distress
N = 1,030	Scale	Crude RR	95% CI	p value	Adjusted RR	95% CI	p value
ADHD-IA symptoms	0-9
Reference group (n = 714)	No
Prenatal distress, only (n = 105)	Yes (SCL-5 score ≥ 9)	1.158	[0.88, 1.52]	.29	1.091	[0.84, 1.42]	.52
Postnatal distress, only (n = 61)	Yes (SCL-5 score ≥ 9)	1.112	[0.82, 1.52]	.50	1.026	[0.76, 1.38]	.87
Postconception distress (n = 55)	Yes (SCL-5 score ≥ 9)	1.429	[1.05, 1.95]	.024	1.154	[0.84, 1.58]	.37
Preconception anxiety/depression, only (n = 59)	Yes	1.094	[0.76, 1.57]	.63	1.132	[0.78, 1.64]	.51
Distress at all time points (n = 36)	Yes	2.068	[1.42, 3.01]	<.001	1.545	[1.01, 2.37]	.045
ADHD-HI symptoms	0-9
Reference group (n = 714)	No
Prenatal distress, only (n = 105)	Yes (SCL-5 score ≥ 9)	1.174	[0.98, 1.40]	.078	1.138	[0.96, 1.35]	.14
Postnatal distress, only (n = 61)	Yes (SCL-5 score ≥ 9)	1.211	[0.99, 1.48]	.063	1.178	[0.97, 1.42]	.09
Postconception distress (n = 55)	Yes (SCL-5 score ≥ 9)	1.322	[1.06, 1.65]	.013	1.146	[0.93, 1.41]	.20
Preconception anxiety/depression (n = 59)	Yes	1.035	[0.80, 1.34]	.80	1.067	[0.82, 1.38]	.63
Distress at all time points (n = 36)	Yes	1.361	[1.03, 1.80]	.031	1.095	[0.8, 1.47]	.55
ODD symptoms	0-8
Reference group (n = 714)	No
Prenatal distress, only (n = 105)	Yes (SCL-5 score ≥ 9)	1.155	[0.96, 1.39]	.14	1.103	[0.91, 1.35]	.33
Postnatal distress, only (n = 61)	Yes (SCL-5 score ≥ 9)	1.213	[0.99, 1.48]	.053	1.129	[0.92, 1.38]	.24
Postconception distress (n = 55)	Yes (SCL-5 score ≥ 9)	1.346	[1.08, 1.68]	.009	1.301	[1.03, 1.64]	.025
Preconception anxiety/depression (n = 59)	Yes	1.136	[0.90, 1.44]	.29	1.152	[0.91, 1.45]	.23
Distress at all time points (n = 36)	Yes	1.258	[0.98, 1.62]	.08	1.171	[0.89, 1.53]	.25
CD symptoms	0-6
Reference group (n = 714)	No
Prenatal distress, only (n = 105)	Yes (SCL-5 score ≥ 9)	1.161	[0.86, 1.56]	.32	1.216	[0.83, 1.54]	.46
Postnatal distress, only (n = 61)	Yes (SCL-5 score ≥ 9)	1.097	[0.71, 1.70]	.68	1.158	[0.76, 1.77]	.50
Postconception distress (n = 55)	Yes (SCL-5 score ≥ 9)	1.997	[1.45, 2.74]	<.001	1.831	[1.33, 2.51]	<.001
Preconception anxiety/depression (n = 59)	Yes	1.367	[0.99, 1.87]	.051	1.413	[1.01, 1.97]	.043
Distress at all time points (n = 36)	Yes	1.668	[1.06, 2.62]	.026	1.369	[0.85, 2.20]	.19
Anxiety symptoms	0-8
Reference group (n = 714)	No
Prenatal distress, only (n = 105)	Yes (SCL-5 score ≥ 9)	1.157	[1.18, 2.06]	.002	1.547	[1.16, 2.06]	.003
Postnatal distress, only (n = 61)	Yes (SCL-5 score ≥ 9)	1.053	[0.71, 1.56]	.80	1.014	[0.66, 1.54]	.95
Postconception distress (n = 55)	Yes (SCL-5 score ≥ 9)	2.124	[1.47, 3.05]	<.001	2.096	[1.43, 3.07]	<.001
Preconception anxiety/depression (n = 59)	Yes	0.965	[0.67, 1.39]	.85	1.007	[0.69, 1.47]	.97
Distress at all time points (n = 36)	Yes	1.825	[1.21, 2.75]	.004	1.882	[1.21, 2.92]	.005

Note. Poisson regression model. Covariates not reported. Composite variable of maternal distress: reference group = 0 (no anxiety/depression at any time point). Prenatal distress, only = 1 (i.e., SCL-5 score ≥ 9, T₁/T₂). Postnatal distress, only = Group 2 (i.e., SCL-5 score ≥ 9, T₃). Postconception distress = Group 3 (SCL-5 score ≥ 9, T₁/T₂/T₃). Preconception anxiety/depression, only = Group 4 (i.e., maternal anxiety/depression previous to conception). Distress at all time points = Group 5 (i.e., maternal anxiety/depression previous to conception, prenatal distress, T₁/T₂, and postnatal distress, T₃). ADHD-IA = ADHD Inattentive. ADHD-HI = ADHD hyperactive–impulsive; ODD = oppositional defiant disorder; CD = conduct disorder; Crude RR = unadjusted relative risk; CI = confidence interval; Adjusted RR = adjusted relative risk; SCL-5 = Symptom Checklist.

Sex Differences

Boys were more likely to have symptoms of ADHD-HI irrespective of maternal distress (Table 2). A negatively directed interaction indicated that girls were less likely to develop CD symptoms when exposed to maternal distress (T₁; p = .014). Separate analyses for boys and girls confirmed the positive association between maternal distress (T₁) and CD symptoms in boys (crude RR = 1.067, 95% CI = [1.02, 1.12], p = .006 vs. girls, crude RR = 1.018, 95% CI = [0.95, 1.10], p = .64; data not shown). However, we found no interactions between the child’s sex and maternal distress at any time point for other outcomes, and neither in analyses stratified by child sex.

Discussion

In this study, prenatal maternal distress during mid-gestation was shown to be significantly, but modestly, associated with symptoms of ADHD-HI, ODD, and CD; the strongest association was found for child anxiety symptoms. We found no relationships between child symptoms and maternal distress during early gestation, postnatally, or previous to conception, except for a small association between maternal mood previous to conception and CD symptoms. Girls, exposed to early gestational distress were less likely to have CD symptoms than boys. Independently of the exposure of maternal distress, girls were also less likely to display symptoms of ADHD-HI. Furthermore, these analyses showed that there was a concomitant and dose-dependent association, where the total effect of maternal distress postconception was the most significant predictor of child symptoms of ODD, CD, and anxiety.

In line with our results, O’Connor and colleagues (O’Connor et al., 2002) found that mid-gestation was a particularly sensitive period for the prediction of ADHD, behavior- and emotional problems in preschoolers, whereas the effects of postnatal maternal emotional symptoms turned out nonsignificant after prenatal maternal symptoms were covaried. However, we found that the magnitude of the prenatal prediction was attenuated by postnatal experiences, which contrast with the findings by O’Connor and colleagues (2002), who reported that the exposure postnatally did not influence the magnitude of the antenatal prediction. Conversely, other studies have found no effects of prenatal maternal emotional complaints once postnatal effects were taken into account (Bekkhus, Rutter, Barker, & Borge, 2011; Kim-Cohen et al., 2005). Methodological differences may account for some of the discrepancies between studies. The case could also be that rather than specific timing effects, maternal symptoms of anxiety and/or depression over time, even at subclinical levels, may constitute an enduring risk of the development of emotional and behavioral problems in children (Giallo et al., 2015). Clavarino and colleagues (2010) reported that antenatal maternal anxiety was associated with persistent attention problems in both 5- and 14-year-old children, but chronic maternal anxiety during the peripartum period and 5 years later was the strongest predictor, which pointed toward an accumulative effect of maternal anxiety. We only investigated the effects of maternal symptoms of distress across pregnancy to 6 months postnatally, but still found a dose-dependent relationship in that the combination of both pre- and postnatal exposure had the strongest effect on preschoolers’ symptoms of ODD, CD, and anxiety. The combined effect of maternal distress at all time points, including preconception anxiety and depression, was associated with symptoms of ADHD-IA. Some studies have found independent effects of antenatal and postnatal maternal depression on toddler neurodevelopment (Koutra et al., 2013) and preschoolers’ behavioral problems (Carter et al., 2001). Maternal depression and anxiety postpartum are suggested to interfere with maternal sensitivity and responsiveness to their infants and thereby increase the risk of disturbed mother–child interactions (Stein et al., 2012). However, a number of studies focusing on postnatal maternal depression did not simultaneously investigate the impact of prenatal maternal mood or postnatal anxiety, which makes comparisons between studies further difficult.

Developmental or adaptive programming models imply that the in-utero exposure instigate an adaptive response, which may be carried forward in development with persisting effects on offspring behavior. O’Donnell and colleagues (2014) examined the relationship between prenatal maternal anxiety and child’s later emotional and behavioral problems. Total problems assessed at five time points (i.e., at the ages of 4, 7, 9, and 13) remained high in the high prenatal maternal anxiety group across development. The effect of the exposure at Week 18 did not significantly differ from that at Week 32 during gestation (O’Donnell et al., 2014), which contrasts our results. However, both animal and human studies show inconsistencies regarding specifically vulnerable periods during gestation (Beydoun & Saftlas, 2008; Glover, 2014), but such findings may be confounded because mothers exposed to stressors early in pregnancy could be affected for a longer period than those exposed at a later time point (O’Connor, Monk, & Fitelson, 2014). The timing of distress could be linked to various defined outcomes, as shown by a large population-based study by Class and colleagues (2014), who found that different patterns of psychopathology emerged following prenatal compared with postnatal maternal distress, where the exposure during third trimester increased the risk of autism spectrum disorders (ASDs) and ADHD, whereas an increased risk of suicide was observed in offspring whose mothers were stressed during the first postnatal year.

In contrast to previous studies (O’Connor et al., 2002; O’Donnell et al., 2014), perinatal maternal symptoms of anxiety and depression did not predict child ADHD-IA symptoms after the adjustments. However, preschoolers’ symptoms of ADHD-HI were significantly predicted by maternal distress. Methodological differences could in part explain some of the observed discrepancies, because the assessment by “The Strength and Difficulties Questionnaire” (SDQ; Goodman, 1997) used in some of these studies (O’Connor et al., 2002; O’Donnell et al., 2014) includes two inattentive and three hyperactive/impulsive symptoms, and the behavior problems include two symptoms of ODD and three diagnostic symptoms of CD. The inclusion of covariates may also play a role. In our study, attenuation of risk relationships between perinatal maternal symptoms of anxiety and depression/distress and most outcomes by maternal and obstetric factors were generally minor for each covariate, but totally, the effects of maternal symptoms on most outcomes were significantly reduced by the covariate adjustments. However, the most marked attenuation by covariates was shown for the prediction of child ADHD-IA symptoms . Maternal depression and/or anxiety previous to conception strongly confounded the effects of maternal distress in ADHD-IA symptoms, but also in CD symptoms, which might represent common genetic effects. A number of obstetric factors have been found associated with ADHD (Pettersson et al., 2015), but also related to pre- and postnatal factors (Henderson, Evans, Straton, Priest, & Hagan, 2003; Seimyr, Edhborg, Lundh, & Sjogren, 2004; Shamberger, 2012). Unadjusted, maternal smoking predicted ADHD and CD, but turned out nonsignificant after the adjustments. However, maternal smoking moderated the effect of early gestational distress on ADHD-HI, as well as postnatal maternal distress for the relationship with CD symptoms. These findings contrast the study by Rodriguez and Bohlin (2005), who found that the association between ADHD symptoms and prenatal maternal distress was independent and not modified by maternal smoking. Preeclampsia moderated the effect of gestational distress on child anxiety symptoms, but showed no independent effect on child symptoms. In contrast to previous results (Littleton, Breitkopf, & Berenson, 2007), we found no modifying and minor attenuation of the effects of perinatal maternal anxiety and depression by gestational length or birth weight. A possible explanation could be small variability, as most children in this sample were born at term and had a standard birth weight. The protective effect of breast-feeding on ADHD symptoms attenuated, but did not modify, the effects of perinatal maternal emotional symptoms, which may suggest independent mechanisms.

Sex differences related to perinatal maternal distress on child outcomes were minor in this sample. Independent of maternal distress, we found that boys were more susceptible for symptoms of ADHD-HI, however, separate analyses showed no sex differences in the relationships between perinatal maternal distress and ADHD, which contrast with some previous findings (Loomans et al., 2011; O’Connor et al., 2002; Rodriguez & Bohlin, 2005; Sciberras et al., 2011). Boys have previously also been found more susceptible to behavior problems than girls ensuing perinatal maternal anxiety (Loomans et al., 2011), but results have been inconsistent (O’Connor et al., 2002; O’Connor et al., 2003). However, we found that child sex moderated the effect of maternal distress on CD symptoms, and separate analyses confirmed that boys, whose mothers experienced early gestational distress, were more likely than girls to develop CD symptoms. Regarding timing effects, previous studies have shown inconsistent results even when the same types of exposure and outcome are scrutinized by gender. O’Connor and colleagues (2003) reexamined the children from the ALSPAC study when aged 7 and found maternal anxiety at 18 weeks of gestation to be the strongest predictor of conduct problems in girls, while the prediction at 32 weeks of gestation remained the strongest one for boys. By contrast, a study by de Bruijn, van Bakel, and van Baar (2009) found third trimester maternal anxiety constituted a risk of externalizing problems in girls, while early gestational maternal anxiety predicted more externalizing problems among boys than girls, which may be in line with our findings of the prediction of CD symptoms by maternal distress during Weeks 1 to 17 of gestation. We found no sex differences related to the exposure of perinatal maternal distress and child symptoms of ODD, which might support the distinction between symptoms of ODD and CD when examining risk factors and correlates. Findings regarding child emotional symptoms have been inconsistent. O’Connor and colleagues (2002) found no sex-difference related to the prediction of emotional problems by prenatal maternal anxiety. Conversely, the study by O’Donnell and colleagues (2014) found girls more susceptible for emotional problems at age 4 than boys. Sandman and colleagues (2013) have suggested that the adaptive flexibility of the female fetus in gestation might render them susceptible to more subtle, but persisting, consequences and to be more vulnerable to emotional or affective problems later in life.

Strengths and Limitations

A strength of this study is the use of a validated, structured diagnostic interview for clinical assessments, which provided detailed information on psychiatric symptoms in 3-year-old children. Furthermore, the data were drawn prospectively from a large birth cohort (MoBa), which gave the opportunity to adjust for a large number of possible confounders. A validated instrument was used to assess maternal symptoms of anxiety and depression at two different time points during pregnancy and 6 months postnatally.

The study also has some methodological limitations. First, the sample selection was a two-step process, first to the MoBa, and then into the preschool ADHD study. Although the original MoBa cohort is large and recruited broadly from the general population, the response rate was only 39%, and the cohort has been shown to have an overrepresentation of mothers with high income and high educational level and an underrepresentation of young mothers, mothers living alone, mothers with more than two children, and mothers smoking during pregnancy (Nilsen et al., 2009). In the preschool ADHD study, with participation rates of only 37.5% in the sampled group and 22.5% in the control group, the participating mothers had even higher education compared with the MoBa. However, because the exposure–outcome associations are not greatly affected by selection in the MoBa (Nilsen et al., 2013), it is uncertain how such selection may affect this study outcome. Because children with severe difficulties may be underrepresented in our sample, the reported risk associations are most likely attenuated. In addition, results can only be generalized to Caucasians because these represented the great majority of participating families of the study.

Second, the selection of participants into the preschool ADHD study was based on high scores on questions about hyperactivity, impulsivity, and attention problems implying that findings cannot be generalized to the general population. However, Rothman, Hatch, and Gallacher (2014) have argued that it is the knowledge of specific conditions and an understanding of mechanisms, rather than representativeness of the study participants, that enhance the generalization. The point is that to understand a mechanism, the variables should be chosen in a way that gives the power for comparison of the variables of interest. However, one could argue that one could detect a mechanism in a specific sample that is not relevant for the larger population.

Third, measures were based on maternal reports, raising the possibility of shared method variance and reporter bias. Depressed mothers may tend to report more symptoms in children; however, maternal self-reports of anxiety and depression were 3 years prior to the assessment of symptoms in children. Moreover, our findings extend to early preschool age, and it is not clear that a single assessment is adequate for operationalizing a programming effect or a lasting developmental effect due to postnatal experiences.

This study used a brief self-report for the assessment of perinatal maternal symptoms of anxiety and depression rather than a more detailed evaluation of maternal mood. The SCL-5 has not previously been validated for pregnant women, but has been used in previous studies on this topic (Bekkhus et al., 2011; de Bruijn et al., 2009). We chose to use the scale as a continuous measure, but in addition, we examined the top 15% of the SCL-5 scores, which is in line with other studies on this topic (Clavarino et al., 2010; O’Connor et al., 2002; O’Connor et al., 2003; O’Donnell et al., 2014). However, we dichotomized the scores above the cutoff to include the reports on maternal anxiety and/or depression previous to conception.

A potential bias for our estimation of timing effects is that the assessment of prenatal maternal symptoms of anxiety or depression did not include the last 10 weeks of gestation. Preconception maternal anxiety and/or depression was reported retrospectively at 17 weeks of gestation and may be subject to potential recall bias. Furthermore, in a nonexperimental study, the observed associations may be due to unmeasured variables. The examination of the relative contribution of genetic versus environmental influence was not possible in this study, neither potential gene–environment interactions. Thus, whether genetic vulnerability is greater in those children exposed to perinatal maternal symptoms is uncertain. The observed link between perinatal maternal symptoms and child symptoms may reflect a known familial link between parental anxiety and depression and child externalizing symptoms rather than the exposure during this period. Furthermore, because maternal mood stayed relatively stable after the child’s birth, child symptoms may reflect parental behavior toward the child after birth, rather than prenatal exposure. In addition, adverse life events/stressors (Class, Lichtenstein, Langstrom, & D’Onofrio, 2011; Huizink et al., 2007), parental symptoms of ADHD, paternal anxiety or depression (O’Donnell et al., 2014), maternal comorbid conditions, or maternal personality disturbances (Mohler, Parzer, Brunner, Wiebel, & Resch, 2006) were not addressed in this study. These conditions could potentially affect maternal emotional state during pregnancy, as well as the mother’s attitude and sensitivity toward her child once born, thus playing a part in the observed associations of child symptoms. ADHD is known to be comorbid with anxiety and depression, which potentially could have modulated the link between parental and child symptoms. Although parental ADHD symptoms were reported in the MoBa questionnaires, these variables were not included because maternal reports were obtained at the same time as children were enrolled into the study, which might have introduced a bias. We excluded paternal reported ADHD symptom score, because of substantial missing data in a number of variables of the paternal reports. Thus, the question of whether maternal symptoms operate through genetic factor, prenatal biological mechanisms, or environmental factors compared with the influence on child symptoms is complicated. Predesigned adoption studies could be more appropriate for these investigations. However, a variety of covariates were adjusted for, and the associations of still more potential risk factors were explored.

Furthermore, the lack of association between perinatal maternal symptoms and child symptoms of CD and ADHD-IA may be due to lack of power because few children in this sample exhibited significant level of such symptoms. The possibility that perinatal maternal symptoms may be associated with other common childhood disorders than those examined by this study should not be ignored. Last, we did not correct for multiple testing, which would set the significance level more close to .004; thus, p values above this level should be interpreted with caution.

Clinical Implications

Prenatal maternal distress during mid-gestation was found to represent the most significant risk for preschool children’s anxiety symptoms, but also a modest risk factor for symptoms of ADHD-HI, ODD, and CD. However, the associations between the combined effects of postconception maternal distress were found more strongly related to child symptoms of ODD, CD, and anxiety than prenatal or postnatal exposure alone, which may suggest concomitant or continuing influence on child development.

Maternal symptoms of anxiety, depression, and distress are quite common mental health problems during the perinatal period, which imply that improving women’s mental health and well-being during childbearing and birth could represent a most feasible strategy for modifying the risks of developing mental health problems in children. Systematic screening should be implemented as early as possible during pregnancy, and individualized intervention commenced.

Footnotes

Acknowledgements

We are most grateful to the participating families in Norway who are taking part in this ongoing study. We also want to thank the clinicians and the research assistants for their work with children and families and the collection of data.

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.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The present study was supported by Grant 39289, from the South Eastern Health Region; funding from the Norwegian Resource Centre for ADHD, Tourette’s Syndrome, and Narcolepsy; and funding from Oslo University Hospital. Data were drawn from the Norwegian Longitudinal ADHD Study, which was supported by funds and grants from the Norwegian Health Directorate, the Norwegian Ministry of Health, the South Eastern Health Region, G&P Sorensen Fund for Scientific Research, and the Norwegian Resource Centre for ADHD, Tourette’s Syndrome, and Narcolepsy. The Norwegian Mother and Child Cohort Study was supported by the Norwegian Ministry of Health and the Ministry of Education and Research, NIH/NIEHS (Contract No. NO-ES-75558), NIH/NINDS (Grant No. 1 UO1 NS 047537-01), and the Norwegian Research Council/FUGE (Grant No. 151918/S10).

Author Biographies

Bothild Bendiksen, MD, is a specialist in child and adolescent psychiatry, a researcher for the Norwegian Longitudinal ADHD Study and at the Oslo University Hospital, Department of Psychiatric Reasearch.

Heidi Aase, PhD, is a psychologist, a principal investigator for the Norwegian Longitudinal ADHD Study and department director of the Norwegian Institute of Public Health, Division of Mental Health.

Lien My Diep, PhD, is a biostatistician at the Oslo University Hospital, Oslo Centre for Biostatistics and Epidemiology.

Elisabeth Svensson, PhD, is an epidemiologist at the University of Aarhus, Denmark, Department of Clinical Epidemiology.

Svein Friis, MD, PhD, is a professor of psychiatry at the University of Oslo and a researcher at the Oslo University Hospital, Department of Psychiatric Research.

Pål Zeiner, MD, Phd, is a specialist in child and adolescent psychiatry, a principal investigator for the Norwegian Longitudinal ADHD Study and a researcher at the Oslo University Hospital, Department of Psychiatric Research.

References

Achenbach

T. M.

Ruffle

T. M.

(2000). The Child Behavior Checklist and related forms for assessing behavioral/emotional problems and competencies. Pediatrics in Review, 21, 265-271.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: Author.

Bekkhus

Rutter

Barker

E. D.

Borge

A. I.

(2011). The role of pre- and postnatal timing of family risk factors on child behavior at 36 months. Journal of Abnormal Child Psychology, 39, 611-621.

Bendiksen

Svensson

Aase

Reichborn-Kjennerud

Friis

Myhre

A. M.

Zeiner

(2017). Co-occurrence of ODD and CD in preschool children with symptoms of ADHD. Journal of Attention Disorders, 21(9), 741-752. doi: 1087054714538655

Beydoun

Saftlas

A. F.

(2008). Physical and mental health outcomes of prenatal maternal stress in human and animal studies: A review of recent evidence. Paediatric and Perinatal Epidemiology, 22, 438-466.

Carter

A. S.

Garrity-Rokous

F. E.

Chazan-Cohen

Little

Briggs-Gowan

M. J.

(2001). Maternal depression and comorbidity: Predicting early parenting, attachment security, and toddler social-emotional problems and competencies. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 18-26.

Class

Q. A.

Abel

K. M.

Khashan

A. S.

Rickert

M. E.

Dalman

Larsson

. . . D’Onofrio

B. M.

(2014). Offspring psychopathology following preconception, prenatal and postnatal maternal bereavement stress. Psychological Medicine, 44, 71-84.

Class

Q. A.

Lichtenstein

Langstrom

D’Onofrio

B. M.

(2011). Timing of prenatal maternal exposure to severe life events and adverse pregnancy outcomes: A population study of 2.6 million pregnancies. Psychosomatic Medicine, 73, 234-241.

10.

Clavarino

A. M.

Mamun

A. A.

O’Callaghan

Aird

Bor

O’Callaghan

. . . Alati

(2010). Maternal anxiety and attention problems in children at 5 and 14 years. Journal of Attention Disorders, 13, 658-667.

11.

de Bruijn

A. T.

van Bakel

H. J.

van Baar

A. L.

(2009). Sex differences in the relation between prenatal maternal emotional complaints and child outcome. Early Human Development, 85, 319-324.

12.

Egger

H. L.

Angold

(2004). The Preschool Age Psychiatric Assessment (PAPA): A Structured Parent Interview for diagnosing psychiatric disorders in preschool children. In Del Carmen-Wiggins

Carter

(Eds.), Handbook of infant, toddler, and preschool mental health assessment (pp. 223-243). New York, NY: Oxford University Press.

13.

Egger

H. L.

Erkanli

Keeler

Potts

Walter

B. K.

Angold

(2006). Test-retest reliability of the Preschool Age Psychiatric Assessment (PAPA). Journal of the American Academy of Child & Adolescent Psychiatry, 45, 538-549.

14.

Giallo

Woolhouse

Gartland

Hiscock

Brown

(2015). The emotional-behavioural functioning of children exposed to maternal depressive symptoms across pregnancy and early childhood: A prospective Australian pregnancy cohort study. European Child & Adolescent Psychiatry, 24, 1233-1244.

15.

Glover

(2011). Annual research review: Prenatal stress and the origins of psychopathology: An evolutionary perspective. Journal of Child Psychology and Psychiatry, 52, 356-367.

16.

Glover

(2014). Maternal depression, anxiety and stress during pregnancy and child outcome; what needs to be done. Best Practice & Research Clinical Obstetrics & Gynaecology, 28, 25-35.

17.

Goodman

(1997). The Strengths and Difficulties Questionnaire: A research note. Journal of Child Psychology and Psychiatry, 38, 581-586.

18.

Henderson

J. J.

Evans

S. F.

Straton

J. A.

Priest

S. R.

Hagan

(2003). Impact of postnatal depression on breastfeeding duration. Birth, 30, 175-180.

19.

Heron

O’Connor

T. G.

Evans

Golding

Glover

(2004). The course of anxiety and depression through pregnancy and the postpartum in a community sample. Journal of Affective Disorders, 80, 65-73.

20.

Hesbacher

P. T.

Rickels

Morris

R. J.

Newman

Rosenfeld

(1980). Psychiatric illness in family practice. Journal of Clinical Psychiatry, 41, 6-10.

21.

Huizink

A. C.

Dick

D. M.

Sihvola

Pulkkinen

Rose

R. J.

Kaprio

(2007). Chernobyl exposure as stressor during pregnancy and behaviour in adolescent offspring. Acta Psychiatrica Scandinavica, 116, 438-446.

22.

Kim-Cohen

Moffitt

T. E.

Taylor

Pawlby

S. J.

Caspi

(2005). Maternal depression and children’s antisocial behavior: Nature and nurture effects. Archives of General Psychiatry, 62, 173-181.

23.

Koutra

Chatzi

Bagkeris

Vassilaki

Bitsios

Kogevinas

(2013). 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. Social Psychiatry & Psychiatric Epidemiology, 48, 1335-1345.

24.

Lahey

B. B.

Van Hulle

C. A.

Rathouz

P. J.

Rodgers

J. L.

D’Onofrio

B. M.

Waldman

I. D.

(2009). Are oppositional-defiant and hyperactive-inattentive symptoms developmental precursors to conduct problems in late childhood? Genetic and environmental links. Journal of Abnormal Child Psychology, 37, 45-58.

25.

Littleton

H. L.

Breitkopf

C. R.

Berenson

A. B.

(2007). Correlates of anxiety symptoms during pregnancy and association with perinatal outcomes: A meta-analysis. American Journal of Obstetrics & Gynecology, 196, 424-432.

26.

Loomans

E. M.

van der Stelt

van

E. M.

Gemke

R. J.

Vrijkotte

den Bergh

B. R.

(2011). Antenatal maternal anxiety is associated with problem behaviour at age five. Early Human Development, 87, 565-570.

27.

Magnus

Irgens

L. M.

Haug

Nystad

Skjaerven

Stoltenberg

(2006). Cohort profile: The Norwegian Mother and Child Cohort Study (MoBa). International Journal of Epidemiology, 35, 1146-1150.

28.

Mohler

Parzer

Brunner

Wiebel

Resch

(2006). Emotional stress in pregnancy predicts human infant reactivity. Early Human Development, 82, 731-737.

29.

Nilsen

R. M.

Suren

Gunnes

Alsaker

E. R.

Bresnahan

Hirtz

. . . Stoltenberg

(2013). Analysis of self-selection bias in a population-based cohort study of autism spectrum disorders. Paediatric and Perinatal Epidemiology, 27, 553-563.

30.

Nilsen

R. M.

Vollset

S. E.

Gjessing

H. K.

Skjaerven

Melve

K. K.

Schreuder

. . . Magnus

(2009). Self-selection and bias in a large prospective pregnancy cohort in Norway. Paediatric and Perinatal Epidemiology, 23, 597-608.

31.

Nock

M. K.

Kazdin

A. E.

Hiripi

Kessler

R. C.

(2007). Lifetime prevalence, correlates, and persistence of oppositional defiant disorder: Results from the National Comorbidity Survey Replication. Journal of Child Psychology and Psychiatry, 48, 703-713.

32.

O’Connor

T. G.

Heron

Glover

(2002). Antenatal anxiety predicts child behavioral/emotional problems independently of postnatal depression. Journal of the American Academy of Child & Adolescent Psychiatry, 41, 1470-1477.

33.

O’Connor

T. G.

Heron

Golding

Glover

(2003). Maternal antenatal anxiety and behavioural/emotional problems in children: A test of a programming hypothesis. Journal of Child Psychology and Psychiatry, 44, 1025-1036.

34.

O’Connor

T. G.

Monk

Fitelson

E. M.

(2014). Practitioner review: Maternal mood in pregnancy and child development—Implications for child psychology and psychiatry. Journal of Child Psychology and Psychiatry, 55, 99-111.

35.

O’Donnell

K. J.

Bugge

J. A.

Freeman

Khalife

O’Connor

T. G.

Glover

(2012). Maternal prenatal anxiety and downregulation of placental 11beta-HSD2. Psychoneuroendocrinology, 37, 818-826.

36.

O’Donnell

K. J.

Glover

Barker

E. D.

O’Connor

T. G.

(2014). The persisting effect of maternal mood in pregnancy on childhood psychopathology. Development and Psychopathology, 26, 393-403.

37.

Pettersson

Sjolander

Almqvist

Anckarsater

D’Onofrio

B. M.

Lichtenstein

. . . Larsson

(2015). Birth weight as an independent predictor of ADHD symptoms: A within-twin pair analysis. Journal of Child Psychology and Psychiatry, 56, 453-459.

38.

Robinson

Mattes

Oddy

W. H.

Pennell

C. E.

van

E. A.

McLean

N. J.

. . . Newnham

J. P.

(2011). Prenatal stress and risk of behavioral morbidity from age 2 to 14 years: The influence of the number, type, and timing of stressful life events. Development and Psychopathology, 23, 507-520.

39.

Rodriguez

Bohlin

(2005). Are maternal smoking and stress during pregnancy related to ADHD symptoms in children? Journal of Child Psychology and Psychiatry, 46, 246-254.

40.

Rothman

Hatch

Gallacher

(2014). Representativeness is not helpful in studying heterogeneity of effects across subgroups. International Journal of Epidemiology, 43, 633-634.

41.

Sandman

C. A.

Davis

E. P.

Glynn

L. M.

(2012). Prescient human fetuses thrive. Psychological Science, 23, 93-100.

42.

Sandman

C. A.

Glynn

L. M.

Davis

E. P.

(2013). Is there a viability-vulnerability tradeoff? Sex differences in fetal programming. Journal of Psychosomatic Research, 75, 327-335.

43.

Sciberras

Ukoumunne

O. C.

Efron

(2011). Predictors of parent-reported attention-deficit/hyperactivity disorder in children aged 6-7 years: A national longitudinal study. Journal of Abnormal Child Psychology, 39, 1025-1034.

44.

Seimyr

Edhborg

Lundh

Sjogren

(2004). In the shadow of maternal depressed mood: Experiences of parenthood during the first year after childbirth. Journal of Psychosomatic Obstetrics & Gynaecology, 25, 23-34.

45.

Shamberger

(2012). Attention-deficit disorder associated with breast-feeding: A brief report. Journal of the American College of Nutrition, 31, 239-242.

46.

Shaw

D. S.

Vondra

J. I.

(1995). Infant attachment security and maternal predictors of early behavior problems: A longitudinal study of low-income families. Journal of Abnormal Child Psychology, 23, 335-357.

47.

Stein

Craske

M. G.

Lehtonen

Harvey

Savage-McGlynn

Davies

Counsell

(2012). Maternal cognitions and mother-infant interaction in postnatal depression and generalized anxiety disorder. Journal of Abnormal Psychology, 121, 795-809.

48.

Strand

B. H.

Dalgard

O. S.

Tambs

Rognerud

(2003). Measuring the mental health status of the Norwegian population: A comparison of the instruments SCL-25, SCL-10, SCL-5 and MHI-5 (SF-36). Nordic Journal of Psychiatry, 57, 113-118.

49.

Stringaris

Goodman

(2009). Longitudinal outcome of youth oppositionality: Irritable, headstrong, and hurtful behaviors have distinctive predictions. Journal of the American Academy of Child & Adolescent Psychiatry, 48, 404-412.

50.

Van Batenburg-Eddes

Brion

M. J.

Henrichs

Jaddoe

V. W.

Hofman

Verhulst

F. C.

. . . Tiemeier

(2013). Parental depressive and anxiety symptoms during pregnancy and attention problems in children: A cross-cohort consistency study. Journal of Child Psychology and Psychiatry, 54, 591-600.

51.

Willcutt

E. G.

Nigg

J. T.

Pennington

B. F.

Solanto

M. V.

Rohde

L. A.

Tannock

. . . Lahey

B. B.

(2012). Validity of DSM-IV attention deficit/hyperactivity disorder symptom dimensions and subtypes. Journal of Abnormal Psychology, 121, 991-1010.