Association of Poor Family Functioning From Pregnancy Onward With Preadolescent Behavior and Subcortical Brain Development | Child Development | JAMA Psychiatry | JAMA Network
[Skip to Navigation]
Sign In
Figure.  Hippocampal Volume as Mediator of the Association Between Prenatal Maternal-Reported Poor Family Functioning and Preadolescent Problem Behavior
Hippocampal Volume as Mediator of the Association Between Prenatal Maternal-Reported Poor Family Functioning and Preadolescent Problem Behavior

Mediation analysis of hippocampal volumes at age 10 years in association with maternal-reported poor family functioning per FAD score during pregnancy with preadolescent problem behavior factor at age 10. B statistics are averaged from 10 imputed data sets. Model is adjusted for child age at brain MRI scan, child sex, total ICV, maternal age, race/ethnicity, education, parity, marital status, maternal psychopathology, smoking and alcohol consumption, and prior child problem behavior when child was aged 1.5 years and harsh parenting when child was aged 3 years. FAD indicates Family Assessment Device; ICV, intracranial volume; MRI, magnetic resonance imaging.

aPath A is the association of prenatal maternal-reported poor family functioning with hippocampal volume at age 10, and path B is for the association of hippocampal volume with preadolescent problem behavior factor. Path C (in black) is the total association between poor prenatal family functioning and preadolescent problem behavior with hippocampal volume not in the model. Path C′ (in red) is the direct association between prenatal maternal-reported poor family functioning and preadolescent problem behavior factor with hippocampal volume in the model.

bThe latent construct of maternal- and paternal-reported child problems. Preadolescent problem behavior factor captures covariation across raters, or the extent to which a given dimension is reflected across parents (ie, a between-rater dimension factor).

Table 1.  Baseline Characteristics
Baseline Characteristics
Table 2.  Associations of Poor Family Functioning With Brain Morphologya
Associations of Poor Family Functioning With Brain Morphologya
1.
Davies  PT, Coe  JL. Family relationship dynamics: a developmental perspective. In: Fiese BH, Celano M, Deater-Deckard K, Jouriles EN, Whisman MA, eds. APA Handbook of Contemporary Family Psychology: Foundations, Methods, and Contemporary Issues Across the Lifespan. American Psychological Association; 2019:165-185.
2.
Cummings  EM, Keller  PS, Davies  PT.  Towards a family process model of maternal and paternal depressive symptoms: exploring multiple relations with child and family functioning.   J Child Psychol Psychiatry. 2005;46(5):479-489. doi:10.1111/j.1469-7610.2004.00368.x PubMedGoogle ScholarCrossref
3.
Davies  PT, Cummings  EM, Winter  MA.  Pathways between profiles of family functioning, child security in the interparental subsystem, and child psychological problems.   Dev Psychopathol. 2004;16(3):525-550. doi:10.1017/S0954579404004651 PubMedGoogle Scholar
4.
Alderfer  MA, Fiese  BH, Gold  JI,  et al.  Evidence-based assessment in pediatric psychology: family measures.   J Pediatr Psychol. 2008;33(9):1046-1061. doi:10.1093/jpepsy/jsm083 PubMedGoogle ScholarCrossref
5.
Pechtel  P, Pizzagalli  DA.  Effects of early life stress on cognitive and affective function: an integrated review of human literature.   Psychopharmacology (Berl). 2011;214(1):55-70. doi:10.1007/s00213-010-2009-2 PubMedGoogle ScholarCrossref
6.
Thompson  RA.  Stress and child development.   Future Child. 2014;24(1):41-59. doi:10.1353/foc.2014.0004 PubMedGoogle ScholarCrossref
7.
O’Connor  TG, Heron  J, Golding  J, Beveridge  M, Glover  V.  Maternal antenatal anxiety and children’s behavioural/emotional problems at 4 years: report from the Avon Longitudinal Study of Parents and Children.   Br J Psychiatry. 2002;180:502-508. doi:10.1192/bjp.180.6.502 PubMedGoogle ScholarCrossref
8.
O’Donnell  KJ, Glover  V, Barker  ED, O’Connor  TG.  The persisting effect of maternal mood in pregnancy on childhood psychopathology.   Dev Psychopathol. 2014;26(2):393-403. doi:10.1017/S0954579414000029 PubMedGoogle ScholarCrossref
9.
Monk  C, Lugo-Candelas  C, Trumpff  C.  Prenatal developmental origins of future psychopathology: mechanisms and pathways.   Annu Rev Clin Psychol. 2019;15(1):317-344. doi:10.1146/annurev-clinpsy-050718-095539 PubMedGoogle ScholarCrossref
10.
Maccari  S, Krugers  HJ, Morley-Fletcher  S, Szyf  M, Brunton  PJ.  The consequences of early-life adversity: neurobiological, behavioural and epigenetic adaptations.   J Neuroendocrinol. 2014;26(10):707-723. doi:10.1111/jne.12175 PubMedGoogle ScholarCrossref
11.
Teicher  MH, Andersen  SL, Polcari  A, Anderson  CM, Navalta  CP, Kim  DM.  The neurobiological consequences of early stress and childhood maltreatment.   Neurosci Biobehav Rev. 2003;27(1-2):33-44. doi:10.1016/S0149-7634(03)00007-1 PubMedGoogle ScholarCrossref
12.
Teicher  MH, Tomoda  A, Andersen  SL.  Neurobiological consequences of early stress and childhood maltreatment: are results from human and animal studies comparable?   Ann N Y Acad Sci. 2006;1071(1):313-323. doi:10.1196/annals.1364.024 PubMedGoogle ScholarCrossref
13.
Qiu  A, Rifkin-Graboi  A, Chen  H,  et al.  Maternal anxiety and infants’ hippocampal development: timing matters.   Transl Psychiatry. 2013;3(9):e306. doi:10.1038/tp.2013.79 PubMedGoogle Scholar
14.
Lupien  SJ, McEwen  BS, Gunnar  MR, Heim  C.  Effects of stress throughout the lifespan on the brain, behaviour and cognition.   Nat Rev Neurosci. 2009;10(6):434-445. doi:10.1038/nrn2639 PubMedGoogle ScholarCrossref
15.
Kolb  B, Harker  A, Mychasiuk  R, de Melo  SR, Gibb  R.  Stress and prefrontal cortical plasticity in the developing brain.   Cogn Dev. 2017;42:15-26. doi:10.1016/j.cogdev.2017.01.001 Google ScholarCrossref
16.
Gould  E, Tanapat  P.  Stress and hippocampal neurogenesis.   Biol Psychiatry. 1999;46(11):1472-1479. doi:10.1016/S0006-3223(99)00247-4 PubMedGoogle ScholarCrossref
17.
Sapolsky  RM, Uno  H, Rebert  CS, Finch  CE.  Hippocampal damage associated with prolonged glucocorticoid exposure in primates.   J Neurosci. 1990;10(9):2897-2902. doi:10.1523/JNEUROSCI.10-09-02897.1990 PubMedGoogle ScholarCrossref
18.
Sapolsky  RM, McEwen  BS, Rainbow  TC.  Quantitative autoradiography of [3H]corticosterone receptors in rat brain.   Brain Res. 1983;271(2):331-334. doi:10.1016/0006-8993(83)90295-0 PubMedGoogle ScholarCrossref
19.
Patel  PD, Lopez  JF, Lyons  DM, Burke  S, Wallace  M, Schatzberg  AF.  Glucocorticoid and mineralocorticoid receptor mRNA expression in squirrel monkey brain.   J Psychiatr Res. 2000;34(6):383-392. doi:10.1016/S0022-3956(00)00035-2 PubMedGoogle ScholarCrossref
20.
Gilbertson  MW, Shenton  ME, Ciszewski  A,  et al.  Smaller hippocampal volume predicts pathologic vulnerability to psychological trauma.   Nat Neurosci. 2002;5(11):1242-1247. doi:10.1038/nn958 PubMedGoogle ScholarCrossref
21.
Hanson  JL, Nacewicz  BM, Sutterer  MJ,  et al.  Behavioral problems after early life stress: contributions of the hippocampus and amygdala.   Biol Psychiatry. 2015;77(4):314-323. doi:10.1016/j.biopsych.2014.04.020 PubMedGoogle ScholarCrossref
22.
Hair  NL, Hanson  JL, Wolfe  BL, Pollak  SD.  Association of child poverty, brain development, and academic achievement.   JAMA Pediatr. 2015;169(9):822-829. doi:10.1001/jamapediatrics.2015.1475 PubMedGoogle ScholarCrossref
23.
Tavassolie  T, Dudding  S, Madigan  AL, Thorvardarson  E, Winsler  A.  Differences in perceived parenting style between mothers and fathers: implications for child outcomes and marital conflict.   J Child Fam Stud. 2016;25(6):2055-2068. doi:10.1007/s10826-016-0376-y Google ScholarCrossref
24.
Lamb  ME. How do fathers influence children’s development? let me count the ways. In: Lamb ME, ed. The Role of the Father in Child Development. John Wiley & Sons Inc; 2010:1-26.
25.
Kooijman  MN, Kruithof  CJ, van Duijn  CM,  et al.  The Generation R Study: design and cohort update 2017.   Eur J Epidemiol. 2016;31(12):1243-1264. doi:10.1007/s10654-016-0224-9 PubMedGoogle ScholarCrossref
26.
White  T, Muetzel  RL, El Marroun  H,  et al.  Paediatric population neuroimaging and the Generation R Study: the second wave.   Eur J Epidemiol. 2018;33(1):99-125. doi:10.1007/s10654-017-0319-y PubMedGoogle ScholarCrossref
27.
Byles  J, Byrne  C, Boyle  MH, Offord  DR.  Ontario Child Health Study: reliability and validity of the general functioning subscale of the McMaster Family Assessment Device.   Fam Process. 1988;27(1):97-104. doi:10.1111/j.1545-5300.1988.00097.x PubMedGoogle ScholarCrossref
28.
Epstein  NB, Baldwin  LM, Bishop  DS.  The McMaster Family Assessment Device.   J Marital Fam Ther. 1983;9(2):171-180. doi:10.1111/j.1752-0606.1983.tb01497.x Google ScholarCrossref
29.
Achenbach  TM, Rescorla  LA.  Manual for the ASEBA Preschool Forms & Profiles: An Integrated System of Multi-Informant Assessment; Child Behavior Checklist for Ages 1 1/2-5; Language Development Survey; Caregiver-Teacher Report Form. University of Vermont; 2000.
30.
Achenbach  TM, Rescorla  LA.  ASEBA School-Age Forms & Profiles. University of Vermont; 2001.
31.
Fischl  B, van der Kouwe  A, Destrieux  C,  et al.  Automatically parcellating the human cerebral cortex.   Cereb Cortex. 2004;14(1):11-22. doi:10.1093/cercor/bhg087 PubMedGoogle ScholarCrossref
32.
Derogatis  LR, Melisaratos  N.  The Brief Symptom Inventory: an introductory report.   Psychol Med. 1983;13(3):595-605. doi:10.1017/S0033291700048017 PubMedGoogle ScholarCrossref
33.
de Beurs E.  Brief Symptom Inventory. Handleiding. Pits Publishers; 2004.
34.
Straus  MA, Hamby  SL, Finkelhor  D, Moore  DW, Runyan  D.  Identification of child maltreatment with the Parent-Child Conflict Tactics Scales: development and psychometric data for a national sample of American parents.   Child Abuse Negl. 1998;22(4):249-270. doi:10.1016/S0145-2134(97)00174-9 PubMedGoogle ScholarCrossref
35.
Benjamini  Y, Hochberg  Y.  Controlling the false discovery rate: a practical and powerful approach to multiple testing.  J R Stat Soc Series B Stat Methodol. 1995;57(1):289-300.
36.
Litman  HJ, Horton  NJ, Hernández  B, Laird  NM.  Incorporating missingness for estimation of marginal regression models with multiple source predictors.   Stat Med. 2007;26(5):1055-1068. doi:10.1002/sim.2593 PubMedGoogle ScholarCrossref
37.
Sánchez  BN, Hu  H, Litman  HJ, Téllez-Rojo  MM.  Statistical methods to study timing of vulnerability with sparsely sampled data on environmental toxicants.   Environ Health Perspect. 2011;119(3):409-415. doi:10.1289/ehp.1002453 PubMedGoogle ScholarCrossref
38.
Valeri  L, Vanderweele  TJ.  Mediation analysis allowing for exposure-mediator interactions and causal interpretation: theoretical assumptions and implementation with SAS and SPSS macros.   Psychol Methods. 2013;18(2):137-150. doi:10.1037/a0031034 PubMedGoogle ScholarCrossref
39.
Weuve  J, Tchetgen  EJ, Glymour  MM,  et al.  Accounting for bias due to selective attrition: the example of smoking and cognitive decline.   Epidemiology. 2012;23(1):119-128. doi:10.1097/EDE.0b013e318230e861 PubMedGoogle ScholarCrossref
40.
Schunk  D.  A Markov chain Monte Carlo algorithm for multiple imputation in large surveys.   Asta Adv Stat Anal. 2008;92(1):101-114. doi:10.1007/s10182-008-0053-6 Google ScholarCrossref
41.
Jabès  A, Nelson  CA.  20 years after “the ontogeny of human memory: a cognitive neuroscience perspective,” where are we?   Int J Behav Dev. 2015;39(4):293-303. doi:10.1177/0165025415575766 Google ScholarCrossref
42.
Casey  BJ, Giedd  JN, Thomas  KM.  Structural and functional brain development and its relation to cognitive development.   Biol Psychol. 2000;54(1-3):241-257. doi:10.1016/S0301-0511(00)00058-2 PubMedGoogle ScholarCrossref
43.
Knudsen  EI.  Sensitive periods in the development of the brain and behavior.   J Cogn Neurosci. 2004;16(8):1412-1425. doi:10.1162/0898929042304796 PubMedGoogle ScholarCrossref
44.
Luby  JL, Barch  DM, Belden  A,  et al.  Maternal support in early childhood predicts larger hippocampal volumes at school age.   Proc Natl Acad Sci U S A. 2012;109(8):2854-2859. doi:10.1073/pnas.1118003109 PubMedGoogle ScholarCrossref
45.
Luby  JL, Belden  A, Harms  MP, Tillman  R, Barch  DM.  Preschool is a sensitive period for the influence of maternal support on the trajectory of hippocampal development.   Proc Natl Acad Sci U S A. 2016;113(20):5742-5747. doi:10.1073/pnas.1601443113 PubMedGoogle ScholarCrossref
46.
Bick  J, Nelson  CA.  Early experience and brain development.   Wiley Interdiscip Rev Cogn Sci. 2017;8(1-2). doi:10.1002/wcs.1387 PubMedGoogle Scholar
47.
Calem  M, Bromis  K, McGuire  P, Morgan  C, Kempton  MJ.  Meta-analysis of associations between childhood adversity and hippocampus and amygdala volume in non-clinical and general population samples.   Neuroimage Clin. 2017;14:471-479. doi:10.1016/j.nicl.2017.02.016 PubMedGoogle ScholarCrossref
48.
Curtis  WJ, Cicchetti  D.  Affective facial expression processing in young children who have experienced maltreatment during the first year of life: an event-related potential study.   Dev Psychopathol. 2011;23(2):373-395. doi:10.1017/S0954579411000125 PubMedGoogle ScholarCrossref
49.
Lieslehto  J, Kiviniemi  V, Mäki  P,  et al; IMAGEN.  Early adversity and brain response to faces in young adulthood.   Hum Brain Mapp. 2017;38(9):4470-4478. doi:10.1002/hbm.23674 PubMedGoogle ScholarCrossref
50.
Nagy  K, Greenlee  MW, Kovács  G.  The lateral occipital cortex in the face perception network: an effective connectivity study.   Front Psychol. 2012;3:141. doi:10.3389/fpsyg.2012.00141 PubMedGoogle ScholarCrossref
51.
Hibar  DP, Adams  HHH, Jahanshad  N, et al. Novel genetic loci associated with hippocampal volume.  Nat Commun. 2017;8(1):13624. doi:10.1038/ncomms13624
52.
Lewis  AJ, Austin  E, Galbally  M.  Prenatal maternal mental health and fetal growth restriction: a systematic review.   J Dev Orig Health Dis. 2016;7(4):416-428. doi:10.1017/S2040174416000076 PubMedGoogle ScholarCrossref
53.
Weinstock  M.  The long-term behavioural consequences of prenatal stress.   Neurosci Biobehav Rev. 2008;32(6):1073-1086. doi:10.1016/j.neubiorev.2008.03.002 PubMedGoogle ScholarCrossref
54.
Li  J, Wang  ZN, Chen  YP,  et al.  Late gestational maternal serum cortisol is inversely associated with fetal brain growth.   Neurosci Biobehav Rev. 2012;36(3):1085-1092. doi:10.1016/j.neubiorev.2011.12.006 PubMedGoogle ScholarCrossref
55.
Hompes  T, Vrieze  E, Fieuws  S,  et al.  The influence of maternal cortisol and emotional state during pregnancy on fetal intrauterine growth.   Pediatr Res. 2012;72(3):305-315. doi:10.1038/pr.2012.70 PubMedGoogle ScholarCrossref
56.
Lindsay  KL, Buss  C, Wadhwa  PD, Entringer  S.  The interplay between maternal nutrition and stress during pregnancy: issues and considerations.   Ann Nutr Metab. 2017;70(3):191-200. doi:10.1159/000457136 PubMedGoogle ScholarCrossref
57.
Luby  J, Belden  A, Botteron  K,  et al.  The effects of poverty on childhood brain development: the mediating effect of caregiving and stressful life events.   JAMA Pediatr. 2013;167(12):1135-1142. doi:10.1001/jamapediatrics.2013.3139 PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    September 16, 2020

    Association of Poor Family Functioning From Pregnancy Onward With Preadolescent Behavior and Subcortical Brain Development

    Author Affiliations
    • 1The Generation R Study Group, Erasmus University Medical Center, Rotterdam, the Netherlands
    • 2Department of Child and Adolescent Psychiatry, Erasmus University Medical Center, Rotterdam, the Netherlands
    • 3Department of Social and Behavioral Sciences, Harvard T. H. Chan School of Public Health, Boston, Massachusetts
    • 4Department of Psychology, Bryn Mawr College, Bryn Mawr, Pennsylvania
    • 5Department of Radiology, Erasmus University Medical Center, Rotterdam, the Netherlands
    • 6Child and Adolescent Mental Health Center, Mental Health Services, the Capital Region of Denmark, Copenhagen, Denmark
    • 7Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark
    JAMA Psychiatry. 2021;78(1):29-37. doi:10.1001/jamapsychiatry.2020.2862
    Key Points

    Question  To what extent is the persistent association of poor prenatal family functioning with preadolescent problem behavior mediated by subcortical brain development?

    Findings  In this population-based cohort study of 2583 children with neuroimaging data, smaller hippocampal volumes were found in preadolescents exposed to prenatal maternal-reported poor family functioning. Smaller hippocampal volumes partially mediated the association of prenatal maternal-reported poor family functioning with preadolescent problem behavior.

    Meaning  Subcortical brain characteristics found after more than 10 years of follow-up may help clinicians understand why poor family functioning is associated with child neurodevelopment and well-being.

    Abstract

    Importance  The association of poor family functioning, a potent stressor, with child behavior is potentially long term and relevant for a person’s well-being later in life. Whether changes in brain development underlie the associations with preadolescent behavior and help identify periods of vulnerability is unclear.

    Objective  To assess the associations of poor family functioning from pregnancy onward with cortical, white matter, and subcortical volumes, and to examine the extent to which, in particular, hippocampal volume mediates the association of prenatal parental environmental exposures with child problem behavior in preadolescence.

    Design, Setting, and Participants  This population-based cohort study, conducted from April 2002 to January 2006, was embedded in Generation R, a multiethnic population-based cohort from fetal life onward. All pregnant women living in Rotterdam, the Netherlands, with an expected delivery date between April 2002 and January 2006 were invited to participate. Of the 8879 pregnant women enrolled during pregnancy, 1266 mothers with no partner data and 490 with missing family functioning data were excluded, as well as 1 sibling of 32 twin pairs. After excluding an additional 657 children with poor imaging data quality or incidental findings, the final sample consisted of 2583 mother-child pairs. Data analysis was performed from March 1, 2019, to June 28, 2019.

    Exposures  Mother- and father-rated poor family functioning was repeatedly measured by the General Functioning subscale of the Family Assessment Device.

    Main Outcomes and Measures  Our primary hypothesis, formulated after data collection but before analysis, was that poor prenatal family functioning would be associated with smaller hippocampal and amygdala volumes in late childhood. High-resolution structural neuroimaging data of children aged 10 years were collected with a single 3-T magnetic resonance imaging system. Child emotional and behavioral problems were assessed with the Child Behavior Checklist.

    Results  Data were available for 2583 children (mean [SD] age, 10.1 [0.6] years; 1315 girls [50.9%]). Data for parents included 2583 mothers (mean [SD] age, 31.1 [4.7] years; 1617 Dutch race/ethnicity [62.6%]) and 1788 fathers (mean [SD] age, 33.5 [5.3] years; 1239 Dutch race/ethnicity [69.3%]). Children exposed to prenatal maternal-reported poor family functioning had smaller hippocampal (B = −0.08; 95% CI, −0.13 to −0.02) and occipital lobe (B = −0.70; 95% CI, −1.19 to −0.21) volumes in preadolescence. There was no evidence for an association of exposure to poor family functioning at mid- or late childhood with brain morphology. Hippocampal volumes partially mediated the association of prenatal maternal-reported poor family functioning with preadolescent problem behavior (B = 0.08; 95% CI, 0.03-0.13), even after adjusting for prior child problems at age 1.5 years. Analyses of combined maternal and paternal family functioning ratings showed similar results, but associations were largely driven by maternal family functioning reports.

    Conclusions and Relevance  In this population-based cohort study, prenatal maternal-reported poor family functioning was associated with a smaller hippocampus in preadolescents. This difference in brain structure may underlie behavioral problems and is a possible neurodevelopmental manifestation of the long-term consequences of poor family functioning for the child.

    Introduction

    Poor family functioning can compromise child development; several studies in the literature refer to a range of negative exposures during childhood that are associated with mental health outcomes.1-3 Poor family functioning often includes, but is not limited to, high levels of conflict and lack of cohesion, disorganization, and poor quality of communication.4 Prior research on child brain development has highlighted the importance and long-term developmental consequences of adverse childhood experiences, often due to poor parenting and parental stress in samples of high-risk children.5 Despite this evidence, it remains unclear (1) why these negative effects persist throughout childhood, (2) at what age children are most vulnerable to poor family functioning, and (3) whether this is generalizable to poor family functioning in the general population. As a potent stressor, poor family functioning interferes with children’s ability to regulate stress physiology and may be associated with disruption in typical brain development.6

    Prenatal stressful life events and maternal anxiety and depression during pregnancy increase children’s risk for socioemotional and cognitive problems.7,8 Research has investigated the biologic correlates and mediators of these findings. These animal and human preclinical studies suggest that the hypothalamic-pituitary-adrenal axis plays a role in mediating the effects of maternal stress on the fetal brain.9-12 Furthermore, brain imaging research suggests that maternal stress is associated with changes in the limbic and frontotemporal structures of children.13 There is also a large amount of literature showing that stress in adults and similarly in children induces the production of stress hormones leading to a modulation of brain function.14 Animal studies suggest that this may be accomplished, in part, by changing the structure of neurons, especially in the hippocampus, amygdala, and prefrontal cortex.15 Overall, preclinical studies during pregnancy and childhood indicate that the hippocampus is highly susceptible to early stressful experiences16,17 because of its high density of glucocorticoid receptors18,19 and persistent postnatal neurogenesis.16

    In a clinical study of monozygotic twins discordant for trauma exposure, Gilbertson et al20 showed that combat veterans with persistent posttraumatic stress disorder (PTSD) had a smaller hippocampus volume than combat veterans without PTSD. However, the non–trauma-exposed identical twins of the combat veterans with PTSD also had a smaller hippocampus. Thus, a smaller hippocampus may also indicate a preexisting familial vulnerability factor that predisposes to pathological stress reactions in the event of a traumatic exposure.

    Several gaps in our understanding remain. First, the period of exposure assessment in prior studies varies, and exposures are rarely assessed repeatedly. Large follow-up studies with repeated measures of family functioning are needed to identify whether periods of specific vulnerability exist. Second, few prospective studies in the general population have been able to demonstrate whether structural brain changes mediate the association between childhood adversities and adjustment problems.21,22 Finally, most studies focus on maternal reports of family functioning only, whereas adding paternal reports of family functioning may capture a different aspect of family functioning or affect children differently.23,24

    We conducted a neuroimaging follow-up study of the relationship between poor family functioning from pregnancy onward and preadolescent brain development. Our primary hypothesis was that poor prenatal family functioning would be associated with smaller hippocampal and amygdala volumes in late childhood. We also postulated that these subcortical volumes would mediate the association of prenatal parental environmental exposures with measures of preadolescent problem behaviors at age 10 years. In the primary analyses, we examined global brain outcome measures, ie, total brain volume, total gray and cerebral white matter volumes, and hippocampal and amygdala volumes. This represents the first step of a hierarchical approach that is followed by secondary analyses only if any associations found in the first step are further tested in substructures.

    Methods
    Participants

    Our research was embedded in the Generation R Study, a multiethnic population-based cohort from fetal life onward.25 Briefly, all pregnant women living in Rotterdam, the Netherlands, with an expected delivery date between April 2002 and January 2006 were invited to participate. The study was approved by the Medical Ethics Committee of the Erasmus Medical Center, Rotterdam. Written informed consent was obtained from all adult participants and from both parents of minors. Participants gave written informed consent for each phase of the study (fetal, preschool, childhood, and adolescence period). In accordance with Dutch law, children must sign their own consent form starting from the age of 12 years onward. Children received oral information about the study. Of the 8879 pregnant women enrolled during pregnancy, we excluded 1266 mothers with no partner data and 490 with missing family functioning data, leaving 7123 eligible mother-child pairs with 4561 actively participating fathers. We randomly excluded 1 sibling of 32 twin pairs. Data from the late-childhood assessment wave (ie, mean child age 10 years) included a research center visit, questionnaires, and a magnetic resonance imaging (MRI) assessment.26 After excluding an additional 657 children with poor imaging data quality or incidental findings, our final sample consisted of 2583 mother-child pairs (eFigure 1 in the Supplement). This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

    Measures
    Family Assessment Device

    Family functioning was assessed using the General Functioning subscale of the Family Assessment Device (FAD), a validated self-report measure of family health and pathology consisting of 12 items (resulting scores range from 1 = not at all to 4 = poor family functioning), with higher scores indicating poor family functioning.27,28 Both mothers and fathers completed this measure at 20 weeks of pregnancy (18-25 weeks’ gestational age) and when their child was aged 10 years (late childhood). In addition, mothers completed the questionnaire when their child was aged 6 years (midchildhood). The FAD uses the Dutch term gezin, which refers only to the nuclear family (ie, siblings and parents). However, even if a pregnant woman already has a child, the wording of the FAD items makes it likely that parents would primarily have their partner in mind (eMethods in the Supplement).

    Child Problem Behavior

    The Child Behavior Checklist for Ages 1.5 to 529 and the Child Behavior Checklist for Ages 6 to 1830 were used to obtain standardized parent reports of children’s emotional and behavioral problems. We used the continuous Total Problems score (the sum of ratings on all problem items; scores range from 0 [not true] to 1 [somewhat or sometimes true] or 2 [very true or often true], with higher scores indicating more emotional and behavior problems) for children aged 10 years as our outcome measure (eMethods in the Supplement).

    Image Acquisition

    All images were acquired using the same sequence on the same 3-T 750w Discovery scanner (GE Healthcare) when children were aged 10 years.26 High-resolution, T1-weighted structural MRI data were acquired using a coronal inversion recovery fast spoiled gradient recalled sequence. Structural MRI data were processed through the FreeSurfer analysis suite, version 6.031 (Athinoula A. Martinos Center for Biomedical Imaging) (eMethods in the Supplement).

    Covariates

    Child age at MRI (based on date of birth) and sex were obtained from birth records. Maternal and paternal age were assessed at intake. Parental race/ethnicity, education, smoking, alcohol consumption, parity, marital status, and parental psychopathology (using the total score of the Brief Symptom Inventory)32,33 were assessed prenatally using self-report questionnaires. Harsh parenting was assessed when the child was aged 3 years using the Parent-Child Conflict Tactics Scale,34 a self-report questionnaire completed by the mother and father (eMethods in the Supplement).

    Statistical Analysis

    Statistical analyses of the data were performed from March 1, 2019, to June 28, 2019. First, we computed descriptive statistics and the correlations between mother- and father-reported poor family functioning scores at different time points (eTable 10 in the Supplement). Then, the prospective associations between maternal and paternal family functioning as assessed at each time point and child brain morphology were determined with separate linear regressions. We ran all models adjusting for all baseline previously mentioned confounders including maternal and paternal psychopathology. The interaction between child sex and poor family functioning was entered into the model in a separate step. In addition, we used structural equation modeling to test prenatal parental family functioning with a latent construct in relation to preadolescent brain morphology. Similarly, a latent construct based on child problem behavior reported by mothers and fathers was constructed (eMethods in the Supplement).

    We used a stepwise hierarchical approach to limit the number of comparisons. Total brain volume, cerebral white and gray matter volumes, and amygdala and hippocampus volumes were examined in relation to poor family functioning. If we observed an association with any of these brain measures, subsequent analyses of substructures were conducted to facilitate interpretation of results obtained with the primary outcome measures (eMethods in the Supplement). A visualization of primary and secondary brain measures is presented in eFigure 3 in the Supplement. False discovery rate was applied to adjust for multiple comparisons.35 We adjusted for multiple hypothesis testing of 5 outcomes—ie, total brain volume, total gray and cerebral white matter volumes, hippocampal and amygdala volumes—and the 2 relevant exposure periods (prenatal and early childhood) in the multiple testing correction (10 comparisons). Furthermore, we tested for potential periods of heightened susceptibility to adversity using repeated measures of poor family functioning measures in relation to brain outcomes36,37 (eMethods in the Supplement).

    Next, we tested whether any subcortical brain structures mediated the association between prenatal maternal-reported poor family functioning and preadolescent problem behavior factor at age 10 years. To this aim, we used a mediation analysis framework providing estimates of the natural direct effect size, the natural indirect effect size, and the total effect size.38 All models were adjusted for baseline confounders and child problem behavior when the child was aged 1.5 years.

    Inverse probability weights39 were tested to correct for any participants lost to follow-up (eMethods in the Supplement). In sensitivity analyses, all microstructural left and right hemispheres were used for their respective volumes (eTables 8 and 9 in the Supplement).

    The unstandardized β coefficients (B) and 95% CIs were calculated. All missing values (maximum percentage, maternal psychopathology = 10.8%) of the potential confounding factors were imputed using multiple imputations.40 Statistical significance was set at a 2-sided P value of less than .05. All analyses were performed using SAS software, version 9.4 (SAS Institute).

    Results

    The descriptive sample characteristics regarding parental socioeconomic factors, parental psychopathology, and child age at the time of MRI scanning are shown in Table 1. Data were available for 2583 children (mean [SD] age, 10.1 [0.6] years; 1315 [50.9%] girls). Data for parents included 2583 mothers (mean [SD] age, 31.1 [4.7] years; 1617 [62.6%] Dutch race/ethnicity) and 1788 fathers (mean [SD] age, 33.5 [5.3] years; 1239 [69.3%] Dutch race/ethnicity).

    As shown in Table 2, prenatal maternal-reported poor family functioning was associated with a decreased total brain volume, cerebral white matter volume, and total gray volume in late childhood (model 1, B = −26.8 [95% CI, −34.6 to −18.9]; B = −9.76 [95% CI, −13.3 to −6.20]; B = −16.7 [95% CI, −21.3 to −12.2], respectively; P < .001), but these associations did not survive correction for multiple testing. Poor prenatal family functioning was associated with a smaller hippocampal volume after adjusting for intracranial volume, an association that remained after correction for multiple testing (B = −0.08; 95% CI, −0.13 to −0.02). Adjusting for harsh parenting also did not meaningfully change this association. We observed no association between poor family functioning and amygdala volume (model 1, B = −0.01 [95% CI, −0.03 to 0.02]; P = .59).

    We observed no associations between mid- or late-childhood poor family functioning scores and any measure of brain morphology in fully adjusted models. Concurrent associations between late-childhood family functioning and brain outcomes are depicted in eTable 1 in the Supplement. Windows of susceptibility results showed the associations of repeated maternal-reported family functioning with hippocampal volume (poor family functioning × exposure period interaction P = .01), but no other brain outcomes varied by the timing of family functioning measurement (eResults in the Supplement).

    Exposure to prenatal maternal-reported poor family functioning was associated with smaller occipital lobe volume (B = −0.70; 95% CI, −1.19 to −0.21). We further explored the nominally significant anatomical findings and present the results of the relation between poor family functioning and the occipital lobe stratified by regions (eTable 3 in the Supplement). These post hoc analyses suggest that children prenatally exposed to poor family adjustment have a smaller lateral occipital lobe (B = −0.47 [95% CI, −0.61 to −0.09]; P = .01). In contrast, we observed no associations between any family functioning score and temporal, frontal, and parietal lobar volumes (eTable 2 in the Supplement). Similarly, no associations were found between family functioning and thalamus, accumbens, caudate, and putamen volumes (eTable 4 in the Supplement).

    After adjusting for socioeconomic factors and paternal psychopathology, we observed no associations between paternal-reported family functioning at either time point and brain morphology (eTables 5 and 6 in the Supplement). We found no interaction by child sex in the association between family functioning and any brain measure. The results using the prenatal parental family functioning factor reflect the common variance in the associations of maternal and paternal family functioning with preadolescent brain outcomes. These results were very similar to those of the unique prenatal maternal-reported associations (eFigure 2 in the Supplement).

    As the Figure illustrates, hippocampal volume partially mediated the association of prenatal maternal-reported poor family functioning with preadolescent problem behavior factor at age 10 years (B = 0.08; 95% CI, 0.03-0.13). The observed indirect association suggests that lower hippocampal volumes account for a portion of the observed preadolescent problem behavior in late childhood. When we adjusted for preexisting child problem behavior at age 1.5 years, we found no meaningful change in mediation results.

    Last, in order to ascertain whether selection bias substantially altered any associations, we weighted complete cases by the inverse of their probability of being a complete case to address a possible source of bias due to selection. Results were essentially unchanged (eTable 7 in the Supplement).

    Discussion

    This cohort study of children from fetal life onward suggests that poor maternal-reported prenatal family functioning is associated with brain development in late childhood. In particular, we observed smaller hippocampal volumes in children exposed to poor family functioning occurring prenatally but not in mid- or late childhood. The association remained when we accounted for parental psychopathology and harsh parenting, indicating a unique association of poor prenatal family functioning with differences in preadolescent brain development. The contribution of prenatal maternal-reported poor family functioning to preadolescent problem behavior was partially mediated by hippocampal volumes. Interestingly, prenatal maternal-reported poor family functioning was associated with smaller occipital lobe volumes. Associations between poor family functioning and brain outcomes did not differ by child sex.

    The vulnerability of the hippocampus to prenatal family functioning is consistent with previous studies reporting that the hippocampus matures rapidly and is functional very early in childhood.41,42 That the association between poor family functioning and hippocampal volumes was observed only from prenatal maternal-reported family functioning and not from mid- or late-childhood family functioning may reflect a sensitive period, which occurs early in life.43 Other research supports this inference. For example, higher levels of early-life maternal support have been linked to increased volume of the hippocampus.44,45 Our key finding, namely the interaction of poor family functioning with child age, suggests that pregnancy is a vulnerable period when development in response to parental care disruptions is maximally dynamic.46

    In contrast to our hypothesis, we were not able to demonstrate an association between poor family functioning and amygdala volumes. The lack of a discernible sensitive period to family functioning for amygdala development is consistent with previous studies of children exposed to adversity, which have found no difference in amygdala volume in adults.47

    The present findings provide evidence for a smaller occipital lobe in children exposed to prenatal maternal-reported poor family functioning. This observation, which was not expected a priori, should be interpreted with caution until it is replicated. However, the face-processing systems relating to occipital regions, in particular the lateral occipital lobe, were found to be particularly vulnerable to early-life adversities.48-50

    Furthermore, we found that the association between prenatal maternal-reported poor family functioning and preadolescent problem behavior was partially mediated by hippocampal volumes. This may suggest that brain morphologic changes precede or may even contribute to behavioral changes. Our results are consistent with the extant literature, showing that smaller hippocampal volumes partially mediated the contribution of early-life stress to higher levels of behavioral problems.21 However, it is likely that the associations in the mediation model are more complex, and they may well be bidirectional. A sample with multiple repeated measures of imaging data starting early in childhood would be necessary to test the directionality between behavior and brain development. Indeed, a twin study in veterans with PTSD showed that a smaller hippocampus may reflect a preexisting vulnerability to stress and thus reverse causality.20 Alternatively, the difference in hippocampal volume could be explained by genetic variation. Recently, a genome-wide association meta-analysis identified a few genetic loci associated with hippocampal volume,51 which could be (indirectly) associated with poor family functioning.

    In addition, associations between paternal-reported family functioning and brain structural measures did not remain after adjustment for sociodemographic factors and paternal psychopathology. Although prenatal parental family functioning factors reflect a common variance across mother- and father-reported family functioning, their association with brain structural measures was largely driven by the maternal report. Thus, the clear association found using maternal-reported functioning during pregnancy suggests that direct maternal physiological changes may underlie the findings. This is consistent with the developmental origins hypothesis that the prenatal or early postnatal environment can be associated with negative health outcomes later in life. Maternal psychological distress may lead to a suboptimal intrauterine environment with long-term consequences for the growth and health of the child.52-54 Intrauterine stress exposure may affect child development via dysregulation of the hypothalamic-pituitary-adrenal axis, but it may also affect brain development through inflammatory responses and changes in the balance of the autonomic nervous system.55 Another potential mechanism is dietary behavior and poor nutrition by which a variation in maternal nutrition (either a surplus or paucity of maternal nutrition) plays multiple roles in the health outcomes of children.56 However, postnatal experiences cannot be ruled out as a mechanism underlying our findings, because the prenatal period could be a marker of exposures in the early postnatal period, such as poor parenting.57 Thus, children of parents with poor family functioning may be more likely to experience a less optimal environment, which underlies the relation with brain developmental differences.

    Parental psychopathology remains another important mechanism potentially underlying our observations. However, when we adjusted for parental psychopathology, we found that the association between poor prenatal family functioning and hippocampal volumes was, if anything, stronger. Thus, our results suggest that poor family functioning and parental psychopathology are closely associated and may predispose each other,13 but higher levels of parental psychopathology did not account for the association of poor family functioning with hippocampal volume.

    Limitations and Strengths

    The current study has several limitations. First, this study has a population-based design, but the relative homogeneity of the population limits its generalizability. Second, we found an association between poor prenatal family functioning and preadolescent brain morphology among children aged 10 years. Although we assessed prenatal family functioning, we cannot establish whether these associations result from strictly prenatal exposures or whether our measure indexes childhood exposure during the period up to age 6 years when parents were reassessed. Third, because poor family functioning was associated with brain findings in children aged 10 years, it is possible that the associations of family functioning reported prenatally had their effects in utero. However, because no scans were obtained before age 10, this cannot be determined. Furthermore, we were unable to examine whether the parental hippocampus is a marker of vulnerability that increases the likelihood of poor family adjustment and whether this propensity is transmitted genetically to the children. Strengths of the present study are the large number of participants and broad spectrum of measured covariates, which enabled us to adjust for multiple confounders. Because of our longitudinal design, we were able to look at possible sensitive periods by leveraging baseline and repeated assessments of poor family functioning reported by both mothers and fathers.

    Conclusions

    In summary, the findings of this cohort study suggest that prenatal maternal-reported poor family functioning is associated with smaller hippocampal and occipital lobe volumes in preadolescents. Importantly, no such association was found for poor family functioning reported later in childhood, ie, at ages 6 and 10 years, suggesting that there is a sensitive period for the associations of poor family functioning during pregnancy with hippocampal and occipital lobe development. The association of maternal-reported poor family functioning during pregnancy with preadolescent problem behavior was partially mediated by hippocampal volume. That the associations between poor prenatal family functioning and hippocampal volumes were found after more than 10 years of follow-up may help clarify why poor family functioning is associated with child neurodevelopment and well-being. This study increases our understanding of how poor family functioning shapes brain and behavioral development and underscores the need to search for effective family interventions.

    Back to top
    Article Information

    Accepted for Publication: July 19, 2020.

    Corresponding Author: Henning Tiemeier, MD, PhD, Department of Social and Behavioral Sciences, Harvard T. H. Chan School of Public Health, 677 Huntington Ave, Kresge Building, Room 619, Boston, MA 02115 (tiemeier@hsph.harvard.edu).

    Published Online: September 16, 2020. doi:10.1001/jamapsychiatry.2020.2862

    Author Contributions: Dr Tiemeier had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Xerxa, Rescorla, Tiemeier.

    Acquisition, analysis, or interpretation of data: Xerxa, Delaney, Hillegers, White, Verhulst, Muetzel, Tiemeier.

    Drafting of the manuscript: Xerxa, Muetzel.

    Critical revision of the manuscript for important intellectual content: All authors.

    Statistical analysis: Xerxa, Delaney, Rescorla.

    Obtained funding: White, Tiemeier.

    Administrative, technical, or material support: Xerxa, White, Muetzel.

    Supervision: Verhulst, Tiemeier.

    Conflict of Interest Disclosures: Dr Rescorla reported receiving a patent to copyright issued and with royalties paid during the course of the submitted work. Dr Hillegers reported receiving grants from Zon MW and grants from the Dutch Ministry of Health during the conduct of the study. Dr Verhulst reported receiving personal fees from Erasmus University Medical Center outside the submitted work and is the contributing editor of the Achenbach System of Empirically Based Assessment, from which he receives remuneration. Dr Muetzel reported receiving grants from the Sophia Foundation and grants from the Erasmus Fellowship during the conduct of the study. No other disclosures were reported.

    Funding/Support: The Generation R Study is supported by the Erasmus Medical Center-Rotterdam, the Erasmus University Rotterdam, the Netherlands Organization for Health Research and Development, the Netherlands Organization for Scientific Research, the Ministry of Health, Welfare and Sport, the Municipal Health Service Rotterdam area, and the Stichting Trombosedienst and Artsenlaboratorium Rijnmond. This study was supported by grant 016.VICI.170.200 from the Netherlands Organization for Scientific Research (Dr Tiemeier), the Consortium on Individual Development (Dr Tiemeier), grant FP7/2007-2013 from the European Union Seventh Framework Program (Dr Tiemeier), and grant 602768 from ACTION: Aggression in Children (Dr Tiemeier).

    Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Meeting Presentation: This work was presented at the biennial meeting of the Society of Research in Child Development; March 23, 2019; Baltimore, Maryland.

    Additional Contributions: We thank the participating parents and their children, general practitioners, hospitals, midwives, and pharmacies for their contributions. There was no financial compensation for these contributions.

    Additional Information: The Generation R Study is conducted by the Erasmus Medical Center, Rotterdam, in close collaboration with the Faculty of Social Sciences of the Erasmus University Rotterdam, the Municipal Health Service, Rotterdam Homecare Foundation, and Stichting Trombosedienst & Artsenlaboratorium Rijnmond.

    References
    1.
    Davies  PT, Coe  JL. Family relationship dynamics: a developmental perspective. In: Fiese BH, Celano M, Deater-Deckard K, Jouriles EN, Whisman MA, eds. APA Handbook of Contemporary Family Psychology: Foundations, Methods, and Contemporary Issues Across the Lifespan. American Psychological Association; 2019:165-185.
    2.
    Cummings  EM, Keller  PS, Davies  PT.  Towards a family process model of maternal and paternal depressive symptoms: exploring multiple relations with child and family functioning.   J Child Psychol Psychiatry. 2005;46(5):479-489. doi:10.1111/j.1469-7610.2004.00368.x PubMedGoogle ScholarCrossref
    3.
    Davies  PT, Cummings  EM, Winter  MA.  Pathways between profiles of family functioning, child security in the interparental subsystem, and child psychological problems.   Dev Psychopathol. 2004;16(3):525-550. doi:10.1017/S0954579404004651 PubMedGoogle Scholar
    4.
    Alderfer  MA, Fiese  BH, Gold  JI,  et al.  Evidence-based assessment in pediatric psychology: family measures.   J Pediatr Psychol. 2008;33(9):1046-1061. doi:10.1093/jpepsy/jsm083 PubMedGoogle ScholarCrossref
    5.
    Pechtel  P, Pizzagalli  DA.  Effects of early life stress on cognitive and affective function: an integrated review of human literature.   Psychopharmacology (Berl). 2011;214(1):55-70. doi:10.1007/s00213-010-2009-2 PubMedGoogle ScholarCrossref
    6.
    Thompson  RA.  Stress and child development.   Future Child. 2014;24(1):41-59. doi:10.1353/foc.2014.0004 PubMedGoogle ScholarCrossref
    7.
    O’Connor  TG, Heron  J, Golding  J, Beveridge  M, Glover  V.  Maternal antenatal anxiety and children’s behavioural/emotional problems at 4 years: report from the Avon Longitudinal Study of Parents and Children.   Br J Psychiatry. 2002;180:502-508. doi:10.1192/bjp.180.6.502 PubMedGoogle ScholarCrossref
    8.
    O’Donnell  KJ, Glover  V, Barker  ED, O’Connor  TG.  The persisting effect of maternal mood in pregnancy on childhood psychopathology.   Dev Psychopathol. 2014;26(2):393-403. doi:10.1017/S0954579414000029 PubMedGoogle ScholarCrossref
    9.
    Monk  C, Lugo-Candelas  C, Trumpff  C.  Prenatal developmental origins of future psychopathology: mechanisms and pathways.   Annu Rev Clin Psychol. 2019;15(1):317-344. doi:10.1146/annurev-clinpsy-050718-095539 PubMedGoogle ScholarCrossref
    10.
    Maccari  S, Krugers  HJ, Morley-Fletcher  S, Szyf  M, Brunton  PJ.  The consequences of early-life adversity: neurobiological, behavioural and epigenetic adaptations.   J Neuroendocrinol. 2014;26(10):707-723. doi:10.1111/jne.12175 PubMedGoogle ScholarCrossref
    11.
    Teicher  MH, Andersen  SL, Polcari  A, Anderson  CM, Navalta  CP, Kim  DM.  The neurobiological consequences of early stress and childhood maltreatment.   Neurosci Biobehav Rev. 2003;27(1-2):33-44. doi:10.1016/S0149-7634(03)00007-1 PubMedGoogle ScholarCrossref
    12.
    Teicher  MH, Tomoda  A, Andersen  SL.  Neurobiological consequences of early stress and childhood maltreatment: are results from human and animal studies comparable?   Ann N Y Acad Sci. 2006;1071(1):313-323. doi:10.1196/annals.1364.024 PubMedGoogle ScholarCrossref
    13.
    Qiu  A, Rifkin-Graboi  A, Chen  H,  et al.  Maternal anxiety and infants’ hippocampal development: timing matters.   Transl Psychiatry. 2013;3(9):e306. doi:10.1038/tp.2013.79 PubMedGoogle Scholar
    14.
    Lupien  SJ, McEwen  BS, Gunnar  MR, Heim  C.  Effects of stress throughout the lifespan on the brain, behaviour and cognition.   Nat Rev Neurosci. 2009;10(6):434-445. doi:10.1038/nrn2639 PubMedGoogle ScholarCrossref
    15.
    Kolb  B, Harker  A, Mychasiuk  R, de Melo  SR, Gibb  R.  Stress and prefrontal cortical plasticity in the developing brain.   Cogn Dev. 2017;42:15-26. doi:10.1016/j.cogdev.2017.01.001 Google ScholarCrossref
    16.
    Gould  E, Tanapat  P.  Stress and hippocampal neurogenesis.   Biol Psychiatry. 1999;46(11):1472-1479. doi:10.1016/S0006-3223(99)00247-4 PubMedGoogle ScholarCrossref
    17.
    Sapolsky  RM, Uno  H, Rebert  CS, Finch  CE.  Hippocampal damage associated with prolonged glucocorticoid exposure in primates.   J Neurosci. 1990;10(9):2897-2902. doi:10.1523/JNEUROSCI.10-09-02897.1990 PubMedGoogle ScholarCrossref
    18.
    Sapolsky  RM, McEwen  BS, Rainbow  TC.  Quantitative autoradiography of [3H]corticosterone receptors in rat brain.   Brain Res. 1983;271(2):331-334. doi:10.1016/0006-8993(83)90295-0 PubMedGoogle ScholarCrossref
    19.
    Patel  PD, Lopez  JF, Lyons  DM, Burke  S, Wallace  M, Schatzberg  AF.  Glucocorticoid and mineralocorticoid receptor mRNA expression in squirrel monkey brain.   J Psychiatr Res. 2000;34(6):383-392. doi:10.1016/S0022-3956(00)00035-2 PubMedGoogle ScholarCrossref
    20.
    Gilbertson  MW, Shenton  ME, Ciszewski  A,  et al.  Smaller hippocampal volume predicts pathologic vulnerability to psychological trauma.   Nat Neurosci. 2002;5(11):1242-1247. doi:10.1038/nn958 PubMedGoogle ScholarCrossref
    21.
    Hanson  JL, Nacewicz  BM, Sutterer  MJ,  et al.  Behavioral problems after early life stress: contributions of the hippocampus and amygdala.   Biol Psychiatry. 2015;77(4):314-323. doi:10.1016/j.biopsych.2014.04.020 PubMedGoogle ScholarCrossref
    22.
    Hair  NL, Hanson  JL, Wolfe  BL, Pollak  SD.  Association of child poverty, brain development, and academic achievement.   JAMA Pediatr. 2015;169(9):822-829. doi:10.1001/jamapediatrics.2015.1475 PubMedGoogle ScholarCrossref
    23.
    Tavassolie  T, Dudding  S, Madigan  AL, Thorvardarson  E, Winsler  A.  Differences in perceived parenting style between mothers and fathers: implications for child outcomes and marital conflict.   J Child Fam Stud. 2016;25(6):2055-2068. doi:10.1007/s10826-016-0376-y Google ScholarCrossref
    24.
    Lamb  ME. How do fathers influence children’s development? let me count the ways. In: Lamb ME, ed. The Role of the Father in Child Development. John Wiley & Sons Inc; 2010:1-26.
    25.
    Kooijman  MN, Kruithof  CJ, van Duijn  CM,  et al.  The Generation R Study: design and cohort update 2017.   Eur J Epidemiol. 2016;31(12):1243-1264. doi:10.1007/s10654-016-0224-9 PubMedGoogle ScholarCrossref
    26.
    White  T, Muetzel  RL, El Marroun  H,  et al.  Paediatric population neuroimaging and the Generation R Study: the second wave.   Eur J Epidemiol. 2018;33(1):99-125. doi:10.1007/s10654-017-0319-y PubMedGoogle ScholarCrossref
    27.
    Byles  J, Byrne  C, Boyle  MH, Offord  DR.  Ontario Child Health Study: reliability and validity of the general functioning subscale of the McMaster Family Assessment Device.   Fam Process. 1988;27(1):97-104. doi:10.1111/j.1545-5300.1988.00097.x PubMedGoogle ScholarCrossref
    28.
    Epstein  NB, Baldwin  LM, Bishop  DS.  The McMaster Family Assessment Device.   J Marital Fam Ther. 1983;9(2):171-180. doi:10.1111/j.1752-0606.1983.tb01497.x Google ScholarCrossref
    29.
    Achenbach  TM, Rescorla  LA.  Manual for the ASEBA Preschool Forms & Profiles: An Integrated System of Multi-Informant Assessment; Child Behavior Checklist for Ages 1 1/2-5; Language Development Survey; Caregiver-Teacher Report Form. University of Vermont; 2000.
    30.
    Achenbach  TM, Rescorla  LA.  ASEBA School-Age Forms & Profiles. University of Vermont; 2001.
    31.
    Fischl  B, van der Kouwe  A, Destrieux  C,  et al.  Automatically parcellating the human cerebral cortex.   Cereb Cortex. 2004;14(1):11-22. doi:10.1093/cercor/bhg087 PubMedGoogle ScholarCrossref
    32.
    Derogatis  LR, Melisaratos  N.  The Brief Symptom Inventory: an introductory report.   Psychol Med. 1983;13(3):595-605. doi:10.1017/S0033291700048017 PubMedGoogle ScholarCrossref
    33.
    de Beurs E.  Brief Symptom Inventory. Handleiding. Pits Publishers; 2004.
    34.
    Straus  MA, Hamby  SL, Finkelhor  D, Moore  DW, Runyan  D.  Identification of child maltreatment with the Parent-Child Conflict Tactics Scales: development and psychometric data for a national sample of American parents.   Child Abuse Negl. 1998;22(4):249-270. doi:10.1016/S0145-2134(97)00174-9 PubMedGoogle ScholarCrossref
    35.
    Benjamini  Y, Hochberg  Y.  Controlling the false discovery rate: a practical and powerful approach to multiple testing.  J R Stat Soc Series B Stat Methodol. 1995;57(1):289-300.
    36.
    Litman  HJ, Horton  NJ, Hernández  B, Laird  NM.  Incorporating missingness for estimation of marginal regression models with multiple source predictors.   Stat Med. 2007;26(5):1055-1068. doi:10.1002/sim.2593 PubMedGoogle ScholarCrossref
    37.
    Sánchez  BN, Hu  H, Litman  HJ, Téllez-Rojo  MM.  Statistical methods to study timing of vulnerability with sparsely sampled data on environmental toxicants.   Environ Health Perspect. 2011;119(3):409-415. doi:10.1289/ehp.1002453 PubMedGoogle ScholarCrossref
    38.
    Valeri  L, Vanderweele  TJ.  Mediation analysis allowing for exposure-mediator interactions and causal interpretation: theoretical assumptions and implementation with SAS and SPSS macros.   Psychol Methods. 2013;18(2):137-150. doi:10.1037/a0031034 PubMedGoogle ScholarCrossref
    39.
    Weuve  J, Tchetgen  EJ, Glymour  MM,  et al.  Accounting for bias due to selective attrition: the example of smoking and cognitive decline.   Epidemiology. 2012;23(1):119-128. doi:10.1097/EDE.0b013e318230e861 PubMedGoogle ScholarCrossref
    40.
    Schunk  D.  A Markov chain Monte Carlo algorithm for multiple imputation in large surveys.   Asta Adv Stat Anal. 2008;92(1):101-114. doi:10.1007/s10182-008-0053-6 Google ScholarCrossref
    41.
    Jabès  A, Nelson  CA.  20 years after “the ontogeny of human memory: a cognitive neuroscience perspective,” where are we?   Int J Behav Dev. 2015;39(4):293-303. doi:10.1177/0165025415575766 Google ScholarCrossref
    42.
    Casey  BJ, Giedd  JN, Thomas  KM.  Structural and functional brain development and its relation to cognitive development.   Biol Psychol. 2000;54(1-3):241-257. doi:10.1016/S0301-0511(00)00058-2 PubMedGoogle ScholarCrossref
    43.
    Knudsen  EI.  Sensitive periods in the development of the brain and behavior.   J Cogn Neurosci. 2004;16(8):1412-1425. doi:10.1162/0898929042304796 PubMedGoogle ScholarCrossref
    44.
    Luby  JL, Barch  DM, Belden  A,  et al.  Maternal support in early childhood predicts larger hippocampal volumes at school age.   Proc Natl Acad Sci U S A. 2012;109(8):2854-2859. doi:10.1073/pnas.1118003109 PubMedGoogle ScholarCrossref
    45.
    Luby  JL, Belden  A, Harms  MP, Tillman  R, Barch  DM.  Preschool is a sensitive period for the influence of maternal support on the trajectory of hippocampal development.   Proc Natl Acad Sci U S A. 2016;113(20):5742-5747. doi:10.1073/pnas.1601443113 PubMedGoogle ScholarCrossref
    46.
    Bick  J, Nelson  CA.  Early experience and brain development.   Wiley Interdiscip Rev Cogn Sci. 2017;8(1-2). doi:10.1002/wcs.1387 PubMedGoogle Scholar
    47.
    Calem  M, Bromis  K, McGuire  P, Morgan  C, Kempton  MJ.  Meta-analysis of associations between childhood adversity and hippocampus and amygdala volume in non-clinical and general population samples.   Neuroimage Clin. 2017;14:471-479. doi:10.1016/j.nicl.2017.02.016 PubMedGoogle ScholarCrossref
    48.
    Curtis  WJ, Cicchetti  D.  Affective facial expression processing in young children who have experienced maltreatment during the first year of life: an event-related potential study.   Dev Psychopathol. 2011;23(2):373-395. doi:10.1017/S0954579411000125 PubMedGoogle ScholarCrossref
    49.
    Lieslehto  J, Kiviniemi  V, Mäki  P,  et al; IMAGEN.  Early adversity and brain response to faces in young adulthood.   Hum Brain Mapp. 2017;38(9):4470-4478. doi:10.1002/hbm.23674 PubMedGoogle ScholarCrossref
    50.
    Nagy  K, Greenlee  MW, Kovács  G.  The lateral occipital cortex in the face perception network: an effective connectivity study.   Front Psychol. 2012;3:141. doi:10.3389/fpsyg.2012.00141 PubMedGoogle ScholarCrossref
    51.
    Hibar  DP, Adams  HHH, Jahanshad  N, et al. Novel genetic loci associated with hippocampal volume.  Nat Commun. 2017;8(1):13624. doi:10.1038/ncomms13624
    52.
    Lewis  AJ, Austin  E, Galbally  M.  Prenatal maternal mental health and fetal growth restriction: a systematic review.   J Dev Orig Health Dis. 2016;7(4):416-428. doi:10.1017/S2040174416000076 PubMedGoogle ScholarCrossref
    53.
    Weinstock  M.  The long-term behavioural consequences of prenatal stress.   Neurosci Biobehav Rev. 2008;32(6):1073-1086. doi:10.1016/j.neubiorev.2008.03.002 PubMedGoogle ScholarCrossref
    54.
    Li  J, Wang  ZN, Chen  YP,  et al.  Late gestational maternal serum cortisol is inversely associated with fetal brain growth.   Neurosci Biobehav Rev. 2012;36(3):1085-1092. doi:10.1016/j.neubiorev.2011.12.006 PubMedGoogle ScholarCrossref
    55.
    Hompes  T, Vrieze  E, Fieuws  S,  et al.  The influence of maternal cortisol and emotional state during pregnancy on fetal intrauterine growth.   Pediatr Res. 2012;72(3):305-315. doi:10.1038/pr.2012.70 PubMedGoogle ScholarCrossref
    56.
    Lindsay  KL, Buss  C, Wadhwa  PD, Entringer  S.  The interplay between maternal nutrition and stress during pregnancy: issues and considerations.   Ann Nutr Metab. 2017;70(3):191-200. doi:10.1159/000457136 PubMedGoogle ScholarCrossref
    57.
    Luby  J, Belden  A, Botteron  K,  et al.  The effects of poverty on childhood brain development: the mediating effect of caregiving and stressful life events.   JAMA Pediatr. 2013;167(12):1135-1142. doi:10.1001/jamapediatrics.2013.3139 PubMedGoogle ScholarCrossref
    ×