Independent Prediction of Child Psychiatric Symptoms by Maternal Mental Health and Child Polygenic Risk Scores

Lawrence M Chen,Irina Pokhvisneva,Marius Lahti-Pulkkinen,Tuomas Kvist,Jessie R Baldwin,Carine Parent,Patricia P Silveira,Jari Lahti,Katri Räikkönen,Vivette Glover,Thomas G O'Connor,Michael J Meaney,Kieran J O'Donnell

DOI: https://doi.org/10.1016/j.jaac.2023.08.018

Abstract:Objective: Prenatal maternal symptoms of depression and anxiety are associated with an increased risk for child socioemotional and behavioral difficulties, supporting the fetal origins of mental health hypothesis. However, to date, studies have not considered specific genomic risk as a possible confound. Method: The Avon Longitudinal Study of Parents and Children (ALSPAC) cohort (n = 5,546) was used to test if child polygenic risk score for attention-deficit/hyperactivity disorder (ADHD), schizophrenia, or depression confounds or modifies the impact of prenatal maternal depression and anxiety on child internalizing, externalizing, and total emotional/behavioral symptoms from age 4 to 16 years. Longitudinal child and adolescent symptom data were analyzed in the ALSPAC cohort using generalized estimating equations. Replication analyses were done in an independent cohort (Prevention of Preeclampsia and Intrauterine Growth Restriction [PREDO] cohort; n = 514) from Finland, which provided complementary measures of maternal mental health and child psychiatric symptoms. Results: Maternal depression and anxiety and child polygenic risk scores independently and additively predicted behavioral and emotional symptoms from childhood through mid-adolescence. There was a robust prediction of child and adolescent symptoms from both prenatal maternal depression (generalized estimating equation estimate = 0.093, 95% CI 0.065-0.121, p = 2.66 × 10-10) and anxiety (generalized estimating equation estimate = 0.065, 95% CI 0.037-0.093, p = 1.62 × 10-5) after adjusting for child genomic risk for mental disorders. There was a similar independent effect of maternal depression (B = 0.156, 95% CI 0.066-0.246, p = .001) on child symptoms in the PREDO cohort. Genetically informed sensitivity analyses suggest that shared genetic risk only partially explains the reported association between prenatal maternal depression and offspring mental health. Conclusion: These findings highlight the genomic contribution to the fetal origins of mental health hypothesis and further evidence that prenatal maternal depression and anxiety are robust in utero risks for child and adolescent psychiatric symptoms. Plain language summary: Depression and anxiety affect approximately 15% of pregnant women, and children exposed to maternal depression or anxiety during pregnancy are at higher risk of developing mental health problems. However, the degree to which shared genetics explains the association between maternal and child mental health is unknown. In this study the authors generated polygenic risk scores (PRS), which provide a single measure of genetic risk for complex traits, to investigate the impact of shared genetic risk on the development of childhood mental health problems. Utilizing two longitudinal studies (n = 6,060), the authors found that PRS only partially explained the association between prenatal maternal depression and childhood mental health problems. These analyses show prenatal maternal depression remained a significant predictor of childhood mental health problems after accounting for shared genetic risk, further highlighting that prenatal maternal mental health is a robust predictor of child and adolescent mental health problems.