Using machine learning techniques on a large dataset of prenatal and perinatal information, researchers identified 17 of 40 potential factors that are remarkably strong predictors of ADHD symptoms in children. By focusing on data collected during pregnancy and childbirth, such as maternal health, substance use, complications during childbirth, and birth metrics, the study aimed to understand how these early life factors may influence the development of ADHD symptoms later in childhood.
A recent study, published by researchers from the Department of Psychiatry at the Royal College of Surgeons in Ireland, explores whether prenatal and perinatal factors (those occurring before and shortly after birth) can help predict symptoms of Attention Deficit Hyperactivity Disorder (ADHD) in childhood.
It also examines whether the accuracy of this type of prediction varies across subgroups of a larger population, using the ABCD (Adolescent Brain Cognitive Development) cohort database from the United States, which includes nearly 10,000 children aged 9 to 10.
To begin, it is helpful to understand what prenatal and perinatal factors are. These factors include a variety of information related to the gestational period and delivery, such as maternal substance use during pregnancy, complications during delivery, newborn weight and health conditions, as well as demographic factors such as parental age and socioeconomic status.
In this study, 40 of these variables known at birth were analyzed as potential predictors of the future development of ADHD symptoms.
ADHD is characterized by symptoms of inattention, impulsivity, and, in some cases, hyperactivity, which can negatively affect a child’s ability to concentrate, organize, and control impulses.
Many studies suggest that a combination of genetic and environmental factors contribute to the development of ADHD. In the current study, the researchers wanted to see if these environmental and biological variables related to birth could predict the onset of these symptoms with any accuracy.
To test this hypothesis, the researchers used a statistical method called “elastic net regression” with five-fold validation. This method helps identify which prenatal and perinatal variables are most powerful in predicting ADHD symptoms. The statistical analysis showed that, among the 40 variables, 17 stood out as consistent predictors of ADHD symptoms.
However, the model’s ability to predict symptoms was moderate, explaining about 8.13% of the variance in ADHD symptoms among children (with a 95% confidence interval). Additionally, the researchers broke down the model into different subgroups to see if the model’s accuracy varied.
Predictive accuracy showed marked differences depending on factors such as gender, race/ethnicity, household income, and parental history of mental health issues. Specifically, prenatal and perinatal factors appeared to be more relevant for some income groups, and certain predictors showed variation between boys and girls.
This in-depth analysis suggests that prenatal and perinatal risk factors may play distinct roles in different population groups. Finally, the study suggests that while it is possible to predict childhood ADHD symptoms from data obtained at birth, this prediction is still limited and needs to be improved.
To confirm and potentially improve the accuracy of these findings, the study recommends replicating the research with prospective follow-up of prenatal and perinatal data.
That is, collecting information on factors over time, from pregnancy through the first years of life, to understand in more detail how these factors may affect the risk of ADHD.
This could help develop more personalized interventions and support strategies for children at risk, enabling earlier monitoring and more effective care throughout the child’s development.
Robust pre/perinatal predictors of ADHD symptoms, in order of strength (mean B-coefficient) in the full sample of pregnant women: male sex, drug use, smoker, UTI during pregnancy, medication use, anemia, race and ethnicity, drug use (non-cannabis), labor complications, severe nausea in pregnancy, pregnancy complications, young maternal age, young paternal age, other complications, black mothers, Asian mothers, delivering more than one child.
READ MORE:
Predicting childhood ADHD-linked symptoms from prenatal and perinatal data in the ABCD cohort
Niamh Dooley, Colm Healy, David Cotter, Mary Clarke and Mary Cannon
Development and Psychopathology, Volume 36 Issue 2, 2023
Abstract:
This study investigates the capacity of pre/perinatal factors to predict attention-deficit/hyperactivity disorder (ADHD) symptoms in childhood. It also explores whether the predictive accuracy of a pre/perinatal model varies for different groups in the population. We used the ABCD (Adolescent Brain Cognitive Development) cohort from the United States (N = 9975). Pre/perinatal information and the Child Behavior Checklist were reported by the parent when the child was aged 9–10. Forty variables that are generally known by birth were input as potential predictors including maternal substance use, obstetric complications, and child demographics. Elastic net regression with 5-fold validation was performed and subsequently stratified by sex, race/ethnicity, household income, and parental psychopathology. Seventeen pre/perinatal variables were identified as robust predictors of ADHD symptoms in this cohort. The model explained just 8.13% of the variance in ADHD symptoms on average (95% CI = 5.6%–11.5%). The predictive accuracy of the model varied significantly by subgroup, particularly across income groups, and several pre/perinatal factors appeared to be sex-specific. Results suggest we may be able to predict childhood ADHD symptoms with modest accuracy from birth. This study needs to be replicated using prospectively measured pre/perinatal data.
コメント