Internalizing/externalizing psychopathologies are identifiable by age 3, with neurodevelopmental risk markers evident in infants. Despite powerful implications for prevention, clinical impact has been minimal. We use innovative computational and epidemiologic data science methods to accelerate clinical translation of neurodevelopmental discovery during infancy towards generalizable risk prediction for preschool psychopathology. Our main objective is generating a pragmatic clinical risk calculator for public health use, the Mental Health Risk Calculator for Young Children (MHRiskCalc-YC). To achieve necessary power and precision, we create the Mental Health, Earlier Synthetic Cohort (MHESC), pooling multiple extramural cohorts at Washington University and Northwestern University to form the first clinically- enriched ?synthetic? neuroimaging cohort for generation of neurodevelopmentally-based clinical risk algorithms (N=1,020, followed from birth-54 mos.). To maximize the risk calculator's clinical and research utility and cost effectiveness, we will generate a series of risk algorithms tailored to envisioned end-users, incorporating input from clinical stakeholders. Algorithms will also establish added value of pre-postnatal environmental factors in risk prediction, a crucial but understudied RDoC element.
Aim 1 optimizes clinical feasibility and cost effectiveness by generating an MHRiskCalc-YC algorithm derived solely from commonly used survey data to optimize feasibility for future use in primary care settings.
Aim 2 optimizes precision of prediction by establishing statistical and clinical incremental utility of more intensive assessment for future use in mental health specialty settings. This algorithm sequentially tests the added predictive value of methods of intermediate-high intensity (from direct assessments to EEG to MRI) for most precise, least burdensome risk prediction.
The Aim 3 algorithm is optimized for future clinical research use in neurodevelopmental consortia, modeling the added value of MRI data to the Aim 1 algorithm. This mirrors ?common? protocols of neuroimaging consortia and will also generate an empirically-derived best practices guide for consortia to optimize timing/ number of neuroimaging assessments. External validity will be established in the Baby Connectome Project (BCP). The MHESC capitalizes on an unprecedented, time-sensitive opportunity to accelerate scientific and clinical impact of multiple extramural activities that have been extensively pre-aligned. The public health impact of an infancy-based clinical risk prediction tool for preschool psychopathology has transformative potential for altering standard of care in early identification and prevention of mental disorders.
Mental disorders are often early-onset chronic diseases, with brain and behavioral risk markers of common internalizing/externalizing problems evident in infancy. To accelerate translation, we will use multi-method imaging, behavioral and data science innovations to generate an infant mental health risk calculator for pragmatic public health use, derived from a clinically enriched ?Mental Health Earlier-Synthetic Cohort? of 1,020 infants using established study samples (characterized from birth-54 months). Clinical prediction algorithms will deliver broadly useful tools for earlier prediction and prevention of mental disorder and parameters to inform when imaging methods will be most informative in large-scale neurodevelopmental studies.
