A precision health approach to medication management in neurodevelopmental conditions: a model development and validation study using four international cohorts

Abstract

Abstract Psychotropic medications are commonly prescribed to children with neurodevelopmental conditions, but responses vary widely, prescribing is largely off-label, and the expertise required is concentrated in specialized programs. We developed artificial intelligence models to predict prescribing patterns of stimulants, anti-depressants, and anti-psychotics. Feasibility was established in research cohorts by predicting cross-sectional medication use from the Child Behaviour Checklist, with training and internal testing in the Province of Ontario Neurodevelopmental network ( N =598) and external testing in the Healthy Brain Network ( N =1,764) and Adolescent Brain Cognitive Development ( N =2,396) studies. Clinical evaluation used electronic medical records (EMRs) from the Psychopharmacology Program ( N =312) at Holland Bloorview Kids Rehabilitation Hospital to predict the medication class prescribed at a follow-up visit (∼3 months later). In all cohorts, the modelled outcome was the clinician’s prescribing decision, which reflects clinician judgement, family preference, tolerability, and access to care in addition to expected effectiveness, and does not directly measure treatment response or clinical benefit. In the research cohorts, internal testing achieved an area under the receiver operating characteristic curve (median [IQR]) of 0.75 [0.73,0.80] for stimulants, 0.83 [0.78,0.87] for anti-depressants, and 0.79 [0.72,0.86] for anti-psychotics, and external testing confirmed generalizability. In the EMR cohort, values were 0.84 [0.81,0.88] for stimulants, 0.82 [0.77,0.87] for anti-depressants, and 0.87 [0.83,0.91] for anti-psychotics. Findings demonstrate that AI can accurately learn expert prescribing patterns and predict medication prescribing decisions, supporting the potential of data-driven tools to guide personalized medication management for neurodevelopmental conditions.

The authors' abstract, as published at the source. Translational Psychiatry, 2026 · DOI ↗