Optimal timing of the General Movement Assessment for early detection of cerebral palsy

Aim To identify trajectories of normal fidgety movements between 10 weeks and 20 weeks corrected age and determine the optimal timing of the General Movement Assessment (GMA) to improve early cerebral palsy detection. Method In this prospective multicenter cohort study, parents submitted up to six serial home videos (median five per infant) between 10 weeks and 20 weeks corrected age using the Baby Moves smartphone application. GMA was performed by certified raters masked to age and clinical history. Group‐based trajectory modelling identified patterns of normal fidgety movement expression over time. Results A total of 438 infants (233 males, 205 females) contributed 2075 videos. Four trajectories were identified: consistent fidgety (22%), later fade (63%), later onset (10%), and low prevalence (5%). The greatest overlap in normal fidgety movements occurred at 14 to 15 +6 weeks corrected age. Approximately 10% of infants (later onset) demonstrated emergence of normal fidgety movements after the currently recommended initial assessment window. The low prevalence trajectory was associated with lower gestational age and greater neonatal morbidity. Interpretation Timing influences fidgety movement classification and may affect early cerebral palsy detection. Assessment at 14 to 15 +6 weeks corrected age may reduce early misclassification.