Abstract
AIM: We aimed to assess a longitudinal trajectory of very preterm infants from corrected 1-5 years and examined how well performances on the Bayley Scales of Infant and (Toddler) Development (BSI(T)D) at 1 and 2 years predict cognitive outcome at preschool age. METHODS: From 2011 to 2018, 499 very preterm infants admitted to the Medical University of Innsbruck were assessed. Cognitive outcomes at 1 and 2 years were measured with BSID-II/BSITD-III, and at 5 years, full scale intelligence quotient (IQ) was evaluated. RESULTS: Of the 499 infants, 345 (69.1%) completed assessments at all three time points. Most had normal cognitive outcomes: 92.2% at 1 year, 80.3% at 2 years and 80.9% at 5 years. The 2-year assessment demonstrated a stronger predictive value than the 1-year assessment, both for identifying cognitive delay at 5 years (< 85 points; AUC 0.88 vs. 0.77) and for predicting full-scale IQ scores in linear regression models (R(2) = 46% vs. 36%). CONCLUSION: The 1-year assessment had limited predictive value, whereas the 2-year assessment more effectively predicted preschool cognitive outcomes. A combination of BSITD at 1 and 2 years gave the best prediction at preschool age.