Identification of neurodevelopmental trajectories in infancy and of risk factors affecting deviant development: a longitudinal birth cohort study.

BACKGROUND We investigate patterns of neurodevelopmental trajectories in infants, using a representative population, and identify risk factors that predict delayed growth. METHODS Participating infants (n = 952; 82.8% of the total sample) were assessed by Mullen Scales of Early Learning at seven time points, from 1 month to 24 months of age. Mothers were recruited in early pregnancy and data on demographic characteristics were collected during pregnancy. Trajectory patterns were investigated using latent class growth analysis, and risk factors for the derived trajectory classes were investigated by multinomial logistic regression. RESULTS Participants were found to be a fairly representative sample with respect to their demographic characteristics. Five classes of high normal (11.5%), normal (49.2%), low normal (21.2%), delayed (14.1%), and markedly delayed (4.0%) were identified. The markedly delayed class was characterized by overall delay from the early developmental stages; notably, such delay first became salient in motor domains and was then exceeded by language domains, especially receptive language. This class was predicted by male sex (odds ratio 4.0; 95% confidence interval 1.7-9.1), small for gestational age (2.8; 1.0-7.5), low placenta-to-birthweight ratio (2.8; 1.2-6.4) and low maternal education (4.7; 1.2-19.0). The delayed class was characterized by gradual downward deviation after the first birthday, and was predicted by male sex (2.5; 1.5-4.2), preterm birth (4.4; 1.6-12.6) and advanced paternal age (1.9; 1.0-3.5). CONCLUSIONS Our results demonstrate that about 1 out of 5 infants exhibits delayed neurodevelopment. Infants with distinct patterns of delayed trajectories and varying risk factors are considered to have different pathophysiological mechanisms.