Early identification of motor problems in very preterm infants : An evaluation of the Structured Observation of Motor Performance in Infants

Sammanfattning: Infants born very preterm are at risk of adverse neurodevelopment. It is important to identify motor problems early to initiate interventions aiming at ameliorating outcomes. Evaluating motor development in high-risk infants is a complex task. There is a need for assessment methods for early identification of abnormal motor performance. The aim of this thesis was to evalute the Structured Observation of Motor Performance in Infants (SOMP-I) method for early identification of motor problems in very preterm children and to investigate early motor performance in relation to neonatal characteristics, cerebral imaging and later outcome. Level of motor development and quality of motor performance was assessed at 2, 4, 6, and 10 months’ of corrected age. Study I validated the revised SOMP-I, and compared early motor performance in 111 very preterm infants with 72 full-term infants. The preterm infants were more delayed and had more quality deficits than the term infants, and the groups had different motor trajectories. We concluded that convergent validity and discriminant validity of the SOMP-I was supported and facilitated early identification of infants with atypical motor development.Study II investigated SOMP-I results in relation to motor outcome (Bayley-III motor index at 2.5 years) in 98 very preterm children. The 28 children with delayed development had significantly poorer SOMP-I scores in infancy. We concluded that level and quality of motor performance were significant markers of later motor problems and quality became more significant with increasing age. Study III investigated early motor performance (SOMP-I), in relation to neurodevelopment and motor competence at 12 years (Movement ABC-2) in 78 very preterm children. At all assessment ages, there were significant associations between SOMP-I and MABC-2 scores. At 6 and 10 months, SOMP-I level and quality scores separately explained unique variance of the MABC-2 scores at 12 years. Study IV explored the relation between neonatal cerebral MRI (morphology, apparent diffusion coefficient, regional brain volumes) and 4-month motor performance (SOMP-I), in relation to 2-year motor outcome in 66 very preterm infants (11 with motor problems). SOMP-I results correlated with several MRI measures and with motor outcome. The level of motor performance had the highest predictive value for motor outcome. Overall conclusion: The two SOMP-I domains, level and quality, explain unique variances towards later motor outcomes, meaning that the two separate domains give added value to the motor assessment and are useful markers of motor outcome in very preterm infants.