Neonatal assessments for the preterm infant up to 4 months corrected age: a systematic review.

AIM: The aim of this study was to systematically review the clinimetric properties of longitudinal neonatal neurobehavioural and neuromotor assessments for preterm infants.

METHOD: Twenty-seven assessment measures were identified. The following eight measures met the study inclusion criteria: Assessment of Preterm Infants' Behaviour (APIB), Neonatal Intensive Care Unit Network Neurobehavioural Scale (NNNS), Test of Infant Motor Performance (TIMP), Prechtl's Assessment of General Movements (GMs), Neurobehavioural Assessment of the Preterm Infant (NAPI), Dubowitz Neurological Assessment of the Preterm and Full-term Infant (Dubowitz), Neuromotor Behavioural Assessment (NMBA), and the Brazelton Neonatal Behavioural Assessment Scale (NBAS). The primary purposes included prediction (TIMP, GMs, Dubowitz), discrimination (all assessments), and evaluation of change (TIMP, NAPI). Measures of assessment were included in the study if they were (1) primarily neurobehavioural or neuromotor assessments that were suitable for use with preterm infants (<37 weeks gestation) up to 4 months corrected age and were discriminative, predictive, or evaluative; (2) standardized procedures designed for serial/longitudinal use; or (3) criterion or norm referenced. However, all assessment tools that were not published in English in a peer-reviewed journal or were primarily neurological assessments, one-time evaluations, screening tools, or not commercially available were not used.

RESULTS: All of the measures included in the review demonstrated adequate content and construct validity. Concurrent validity was reported for APIB, NNNS, Dubowitz, and GMs. Predictive validity was high for GMs with studies reporting up to 100% senstivity for predicting cerebral palsy at the age of 12 to 24 months. Interrater reliability was strong for the TIMP (intraclass correlation=0.95), GMs (K=0.8), and moderate for the NAPI (r=0.67-0.97). Clinical utility was variable for ease of scoring, interpretability, cost, and access.

INTERPRETATION: In the absence of a criterion standard for neonatal neuromotor assessments, the NNNS and APIB have strong psychometric qualities with better utility for research. Similarly, the GMs, TIMP, and NAPI have strong psychometric qualities but better utility for clinical settings. The GMs has best prediction of future outcome and the TIMP has best evaluative validity.