Purpose: To determine the interrater and test-retest reliabilities and construct validity of the Premie-Neuro, a standardized neurologic assessment tool for preterm infants.
Subjects: Thirty-four preterm infants (mean gestational age at birth 29 ± 3.7 weeks, mean birth weight 1343.2 ± 696.3 g) participated in the study.
Design: A prospective repeated-measures design was used to assess the reliability and validity of the Premie-Neuro.
Methods: The Premie-Neuro was administered twice on consecutive days and then weekly through 37-weeks postmenstrual age or hospital discharge. At discharge, infants' medical histories were reviewed and a Neurobiologic Risk Score (NBRS) was used to determine risk for poor neurodevelopmental outcomes.
Main outcome measure: Premie-Neuro raw scores and classifications were analyzed to determine the tool's reliability. Construct validity was measured by determining whether the Premie-Neuro could discriminate between infants identified as high-risk or low-risk for neurodevelopmental delays by using a NBRS of 5 as the cutoff for high- and low-risk infants.
Results: The intraclass correlation coefficients for interrater and test-retest reliability varied from 0.391 to 0.556 and from 0.493 to 0.592, respectively. Analysis of variance revealed that the Premie-Neuro raw scores for infants with NBRS > 5 were significantly worse than those for infants with NBRS < 5 (P = .000-.010).
Conclusions: The Premie-Neuro is a valid assessment tool for discriminating between preterm infants at high and low risk for neurodevelopmental delay. Interrater reliability of the Premie-Neuro was poor, and test-retest reliability of the Premie-Neuro was fair to moderate. The Premie-Neuro may be acceptable for assessing groups of infants, but there is no evidence that reliability is sufficient for clinical decision-making for individual infants. More research needs to be done to improve the reliability of the Premie-Neuro and assess other facets of the Premie-Neuro's reliability.