476 - Interrater Agreement of Neonatal Encephalopathy Examination and Scoring System

Poster Number: 476
Publication Number: 476.2753

Background: Therapeutic hypothermia (TH) is the standard of care for hypoxic-ischemic encephalopathy. Its implementation relies on clinically assessing neonatal encephalopathy (NE). While several variations of the assessment initially reported by Sarnat and Sarnat are in practice, their interrater agreement has not been assessed.

Objective: To assess the coherence of the NE assessment by healthcare practitioners with varying levels of experience.

Design/Methods: Infants born >34 weeks gestational age and assessed by a trained physician for NE in a single level three NICU, were video recorded ensuring confidentiality. The total score was calculated as the sum of ordinal scores (3 levels) across 8 subscales. Reference scores were determined by the consensus opinion of the research team. Thereafter, two de-identified videos, one of mild and another of moderate NE, were circulated via virtual platforms along with a short clinical vignette and a scoring survey over multiple centers in America and Ireland. The participants could opt to watch a 25-minute teaching video (by ME,TI) and score again. The degree of consistency was analyzed by comparing the difference between the measured total score and the reference, by NE status (mild v moderate) using independent-sample Mann-Whitney U test, and by experience, using Independent-Samples Kruskal-Wallis Test.

Results: Of the 110 survey responses received, 101 (91.8%) were physicians and 93 (84.5%) were from level 3 or 4 units. 37 (33.6%) had >10 years of experience in their current role. There was wide variation in scores and tendency to overscore (Figure 1). Level of consciousness, spontaneous activity and neonatal reflexes had less agreement as opposed to tone and autonomic functions. The divergence from the reference score was significantly greater for mild compared with moderate NE (p=0.034)(Figure 2). While the difference did not significantly change by experience category (p = NS), the variability of the scores decreased with greater experience (Figure 3). 73 watched the teaching and scored again. Assessment of NE post-teaching showed similar results with no significant improvement in interrater agreement.

Conclusion(s): Assessment of NE varies among practitioners, especially in milder cases. A single teaching session may not suffice in increasing the skills and continuing education with active feedback and case-based teaching should be considered. Concomitant observation by multiple health professionals with expertise may be useful in confounding cases and deciding management in infants with NE.