Inter-observer agreement of the General Movements Assessment with infants following surgery.

BACKGROUND The General Movements Assessment (GMA) is a validated and reliable method of identifying infants at risk of adverse neurodevelopmental outcomes, however there is minimal data available on the use of the GMA with infants following surgery. AIMS The aim of this study was to investigate the inter-observer agreement for the GMA with this infant population. STUDY DESIGN Reliability and agreement study. SUBJECTS This was a prospective cohort study of 190 infants (male n=112) born at term (mean 38weeks, SD 2weeks). OUTCOME MEASURES A GMA was conducted in the Neonatal Intensive Care Unit (NICU) following either cardiac surgery (n=92), non-cardiac surgery (n=93) or both types of surgery (n=5), and then again at three months of age. All videos were independently assessed by three advanced trained clinicians. Agreement and reliability statistics were calculated between each pair. RESULTS We found moderate to substantial levels of agreement in the writhing period (66-77%, AC1=0.53-0.69). For fidgety general movements, agreement was classified as almost perfect, ranging from 86 to 89% (AC1=0.84-0.88). CONCLUSIONS The GMA has high levels of inter-observer agreement when used with infants who have undergone surgery in the neonatal period, making it a valid, complementary assessment tool. Research is now underway to determine the ability of the GMA to predict neurodevelopmental outcomes in this population.