ABSTRACT
Objective To compare the validity of general movements ( GM ) assessment with that of a 52 item neurological motor examination in predicting cerebral palsy among high risk neonates.Methods Eighty-four infants aged 12 months at high risk were evaluated simultaneously with GM assessment and the 52 item neurological motor examination.The utility of the two methods in predicting cerebral palsy was then compared.Results There were 4 cerebral palsy patients among the 84 high risk neonates.The GM assessment's predictive validity for Monotonous movements was 78.57％,its sensitivity was 25％,its specificity was 81.25％,its positive predictive value was 6.25％,its negative predictive value was 95.59％,and its kappa value was 0.215.The comparable values for the Cramped-synchronised scale were validity 96.43％,sensitivity 50％,specificity 98.75％,positive predictive value 66.67％,negative predictive value 97.53％,and kappa value O.553.The values for the absence of Fidgety movements were validity 98.81％,sensitivity 75％,specificity 100％,positive predictive value 100％,negative predictive value 98.77％,and kappa value 0.851.For the entire 52 items of the neurological motor examination the corresponding values were validity 51.19％,sensitivity 100％,specificity 48.75％,positive predictive value 8.89％,negative predictive value 100％,and kappa value 0.083.Conclusions GM assessment can predict the severe neurological deficits of cerebral palsy among high risk neonates by the age of 12 months.The 52-item neurological motor examination can detect mild and moderate neurological deficits.GM assessment combined with the 52-item neurological motor examination is a desirable and effective way to assess high risk infants.