In a follow-up study of 122 full-term infants in whom postasphyxial encephalopathy occurred the incidence of death or severe handicap was 1 in 1000 deliveries. The abilities of two methods of diagnosing intrapartum asphyxia to predict outcome at a median age of 2.5 years were compared. A decision matrix calculation was undertaken to assess the sensitivity and specificity of low Apgar score and postasphyxial encephalopathy. A 10 min Apgar score less than or equal to 5 was the most sensitive of six different Apgar ratings in predicting adverse outcome (sensitivity 43%, specificity 95%) but even this was much less sensitive than the presence of moderate or severe encephalopathy in predicting death or severe handicap (sensitivity 96%).