The role of possible clinical factors that might influence the efficacy of treatment with delayed head cooling and mild systemic hypothermia for neonatal encephalopathy was determined in a total of 218 term infants treated at University College, London, UK; University of Auckland, New Zealand; and other centers in the UK, USA, and Canada. Infants with moderate to severe encephalopathy plus abnormal amplitude-integrated EEG recordings were assigned randomly to head cooling for 72 hours, starting within 6 hours after birth, or conventional care without cooling. Analysis of clinical data at 18 months showed that infants with a lower encephalopathy grade, lower birth weight, greater amplitude-integrated EEG, absence of seizures, and higher Apgar score had significantly better outcomes. Gender and gestational age were not significantly associated. In multivariate analysis, each of the clinical factors except the Apgar score was predictive of a good prognosis. Larger infants with birth weights -/> 25th percentile showed a greater frequency of favorable outcomes with cooling but less favorable outcomes for the control group. The encephalopathy grade was the single most predictive factor of outcome. Pyrexia (-/>38C) in control infants was associated with adverse outcomes; 34 control patients had rectal temperatures of ->38C during the 76-hour monitoring period, and 28 (82%) had unfavorable outcomes; of 76 without pyrexia, 45 (59%) had unfavorable outcomes (P=0.028).