We prospectively studied 653 consecutive head-injured children (<14 years old) treated over a 54-month period (1984–88) at the Department of Neurosurgery of the Rambam Medical Center (Haifa, Israel). Demographic and clinical data were collected, the patients were divided into five age groups (birth to 2 years, 169; 3–5 years, 194; 6–9 years, 164; 10–12 years, 77; and 13–14 years, 49), and the information relative to each was then compared. All patients (except three who died on the operating table) underwent computed tomography (CT) scans; 225 (34.6%) had intracranial pathology, e.g. focal mass lesions, diffuse axonal injury, and subarachnoid haemorrhage. The rate of detected pathology increased with age. Skull fracture was documented in 468 (72%) patients. Craniotomies were done on 114 (17.5%) patients. After 3 months, the patients were classified as having good recovery (84.8%), moderate disability (5.5%), or severe disability (2.3%); 0.9% were in a vegetative state. The mortality was 6.6% (43 patients); of these, 39 (90.7%) had admission Glasgow Coma Scale scores below 8. In our area the annual incidence of neurosurgical hospitalization due to head injury in the pediatric group was 37.6 per 100,000 inhabitants per year. This study substantiates the findings of other series on the effects on prognosis of factors such as associated trauma, admission Glasgow Coma Score, mass lesions with persistent intracranial pressure elevation, or diffuse axonal injury. This study also indicates that age has a bearing on severity of injury and outcome, even within the pediatric patient population, and that the presence of a skull fracture is not a reliable indicator of intracranial pathology. The obvious benefit of easy access to CT is illustrated.