Study objective We derive a prediction rule to identify children at very low risk for intra-abdominal injuries undergoing acute intervention and for whom computed tomography (CT) could be obviated. Methods We prospectively enrolled children with blunt torso trauma in 20 emergency departments. We used binary recursive partitioning to create a prediction rule to identify children at very low risk of intra-abdominal injuries undergoing acute intervention (therapeutic laparotomy, angiographic embolization, blood transfusion for abdominal hemorrhage, or intravenous fluid for ≥2 nights for pancreatic/gastrointestinal injuries). We considered only historical and physical examination variables with acceptable interrater reliability. Results We enrolled 12,044 children with a median age of 11.1 years (interquartile range 5.8, 15.1 years). Of the 761 (6.3%) children with intra-abdominal injuries, 203 (26.7%) received acute interventions. The prediction rule consisted of (in descending order of importance) no evidence of abdominal wall trauma or seat belt sign, Glasgow Coma Scale score greater than 13, no abdominal tenderness, no evidence of thoracic wall trauma, no complaints of abdominal pain, no decreased breath sounds, and no vomiting. The rule had a negative predictive value of 5,028 of 5,034 (99.9%; 95% confidence interval CI 99.7% to 100%), sensitivity of 197 of 203 (97%; 95% CI 94% to 99%), specificity of 5,028 of 11,841 (42.5%; 95% CI 41.6% to 43.4%), and negative likelihood ratio of 0.07 (95% CI 0.03 to 0.15). Conclusion A prediction rule consisting of 7 patient history and physical examination findings, and without laboratory or ultrasonographic information, identifies children with blunt torso trauma who are at very low risk for intra-abdominal injury undergoing acute intervention. These findings require external validation before implementation.