Background. The majority of childhood acute myeloid leukemia (AML) patients lack a matched-related bone marrow transplant (BMT) donor in first remission. Procedure. Disease-free survival (DFS), overall survival (OS), relapse-free survival (RFS), and post-relapse outcome were evaluated for children with de novo AML on CCG 213 and the standard timing (ST) and intensive timing (IT) induction arms of CCG 2891 who were randomized to (intent-to-treat, ITT) or who received (as-treated, AT) only chemotherapy intensification. Results. Outcomes at 8 years post-induction in ITT analysis of chemotherapy intensification were as follows: 31% DFS, 43% OS on CCG 213; 34% DFS, 51% OS on CCG 2891 ST; 48% DFS, 56% OS on CCG 2891 IT. All toxic deaths during and following Capizzi II chemotherapy intensification on both protocols were in patients >3 years of age (P ≤ 0.001). Black race was a significant poor prognostic factor for OS (P = 0.008, hazard ratio: 1.74, 95% CI: 1.15-2.61). Overall 48% of patients on both trials relapsed and 19.1% of patients who relapsed on these trials survived. CR1 >12 months portends a much better OS for patients who relapse. Post-relapse treatment included BMT in 47% of patients. Conclusions. OS on CCG 2891 was superior to CCG 213 but equivalent between ST and IT arms due to better salvage rates post-relapse in ST patients. Overall survival for childhood AML in the absence of BMT in CR1 is influenced by duration of CR1 and by race.