To improve the complete response rate (CRR) and reduce the recurrence rate of newly diagnosed non-elderly acute myeloid leukemia (AML), the present study compared the clinical efficacy of decitabine with cytarabine (A) and daunorubicin (D)-based remission induction therapy with D + A-based remission induction therapy. A total of 81 patients with newly diagnosed non-elderly AML (non-M3) were enrolled in the present study, and divided into the observation group [decitabine with D + A, demethoxydaunorubicin (I) + A or homoharringtonine (H) + A] and the control group (D + A, I + A or H + A). The observation group displayed a 91.4% CRR [95% confidence interval (CI), 81.7-100%] and the control group displayed a 69.6% CRR (95% CI, 55.8-83.4%). The 2-year overall survival (OS) rate was improved in the observation group compared with the control group (P=0.008). Patients aged <60 years displayed a 92.9% CRR in the observational group and a 71.1% CRR in the control group (P<0.05). Patients with undetected methylation gene mutations displayed an improved CRR in the observation group compared with the control group (92.9 vs. 71.4%; P=0.028). Furthermore, relapse-free survival (P=0.041) and OS (P=0.007) were significantly extended in the observation group compared with the control group. The present study suggested that the administration of decitabine with DA, IA or HA as an induction therapy improved the clinical efficacy and reduced the recurrence rate in patients with AML.