BACKGROUND: In a recent trial of artemisinin-naphthoquine (artemisinin-NQ) and artemether-lumefantrine (AM-LM) therapy in young children from Papua New Guinea (PNG), there were no treatment failures in artemisinin-NQ-treated children with Plasmodium falciparum or Plasmodium vivax compared with 2.2% and 30.0%, respectively, in AM-LM-treated children during 42 days of follow-up. To determine whether, consistent with the long elimination half-life of NQ, this difference in efficacy would be more durable, clinical episodes of malaria were assessed in a subset of trial patients followed for six months post-treatment. METHODS: For children completing trial procedures and who were assessable at six months, all within-trial and subsequent clinical malaria episodes were ascertained, the latter by clinic attendances and/or review of hand-held health records. Presentations with non-malarial illness were also recorded. Differences between allocated treatments for pre-specified endpoints were determined using Kaplan-Meier survival analysis. RESULTS: Of 247 children who were followed to Day 42, 176 (71.3%) were included in the present sub-study, 87 allocated to AM-LM and 89 to artemisinin-NQ. Twenty children in the AM-LM group (32.8%) had a first episode of clinical malaria within six months compared with 10 (16.4%) in the artemisinin-NQ group (P = 0.033, log rank test). The median (interquartile range) time to first episode of clinical malaria was 64 (50-146) vs 116 (77-130) days, respectively (P = 0.20). There were no between-group differences in the incidence of first presentation with non-malarial illness (P = 0.31). CONCLUSIONS: The greater effectiveness of artemisinin-NQ over conventional AM-LM extends to at least six months post-treatment for clinical malaria but not non-malarial illness. TRIAL REGISTRATION: Australian New Zealand Clinical Trials Registry ACTRN12610000913077 .