Interventions for prevention and treatment of childhood obesity typically target increases in physical activity and, more recently, reductions in physical inactivity (sedentary behaviour such as television viewing). However, the evidence base for such strategies is extremely limited. The main aim of the present review was to update the systematic review and critical appraisal of evidence in the light of the recent rapid expansion of research in this area. Randomised controlled trials (RCT) that targeted activity or inactivity, that followed up children or adolescents for at least 1 year and that included an objective weight-related outcome measure were included. Trials were appraised using previously published criteria (Harbour & Miller, 2001), and literature search strategies described previously (Reilly et al. 2002) were updated to May 2002. A total of four new RCT, two new systematic reviews and one meta-analysis were identified. The evidence base has increased markedly since the completion of earlier reviews, although high-quality evidence is still lacking. The evidence on childhood obesity prevention is not encouraging, although promising targets for prevention are now clear, notably reduction in sedentary behaviour. There is stronger evidence that targeting activity and/or inactivity might be effective in paediatric obesity treatment, but doubts as to the generalisability of existing interventions, and the clinical relevance of the interventions is unclear. Further research in settings outside the USA is urgently needed, and two ongoing RCT in Scotland are summarised.