Financial disclosure or conflict of interest for all authors: NoneWord count: 3428 2/25 ABSTRACT To determine whether the youngest age groups are less likely to be included in paediatric randomised controlled trials (PRCTs) than older children, we conducted a PubMed search using the keyword "randomised controlled trial" and the limit "all child: 0 -18 years". We retrieved 417 articles published in 2006 in 34 leading journals classified as general medical journals, paediatric medical journals, or specialist medical journals. We arbitrarily selected 144 articles, at random. For each study, we evaluated population age characteristics (central tendency, range, and dispersion), study design, sample size and topic. Of the 144 studies, only 82 were first reports of paediatric randomised controlled trials (PRCTs). Among the other studies, many were done in adults. Of the 82 PRCTs, only 11% included newborns and 26% infants; 59% included children and 39% adolescents. Using the same search strategy to retrieve PRCTs in the same journals in the last four months of 2009 retrieved 66 PRCTs, of which 17% included newborns, 24% infants, 61% children and 55% adolescents. The three health conditions most often reported were respiratory diseases, infectious diseases, and mental and behavioural disorders. In 34 leading journals, PRCTs were significantly less likely to include newborns and infants than older paediatric patients. Given the huge impact of PRCTs on paediatric health, additional efforts are needed to promote studies in newborns and infants, as well as studies of the impact of recent European and American regulations designed to encourage paediatric drug trials.