BACKGROUND: Otitis media with effusion (OME) is the most common cause of acquired hearing loss in childhood and has been associated with delayed language development and behavioural problems. This condition has a prevalence of about 20% at the age of two years, a time of rapid language development. It is most often asymptomatic. Effective treatment exists for clearing effusions. Some have argued, therefore, that children should be screened and treated early if found to have clinically important OME. However, there is a high rate of spontaneous resolution of effusions and for some children, effusions may represent a physiological response that does not reduce hearing significantly or impact negatively on language development or behaviour. Previous reviews of the effect of screening and treatment have included studies using non-randomised designs. OBJECTIVES: The aim of this review was to assess evidence from randomised controlled trials about the effectiveness, on language and behavioural outcomes, of screening and treating children with clinically important OME in the first four years of their life. The focus was on the first four years of life because this is the time of most rapid language development. The consequences of hearing loss are likely to be most serious during this time. In addition, children of this age are least likely to be able to report or seek help for impaired hearing, particularly if these problems have a slow onset and are subtle. SEARCH STRATEGY: We searched the Cochrane Controlled Trials Register (the Cochrane Library Issue 1, 2002), MEDLINE (1966-2002) and EMBASE (1974-2002) (all in February 2002) and reference lists of all studies. We also contacted the first authors of the studies we included in this review. SELECTION CRITERIA: 1. Randomised controlled trials evaluating interventions for OME among children with OME identified through screening. 2. Comparison of outcomes for children randomised to be screened for OME and outcomes for children who were not randomised to be screened for OME. DATA COLLECTION AND ANALYSIS: Two reviewers independently extracted data and assessed trial quality. MAIN RESULTS: We identified no trials comparing outcomes for children randomised to be screened for OME with outcomes for children who were not randomised to be screened for OME. We identified three trials evaluating interventions for OME among children with OME identified through screening. From these trials, we found no evidence of clinically important benefit in language development from screening and treating children with clinically important OME. Although there was a beneficial effect on the resolution of OME and improved hearing in the short-term (six months), this effect largely disappeared in the long-term (12 months). REVIEWER'S CONCLUSIONS: The identified randomised trials do not show an important benefit from screening of the general population of asymptomatic children in the first four years of life for OME on language development and behaviour.