Buschmann’s randomised controlled trial in which 2-year-old children with specific expressive language delay were exposed to an intervention with a parent based programme purports to demonstrate significant improvement and challenges the complacency of the “watch and wait” policy for such children.1 Do these findings plug one of the many gaps in our evidence base with which we try to judge the role of assessment and interventions in the screening and surveillance of language acquisition difficulties? 2 3 Do they run counter to the conclusions of the UK National Screening Committee Child Health Subgroup report on speech and language delay in May 2005 in which the clinician is assured that there is a good prognosis for isolated expressive language delay presenting under the age of 3 and that watchful waiting is a reasonable approach unless the parent is very worried? 4 5

Firstly we need to examine the rigor and design of this trial and then we can return to the question as to how relevant its findings are to everyday clinical practice. The trial benefits from some strong design features such as blinded randomisation and outcome assessments, but it is not perfect and two concerns that might affect translation into everyday practice are the failure to analyse according to intention to treat and the lack of detail as to how potential participants were identified.

Participating children and their families were referred by clinicians delivering a population-wide free routine developmental check-up in Germany. The authors made the important observation that good outcomes for treated children were not frustrated by lower maternal education and socio-economic status, as these can be critical factors in language delay and …