The recent article in this journal regarding attention deficit hyperactivity disorder (ADHD) treatment1 provides an interesting snapshot of clinical practice. I write to extend discussion of their findings. The authors drew two potentially contradictory conclusions. The first was that practice falls in line with the recent draft NHMRC guidelines. Secondly, they identified considerable variation in prescription and treatment practice by profession, sector, duration of time in practice, practice size and focus, and location. If these conclusions are concurrently valid, the guidelines must encompass a significant degree of practice diversity. Is this ideal? Variability in practice can reflect local opportunities and limitations2 and accordingly may have adaptive advantages. Variation in practice may also be acceptable if clinical outcomes are relatively independent of practice activities and pathways. In other chronic medical conditions, however, where medical intervention is part of a comprehensive longitudinal strategy, I believe such variability would be considered undesirable and potentially detrimental to child health. With conditions such as diabetes and cystic fibrosis we have a variety of markers by which to monitor the cumulative impact of the disorder. By evaluating practice against these measures, particularly over long periods of time, consensus has emerged regarding which practices optimise not only short-term but particularly long-term health and well-being. As a result, variability in practice is reduced. For ADHD this is a complex challenge. The main clinical measure of disorder and related intervention for ADHD is behavioural symptom profile as determined by observational rating scales. This provides only limited information regarding current and cumulative impact of the disorder. The challenge further includes the scope of impact of ADHD in the lives of developing children. Variability around what clinicians consider to be their scope of care and related outcomes is clearly reflected in this study. The impairment attributable to ADHD relates not only to the symptoms but also to the demands of the context in which they manifest. Long-term well-being relates to a diverse set of secondary consequences across child development, mental health and social function3 rather than just the primary pathology. This situation may never change while we continue to manage children with diagnoses taken from the Diagnosis and Statistical Manual of Mental Disorders IV alone and measures of symptom expression. What are the short- and long-term measures of outcome against which we more effectively consider best practice?