BACKGROUND: The rated severity of child mental health problems depends on who is doing the rating, whether child, carer or clinician. It is important to know how these ratings relate to each other. AIMS: To investigate to what extent clinicians’ views are associated with carers’ and young people’s views in routine care in the United Kingdom. METHOD: Ratings of clinician and parent/child viewpoints from a large Child and Adolescent Mental Health Services (CAMHS) sample ( ns 1773-47,299), as measured by the Children’s Global Assessment Scale (CGAS) and Strengths and Difficulties Questionnaire (SDQ) respectively, were analysed. The parent SDQ added value score (AVS), which adjusts for regression to the mean and other non-treatment change, was also included in the analyses. RESULTS: Small-to-medium correlations were found between family and clinician ratings; however, ratings diverged for the lowest-function CGAS bands. Regression analyses showed that pro-social ratings from both child and parent contributed to clinician ratings. Knowing child-reported emotional problem severity made parent ratings of emotions irrelevant to clinician judgements. There was a positive association between SDQ AVS and CGAS; as hypothesised, CGAS showed more change than the SDQ AVS, suggesting that clinicians over-estimate change. CONCLUSION: This study shows the importance of multi-informant data gathering and the integration of multiple views by clinicians when monitoring outcomes.