Background <p>Geographical variation in Attention-Deficit/Hyperactivity Disorder (ADHD) diagnoses remains poorly understood. Previous research has found that the variation in ADHD rates between Child and Adolescent Mental Health Services (CAMHS) is not attributable to ADHD symptom load in the catchment areas of the CAMHS. This study aimed to investigate if geographical variation in rates of ADHD-diagnosis per population and per patient, as well as referral rates to CAMHS, were associated with catchment area characteristics.</p> Methods <p>We used data from the Norwegian Patient Registry, covering everyone aged 5–18 in contact with CAMHS 2009–2011, and catchment area level data from Statistics Norway, including population size, centrality, socioeconomic position, welfare reliance, and general population health. Spearman’s Rho was calculated to assess the strength and direction of monotonic correlation.</p> Results <p>Referral rates to CAMHS per population were lower in catchment areas with higher proportion of non-European population; more urban areas; fewer adults receiving disability benefits; and more economic inequality. Higher rates of ADHD per population were seen in areas characterised by lower socio-economic position. The rate of ADHD diagnosis per patient in CAMHS ranged from 5 to 27% and was not associated with catchment area characteristics.</p> Conclusion <p>This study found that the proportion of ADHD diagnoses per population was moderately related to catchment area characteristics, following a social gradient in health. However, the proportion of ADHD diagnoses among patients in CAMHS was unrelated to area characteristics. This suggests that other factors, such as variation in clinician attitudes or local practice styles, may contribute to the strong variation in ADHD diagnoses.</p>