Geographic analysis of the variation in the incidence of ADHD in a country with free access to healthcare: a Danish cohort study

Background: The prevalence of citizens diagnosed with Attention Deficit Hyperactivity Disorder (ADHD) has risen dramatically over the past decades in many countries, however, with large variations. Countries such as Denmark with centrally organized well fare systems, free access to health services and individual tracking based on unique personal identification may in particular contribute to our understanding of the reasons for this increase. Based on Danish registers we aimed to examine the geographical patterns of the distribution of ADHD diagnosis and medication use and explore the association with access to diagnostic services, diagnostic culture, neighbourhood socioeconomic status and municipal spending on health care for children. Methods: We combined information on registered diagnosis of ICD-10 Hyperkinetic Disorder and ADHD medication use in a Danish register-based cohort of children born between 1990 and 2000. We mapped incidence proportions of diagnoses and medication use within the 98 Danish Municipalities. Global and local clustering of ADHD was identified using spatial analysis. Information on contextual factors in the municipalities was obtained from national registers. The associations between the incidence of ADHD and contextual factors were analysed using Bayesian spatial regression models. Results: We found a considerable variation in the incidence of ADHD across the municipalities. Significant clustering of both high and low incidence of ADHD was identified and mapped using the local Moran's I. Clustering of low incidence of diagnosis and medication use was observed in less populated areas with limited diagnostic resources and in contrast clustering of high incidence in densely populated areas and greater diagnostic resources. When considering the spatial autocorrelation between neighbouring municipalities, no significant associations were found between ADHD and access to diagnostic services, different diagnostic culture, socioeconomic status at municipality level or the municipal spending on health care for children. Conclusions: A large geographical variation of ADHD in the municipalities was observed despite tax-financed and free access to healthcare. Although not statistically significant, results indicate that accessibility to diagnostic resources might explain some of the variation in ADHD incidence. In contrast to US studies the observed variation was not statistically associated to contextual factors in terms of SES, municipal spending on health care for children or differences in diagnostic practices.