Examine the hospital admission risk in young children who are subsequently diagnosed with attention deficit hyperactivity disorder (ADHD). We conducted a population-based, record linkage study. Records of all non-Aboriginal children under 18 years who met the DSMIV/ICD10 criteria for ADHD and prescribed stimulant medication in Western Australia between 2003 and 2007 (n = 11,902) were linked to two other health data systems—the hospital morbidity data system and the midwives notification system (MNS). The non-ADHD reference population (n = 27,304) was randomly selected from the MNS. Compared with controls, children under 4 years who subsequently were diagnosed and treated for ADHD were 70 % [odds ratio (OR) 1.70; 95 % confidence intervals (CI) 1.62–1.77] more likely to be admitted to hospital under 4 years of age. There was an increased risk for injury or poison (OR 1.73; 95 % CI 1.59–1.88), respiratory disease (OR 1.49; 95 % CI 1.40–1.59), ear disease (OR 2.03; 95 % CI 1.86–2.21), infectious diseases (OR 1.68; 95 % CI 1.53–1.85) and neurological conditions (OR 2.03; 95 % CI 1.68–2.44). Admissions under 4 years of age for head injuries, burns, poisons, all other injuries, diseases of the tonsils and adenoids, asthma and early infections were all more common amongst children subsequently diagnosed with and treated for ADHD. There is significant early hospital morbidity for children subsequently diagnosed with ADHD. Multiple aetiologies and causal pathways need to be considered where some of these may include early infections, inflammatory conditions, epilepsy and injuries. Future studies should look at which of these conditions may be on the causal pathway or likely early markers for ADHD.