Background: Little is known about the predictive validity of hyperkinetic disorder (HKD) as defined by the Diagnostic Criteria for Research for mental and behavioral disorders of the tenth edition of the International Classification of Diseases (ICD-10; World Health Organization, 1993), particularly when the diagnosis is given to younger children. Methods: The predictive validity of HKD was evaluated over a 6-year period and compared to the predictive validity of DSM-IV attention-deficit/hyperactivity disorder (ADHD) in 95 4-6-year-old children who met full criteria for at least ADHD and 122 demographically-matched nonreferred comparison children. Diagnoses were based on structured assessments of both parents and teachers. Results: All children who met full criteria for HKD also met full DSM-IV criteria for ADHD, but only 26% of ADHD children met criteria for HKD. Children who met criteria for HKD (N = 24), children who would have met criteria for HKD but were excluded from the diagnosis because they concurrently met criteria for an anxiety disorder or depression (N = 16), and the remaining children who met DSM-IV criteria for ADHD (N = 55) all exhibited significantly more symptoms of ADHD and greater social and academic impairment during years 2-7 than nonreferred comparison children. Unlike the two other diagnostic groups, however, children who met strict criteria for HKD were not more likely than comparison children to be injured unintentionally or to be placed in special education. Conclusions: Both ICD-10 HKD and DSM-IV ADHD exhibit predictive validity over 6 years, but ICD-10 HKD appears to under-identify children with persistent ADHD symptoms and related impairment. Children who met criteria for DSM-IV ADHD but not HKD exhibited at least as much functional impairment over time as hyperkinetic children.