Attention deficit hyperactivity disorder (ADHD) occurs in 3–5% of school-aged children (Shaffer et al., 1996) and accounts for as many as 30–50% of child referrals to mental health services (Popper, 1988). Children meeting full (DSM-IV: American Psychiatric Association, 1994) criteria (early onset, 6-month duration and symptoms in more than one setting) have substantial impairment in peer, family and academic functioning (Hinshaw, 1992). Long-term outcome studies indicate that the syndrome persists into adulthood in most cases (Barkley et al., 1990; Mannuzza et al., 1993; Weiss and Hechtman, 1993), with increased risk for substance abuse and delinquency-related outcomes (Satter-field et al., 1987; Mannuzza et al., 1993). Although a very large literature (Pelham and Murphy, 1986; Swanson, 1993; Spencer et al., 1996; Hinshaw et al., 1998) has documented the beneficial effects of medication (principally stimulants), psychosocial treatments (principally behaviour therapy) and their combination, this body of research has suffered from significant limitations, including the brief duration of treatment (typically days to months), small sample sizes and reliance on a restricted range of outcome measures. Despite nearly 30 years of research on ADHD treatments, few controlled studies have examined the effectiveness of long-term treatments. Two recent exceptions (Hechtman and Abikoff, 1995; Gillberg et al., 1997) were of 1–2 years duration, and suggested that stimulant effects may persist when the stimulant is taken faithfully.