Background It is estimated that 20–30% of prisoners meet diagnostic criteria for attention deficit hyperactivity disorder (ADHD). Methylphenidate reduces ADHD symptoms, but its effect among prisoners remains uncertain. Objectives The primary objective was to estimate the efficacy of osmotic release oral system (OROS) methylphenidate in reducing ADHD symptoms in male prisoners aged 16–25 years who met diagnostic criteria for ADHD. Secondary objectives investigated change for associated clinical and behavioural problems and the role of ADHD symptoms in mediating change in behaviour. Design A Phase IV, 8-week, parallel-arm, double-blind, randomised, placebo-controlled trial of OROS-methylphenidate, compared with placebo, in young male adult prisoners with ADHD. Participants were randomised in a 1 : 1 ratio of OROS-methylphenidate to placebo, stratified by prison. Setting Participants were recruited from Her Majesty’s Prison and Young Offender Institution Isis (London, England) and Her Majesty’s Young Offender Institution Polmont (Falkirk, Scotland). Participants The participants were 200 male prisoners with ADHD aged 16–25 years who met the diagnostic criteria for ADHD. Exclusion criteria included moderate or severe learning disability; serious risk of violence to researchers; current major depression, psychosis, mania or hypomania, or a past history of bipolar disorder or schizophrenia; and drug-seeking behaviour that was of sufficient severity to affect the titration protocol. Intervention The intervention was overencapsulated OROS-methylphenidate (18 mg) or placebo capsules. Trial medication was titrated weekly for 5 weeks against symptom reduction and adverse effects to a final dose of one to four capsules per day, followed by a stable dose for 3 weeks. Main outcome measures The primary outcome was ADHD symptoms at 8 weeks using the investigator-rated Conners’ Adult ADHD Rating Scale-Observer. There were 13 secondary outcomes, including measures of emotional dysregulation, general psychopathology, reports of behaviour by prison staff and engagement with educational activities. Results For the primary outcome, the estimated improvement between the OROS-methylphenidate and placebo arms was 0.57 points on the Conners’ Adult ADHD Rating Scale-Observer (95% confidence interval –2.41 to 3.56) at 8 weeks, with a standardised effect size of 0.06. The difference was not statistically significant and was smaller than the difference the trial was powered to detect. Responder rate, defined as a 20% reduction in the Conners’ Adult ADHD Rating Scale-Observer score, was 48.3% for the OROS-methylphenidate arm and 47.9% for the placebo arm. None of the 13 secondary outcomes that could be formally compared between the trial arms showed a significant effect and no mediators of change in behaviour were identified. Limitations Low adherence to trial medication and low medication dose might have affected the results. Conclusion OROS-methylphenidate was not found to have an effect, compared with placebo, on the primary and secondary outcomes investigated. The findings indicate that ADHD symptoms do not respond to a standard treatment for ADHD following titration to low doses in young adults in prison. The findings do not support the routine treatment with OROS-methylphenidate of young adult prisoners meeting diagnostic criteria for ADHD. Future research Investigations of adequate, maintained dosing, non-pharmacological interventions and community studies are suggested. Trial registration This trial is registered as ISRCTN16827947 and EudraCT 2015-004271-78. Funding This project was funded by the Efficacy and Mechanism Evaluation (EME) programme, a MRC and National Institute for Health and Care Research (NIHR) partnership. Janssen-Cilag Ltd supplied OROS-MPH (Concerta-XL). This will be published in full in Efficacy and Mechanism Evaluation; Vol. 9, No. 6. See the NIHR Journals Library website for further project information.