There is a need for effective interventions to address the social difficulties of children with ADHD. This randomised controlled trial examined the effectiveness of a play-based intervention for improving the social play skills of children with ADHD in peer-To-peer interactions. Children with ADHD (5 to 11 years) were randomised to an intervention-first (n = 15) or waitlist control-first group (n = 14). Participants allocated to the control-first group received the intervention after a 10-week wait period. Children invited a typically-developing playmate and parents of children with ADHD participated. The intervention involved: six clinic play-sessions, weekly home-modules and a one-month home follow up. The Test of Playfulness (ToP) was scored by a blinded rater. Parent reported treatment adherence was used to assess treatment fidelity. Between group statistics were used to compare the change of the intervention-first (10-week intervention period) and control-first (10-week wait period) groups. Once all children had received the intervention, repeated measures ANOVA, post hoc Least Significance Difference tests and Cohen's-d were used to measure effect. Changes in ToP social items were analysed using Friedman's ANOVA. Linear regression analyses were used to identify variables that predicted change. The control-first group did not change during the wait period. The change in the intervention-first group was significantly greater than the change in the control-first group (during the wait period). When the data from the two groups were combined, the mean ToP scores of the children with ADHD (n = 29) improved significantly following the intervention, with a large effect from pre to post intervention and from pre intervention to follow up. Children maintained treatment gains at follow up. All ToP social items improved significantly following the intervention. The findings support the use of play involving parent and peer mediated components to enhance the social play skills of children with ADHD.