Background: Our professional development plan aimed to improve the primary care management of acute asthma, which is known to be suboptimal. Methods: We invited 59 general practices in Grampian, Scotland to participate. Consenting practices were randomised to early and delayed intervention groups. Practices undertook audits of their management of all acute attacks (excluding children under 5 years) occurring in the 3 months preceding baseline, 6-months and 12-months study time-points. The educational programme [including feedback of audit results, attendance at a multidisciplinary interactive workshop, and formulation of development plan by practice teams] was delivered to the early group at baseline and to the delayed group at 6 months. Primary outcome measure was recording of peak flow compared to best/predicted at 6 months. Analyses are presented both with, and without adjustment for clustering. Results: 23 consenting practices were randomised: 11 to early intervention. Baseline practice demography was similar. Six early intervention practices withdraw before completing the baseline audit. There was no significant improvement in our primary outcome measure (the proportion with peak flow compared to best/predicted) at either the 6 or 12 month time points after adjustment for baseline and practice effects. However, the between group difference in the adjusted combined assessment score, whilst non-significant at 6 months (Early: 2.48 (SE 0.43) vs. Delayed 2.26 (SE 0.33) p = 0.69) reached significance at 12 m (Early:3.60 (SE 0.35) vs. Delayed 2.30 (SE 0.28) p = 0.02). Conclusion: We demonstrated no significant benefit at the a priori 6-month assessment point, though improvement in the objective assessment of attacks was shown after 12 months. Our practice development programme, incorporating audit, feedback and a workshop, successfully engaged the healthcare team of participating practices, though future randomised trials of educational interventions need to recognise that effecting change in primary care practices takes time. Monitoring of the assessment of acute attacks proved to be a feasible and responsive indicator of quality care.