The objective of this study was to determine whether the Arthritis Self-Management Programme (ASMP) improves perceptions of control, health behaviours and health status, and changes use of health care resources. The design was a pragmatic randomized controlled study; participants were allocated to ASMP (Intervention Group) or a 4-month waiting-list Control Group. The Intervention Group completed a 12-month follow-up. In total, 544 people with arthritis were recruited from the community--311 in the Intervention Group and 233 in the Control Group. Main outcome measures included: arthritis self-efficacy, health behaviours (exercise, cognitive symptom management, diet and relaxation) and health status (pain, fatigue, anxiety, depression and positive affect). At 4 months follow-up, the ASMP had a significant effect on arthritis self-efficacy for other symptoms and pain subscales. Performance of a range of health behaviours (cognitive symptom management, communication with physicians, dietary habit, exercise and relaxation) was significantly greater among the Intervention Group. The Intervention Group were significantly less depressed and had greater positive mood. In addition, trends towards decreases on fatigue and anxiety were noted. Physical functioning, pain and GP visits remained stable at 4 months. A similar pattern of findings was found at 12 months follow-up for the Intervention Group. Furthermore, a significant improvement was found on pain and visits to GPs had decreased. Apart from a small improvement on physical functioning among the Intervention Group participants with osteoarthritis 12 months, all effects were independent of the type of arthritis. The findings suggest that the ASMP is effective in promoting improvements in perception of control, health behaviours and health status, when delivered in UK settings.