180 shoulders of patients with a non-traumatic, isolated, full-thickness supraspinatus tear were randomized to receive physiotherapy either alone, with acromioplasty, or with both acromioplasty and surgical repair of the tear. The purpose of the study was to compare the efficacy of treatments in terms of clinical outcome, radiographic outcome, and cost. No significant differences between groups in Constant shoulder score for function, VAS pain, or patient satisfaction were noted between groups at 2 year follow-up. Patients who had undergone surgical repair had significantly smaller rotator cuff tears on MRI at 2 year follow-up compared to the other groups. No significant difference was observed in Constant score between those with full-thickness tear and those with a healed tear at 2 years. Costs were significantly different between groups, being lowest in the physiotherapy group and highest in the group receiving all 3 treatments.