Five hundred three patients with unilateral primary frozen shoulder were randomized to receive manipulation under anaesthesia with steroid injection (n=201), arthroscopic capsular release with or without steroid injection (n=203), or early structured physiotherapy with steroid injection (n=99). The primary outcome was the Oxford Shoulder Score (OSS) at 12 months. Secondary outcomes included the QuickDASH, pain intensity (Numeric Rating Scale), perceived extent of recovery, and quality of life (EQ-5D-5L). Outcomes were assessed at baseline, 3, 6, and 12 months. Overall, the results revealed that all three treatments resulted in substantial improvements in OSS, but none were clinically superior. Arthroscopic capsular release had slightly higher OSS at 12 months but also more adverse events and higher costs. Manipulation under anaesthesia was the most cost-effective. These findings suggest that less invasive options like manipulation or physiotherapy can be considered first-line, with surgical options reserved for refractory cases.