115 older patients (≥ 50 years) who underwent a rotator cuff repair were randomized to receive, as part of a 12-week rehabilitation program, augmented reality (AR)-based digital rehabilitation for the last 6 weeks (n=58) or conventional rehabilitation for the full 12 weeks (brochure-based exercises; n=57). The primary outcome of interest was the Simple Shoulder Test (SST). Secondary outcomes of interest included a variety of patient-reported measures for function, pain, quality of life, and overall health perception. Outcomes were assessed up to 24 weeks post-operation. Patients in the AR group reported significantly better improvement in SST scores (12 weeks), DASH scores (24 weeks), SPADI scores (6, 12, 24 weeks), and EQ-5D-5L scores (12, 24 weeks) compared to the control group. In the subgroup analysis by tear size, the medium-size group showed significant differences between groups in those outcomes. Patient satisfaction was high in the AR group. No group differences in objective functional outcomes (range of motion, muscle strength, hand grip strength) were observed between the two groups.