Thirty-eight patients with primary frozen shoulder were randomized to graded motor imagery (GMI) plus multimodal physiotherapy (n=19) or multimodal physiotherapy alone (n=19). The primary outcome was SPADI; secondary outcomes were shoulder ROM, Numeric Pain Rating Scale (activity), Q-DASH, and SF-12; assessments occurred at baseline, 6 weeks (end of supervised treatment), and 8 weeks (post-treatment follow-up). Both groups attended twelve 45-minute sessions (twice weekly for 6 weeks); the GMI arm performed left–right discrimination, motor imagery, and mirror therapy alongside a reduced-dose multimodal program, with matched total session time and prescribed home practice in both groups. Overall, the results of the study revealed greater 8-week improvements with GMI for SPADI (MD 10; p=0.01) and Q-DASH (MD 9; p=0.01), with small between-group differences in ROM and no clinically meaningful advantage for activity pain. These findings suggest short-term functional gains from integrating GMI into multimodal care, though effects on pain and ROM were limited.