Aim The aim of this prospective study was to evaluate the effect of corticosteroids intra‐articular injections on pain in patients with rotator cuff tear (RCT), and to identify predictors for pain outcomes. Methods A total of 60 patients with RCT were enrolled. All patients underwent rehabilitation; 20 patients received a single intra‐articular injection of 40 mg triamcinolone acetonide and 20 patients had a repeat injection at a 21‐day interval. Backward stepwise regression analysis was used to predict effectiveness and improvement of pain. The independent variables were age, sex, symptom duration, tear size, passive range of motion (ROM), active ROM, non‐steroidal anti‐inflammatory drugs request, pain at rest, number of triamcinolone injections and severity of osteoarthritis at admission. Results At 3 and 6 months, patients who received triamcinolone had higher effectiveness and improvement in pain during activities and pain at night than those of control group. At the 3‐month interval post‐therapy, active ROM was the only predictor for effectiveness in pain during activity, effectiveness in pain at night and improvement in pain at night. Six months after therapy, active ROM was a predictor for improvement in pain at night. Age was a predictor for effectiveness in pain at night, whereas tear size of RCT was a predictor for effectiveness and improvement in pain during activity. Conclusions Corticosteroids can relieve pain in RCT. Active ROM is the most important predictor of pain outcomes. This finding can be useful to physicians when deciding on the type of patients who might best benefit from intra‐articular injections of corticosteroids. Geriatr Gerontol Int 2013; ●●: ●●–●●.