126 patients with frozen shoulder were randomized to dual glenohumeral and subacromial injection of betamethasone once per week for 5 weeks, to glenohumeral injection alone once per week for 5 weeks, or to oral non-steroidal anti-inflammatory drugs (NSAIDs) for 5 weeks. Patients were assessed for pain, function, and range of motion over the first 6 months post-treatment. Results demonstrated significantly lower pain in the injection group compared to the oral NSAIDs group. Function and range of motion were reported to be significantly greater during follow-up in the dual glenohumeral and subacromial injection group compared to the other two groups.