Randomised controlled trials (RCTs) of participants who received at least one application of manual therapy for disorders of the shoulder (including fractures, dislocations, degenerative/osteoarthritis and orthopaedic surgery) were eligible for inclusion. Studies that applied manual therapy to the shoulder girdle, cervical or thoracic spine were included. Multi-modal interventions were included if manual therapy could be extracted from the other interventions. Studies had to report pain, range of motion, functional outcomes, patient satisfaction or recovery rates. Studies were excluded if they were in participants with rheumatoid arthritis, had shoulder symptoms of spinal origin or had neurological disorders such as stroke.
Included studies were in participants with adhesive capsulitis, shoulder impingement syndrome, or non-specific shoulder pain or dysfunction. Manual therapy interventions included Maitland joint mobilisations of the shoulder girdle, mobilisation of the upper quarter, manipulation and soft tissue massage. Studies that assessed manual therapy in isolation or in combination with other interventions (exercise, hot packs or medical care) were included. Control group interventions included exercise, stretching and no treatment. Mean age of participants ranged from 44 to 65 years.
Two authors independently preformed the study selection. Disagreements were resolved through consultation with a third reviewer.