Background: Frozen shoulder syndrome (FSS) or adhesive capsulitits, a condition of uncertain etiology characterized by a progressive loss of both active and passive shoulder motion, is one of the most common musculoskeletal disorders encountered in daily orthopaedic practice and remains challenging to treat. To regain the normal extensibility of shoulder capsule and tight muscular tissues, passive stretching of the shoulder capsule by means of mobilization techniques has been recommended, but limited data for supporting the use of these treatments are available. Due to the performance of techniques (mid-range/end-range mobilizations combined with or without interscalene brachial plexus block), a lack of objective and subjective outcome criteria, an inappropriate research design (case report and clinical trial without control), and utilization of other treatment modalities (home exercises and hot/cold packs), it is not possible to draw firm conclusions about the efficacy of mobilization in patients with FSS. Objective: The aim of our study is to investigate the effect of mobilization treatment and to determine whether a difference of treatment efficacy exists among three mobilization techniques (mid-range mobilization, end-range mobilization, and mobilization with movement) in patients with FSS. Design: We will carry out a crossover multiple-treatment trial on two groups. In group one, an A-B-A-C (A: mid-range mobilization, B: end-range mobilization, C: mobilization with movement) multiple-treatment design will be used. In group two, an A-C-A-B multiple-treatment design will be used. There will be three weeks of each phase. An independent trained outcome assessor, blinded to treatment allocation, will evaluate the participates at baseline and at 3-week intervals for 12 weeks. Outcome assessment include pain perception, disability assessment, and Shoulder complex kinematics and associated muscular activity. Analysis of covariance (ANCOVA) will be undertaken using the follow up data at 3, 6, 9 and 12 weeks for each of the outcomes, with adjustment for the baseline values of the outcome of interest. Independent t-Tests will be conducted to compare change of outcome variables between two groups (A-B in one group vs. A-C in the other group, A-C in one group vs. A-B in the other group).

We will carry out a multiple-treatment trial on two groups. The multiple-treatment trial involves the application of two or more treatments in one single-subject [Portney and Watkins, 2000; Barlow and Hersen, 1984]. It is used to compare the effects of two or more treatments. Consenting patients will be randomized by computer generated permuted block randomization of 5 to receive different mobilization treatments. Allocation concealment will be ensured and the study biostatistician will keep the assignment scheme. In group one, an A-B-A-C (A: mid-range mobilization, B: end-range mobilization, C: mobilization with movement) multiple-treatment design will be used. In group two, an A-C-A-B multiple-treatment design will be used. The two groups used here are intended to counterbalance the order effects of treatments. There will be three weeks of each phase. The differences in outcomes across the four phases of the study will be examined. An independent trained outcome assessor, blinded to treatment allocation, will evaluate the participates at baseline and at 3-week intervals for 12 weeks.

Inclusion Criteria:

• (1) having a painful stiff shoulder for at least 3 months, (2) having a limited ROM of a shoulder joint (ROM losses of 25% or greater compared with the noninvolved shoulder in at least 2 of the following shoulder motions: glenohumeral flexion, abduction, or medial/lateral rotation), and (3) the consent of the patient's physician to participate in the study.

Exclusion Criteria:

• Additional specific exclusion criteria will be patients with (1) diabetes mellitus, (2) a history of surgery on the particular shoulder, (3) rheumatoid arthritis, (4) a painful stiff shoulder after a severe trauma, (5) fracture of the shoulder complex, (6) rotator cuff rupture, or (7) tendon calcification.