The Mulligan's MWM group received mobilization with movement in the glenohumeral joint, as described by Mulligan. The participant was seated, and the physiotherapist was placed on the opposite side to the affected shoulder with the thenar eminence of one hand over the anterior aspect of the humeral head, and the other hand over the spine of the scapula posteriorly of the affected shoulder. The hand over the humeral head performed a posterolateral glide, while the other hand stabilized the scapula. This maneuver was sustained while the participant asked to raise the arm in scapular plan (as explained above) without pain. The physiotherapist previously determined the amount of posterolateral pressure which was the best to abolish the pain during shoulder scaption for each participant.

In the Placebo group, no effective intervention was performed, the participant was asked to raise the affected arm in scapular plane without pain, while the physiotherapist placed their hands-on shoulder, with the same manual contacts described in Mulligan's MWM group, but without performing any pressure.

The Mulligan's MWM group received mobilization with movement in the glenohumeral joint, as described by Mulligan. The participant was seated, and the physiotherapist was placed on the opposite side to the affected shoulder with the thenar eminence of one hand over the anterior aspect of the humeral head, and the other hand over the spine of the scapula posteriorly of the affected shoulder. The hand over the humeral head performed a posterolateral glide, while the other hand stabilized the scapula. This maneuver was sustained while the participant asked to raise the arm in scapular plan (as explained above) without pain. The physiotherapist previously determined the amount of posterolateral pressure which was the best to abolish the pain during shoulder scaption for each participant. The participants performed three sets of 10 trials with a resting time of 30 seconds between each set. After intervention was allowed a resting time of three minutes.

The participant was asked to rate the pain intensity in Neer impingement signal and Hawkins-Kennedy impingement test, using a 100 mm visual analogue scale.

Pressure algometry was performed to assess PPT on the lateral aspect of shoulder. A hand-held electronic pressure algometer Force One FDIX (Wagner Instruments, Greenwich CT, United States of America) with a 1 cm2 rubber-tipped probe area was applied perpendicular to the skin at a rate of 1 kg per second. The PPT was measured over the most sensitive point to manual palpation of the lateral aspect of shoulder. This region was marked with a permanent marker so that the same point could be used pre- and post-intervention. The researcher instructed the participant to say "ouch" as soon as sensation changed from feeling pressure to feeling pain. Three trials were performed, with a resting time of 30 seconds and a subsequent analysis of its average.

Photogrammetry was performed to assess ROM of shoulder scaption until onset of pain. The ROM was recorded by photography, using a digital camera (30 frames/ second) placed in a perpendicular position to the plane of motion, at a distance of 3 meters from a flat surface. The photography was subsequently analyzed by a postural analysis software (SAPo version 0.69, open source software). To ensure consistent reference points, spherical markers were placed on the skin: one over the acromion and the last over the lateral epicondyle. The participant was asked to raise the arm until feeling pain, with elbow fully extended and thumb pointed up along a flat vertical surface positioned 30º anterior to the frontal plane.

Surface electromyography (sEMG) was performed to assess the muscle activity of trapezius (upper, middle and lower) and SA, during concentric and eccentric phases of shoulder scaption. The muscle activity was recorded by BioPlux research device (Plux wireless biosignals S.A., Arruda dos Vinhos, Portugal) with 12-bit analog channels and a sampling frequency of 1000 Hz, using double differential electrode leads. To perform sEMG, the participant's hair was shaved, an abrasive cream was used to remove dead cells from the skin's surface, and the skin was cleaned with isopropyl alcohol (70%) to remove oil and remaining dead cells. An electrode impedance checker (Noraxon Corporate, Scottsdale AZ, United States of America) was used to ensure that impedance levels were less than 5 KΩ, which was considered to signify the satisfactory acquisition of the sEMG signal.