Dance intervention for 3 months, 2 times/week (1 instructor lecture session and 1 home session following a prepared video), 2 hours per week. Dance is a safe and effective form of activity, which has been used in PD population (Delabary et al., 2017; Sharp and Hewitt, 2014). Participants take part in a dance class specifically for people with PD lead by a qualified dance instructor. The dance class typically include a warm-up, dance related activities (specific to the genre of the class) and a cool-down.

Controls, not alter their personal lifestyle, but undergo the same testing as the exercise intervention group.

The dance program consists of 3 months of dance classes. Classes are specifically designed for people with Parkisnon's and lead by a qualified dance instructor experienced in teaching this population. The in-person dance classes are conducted in an appropriate room with chairs and ballet barres once a week. In addition, the participants received a CD containing a video of the recorded dance routine to be performed at home once a week. The class typically included a warm-up (10-15 min.), barre ballet exercises (10-15 min.), general dance-related activities (15-20 min), and a cool-down (10-15 min).

The beginning of the turning phase started from the last heel-strike of the regular straight line walking pattern. The end of the turn was the final heel-off at the end of the turning motion, where the participant was ambulating in the opposite direction and prior to the cyclic straight line gait pattern resuming. The exact time points were taken from the peak anterior-posterior linear velocity of the lower leg data corresponding with the specific heel-strike and heel-off instances. The number of steps and the total time taken to complete the 180 turning phase of the TUG test were measured.

The girdle dissociation is a selected biomechanics parameter. To obtain the value for the girdle dissociation, the difference between the orientation of the pelvis and the affected shoulder in the transverse plane, was calculated for each data point throughout the 180 turning phase of the TUG test. The range between the maximum and minimum dissociation angles was reported and further analysed.