People with Parkinson's Disease (PD) and Freezing of Gait (FOG) reportedly turn using an 'en-bloc' strategy, where the head and pelvis rotate together, unlike the head-leading movement seen in healthy adults. However, previous research relies on discrete maximum separation angles in 180° walking turns, despite recommendations to use 360° on-the-spot turns to better induce FOG. Moreover, existing studies do not capture the time-varying coordination of body segments throughout the turn. Our study aimed to investigate head-pelvis during 360° on-the-spot turns in people with PD and FOG (PD + FOG), PD without FOG (PD-FOG), and healthy controls (HC). Fifteen PD + FOG, fourteen PD - FOG (tested ON medication), and seventeen healthy controls (HC) completed repeated 360° turns while kinematic data were collected using marker-based motion capture. Head and pelvis angles were calculated across strides in the first, middle, and final sections of the turn. Coordination was quantified using vector coding, alongside discrete measures of angular difference and turning performance. Head-pelvis angular difference did not differ between groups. However, PD + FOG showed increased coordination variability compared to HC (4.9°, p = 0.005) and PD - FOG (3.5°, p = 0.047); an observation no longer significant after adjusting for Motor Section (III) of the Movement Disorder Society-Unified PD Rating Scale (MDS-UPDRS) motor scores (p = 0.249) and Mini Balance Evaluation Systems Test (MiniBEST) score (p = 0.051). PD + FOG also took more steps than PD - FOG (34.0, p = 0.008) and HC (6.5, p < 0.001), even after adjusting for covariates (MDS-UPDRS: p = 0.037; MiniBEST: p = 0.003). These findings suggest that people with PD do not necessarily turn more en-bloc than healthy controls. While head-pelvis coordination variability is higher in PD + FOG, this does not seem to be linked to FOG pathology per se, but rather balance deficits associated with disease severity. The impact of balance training on coordination variability and turn performance should be evaluated. Increased step count seems to be related to FOG, which could be a factor that might provoke FOG, but may serve as a compensatory strategy to promote postural stability. Future work should investigate how stride length/frequency modulation during turning affects postural stability and FOG occurrence.