Relating movement behaviours and non-motor characteristics in people with Parkinson's disease: A compositional data analysis approach

Background People with Parkinson's disease (PwPD) have unhealthier movement behaviours (less moderate-to-vigorous physical activity (MVPA) and more sedentary behaviour (SB)), than healthy older adults. Associations across movement patterns and non-motor characteristics are poorly understood. Objectives To investigate associations between relative time spent in MVPA, light-intensity physical activities (LIPA) and SB, and non-motor characteristics among PwPD, and to investigate theoretical changes in non-motor characteristics when time in different movement behaviours is reallocated. Methods Baseline data from 119 participants in the STEPS randomised controlled trial was used. Movement behaviours were measured by ActiGraph GT3X accelerometers. Compositional data analysis assessed relative time in MVPA, LIPA and SB. Linear regression assessed associations between MVPA, LIPA and SB and self-reported anxiety and depression (HADS), executive function (TMT IV), self-efficacy for exercise (S-ESES) and activities-specific balance confidence (ABC). Isotemporal substitution modelling investigated theoretical changes in outcomes when time in MVPA, LIPA and SB were reallocated. Results Better executive function was associated with more relative time in MVPA and less in LIPA. Higher exercise-self-efficacy was associated with more relative time in MVPA and less in SB. Better balance confidence related to more relative time in MVPA. Reallocating time showed that losing 20 min MVPA had a worse theoretical impact for these outcomes than the benefit of gaining 20 min. Conclusions The observed relationships between MVPA and executive function, balance confidence, and exercise-self-efficacy suggests particular importance of maintaining MVPA in PwPD. These findings can be utilized clinically by communicating the importance of maintaining time in MVPA among PwPD.