Demographic data.

Introduction
People with Parkinson’s disease (PwPD) present motor alterations which can impact daily life tasks that require speed and/or accuracy of movement.
Objective
A sub analysis of NCT01439022, aiming to estimate the extent to which two different exercise training protocols (global and handwriting upper limb exercise training) impact reaction time, travel speed, and accuracy in PwPD.
Methods
Seventy PwPD, right-side dominant were randomised 1:1 into two six-month training protocol groups; 35 PwPD performed global exercise training and 35 performed specific training (handwriting upper limb exercise movements). Assessments of speed-accuracy and trade-off were carried out at baseline, after 3 and 6 months of training, and at a 12-month follow-up. The current study used data from a previous publication of a randomised controlled trial that included a 6-month self-managed community exercise programme for PwPD. For the present study we included only the participants who completed the Fitts’ task during the baseline assessment.
Results
In the upper limb assessments, no main effects were found for the number of touches, but the exercise group showed a marginal increase over time on the left side. Error averages on the left side decreased significantly for the exercise group from baseline to 6 and 12 months. The exercise group also presented a lower Error CoV and the Reaction Time CoV increased on the right side. Significant findings for Fitts r on the left side indicated lower values for the exercise group, with improvements continuing at 12 months.
Conclusion
We report the potential of global exercise interventions to facilitate improvements in reaction time and travel speed, as well as other motor control metrics, with lasting effects at 12 months, particularly on the non-dominant side.