# Does cognition affect supervised and unsupervised mobility differently in people with Parkinson’s disease? A cross-sectional study

## TL;DR

This study finds that cognitive functions, especially executive functions, are more strongly linked to supervised mobility tasks than to unsupervised mobility in people with Parkinson’s disease.

## Contribution

The study is the first to compare the impact of cognition on supervised versus unsupervised mobility in Parkinson’s disease using objective measures.

## Key findings

- Cognitive functions explained 12-15% of variance in supervised mobility tasks but only 5.5% in unsupervised mobility.
- The Word-Colour Stroop Test was a significant predictor of both supervised and unsupervised mobility.
- Cognitive predictors lost significance when clinical and demographic variables were included.

## Abstract

The co-existence of cognitive and mobility impairments is frequent in people with Parkinson’s disease (PwPD). Cognitive functions are associated with different types of supervised mobility, but much less is known about unsupervised mobility.

Mild-to-moderate PwPD, without severe cognitive impairment and not using walking aids, performed three hospital-based supervised mobility tasks: forward walking, instrumented timed-up-and-go (iTUG) with a lower-back-mounted inertial sensor and a 3-meter backward walking test. Forward walking speed (FWS), backward walking speed (BWS) and mean angular velocity (MAV) of the 180° turn during iTUG were calculated. Unsupervised mobility was monitored through average daily steps (AvDS) measured over 5 days by means of a commercial smartwatch (Garmin Vivosmart 4) in free-living conditions. Cognitive performance was evaluated using the Montreal Cognitive Assessment (MoCA) and the Word-Colour Stroop Test (WCST). Multivariable linear regression models with backward elimination method were used to evaluate how much variance of each mobility measure was explained by cognitive functions and the effect of other clinical and demographic variables.

A total of 105 PwPD were recruited. Regarding supervised mobility, multivariable regressions revealed that cognitive functions explained 12.3% (P < 0.001), 12.4% (P < 0.001) and 15.0% (P < 0.001) variance of FWS, MAV and BWS, respectively. For unsupervised mobility, cognitive functions explained only 5.5% (P = 0.009) of AvDS variance. WCST was included in all regression models and was a significant predictor of all mobility parameters, either supervised or unsupervised. MoCA was included only in the model for supervised BWS and was not a significant predictor. When considering clinical and demographic variables, neither MoCA nor WCST survived as significant predictors.

Our findings support the role of executive functions in PwPD mobility. Cognition correlated with supervised mobility more than unsupervised, likely due to the role of individual and environmental factors. These findings suggest that a comprehensive approach to PD should include cognitive assessments alongside traditional motor evaluation. Mobility evaluation in PwPD should be performed in both supervised and unsupervised conditions.

Cognition, particularly executive functions, correlates with supervised more strongly than with unsupervised mobility in people with Parkinson’s disease (PD).

In supervised conditions, cognitive functions are associated with backward walking more strongly than with forward walking and turning.

When clinical and demographic variables are included, cognitive predictors lose statistical significance.

Motor-cognitive evaluation should be integrated in PD clinical assessment and management.

Mobility in PD should be assessed under both supervised and unsupervised conditions.