A training approach to improve stepping automaticity while dual-tasking in Parkinson's disease: A prospective pilot study

Abstract

Background: Deficits in motor movement automaticity in Parkinson's disease (PD), especially during multitasking, are early and consistent hallmarks of cognitive function decline, which increases fall risk and reduces quality of life. This study aimed to test the feasibility and potential efficacy of a wearable sensor-enabled technological platform designed for an in-home music-contingent stepping-in-place (SIP) training program to improve step automaticity during dual-tasking (DT).

Methods: This was a 4-week prospective intervention pilot study. The intervention uses a sensor system and algorithm that runs off the iPod Touch which calculates step height (SH) in real-time. These measurements were then used to trigger auditory (treatment group, music; control group, radio podcast) playback in real-time through wireless headphones upon maintenance of repeated large amplitude stepping. With small steps or shuffling, auditory playback stops, thus allowing participants to use anticipatory motor control to regain positive feedback. Eleven participants were recruited from an ongoing trial (Trial Number: ISRCTN06023392). Fear of falling (FES-I), general cognitive functioning (MoCA), self-reported freezing of gait (FOG-Q), and DT step automaticity were evaluated.

Results: While we found no significant effect of training on FES-I, MoCA, or FOG-Q, we did observe a significant group (music vs podcast) by training interaction in DT step automaticity (P<0.01).

Conclusion: Wearable device technology can be used to enable musically-contingent SIP training to increase motor automaticity for people living with PD. The training approach described here can be implemented at home to meet the growing demand for self-management of symptoms by patients.

Figure 1

Overview of study design. Schematic of the study design (Panel A). Participants were tested before (“pre”) and after (“post”) a 4-week in-home Ambulosono intervention program. Participants were asked to use the device in-home a minimum of 3 times per week, for 10 to 20 minutes per training session. Pre- and post assessments included The Falls Efficacy Scale International (FES-I) survey for fear of falling, the Freezing of Gait Questionnaire (FOG-Q) for freezing, and The Montreal Cognitive Assessment (MoCA) for general cognitive functioning. The dual-tasking test (DT test) consisted of measuring step-height decrement while SIP and simultaneously performing cognitive tasks (shown in Panel B). The order was standardized as done previously.^[4] The control group was identical to the Ambulosono music group with the exception that instead of contingent music playback, they received contingent auditory feedback in the form of a CBC podcast (CBC Quirks and Quarks segment: The Sloth's Pharmaceutical Fur).

Figure 2

Basic concept of Ambulosono Music-Contingent Training. The basic concept of the Ambulosono technological platform is to computationally link auditory musical reward stimulation to scalable motor action. Based on an individuals’ mono-task SIP step height, a predefined step-height training threshold (red dotted line) can be set for auditory feedback. Therefore, during training, if step height (black trace) is above the threshold (red dotted line), music will play (green regions), while if step height is below the threshold, music does not play. This allows participants to use real-time step measurements and analytics that provide biofeedback to help self-motivate and self-activate reward networks with training.

Figure 3

Improvements in Dual-Tasking Step Automaticity Following Ambulosono Training. Plotted are the fixed effects parameter estimates (i.e., mean and standard error) from the linear mixed-effect model. Colors represent average changes in automaticity ratio (DT/MT) for each DT condition (i.e., immediate word recall, red; delayed word recall, green; serial subtract 5, yellow; serial subtract 7, purple). While there was no significant DT condition by training interaction (F_(3,69) = 0.07, P = 0.98), there was a significant group (i.e., music vs podcast) by training interaction (F_(1,69) = 9.20, P = 0.003).