A digital cueing intervention for Parkinsonian gait: Laboratory-based clinical validation and acute gait responses

Falls are common in people with Parkinson’s (PwP). There is a need for a scalable, pragmatic, personalisable, and single device intervention to improve gait characteristics related to fall risk. CuePD is an app-only approach to deliver personalised auditory cues to retrain gait in PwP. This study aimed to clinically validate and evaluate effectiveness of CuePD in PwP and to explore their perceptions of CuePD. Sixty/60 PwP performed cued walks via metronome, instrumental and vocal music personalised to their baseline walking speed (cadence). PwP chose music from e.g., pop, rock genres. Cued walks were +10% on baseline cadence. We examined effectiveness of each cue and any carryover effects across non-cued walks. CuePD clinical validation (e.g., intraclass correlation coefficients, ICC(2,1)) was established against a reference standard. CuePD effectiveness was examined via improvement (e.g., standardised response means) of gait characteristics indicative of fall risk (stride length, walking speed, step time coefficient of variation). CuePD quantified fall relevant mean gait characteristics against a reference system with good-to-excellent agreement (ICC(2,1) 0.849 – 0.986), though agreement was poor for several variability and asymmetry measures. Vocal music was the most preferred cue type, improving gait characteristics by increasing stride length (7cm, SRM of 0.873) and gait speed (0.115 m/s, SRM of 1.264), while reducing stride time coefficient of variation (-0.418%, SRM of -0.505). CuePD received positive feedback and is an app-only approach that is clinically valid to quantify mean gait characteristics and has demonstrated acute improvements in gait characteristics related to fall risk. Findings suggest that the personalised approach may have pragmatic utility beyond the lab, but ecological validity needs to be established.