JMIR Publications

ABSTRACT

Background:

Clinical assessment of tremor in Parkinson's disease (PD) is limited by human vision, leading to potential under-reporting of tremor occurrences. Utilizing technology-based solutions, such as smartwatches, may contribute to the enhanced documentation of tremors in the routine of individuals with Parkinson's disease.

Objective:

The objective of this study was to test whether a smartwatch-based system utilized in a clinical setting could detect clear tremor in Parkinson’s disease cases where no tremor had been previously documented in routine care.

Methods:

A total of 219 participants, including 142 individuals with Parkinson’s disease and 77 healthy controls, were included in this study. During routine visits at the outpatient clinic of movement disorders at a tertiary care hospital, each participant underwent a 20-second resting assessment using smartwatches positioned on both wrists. The smartwatch sensor recordings were then used to calculate a tremor occurrence score. This score was compared to the routine documentation of tremor as part of UPDRs (Unified Parkinson's Disease Rating Scale) forms and free-text discharge letters.

Results:

In 78 out of 142 Parkinson’s disease participants (55 %) a tremor was documented in routine documentation, whereas in 64 Parkinson’s disease participants (45 %) a tremor was not documented. Among them, 16 PD participants (25 %) showed subtle but clear tremor in smartwatch-based analysis with average frequency of 6.20 Hz and average amplitude of 0.07g.

Conclusions:

The findings suggest that a smartwatch-based system can be beneficial in enhancing tremor documentation, particularly in the detection of subtle or low amplitude tremor that might have been missed during routine clinical assessments. Consumer-grade devices such as smartwatches can increase documentation completeness of tremor and provide additional details such as tremor frequency and amplitude for fine-grained assessment or monitoring. Clinical Trial: NCT03638479