Wearable sensor measurements in relation to clinical characteristics and mortality in patients with Parkinson’s disease

TL;DR

Wearable sensors in Parkinson's patients show that high bradykinesia scores are linked to higher mortality and less treatment, suggesting they could be a useful clinical marker.

Contribution

This study establishes a link between wearable sensor data and mortality in a population-representative sample of Parkinson’s disease patients.

Findings

High bradykinesia scores (BKS > 25) correlate with increased mortality and less intensive treatment.

Sensor-based assessments reflect clinical severity and patient-reported outcomes in PD patients.

BKS is a potential marker of undertreatment and mortality risk in Parkinson’s disease.

Abstract

Parkinson’s disease (PD) is highly variable between patients, and regular assessments are needed to adjust symptomatic treatment. Wearable sensor measurements can complement clinical examinations and patient-reported outcome measures in the management of PD, but their clinical usefulness is yet to be established. Previous studies have described wearable sensor measurements from selected patients, often with advanced disease, but not PD patients in general. We sought to objectively describe daily-life movement characteristics of population-representative patients with PD using the wrist-worn Personal KinetiGraph (PKG), and to relate these sensor measurements to clinical data and proposed PKG-based treatment targets.

Individuals in a population-based random sample of patients with PD were evaluated with clinical assessments, patient-reported outcome measures and six-day PKG recordings. PKG outcomes included bradykinesia score (BKS) and dyskinesia score (DKS), reflecting symptom severity. Mortality within eight years was recorded.

The study included 286 patients with a median age of 73 years and a median time since diagnosis of five years. Clinical and patient-reported variables expected to increase with disease or symptom severity were positively correlated with bradykinesia score (BKS) and negatively correlated with dyskinesia score (DKS). Patients with BKS > 25 had lower levodopa-equivalent daily dose, higher self-reported symptom burden and lower health-related quality of life compared to patients with BKS ≤ 25. Age- and sex-adjusted mortality rates were 1.8 times higher in patients with BKS > 25.

Sensor-based motor assessments with PKG reflect clinical outcome measures in population-representative PD patients. A high degree of sensor-assessed bradykinesia (BKS > 25) at baseline was associated with higher mortality, shorter disease duration and less intensive treatment. The results suggest that high BKS is a risk factor for mortality, possibly as a marker of undertreatment.

ClinicalTrials.gov identifier NCT03130595, registration date 22 April 2017.

The online version contains supplementary material available at 10.1186/s42466-026-00474-8.

Figures