Objective: The purpose of this study was to determine the reliability and clinical validity of smartphone- and smartwatch-based, passively acquired measures of motor behavior in daily life of individuals recently diagnosed with Parkinson’s disease (PD).

Background: Motor functioning in daily life is the most ecologically valid reflection of motor disease severity, but is difficult to quantify objectively. Smartphone- and smartwatch-based sensors may represent a passive and objective means towards this aim.

Method: Data from 316 individuals with early PD (<2y) participating in a phase II clinical trial (PASADENA, NCT03100149) were analyzed. Two-week averaged sensor data (centered around the baseline visit) and MDS-UPDRS scores were corrected for age and gender using regression analysis. Intraclass correlation coefficients (ICC) were calculated over two consecutive 14-day averages. Spearman’s rank order correlations and logistic regression tested for associations between digital sensor data and MDS-UPDRS scores.

Results: Strong evidence of test-retest reliability was demonstrated (all ICC’s>.7; median ICC=.89, all p<.001). Hand gesture features negatively correlated with MDS-UPDRS total, parts 1, 2 and 3, bradykinesia and rigidity subscores (all rs[≥263] ≤.-14, p<.05), and positively correlated with tremor subscore (rs[≥263] ≥.14, p<.05). Measures of gait negatively correlated with MDS-UPDRS total, parts 2 and 3, bradykinesia, rigidity, and tremor subscores (all rs[≥264] ≤.-15, p<.05), and positively with the axial symptom score (rs[≥264] ≥.13, p<.05). Hand gesture and gait features predicted self-reported and clinician-evaluated motor impairment on MDS-UPDRS item level: e.g. individuals with impairment on left hand pronation/supination movement showed less gesture power (β=-0.19, z=-5.8, p<.001) and individuals reporting problems when getting out of their bed, a car, or a deep chair showed reduced frequency of turns while walking in everyday live (β=-.05, z=-2.3, p<.05).

Conclusion: These findings provide support for the preliminary reliability and validity of passively monitored daily motor behavior with the Roche PD Mobile Application v2. These findings are notable in the context of the overall low levels of impairment in this PD population.

« Back to MDS Virtual Congress 2020

MDS Abstracts - https://www.mdsabstracts.org/abstract/passively-measuring-motor-behavior-in-daily-life-preliminary-reliability-and-validity-in-individuals-recently-diagnosed-with-parkinsons-disease/