Reliability seen in using smartwatch to assess patients’ motor function
Tasks assigned by Verily Smart Watch also capture 'on' and 'off' states, study says
Motor abilities assessed via multiple sensors on a smartwatch correlated well with motor function as rated by clinicians, and could distinguish between “on” and “off” medication states for Parkinson’s disease patients, according to a study in Japan.
“This study has shown that wearable devices can provide reliable and objective data from continuous monitoring of motor signs and symptoms in patients’ daily lives,” the researchers wrote. But they noted that more work is needed, particularly in developing “reliable digital biomarkers” and in assigning motor tasks “more suitable for long-term monitoring.”
Findings in hospitalized patients “will be used to design future studies to assess levodopa effect sizes and treatment response in home-based, naturalistic settings,” the team added.
The study, “Analytical and clinical validity of wearable, multi-sensor technology for assessment of motor function in patients with Parkinson’s disease in Japan,” was published in Scientific Reports.
Importance seen in home monitoring of Parkinson’s motor symptoms
To evaluate disease progression, patients typically undergo motor function assessments at doctor appointments held intermittently throughout the year. At these visits, doctors look into hallmark Parkinson’s motor symptoms, such as bradykinesia (slowed movements), walking abnormalities, and tremors.
“However, assessments of motor symptoms are based on subjective clinician ratings and, because they are performed in a clinic, can only capture a ‘snapshot’ of the motor function that patients experience in their daily lives,” the researchers wrote, with repeat visits required when adjusting the dose of medications like levodopa, a standard treatment.
Wearable technologies have emerged as a potential solution, allowing for continuous, at-home monitoring of patients’ movements.
One such technology is the Verily Study Watch from Verily Life Sciences. This multisensor smartwatch is worn around the wrist and passively collects data related to a person’s daily movements. In a previous study, it was able to distinguish Parkinson’s patients from people without the disease based on recorded movement.
It also contains a structured, seven-item virtual motor examination (VME), assigning tasks that are used to assess tremors, bradykinesia, gait, and postural imbalance. People wearing the watch are prompted on its screen to move through these activities, and composite motor scores are generated from their performance.
Scientists, largely at Juntendo University Faculty of Medicine in Tokyo, conducted a clinical trial to examine the watch’s reliability in monitoring motor function among 96 Parkinson’s patients in Japan (54% women), with a mean age of 62.3 and evident Parkinson’s symptoms for a mean of 12.5 years.
Participants were asked to wear the watch on their dominant arm for up to 23 hours a day, with most (90.6%) choosing to wear it on their left arm, the study noted.
Patients using the watch did seven motor tasks twice daily for five days
All were being treated with levodopa but experiencing off periods, also called motor fluctuations, times when motor symptoms return between scheduled levodopa doses. These people required hospitalization to monitor their motor symptoms.
During the main part of the trial, hospitalized patients used the watch to perform the VME twice daily for five days. During this time, they also were evaluated by a trial investigator for motor function, as assessed using the Movement Disorder Society — Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) part III.
These evaluations were conducted both during an on medication state — when levodopa had motor symptoms under control — as well as off medication state, periods between doses when motor fluctuations occur.
Their five-day hospital stay was preceded by an optional, one-month period during which 35 patients performed the VME twice daily at home. After leaving the hospital, 84 people completed another month of VME assessments at their home.
Patient-reported outcomes were recorded before and after the study, including questions about the devices’ usefulness and feasibility.
Across the entire study, the mean percentage of VMEs completed was 60.9%, and most were done within 10 minutes.
VME tasks skipped most often were the final two requested — the “up and go” and “postural stability” tasks, the researchers reported, noting they also are the most difficult ones assigned. No difference in skipping was seen between on and off medication states.
Most participants (72.8%) reported being either satisfied or neutral regarding smartwatch comfort, and most found it easy to use. Still a majority, 62%, said they would not wear the watch longer than the study period. Reasons for this decision were not asked.
In general, composite motor scores obtained from the digital examination correlated well with neurologist-rated motor function during the patients’ hospital stay, the researchers wrote. The closest correlation was observed with the overall motor score, with a weaker relationship between individual components (i.e., bradykinesia, tremor, gait).
Wearable device ‘sensitive’ to patients’ medication state
Notably, digital measurements were sensitive enough to detect changes in medication state, both when the VME was conducted with and without a neurologist’s supervision.
“Although use of wearable devices for the detection of … the OFF state is challenging, this study suggested that detection of the OFF state may be feasible by combining a VME with passive movement monitoring in an inpatient [hospital] setting,” the researchers wrote.
Evaluations of VME data on a weekly — instead of daily — basis during home use found the watch to have “average-to-good” reliability across measures of bradykinesia, gait, tremor, and overall motor function.
Most digital measurements collected during the study’s home-based portions correlated weakly with patients’ self-reported daily functioning.
“These results show that unsupervised digital measurements of motor features with wrist-worn sensors are sensitive to medication state and are reliable in naturalistic settings,” the researchers wrote.
As a “proof-of-principle” study, however, its findings are preliminary. “Further research is required to confirm the use of digital measurements in home-based settings and as monitoring biomarkers,” they added.
Scientists with Verily Life Sciences and Takeda Pharmaceuticals took part in this research.
About the Author
