Wrist-worn device shows good accuracy for measuring off time
Parkinson KinetiGraph tracks symptom fluctuations, medication side effects
The Parkinson KinetiGraph (PKG), a wrist-worn device resembling a watch, offers a moderate level of measurement for tracking symptom fluctuations and medication side effects in individuals with Parkinson’s disease.
That’s according to the study “Application of single wrist-wearable accelerometry for objective motor diary assessment in fluctuating Parkinson’s disease,” which was published recently in npj Digital Medicine. The work was funded by Global Kinetics, the company that sells the PKG system.
Although there is currently not a cure for Parkinson’s disease, there are a variety of disease treatments that can help to manage symptoms, with the gold standard being levodopa and its derivatives.
While these therapies can be effective for controlling Parkinson’s symptoms, many patients — especially individuals with more advanced disease — will experience off periods, when symptoms aren’t fully controlled between scheduled doses of medication. Also, long-term use of levodopa and similar medications almost always leads to dyskinesia, a side effect marked by sudden, uncontrolled movements.
As Parkinson’s treatment continues to advance, scientists are working to develop new therapies that can help minimize off periods and manage dyskinesia. In clinical trials testing such therapies, it’s necessary to have accurate ways to measure the amount of off time and dyskinesia that patients are experiencing. Traditionally, this has been done via patient-kept diaries. However, as with any records kept by human beings, these diaries may be vulnerable to biases and errors.
As technology like smartphones has advanced in recent years, several systems like the PKG have been developed to track Parkinson’s symptoms using a simple sensor worn on the wrist. Theoretically, these devices could allow for more objective measures that aren’t as vulnerable to human error.
Previous studies have shown that measurements taken by PKG and similar systems tend to line up reasonably well with traditional diary-based measurements. But prior research has looked mostly at whether the devices can assess the total amount of off time over the course of an entire day.
Ccientists at Global Kinetics and other institutions wanted to see whether PKG-based measurements also would line up with more granular measures assessing symptom changes over time.
“Our main aim is to establish whether [measurements using the PKG device] can detect the clinical state at a point in time using simultaneous half-hourly performed clinical examinations by experienced raters with the [Parkinson’s] home diary as the main comparator,” the scientists wrote.
Parkinson KinetiGraph times converted for this study
The team noted that, while the PKG system used in this study is commercially available, the commercial version is designed to take measures day-by-day. For this study the researchers converted PKG measurements into half-hour chunks. These measurements, referred to as accelerometer-based digital Parkinson’s Motor Diary (adPMD) measures, aren’t available with the commercial version of the device.
The study, conducted at centers in Germany and Sweden, included data from 63 people with advanced Parkinson’s who had two days’ worth of adPMD data available, with matched data from patient-kept diaries and ratings taken by clinical experts.
In total, there were adPMD measures for 2,384 half-hour periods. Among these, 40.9% were classified as off time, and another 20.4% were classified as times with dyskinesia. About a third (32.7%) of the adPMD measurements were classified as on time — when symptoms are well-controlled without dyskinesia — and the remaining 6% of measurements were classified as being when patients were asleep.
The researchers constructed statistical models to see how well these labels lined up with what was documented by patients and clinical experts.
Results showed the adPMD measures matched up with more than half (55%) of expert ratings and 40% of patient ratings. The sensor-based measures showed particularly good agreement at identifying periods of off time: adPMD scores matched with 65% of clinician-labeled and 57% of patient-rated off times.
Statistical models generally showed meaningful correlations between adPMD scores and ratings of off time and dyskinesia, but not for on time.
“We demonstrated moderate validity for Off time and Dyskinetic time, but only limited correlation/validity for the other measures of motor fluctuations,” the scientists wrote.
In further analyses where the researchers calibrated adPMD data based on initial clinician measurements, there was better agreement between adPMD data and clinician measurements at identifying dyskinesia, but at the cost of less accuracy for detecting off periods.
More PKG device studies needed
Overall, these data support the use of the PKG device particularly to assess off episodes, though the researchers stressed that additional research is needed to determine whether this can be used reliably in clinical trials.
“Future studies in larger patient cohorts are warranted to confirm our data on thresholding for PKG data transfer into motor state diary data and to evidence the adPMD using the PKG device as a suitable trial endpoint,” the scientists concluded.