Pauline Anderson

June 08, 2017

VANCOUVER — Researchers are using smartphones to collect and analyze data on motor symptoms and mobility of patients with early Parkinson's disease (PD).

The technology reflects how well patients with PD are functioning and might eventually be used in clinical trials to determine whether therapeutic interventions are working, Michael Lindemann, PhD, Roche Pharma Research and Early Development, Basel, Switzerland, told Medscape Medical News.

Dr Lindemann presented the research during a poster session here at the International Congress of Parkinson's Disease and Movement Disorders (MDS) 2017.

As part of a phase 1 clinical trial, the current analysis included 44 mostly male patients with PD at eight centers in the United States (mean age, 58.5 years), who were at a relatively early stage in their disease (mean Hoehn & Yahr stage of 1.9). Their total Movement Disorders Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS) score was 42.2.

These patients were provided with smartphones, which they used to carry out active assessments for 24 weeks at home. They were prompted to perform six active tests daily that targeted postural tremor, rest tremor, sustained phonation, balance, gait, and dexterity.

The research also included 35 healthy age- and sex-matched controls without PD at a site in Switzerland who performed the same active assessments daily for 6 weeks.

Passive Data

In addition to the active assessments, study participants were asked to carry the smartphone as part of their daily routine to collect passive data.

"Once patients completed the active test workflow, which they did every morning, we asked them to put the phone in their pocket and not worry about it; just to carry it around," said Dr Lindemann.

Study patients recharged their device at night. They regularly uploaded the data to a server using any WiFi network.

The researchers found that participants were keen on using the device even though they were not able to see the data. For the active testing, adherence was 61%, with patients performing the tests on average at least three times a week.

"Patients are motivated to do the assessments over a long period of time," commented Dr Lindemann.

For the passive monitoring, adherence was maintained over 6 months, starting from about 6 hours per day at the beginning of the study to about 4 hours per day at week 25.

The study also showed that the technology actually works. The active test data agreed with physician scores on corresponding MDS-UPDRS items.

"We have very detailed clinical data and have been able to compare the sensor data with the standard MDS-UPDRS and the agreement is very good," said Dr Lindemann.

He noted that the system collects much more data than would be possible in a clinical setting.

"In clinical standard of care, a patient would see the doctor usually once every 6 months," he said. "Even in a clinical trial, an assessment like the MDS-UPDS would at best be conducted monthly compared to our approach, which enables daily data collection and therefore a better understanding of disease progression and fluctuation."

Resting Tremor

The system was able to detect resting tremor in patients with PD who had no tremor reported by clinicians.

"We see something with the sensor data that you don't see in the clinic," said Dr Lindemann.

"While this may suggest that the system is more precise than the assessment by the doctor, keep in mind that the physician sees the patient only for a short time on a single day while we use 4 weeks of sensor data around each of these in-clinic visits. It may well be that the tremor simply wasn't there during the isolated short visit in the clinic."

In addition, a human activity recognition model trained on public data allowed reliable identification of six activities of daily living (walking, using stairs, jogging, sitting, standing, and lying down) from the passive monitoring data. High agreement is reported between gait features, such as step frequency for active gait tests, and passive monitoring of gait.

This, said Dr Lindemann, further validates the approach.

He also noted that the passive monitoring data showed the impact of PD on everyday motor behavior.

"Mobility features, such as gait activity and sit-to-stand transitions, were significantly reduced in our early-stage PD patients compared to age-matched controls, demonstrating that this technology and approach can measure the effects of Parkinson's disease on patients' activities of daily life and involves virtually no burden for them."

Asked to comment, Alberto J. Espay, MD, associate professor of neurology, director and endowed chair, James J. and Joan A. Gardner Center for Parkinson's Disease and Movement Disorders, University of Cincinnati, Ohio, said this new research is important.

"It's relevant because it harnesses the power of a ubiquitous tool, the smartphone, to collect data heretofore unavailable to clinicians caring for patients with Parkinson's disease and other movement disorders."

However, Dr Espay noted one caveat. He questioned whether the changes that were captured are truly relevant to patients.

"Big data does not mean good data and the interpretation could potentially lead clinicians astray," he said.

International Congress of Parkinson's Disease and Movement Disorders (MDS) 2017. Poster 541. Presented June 6, 2017.

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.