Currently, different Parkinson’s disease (PD) rating scales exist to assess symptoms and quality of life. The two most widely used and accepted rating scales are the Movement Disorder Society-sponsored Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) and the Hoehn and Yahr Scale (1). Though these rating scales have no doubt offered insight and helped classified PD stages since their introduction and revision (MDS-UPDRS, 1987; Hoehn and Yahr, 1967), a lack of accessibility presents a limitation. Specifically, both these scales require testing carried out by a healthcare professional. In patients who progress quickly, do not have access to health care, or have low understandings of health literacy, these rating scales are not ideal. However, a recent 2018 study may offer some hope to remediate this lack of accessibility.
Among the vast research providing insights and expanding knowledge on PD, researchers have proven the use of smartphones paired with machine learning to be able to develop a PD severity score (2). This score could help keep track of symptom progression and consequently help identify clinical care needs. Specifically, the study by Zhan et al developed a smartphone application using smartphone sensors, named HopkinsPD, to assess five important activities impacted by PD: voice, finger tapping, gait (that is, patterns observed as one walks), balance and reaction time. A set of activities, such as speaking into the smartphone’s microphone, can be done at any time of day by an individual with the application. The app also allows users to enter their use of medications/therapy, offering insight on the intensity of symptoms around times of treatment. To create a rating scale score based on the data able to be provided by the app, a rank-based machine-learning algorithm -specifically, a disease severity score learning algorithm- was used. The developed rating scale is dubbed the mobile Parkinson disease score (mPDS), whose scale is from 0 to 100. Higher scores indicate greater symptom severity and concern.
In addition to the increased accessibility and real-world setting data collection offered by HopkinsPD, the mPDS takes into consideration the different amounts of burden caused by each assessed activity, offering a more well-rounded rating system of motor symptoms. For example, the MDS-UPDRS part III contains a biased process geared towards tremor-presenting individuals, as only 5 of 33 of the scale’s activities investigate gait or balance. In comparison, the mPDS’ focus on gait and balance is of 56.6%. The treatment reporting feature of the app also offers promise in assessing how effective medications are at improving symptoms, as the app can collect data at several points throughout the day from treatment intake. Further, the concept of using a smartphone app to assess PD severity allows for a more specific view of an individual’s symptoms, as they can be assessed multiple times a day, every day, in comparison to visiting a clinic once every few weeks. This also eliminates the possibility of physician bias. Physicians and PD specialists may instead use the data collected by the application and the resulting mPDS score to determine the best treatment plan for their patient, and evaluate the impact of said treatment on symptom progression.
Of course, the promise presented by this study does not come without limitations. The mPDS score does not take into consideration the neuropsychiatric symptoms of PD, such as depression and anxiety. Adding questionnaires on these symptoms within the HopkinsPD app would be of value to ensure a more representative rating score for PD. Additionally, though smartphones are widely used, there is a lack of accessibility for those without access to these devices. However, in contrast to trouble with access to care and long wait times for special
ist appointments, this lack of accessibility may not be as dramatic and the app thus offers a better alternative in this respect. Further, the app was designed to be understood by a grade seven understanding of health literacy, thus also increasing the app’s accessibility. Thus, overall, HopkinsPD and the mPDS rating scale is an exciting tool in the therapeutic development of PD.
Bibliography
Parkinson's Europe. Rating scales. www.parkinsonseurope.org. Published March 2017. https://www.parkinsonseurope.org/about-parkinsons/symptoms/rating-scales/
Zhan A, Mohan S, Tarolli C, et al. Using Smartphones and Machine Learning to Quantify Parkinson Disease Severity. JAMA Neurology. 2018;75(7):876. doi:https://doi.org/10.1001/jamaneurol.2018.0809
Comments