Does a Sway-Based Mobile Application Predict Future Falls in People With Parkinson Disease?

Area of Science:

• Neurology
• Biomedical Engineering
• Gerontology

Background:

• Falls are a significant concern for individuals with Parkinson disease (PD), impacting mobility and quality of life.
• Accurate prediction of falls is crucial for implementing timely interventions and preventing injuries.
• Current clinical measures for fall risk assessment in PD have limitations.

Purpose of the Study:

• To evaluate the predictive capability of the Sway mobile application for falls in individuals with Parkinson disease (PD).
• To compare the sensitivity and specificity of Sway in identifying fallers against established clinical assessments.

Main Methods:

• An observational, cross-sectional study was conducted with 59 participants with idiopathic PD (Hoehn and Yahr stages I-III).
• Assessments included the Sway app, Movement Disorders Society-Unified Parkinson's Disease Rating Scale (MDS-UPDRS) motor exam, Mini-BESTest, and Activities-specific Balance Confidence (ABC) Scale.
• Participants' 6-month fall history was recorded, and logistic regression was used to analyze fall prediction.

Main Results:

• A predictive model including fall history, ABC Scale, and Sway explained 61% of fall variance, correctly classifying 85% of fallers.
• However, only past fall history and the ABC Scale were statistically significant predictors (P<.02).
• The ABC Scale (AUC=0.76) and Mini-BESTest (AUC=0.72) showed higher accuracy in predicting falls than Sway (AUC=0.65).

Conclusions:

• The Sway application did not enhance the accuracy of predicting future falls in individuals with PD when compared to existing clinical measures and self-reported fall history.
• Past fall history and the ABC Scale remain more reliable indicators of future fall risk in this population.