SCREENING RISK OF FALLS IN PEOPLE WITH PARKINSON’S DISEASE USING A DIGITAL APP: A FEASIBILITY STUDY

T. Capato^1,2, J. Miranda¹, V. Grabois¹, F. Santos¹, F. Almeida¹, R. Carra¹, R. Cury¹, M.J. Teixeira¹, E. Barbosa¹

¹University of São Paulo, Neurology, São Paulo, Brazil, ²Radboudumc, Neurology, Nijmegen, Netherlands

Background: Digital technologies promise to change research and treatment monitoring in Parkinson's Disease (DP). Center of Body Mass (COM) objective measurements such as accelerometers and gyroscopes data extracted from a smartphone have been studied due to their practicality in stabilometry assessments. However, it is unknown which is the better digital system to screening the risk of falls.

Purpose: To verify the feasibility of the TechBalance-App to assess the risk of falls in PD patients in a single clinical setting assessment.

Methods: In this observational cohort study. We collected a single clinical setting assessment with 100 PD patients (1-4 H&Y stage). TechBalance-App consists on a self-report of falls, by the clinical questionnaire and an objective measure collected by Stabilometric variables. Participants completed the TechBalance-App's interview (Risk of falls questionary and Fragility scale). After the interview, patients performed motor tests (MBEST and TUG) while data COM were collected by using a gyroscopic and an accelerometer allocated in a smartphone attached to the patient’s body by a belt. The data collected by this smartphone-based digital assessment is processed on a digital platform. The scores of the questionnaires were compared between fallers and non-fallers by the ANOVA test for independent measures and by the ability in discriminate fallers by the Area Under de Curve (AUC) of the ROC curve and traditional classification measures.

Results: TechBalance-App use was acceptable for all participants who completed the assessments (59% man); mean age 68,1. According fragility scale, 67% of patients were in fragile profile. TechBalance-App demonstrated to be feasible to obtain stabilometry parameters, whereas the App generates a score between low (2%), medium (75%), high (16%), and super high (7%) risk of falls. We found a difference between the H&Y severity PD stage and the App's scores (p=0.005). In addition, the risk of falls scores significantly related to corresponding MDS-UPRDS III item number 3.12. In addition, the risk of falls scores also showed a significant correlation with MBEST(p=0.005) and TUG (p= 0.003).

Conclusions: Our smartphone-based digital assessment results indicate that TechBalance-App provides sensitive fall rates and is a feasible app to screening the risk of falls in people with PD. Further studies should investigate the test-retest reliability, validity, and clinically meaningful to be used as an outcome in future PD trials and others movement disorders.

Implications: Inertial sensors have been proposed for postural balance measurements due to their portability and low-cost characteristics. The mobile stabilometry proposed by TechBalance-App may be promising as a balance measure due to its correlated with balance and gait tests, is portable and has a stable data characteristics.

Funding acknowledgements: We also thank the support of University of São Paulo and PHYSICAL Parkinson’s Disease and Movement Disorders Rehabilitation Center

Keywords:
Parkinson's disease
Falls
Digital technology

Topics:
Neurology: Parkinson's disease
Neurology
Innovative technology: information management, big data and artificial intelligence

Did this work require ethics approval? Yes

Institution: University of São Paulo

Committee: Comissāo de Ética para Análise de Projetos de Pesquisa

Ethics number: 4282414