Objective: To build a fall prediction model that includes cognitive and motor outcomes in a subset of patients with Parkinson’s disease (PwPD) enrolled in the CYClical Lower Extremity Exercise for Parkinson’s disease II (CYCLE-II) clinical trial.

Background: Accurate prediction of falls in PwPD can profoundly impact effective prevention efforts and facilitate tailored treatment. Historically, fall risk models have relied heavily on motor features of PD, despite the interplay between cognitive and motor processes needed for community locomotion.

Method: In sum, 123 participants with PD (65.3±8.3 years, 81 (66%) males, disease duration 4.9±4.1 years) completed a comprehensive motor and cognitive assessment and were followed for 12-months. Falls were recorded via a fall diary and verified during biweekly study coordinator calls. A LASSO logistic regression model was developed to assess thirty-seven demographic, motor, and cognitive variables and identify key fall predictors. To explore the practical implementation of predicting falls in a clinical setting, a second model was developed using a subset of nine candidate measures conducive for retrieval from electronic medical records (EMR).

Results: Over 12 months, 72 participants (58.5%) fell at least once; with 33.1% occurring during walking, 34.4% resulting in injuries, and 13.1% resulting in injuries that required medical care. The initial prediction model, incorporating 37 candidate variables from the CYCLE-II trial, identified 8 outcomes that predicted falling, with an AUC of 0.68. In the second model, using EMR data, three outcomes, two of which were cognitive, achieved an AUC of 0.67. Those whose speed of information processing and spatial memory were worse were more likely to fall over the 12-month period.

Conclusion: Model results indicate impaired cognitive performance and longer disease duration are powerful predictors in identifying a future fall in PwPD. The link between cognitive performance and potential for falling reinforces the strong interplay between gait and cognition. It is recommended that cognitive screening be part of routine clinical care of PwPD to aid in the identification of those at greater risk of falling.

« Back to 2024 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/spatial-memory-and-processing-speed-as-cornerstones-in-predicting-falls-in-parkinsons-disease/