Accuracy of fall prediction in parkinson disease: Six-month and 12-month prospective analyses

Ryan P. Duncan, Abigail L. Leddy, James T. Cavanaugh, Leland E. Dibble, Terry D. Ellis, Matthew P. Ford, K. Bo Foreman, Gammon M. Earhart

Research output: Contribution to journalArticle

67 Scopus citations

Abstract

Introduction. We analyzed the ability of four balance assessments to predict falls in people with Parkinson Disease (PD) prospectively over six and 12 months. Materials and Methods. The BESTest, Mini-BESTest, Functional Gait Assessment (FGA), and Berg Balance Scale (BBS) were administered to 80 participants with idiopathic PD at baseline. Falls were then tracked for 12 months. Ability of each test to predict falls at six and 12 months was assessed using ROC curves and likelihood ratios (LR). Results. Twenty-seven percent of the sample had fallen at six months, and 32% of the sample had fallen at 12 months. At six months, areas under the ROC curve (AUC) for the tests ranged from 0.8 (FGA) to 0.89 (BESTest) with LR+ of 3.4 (FGA) to 5.8 (BESTest). At 12 months, AUCs ranged from 0.68 (BESTest, BBS) to 0.77 (Mini-BESTest) with LR+ of 1.8 (BESTest) to 2.4 (BBS, FGA). Discussion. The various balance tests were effective in predicting falls at six months. All tests were relatively ineffective at 12 months. Conclusion. This pilot study suggests that people with PD should be assessed biannually for fall risk.

Original languageEnglish
Article number237673
JournalParkinson's Disease
DOIs
StatePublished - Jan 12 2012

Fingerprint Dive into the research topics of 'Accuracy of fall prediction in parkinson disease: Six-month and 12-month prospective analyses'. Together they form a unique fingerprint.

  • Cite this