Utility of an obstacle-crossing test to classify future fallers and non-fallers at hospital discharge after stroke: A pilot study.

Journal Article (Journal Article)

BACKGROUND: Existing clinical assessments of balance and functional mobility have poor predictive accuracy for prospectively identifying post-stroke fallers, which may be due to a lack of ecological complexity that is typical of community-based fall incidents. RESEARCH QUESTION: Does an obstacle-crossing test at hospital discharge predict fall status of ambulatory stroke survivors 3 months after discharge? METHODS: Ambulatory stroke survivors being discharged home completed an obstacle-crossing test at hospital discharge. Falls were tracked prospectively for 3 months after discharge. Logistic regression examined the relationship between obstacle-crossing at discharge (pass/fail) and fall status (faller/non-faller) at 3 months post discharge. RESULTS: 45 participants had discharge obstacle test and 3-month fall data. 21 (47 %) participants experienced at least one fall during follow-up, with 52 % of the falls occurring within the first month after discharge. Of the 21 fallers, 14 failed the obstacle-crossing test (67 % sensitivity). Among the 24 non-fallers, 20 passed the obstacle-crossing test (83 % specificity). The area under the receiver operating characteristic curve was 0.75 (95 % CI 0.60-0.90). Individuals who failed the obstacle-crossing test were 10.00 (95 % CI: 2.45-40.78) times more likely to fall in the first 3 months after discharge. The unadjusted logistic regression model correctly classified 76 % of the subjects. After adjusting for age, sex, days post stroke, and post-stroke disability, the odds ratio remained significant at 6.93 (95 % CI: 1.01-47.52) and correctly classified 79.5% of the participants. SIGNIFICANCE: The obstacle-crossing test may be a useful discharge assessment to identify ambulatory stroke survivors being discharged home who are likely to fall in the first 3 months post discharge. Modifications to improve the obstacle-crossing test sensitivity should be explored further.

Full Text

Duke Authors

Cited Authors

  • Feld, JA; Goode, AP; Mercer, VS; Plummer, P

Published Date

  • July 2022

Published In

Volume / Issue

  • 96 /

Start / End Page

  • 179 - 184

PubMed ID

  • 35667230

Pubmed Central ID

  • PMC9535661

Electronic International Standard Serial Number (EISSN)

  • 1879-2219

Digital Object Identifier (DOI)

  • 10.1016/j.gaitpost.2022.05.037


  • eng

Conference Location

  • England