Inter-observer reliability and concurrent validity of reactive balance strategies after stroke

Shirley Handelzalts, Flavia Steinberg-Henn, Nachum Soroker, Michael Schwenk, Itshak Melzer

Research output: Contribution to journalArticlepeer-review

3 Scopus citations


Background: Falls are a common complication in persons with stroke (PwS). Reliable assessment of balance responses to unexpected loss of balance has the potential to identify risk for falls. Objectives: To examine inter-observer reliability of balance responses to unannounced surface perturbations in PwS and to explore the concurrent validity of a balance recovery assessment protocol. Methods: Two observers evaluated balance recovery strategies and fall threshold (a fall into a harness system) in 15 PwS and 15 healthy adults who were exposed to forward, backward, right, and left unannounced surface translations in six increasing intensities while standing. Results: Observer agreement was 100% for the fall threshold. Kappa coefficients for step strategies were 0.960-0.988 in PwS and 0.886-0.938 in healthy adults, 0.905-0.988 for arm reactions in PwS and 0.754-0.926 in healthy adults. Significant correlations were found between fall threshold and Berg Balance Scale (r = 0.691), 6-minute walk test (r = 0.599), and fall efficacy scale-international (r = -0.581). Conclusions: A trained examiner can reliably classify reactive balance responses to surface perturbations. The high frequency of falls observed in PwS highlights the importance of assessing reactive balance responses to different directions and intensities of surface translations.

Original languageEnglish
Pages (from-to)773-778
Number of pages6
JournalIsrael Medical Association Journal
Issue number12
StatePublished - 1 Jan 2019


  • Compensatory step
  • Fall threshold
  • Falls
  • Step threshold

ASJC Scopus subject areas

  • General Medicine


Dive into the research topics of 'Inter-observer reliability and concurrent validity of reactive balance strategies after stroke'. Together they form a unique fingerprint.

Cite this