Hemiplegic gait: Relationships between walking speed and other temporal parameters

Elliot J. Roth*, Charles Merbitz, Kenneth Mroczck, Sheila A. Dugan, W. Warren Suh

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

160 Scopus citations


It has been asserted that speed alone is an effective indicator of the degree of gait abnormality. To determine the validity of this assertion, relationships between velocity and 18 other temporal gait parameters were determined in 25 patients with a first hemispheric stroke resulting in hemiplegia or hemiparesis of at least one month duration. Gait characteristics were recorded using footswitches connected to a portable computerized monitoring device. Velocity was found to be significantly correlated with cadence, mean cycle duration, mean cycle length, hemiplegic limb stance phase duration, nonhemiplegic limb stance phase duration and percent, nonhemiplegic limb swing phase percent, double support phase duration and percent, hemiplegic limb swing/stance phase ratio, nonhemiplegic limb swing/stance phase ratio, and swing phase symmetry ratio but not with the hemiplegic limb stance phase percent, hemiplegic limb swing phase duration and percent, nonhemiplegic limb swing phase duration, stance phase symmetry ratio, and overall asymmetry ratio. Velocity is related to most, but not all, of the other temporal measures of hemiplegic gait. A comprehensive gait evaluation should also include characterization of the degree of asymmetry and descriptions of individual phase durations and proportions (particularly hemiplegic stance and swing percentages).

Original languageEnglish (US)
Pages (from-to)128-133
Number of pages6
JournalAmerican Journal of Physical Medicine and Rehabilitation
Issue number2
StatePublished - Jan 1 1997


  • Gait
  • Hemiplegia
  • Stroke

ASJC Scopus subject areas

  • Physical Therapy, Sports Therapy and Rehabilitation
  • Rehabilitation


Dive into the research topics of 'Hemiplegic gait: Relationships between walking speed and other temporal parameters'. Together they form a unique fingerprint.

Cite this