Background and purpose: The purpose of this study was to determine the reliability and validity of measurements obtained with a seven-item modified version of the Gait Abnormality Rating Scale (GARS-M), an assessment of gait designed to predict risk of falling among community-dwelling, frail older persons.
Subjects: Fifty-two community-dwelling, frail older persons, with a mean age of 74.8 years (SD = 6.75), participated.
Methods: A history of falls was determined from self-report or by proxy report. The GARS-M was scored from videotapes of subjects walking at self-selected paces. Gait characteristics were recorded during a timed walk on a 6-m brown-paper walkway.
Results: Scores obtained by three raters for 23 subjects demonstrated moderate to substantial intrarater and interrater reliability. Concurrent validity, as assessed by Spearman rank-order correlation coefficients, was demonstrated for the relationship between GARS-M scores and stride length (r = -.754) and for the relationship between GARS-M scores and walking speed (r = -.679). Mean GARS-M scores distinguished between frail older persons with and without a history of recurrent falls (mean GARS-M scores of 9.0 and 3.8, respectively).
Conclusion and discussion: The GARS-M is a reliable and valid measure for documenting gait features associated with an increased risk of falling among community-dwelling, frail older persons and may provide a clinically useful alternative to established quantitative gait-assessment methods.