Measurement of Gait Speed of Older Adults is Feasible and Informative in a Home-care Setting

Journal of Geriatric Physical Therapy, 2009 by Bohannon, Richard W

ABSTRACT

Purpose: Although gait speed is widely recommended as a measure of activity limitation, it is not routinely used clinically with older adults. This retrospective study was undertaken to determine whether the measurement of gait speed is feasible and informative in a home care setting.

Methods: The therapy records of 27 ambulatory patients were examined for gait speed measures and other relevant data.

Results: Gait speed was documented for all patients. It was significantly lower than that of age and sex matched normals. A wide range of speeds were noted for patients who required total assistance or were completely independent according to Functional Independence Measure criteria or who were able to walk at least 150 feet.

Conclusions: Measurement of the gait speed of older adults is feasible in a home care setting. Its sensitivity to limitations not revealed by other measures provides support for broader use.

Key Words: gait, measurement, aging

INTRODUCTION

Gait speed is frequendy used as a measure of activity limitation in studies of older adults.1-4 Such use is not surprising given the importance of speed for ambulation in the community5 and speed s power as a predictor of outcome among older adults.4,6 The routine measurement of gait speed has been advocated among older adults. Studenski et al,4 and more recendy Duncan and Tilson,7 have even suggested that it be considered a "vital sign" universally applicable to older adults. Nevertheless, therapists do not necessarily incorporate the measure into everyday practice.8 The present communication, therefore, was initiated to demonstrate that the measurement of gait speed of older adults in a home care setting is both feasible and informative.

METHODS

The institutional review board of the University of Connecticut granted exemption to the research as it involved the retrospective retrieval of data from initial therapy records maintained by the author on all of his personal patients. The therapy records of all 39 patients were examined. Twelve of the records were excluded because the patients were either less than 60 years of age or nonambulatory.

Data gathered or derived from the 27 remaining patients' records were demographics (diagnosis, age, and gender) and 3 gait-related measures (speed, independence, and maximum distance). Gait speed was determined as patients walked past a steel measuring tape extended to lengths between 7 and 20 feet, as allowable by each patient's habitat. Patients were asked to walk at their own comfortable speed. They began walking a few feet before the tape and walked past its end. Timing began and ended as their mid-sagittal line crossed the beginning and end of the tape, respectively. Speed was derived by dividing the length of tape traversed by the time required to traverse it. Gait independence was fated based on information about assistance, device use, and distance documented in patients' records. The 1 (total assistance) to 7 (complete independence) scoring system of the original Functional Independence Measure (FIM) was used for this purpose.9 According to the original FIM system, a score of 1 is assigned for any distance less than 50 feet (regardless of actual assistance provided) and a score of 2 is the maximum that can be realized unless the individual reaches 150 feet (regardless of actual assistance). Scores of 3, 4, 5, or 6 are assigned to individuals requiring assistance (including supervision or a device) over distances of 150 feet or more. A score of 7 represents walking at least 150 feet without assistance (including device or supervision). Maximum gait distance was indicated by the total distance covered without stopping to rest. Depending on the habitat, the total distance walked might involve a loop or a back-and-forth course. Distance was determined by measuring tape, official documentation, or examiner striding (a distance within a few inches of 3 feet). Based on habitat, time constraints and patient factors, the distance was sometimes truncated at 150 feet. All statistical analysis was conducted using the Statistical Package for Social Sciences (SPSS, version 11.0). Gait speed was described for the patients as a whole as well as for patients within specific strata. Those strata were FIM scores 1, 2-6, and 7 and walking distances less than 150 feet and 150 feet or more. A paired T-test was used to compare the patients' actual walking speed with expected walking speed; that is, the mean speed of age and gendermatched individuals from the National Health and Nutrition Survey (NHANES) study.10

RESULTS

The age of patients of this retrospective study was a mean (standard deviation) 75.4 (9.2) with a range of 60-88 years. Thirteen were men and 14 were women. They presented with 13 different primary diagnoses, but fracture (n=4), osteoarthritis (n=4), stroke (n= 3), and congestive heart failure (n=3) were most common.

Gait speed was recorded for all 27 ambulatory patients. The distances over which speed was measured ranged from 7 to 20 feet. The patients' gait speed was a mean (standard deviation) 1.2 (.8) with a range of .1-2.9 ft/sec. Their expected speed based on the NHANES study was 2.9 (.3) with a range 2.5-3.4 ft/sec (Figure 1). All but 2 patients walked more slowly than expected. Actual and expected speeds differed significantly (T= -11.193, p