Population Representative Gait Speed and Its Determinants

    loading  Checking for direct PDF access through Ovid



Although gait speed is an important measure to obtain from older adults, much is uncertain about the measure. Therefore, using a large population-based sample, I sought to: (1) gain an indication of “normal” gait speed, (2) compare speed measured over 8 and 20 feet, (3) determine the relationship of several independent variables with gait speed.


The National Health and Nutrition Examination Survey (NHANES) 2001–2002 provided the following data for individuals (≥ 50 years, n = 1923): gait speed over 8 and 20 feet, age, stature, adiposity, and isokinetic knee extension force. Estimates of normal gait speed were calculated for gender and age strata. Speeds over 8 and 20 feet were compared using a paired t-test and a Pearson correlation. The relationship between the independent variables and gait speed were examined using Pearson correlations and forward multiple regression analysis.


Mean gait speeds were 3.26 ft/sec and 3.29 ft/sec over 8 and 20 feet, respectively. Speeds over the distances differed significantly (p = .002) but not meaningfully; they were correlated strongly (r = .933, p < .001). Age, stature, knee extension force, and waist circumference correlated significantly (p < .004) with gait speed over both distances. Age, gender, knee extension force, waist circumference, and stature were shown by multiple regression to predict walking speed over both 8 feet (R = .459, p < .001) and 20 feet (R = .506, p < .001).


The estimates of normal gait speed provided are slower than many of those previously published but may be more representative of the population. As gait speed measured over 8 and 20 feet are comparable, using the shorter distance appears justifiable. Individuals walk faster who are younger, male, taller, stronger, and have smaller waists.

Related Topics

    loading  Loading Related Articles