OBJECTIVE: To quantify habitual walking speed and estimate the prevalence of low habitual walking speed (<0.8m/s and <0.5m/s) in nursing home residents; and secondarily to gain some insight into whether demographic, health, and functional outcomes could predict the nursing home residents' walking speed. DESIGN: Cross-sectional study. SETTING: Eleven nursing homes. PARTICIPANTS: Nursing home residents (N=102 [37%] of 273 eligible, randomly selected residents from 11 nursing homes consented to participate in this study). INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: The primary outcome was habitual walking speed assessed over a distance of 2.4m. Secondary outcomes including body composition, muscle strength, balance and physical performance as assessed via the Short Physical Performance Battery, and historical and current demographic and health measures were all assessed as potential predictors of walking speed. RESULTS: Mean walking speed was .37±.26 m/s, meaning that 97% and 75% of participants had walking speeds <0.8m/s and <0.5m/s, respectively. Multivariable linear regression identified physical activity status before 50 years of age and daily sitting time as independent predictors of walking speed (r(2)=.25, P<.05), although this regression only accounted for 25% of the variance in walking speed. CONCLUSIONS: Almost all participants in this study had below-normal walking speed, a known clinical predictor of physical performance. Because walking speed is a clinical marker of many age-related adverse outcomes in older age, efforts to increase or at least maintain walking speed in nursing home residents should be considered. Some evidence suggests that progressive resistance training may offset these declines in walking speed.