Abstract
OBJECTIVES: To compare the relative predictive power of handgrip and leg extension strength in predicting slow walking. DESIGN: Report of correlative analysis from two epidemiological cohort studies. SETTING: Foundation of the National Institutes of Health Sarcopenia Project. PARTICIPANTS: Men and women aged 67 to 93 (N=6,766). MEASUREMENTS: Leg strength, handgrip strength, and gait speed were measured. Strength cutpoints associated with slow gait speed were developed using classification and regression tree analyses and compared using ordinary least squares regression models. RESULTS: The cutpoints of lower extremity strength associated with slow gait speed were 154.6 N-m in men and 89.9 N-m in women for isometric leg extension strength and 94.5 N-m in men and 62.3 N-m in women for isokinetic leg extension strength. Weakness defined according to handgrip strength (odds ratios (OR)=1.99 to 4.33, c-statistics=0.53 to 0.67) or leg strength (ORs=2.52 to 5.77; c-statistics=0.61 to 0.66) was strongly related to odds of slow gait speed. Lower extremity strength contributed 1% to 16% of the variance and handgrip strength contributed 3% to 17% of the variance in the prediction of gait speed depending on sex and mode of strength assessment. CONCLUSION: Muscle weakness of the leg extensors and forearm flexors is related to slow gait speed. Leg extension strength is only a slightly better predictor of slow gait speed. Thus, handgrip and leg extension strength appear to be suitable for screening for muscle weakness in older adults.