OBJECTIVES:  To examine physical function by muscle strength in the US elderly population. METHODS:  Individuals aged 50 and above from the National Health and Nutrition Examination Survey (NHANES) 1999-2002databases were selected. Physical function was measured via 20-foot walking speed [WS] in meters/second, having difficulty walking up 10 steps (yes/no), and having difficulty standing up from armless chairs (yes/no). Muscle strength was measured via the isokinetic quadriceps strength in newtons. Two study cohorts were constructed based on age and gender (10 groups) adjusted muscle strength (high strength = above the median within each age and gender group). Demographics, comorbidities, and body mass index (BMI) were compared between the high and low strength groups. The association between high/low muscle strength and 20-foot WS was analyzed via multiple linear regression that utilized survey strata and weighting and controlled for between-group differences in demographics, BMI, and comoribidites. Logistic regressions were employed to examine the relationship between having difficulty walking up 10 steps or standing up from armless chairs and high/low muscle strength, respectively.  RESULTS:   The study included 646 individuals with 320 in the high and 326 in the low strength cohorts.  More individuals with low muscle strength had diabetes, arthritis, and osteoporosis than those with high muscle strength. Mean BMI was also higher in the high strength cohort. Controlling for differences in demographics, comorbidities, and BMI, individuals in the high strength cohort had significantly faster 20-foot WS speed (0.10, p<0.05) and were less likely to have difficulty standing up from armless chairs (Odds Ratio = 0.32, p<0.05) compared with those in the low strength cohort. CONCLUSIONS: Among individuals aged 50 and above in the US, high muscle strength is positively correlated with WS but negatively correlated with having difficulty standing up from armless chairs.