Objective: To examine whether low serum 25-hydroxyvitamin D (25OHD) concentration were associated with low muscle strength while taking into account the effects of potential confounders among a cohort of community-dwelling women aged 75 years and older. Design: Cross-sectional study corresponding to the baseline assessment of the EPIDOS study. Setting: Five French cities including Amiens, Lyon, Montpellier, Paris and Toulouse. Participants: Randomized sample of 440 women included in the EPIDOS study. Measurement: Maximal isometric voluntary contraction strength of the lower limb and hand with computerized dynamometers, serum 25OHD and parathyroid hormone concentration. Age at baseline evaluation, number of chronic diseases, body mass index (BMI), use of calcium drug, practice of a regular physical activity, serum calcium concentration and clearance of creatinine were used as covariables. Subjects were separated into 3 groups based on serum 25 OHD levels with the following cut-off values: < 15 ng/ml, 15-30 ng/ml and > 30 ng/ml. Results: More than 90% of women had a serum 25OHD insufficiency (i.e. < 30ng/ml) and 40.2% had a related secondary hyperparathyroidism. The mean value of muscle strength was not different among the 3 groups of women (174.9 +/- 53.2 for serum 25OHD < 15 ng/ml versus 175.9 +/- 52.6 for serum 25OHD 15-30 ng/ml versus 173.4 +/- 53.1 for serum 25OHD > 30 ng/ml with P=0.946 for quadriceps, and 56.1 +/- 13.2 for serum 25OHD < 15 ng/ml versus 57.1 +/- 13.5 for serum 25OHD 15-30 ng/ml versus 61.1 +/- 12.7 for serum 25OHD > 30 ng/ml with P= 0.064 for handgrip). There was no significant association between serum 25OHD concentration and quadriceps strength (crude beta = 0.03 with P = 0.891 and adjusted beta = -0.04 with P = 0.837). Univariate linear regression showed a significant association between serum 25OHD concentration and handgrip strength (crude beta = 0.16 with P = 0.049) but not while using an adjusted model (adjusted beta = 0.13 with P = 0.106). Conclusions: The findings of this study do not support the hypothesis of a relationship between low serum 25OHD concentration and low muscle strength. Further research is needed to corroborate and explain this finding.