Low fat-free mass may be an independent risk factor for malnutrition that results in an increased length of hospital stay (LOS). The objectives were to compare differences in fat-free mass and fat mass at hospital admission between patients and healthy control subjects and to determine the association between these differences and the LOS. Patients (525 men, 470 women) were prospectively recruited at hospital admission. Height-corrected fat-free mass and fat mass (fat-free-mass index or fat-mass index; in kg/m2) were determined in patients at admission by bioelectrical impedance analysis and were compared with values for sex-, age-, and height-matched control subjects. Patients were classified as well-nourished, moderately depleted, or severely depleted on the basis of a Subjective Global Assessment questionnaire and a body mass index (in kg/m2) < or > 20. Low fat-free mass was noted in 37% and 55.6% of patients hospitalized 1-2 d and > 12 d, respectively. The odds ratios were significant for fat-free-mass index and were higher in patients with a LOS of > 12 d [men (odds ratio: 5.6; 95% CI: 3.1, 10.4), women (4.4; 2.3, 8.7)] than in those with a LOS of 1-2 d [men (3.3; 2.2, 5.0), women (2.2; 1.6, 3.1)]. Severe nutritional depletion was significantly associated only with a LOS > 12 d. Fat-free mass and fat-free-mass index were significantly lower in patients than in control subjects. Because the fat-free-mass index is significantly associated with an increased LOS, provides nutritional assessment information that complements that from a Subjective Global Assessment questionnaire, and is a more sensitive determinant of the association of fat-free mass with LOS than is a weight loss > 10% or a body mass index < 20, it should be used to evaluate nutritional status.