The prevalence of malnutrition and its predictive value for the incidence of complications were determined in 155 patients hospitalized for internal or gastrointestinal diseases. At admission, 45% of the patients were malnourished according to the Subjective Global Assessment (physical examination plus questionnaire), 57% according to the Nutritional Risk Index [(1.5 x albumin) + (41.7 x present/usual weight)], and 62% according to the Maastricht Index [(20.68 - (0.24 x albumin) - (19.21 x transthyretin (prealbumin) - (1.86 x lymphocytes) - (0 04 x ideal weight)]. Crude odds ratios for the incidence of any complication in malnourished compared with well-nourished patients during hospitalization were 2.7 (95% CI: 1.4, 5.3) for the Subjective Global Assessment, 2.8 (1.5, 5.5) for the Nutritional Risk Index, and 3.1 (1.5, 6.4) for the Maastricht Index. Odds ratios were reduced to 1.7 (0.8, 3.6), 1.6 (0.7, 3.3), and 2.4 (1.1, 5.4), respectively, after a multivariate analysis that included disease category and disease severity. Because the confounding factors adjusted for are not only a measure of the severity of the disease but may also be influenced by malnutrition itself, the actual risk for complications due to malnutrition could be higher than the adjusted odds ratios. In conclusion, malnutrition was frequent in patients with gastrointestinal disease and other internal diseases at the time of admission. The severity of malnutrition in the patients predicted the occurrence of complications during their hospital stay and this association was not completely explained by confounding factors.