OBJECTIVES: The aim of this study was to explore the prevalence of eating difficulties and malnutrition among persons in hospital care and in special accommodations. DESIGN: The cross-sectional observational study was performed in Nov. 2005. SETTING: Hospitals and special accommodations. PARTICIPANTS: Out of 2,945 persons, 2,600 (88%) agreed to participate (1,726 from special accommodations and 874 from hospitals). In total all special accommodations in six municipalities and six hospitals were involved. MEASUREMENTS: Risk of undernutrition was estimated as at least two of: body mass index below recommendation, weight loss and/or eating difficulties. Overweight was graded based on body mass index (if 69 years or younger: 25 or above: if 70 years or older: 27 or above). RESULTS: The mean age of those living in hospitals was 69 years and 53% were women, while the corresponding figures for those in special accommodations were 85 years and 69% women. In hospitals and special accommodations, eating difficulties were common (49% and 56% respectively) and about one quarter had a body mass index (BMI) below the limits (20% and 30% respectively) and one-third above the limit (39% and 30% respectively) thus only about 40% had a BMI within the limits. Both in hospitals and in special accommodations 27% were considered to have a moderate or high risk of undernutrition. CONCLUSION: Only about 40% in special accommodations and hospital care have a BMI within the recommended limits. As both low and high BMI are frequent in both settings, the focus of care should not only be on undernutrition but also on overweight. Using the Swedish criteria for defining risk of undernutrition seems to give a slightly lower prevalence than has been shown in previous Swedish studies, but this can be due to an underestimation of the occurrence of eating difficulties.