BackgroundDisease-related malnutrition is a major health problem in the elderly population, but it has until recently received very little attention, especially are management issues under-explored. By identifying residents at the risk of undernutrition (UN), appropriate nutritional care can be provided.ObjectiveTo investigate if study circles and policy documents improve the precision in nutritional care and decrease the prevalence of low or high body mass index (BMI).DesignPre and post-intervention study.SettingSpecial accommodations (nursing homes) within six municipalities were involved.ParticipantsIn 2005, 1,726 (90.4%) of 1,910 residents agreed to participate and in 2007, 1,526 (81.8%) of 1,866 residents participated.InterventionStudy circles in one municipality, having a policy document in one municipality and no intervention in four municipalities.MeasurementsRisk of UN was defined as involving any of: involuntary weight loss; low BMI; and/or eating difficulties. Overweight was defined as high BMI.ResultsIn 2005 and 2007, 64% and 66% of residents, respectively, were at the risk of UN. In 2007, significantly more patients in the study circle municipality were accurately provided protein and energy enriched food (PE-food) compared to the no intervention municipalities. There was a decrease between 2005 and 2007 in the prevalence of low BMI in the study circle municipality, but the prevalence of overweight increased in the policy document municipality.ConclusionsStudy circles improve the provision of PE-food for residents at the risk of UN and can possibly decrease the prevalence of low BMI. It is likely that a combination of study circles and implementation of a policy document focusing on screening and on actions to take if the resident is at UN risk can give even better results.