OBJECTIVE The aim of the study was to examine the prevalence of physical restraint use in cognitively impaired nursing home residents, the manner in which restraints are used, reasons for using them, and relationships between residents' characteristics and use of physical restraints. METHODS A point prevalence study was conducted on the use of physical restraints among all residents cared for in two Dutch nursing homes and one nursing home unit (n = 265). Data about the nursing home residents and the use of restraints were collected by means of a questionnaire, which was filled in by the nurses. The response rate was 98%. The mean age of residents was 81 years (sd = 8.6), 74% of whom were female. RESULTS One or more restraints were used with 49% of the residents. The most frequently used physical restraints were bed rails, a waist belt, and a chair with a table. In almost all situations (90%), residents were continuously restrained and restraints were used for longer than 3 months. The most common reason (90%) for use of restraints was to prevent falls. Logistic regression analysis revealed that use of restraints was highly associated with poor mobility, care dependency and risk of falling in the opinion of nursing staff. CONCLUSIONS The results of this study are comparable with those of other studies. However, since recent studies have reported that physical restraints are inadequate to prevent falls, recommendations are made to re-evaluate critically the use of restraints and to conduct future research into a responsible and safe way of decreasing the use of physical restraints.