Paper
Thursday, July 12, 2007
Physical restraint use among nursing home residents - a multicentre observational study
Sascha Köpke, RN1, Gabriele Meyer, PhD, RN2, and Ingrid Mühlhauser, PhD, MD1. (1) Unit of Health Sciences and Education, University of Hamburg, Hamburg, Germany, (2) Institute for Public Health and Nursing Research (IPP), University of Bremen, Bremen, Germany
| Learning Objective #1: state valid data on physical restraint use and psychotropic medication in nursing homes in Hamburg, Germany. |
| Learning Objective #2: reflect the most valid method of data collection on prevalence and incidence of restraint use. |
Background: High quality epidemiological data on restraint use in nursing home residents in Germany are lacking. We examined prevalence and incidence of restraints, legal justification of devices, and associations between residents' and institutional characteristics and restraints in a representative sample of nursing home residents.
Methods: 30 nursing homes were recruited. Prevalence was obtained by direct observation of restraints on three occasions on one day. Justification of devices was collected by medical chart review. Incidence data were collected prospectively using a standardised documentation sheet.
Results: A total of 2,367 residents were included. Mean age was 86 years and 82% were female. Centre prevalence ranged from 4 to 59%. Cluster-adjusted prevalence of residents with at least one device was 26%. Bedrails were most often used (25%). Fixed tables, belts and other restraints were observed in 2%, 3%, and 2% of residents, respectively. 54% of devices were justified according to current legal standards in Germany with authorisation by judge (39.5%) or residents’ written consent (15%). 51% of residents had a prescription of at least one psychotropic medication, with neuroleptics most commonly prescribed (34%). 12-month incidence of residents with new devices was 11%. Relative frequency of resident days with at least one device ranged from 5 to 65% between centres. 66% of residents with devices were restrained on more than 75% of observation days. High level of care dependency, severe cognitive impairment, and history of fracture were significantly associated with the use of restraints. None of the institutional characteristics examined was associated with increased use of restraints.
Conclusion: Rates of physical restraints were lower than previously reported. However, restraints are still routine nursing care and the large centre differences on restraint use and justification indicate a reduction approach. Therefore, an evidence-based practice guideline is in preparation, which will be evaluated within a randomised-controlled trial.