Use of physical restraints in nursing homes: a multicentre cross-sectional study

Abstract

Background: Although many countries have implemented strict legal rules, the prevalence of physical restraints in nursing homes seems to remain high. In Switzerland, data related to the frequency of physical restraints are scarce and little is known about associations with resident and nursing home characteristics. The aim of this study was to investigate the prevalence and types of physical restraints in nursing homes in two Swiss cantons and to explore whether resident-related and organisational factors are associated with the use of physical restraints. Methods: We conducted a multicentre cross-sectional study. Twenty nursing homes with 1362 residents from two culturally different cantons were included. Data on physical restraints and residents' characteristics were extracted from residents' records (11/2013 to 2/2014). Organisational data were collected by questionnaires addressing nursing home directors or nursing managers. Sample size calculation and outcome analysis took cluster-adjustment into account. Descriptive statistics and multiple logistic regression analysis with nursing homes as random effect were used for investigation. Results: The prevalence of residents with at least one physical restraint was 26.8 % (95 % confidence interval [CI] 19.8-33.8). Centre prevalence ranged from 2.6 to 61.2 %. Bilateral bedrails were most frequently used (20.3 %, 95 % CI 13.5-27.1). Length of residence, degrees of care dependency and mobility limitation were significantly positively associated with the use of physical restraint, but none of the organisational characteristics was significantly associated. Conclusion: Approximately a quarter of the nursing home residents included in our study experienced physical restraints. Since variation between nursing homes was pronounced, it seems to be worthwhile to explore nursing homes with particularly low and high use of physical restraints in future research, especially by using qualitative methods. There is a need for effective interventions aiming at restraint-free nursing care. Development of interventional approaches should consider specific residents' characteristics associated with restraint use.