Factors associated with use of restraints on older adults with home care: A secondary analysis of a cross-sectional survey study

Background: Although there is evidence that use of restraints in home care is increasing, research into the factors associated with restraints in this setting is scarce. Objective: To gain insight into the factors associated with restraints in older adults receiving home care. Design: A secondary analysis of a cross-sectional survey about restraint use in home care. Settings: Older adults receiving home care in Belgium. Participants: 8000 subjects were randomly selected from a total of 45,700 older adults. The mean age of the sample (n = 6397) was 80.6 years, 66.8% were women and 46.4% lived alone. Methods: A cross-sectional survey of restraint use on older adults receiving home care from a nursing organisation in Belgium was completed by the patients' primary care nurses. A binary logistic regression model with generalised estimating equations was used to evaluate factors associated with restraint use. Additional analyses focused on the subgroups with and without an informal caregiver and living alone / with others. Data from 6397 participants were analysed in detail. Results: Multivariate logistic regression indicated that restraint use was associated with supervision [OR = 2.433, 95% CI = 1.948-3.038]; dependency in activities of daily living (i.e. eating [OR = 2.181, 95% CI = 1.212-3.925], transfer [OR = 2.131, 95% CI = 1.191-3.812] and continence [OR = 1.436, 95% CI = 0.925-2.231]; perceived risk of falling in the nurses' clinical judgement [OR = 1.994, 95% CI = 1.710-2.324], daily behavioural problems [OR = 1.935, 95% CI = 1.316-2.846] and less than daily behavioural problems [OR = 1.446, 95% CI = 1.048-1.995]; decreased well-being of the informal caregiver [OR = 1.472, 95% CI = 1.126-1.925], the informal caregiver's dissatisfaction with family support [OR = 1.339, 95% CI = 1.003-1788]; patient's cognitive impairment [OR = 1.398, 95% CI = 1.290-1.515]; and poly pharmacy [OR = 1.415, 95% CI = 1.219-1.641]. The nurses' perception of risk of falling, cognitive impairment (observed with the Cognitive Performance Scale) and supervision are the only variables consistently associated with restraint use across all the analyses. Conclusion: The study results provide insight into new and context-specific factors associated with restraint use in home care (e.g. supervision, informal caregiver's decreased well-being and dissatisfaction with family support). These insights could support the development of interventions to reduce restraint use in home care.