A Multivariate Analysis of Care Workers’ Perceptions of Risk and Quality of Life

Abstract

This study aimed to examine the relationship between care workers’ perceptions of quality of life (QOL) enhancement and the risk of care-related accidents and incidents in long-term care facilities. A questionnaire survey was mailed to 580 facilities in Tokyo, yielding 384 responses, of which 320 valid cases were included in the analysis. Twenty-one items assessing perceptions of QOL and risk were rated on a five-point Likert scale. The dependent variable was the belief that “enhancing QOL increases the risk of accidents or incidents.” Data were analyzed using a decision tree model based on the CHAID method and a multilayer perceptron (MLP) neural network. The decision tree model achieved an overall classification accuracy of 83.8%, with the primary splitting variable being the perception of being conflicted between prioritizing QOL improvement and risk avoidance. The MLP demonstrated high predictive performance, with accuracies of 96.5% for the training dataset and 94.6% for the test dataset. The findings indicate that care workers who place strong importance on improving QOL tend to recognize the associated risks in a realistic and nuanced manner. These results suggest the need for educational programs that foster balanced decision-making between QOL enhancement and risk avoidance, and that promote evidence-based care practices in long-term care settings.