Purpose: This study aims to develop a "Fall risk assessment tool and preventive measures", which will be effective for patients with neurological disorders, and statistically establish the validity of the items included.
Methods: Subjects were 1,002 cases in total of 366 inpatients of hospitals specializing in neurological diseases. Their fall risks were assessed by nurses of the hospital using a draft assessment tool. We had fifteen nurses complete a questionnaire about the usability of the assessment tool. In the analysis, we calculated the odds ratio for the items of the tool for two groups: with and without falls, and carefully selected items by considering the clinical significance reported in the questionnaires.
Results: Based on the statistical verification of the assessment tool and the survey results, we excluded one item with a low odds ratio, "Taking sleeping pills/tranquilizers", and added one item with a statistically significant difference in the univariate analysis, "Memory disorder". As a result, we completed a seven-item fall risk assessment tool, covering "Experience of falls", "Gait disturbance", "vertigo/standing dizziness", "Unable to activate the nurse call button", "Wandering/hyperactivity", "Intuition of nurses", and "Memory disorders". The sensitivity, specificity, and Area Under the Curve (AUC) of the assessment tool were 87.5%, 34.3%, and 0.768, respectively.
Discussion: The findings show that the assessment tool developed here has a similar level of predictive accuracy as existing tools. It is however considered that inclusion of clinical concerns has made the assessment tool easier to use in clinical settings.
View full abstract