1Department of Children's Nursing, Seoul National University Hospital, Seoul, Korea. 2College of Nursing and The Research Institute of Nursing Science, Seoul National University, Seoul, Korea. email@example.com
PURPOSE: This study was conducted to identify risk factors in hospitalized children, and to develop and validate a fall-risk assessment tool for hospitalized children.
METHODS: A retrospective chart review was performed at one university children's hospital, and an analysis was done of the characteristics of all patients who fell during a 44-month period (n=48). These patients were compared with another 149 hospitalized children who did not fall.
RESULTS: Significant predictors of falls as identified in a multivariate logistic regression analyses were age of less than 3 years old, neurological diagnosis including epilepsy, children's dependency of ADL, physical developmental delay, multiple usage of fall-risk-increasing drugs. The respective odds ratios ranged from 2.4 to 7.1 with 95% confidence interval (p<0.05). Accordingly, defining patients with either 5 risk factors as fall-prone hospitalized children provided a sensitivity of 93.6% and specificity of 16.2%.
CONCLUSION: The results show that this tool has an acceptable level of sensitivity to assess the risk factors of fall in hospitalized children even though the specificity was low, suggesting that this tool may enable nurses to predict the risk level of childhood falls, and develop preventive strategies against pediatric falls in children's units.