Evaluation of the Morse Fall Scale: Applicability in Chinese hospital populations

Background: Patient falls have been identified as a significant health problem in the general hospital patient population. The Morse Fall Scale (MFS) is an individualized criterion-referenced assessment too] designed for measuring the likelihood of patient falls in hospitals. Despite the scale has demonstrated high validity and reliability in the previous researches, this study is to provide additional validity tests to determine the applicability in the Chinese hospital population. Objectives: To examine the predictive power of the MFS to predict patient falls. To conduct reliability tests on internal consistency, item analysis, inter-rater and test-retest reliability. Design: The study was a cross-sectional study. Settings: The medical and geriatric units of three rehabilitation hospitals in Hong Kong participated in the study. Participants: A convenience sample of 954 Chinese patients was recruited sequentially upon admission in the hospital ward units. They ranged in age from 17 to 100 years, with a mean of 70.2 years. Methods: The patients were assessed for fall risk using the MFS on admission. Data was collected on the number of patients who fell rather than the number of falls. Results: The scale had a sensitivity of 31% and a specificity of 83% when the cut-off point was determined at 45. The field test demonstrated excellent inter-rater reliability with an ICC value of 0.97 (95% CI 0.94-0.98). Repeatability was high with an ICC of 0.98 (95% CI 0.98-0.99). The evaluation revealed a low Cronbach's alpha coefficient and a low to moderate item-to-scale correlation. Conclusions: The evidence collected in this study has shown both the positive and negative aspects of using the MFS in assessing the fall risk of Chinese patients during rehabilitation. The discriminative validity and internal consistency reliability provide researchers and clinicians with a major step in further developing or modifying the scale. (c) 2006 Elsevier Ltd. All rights reserved.