Validity of the Morse Fall Scale implemented in an electronic medical record system

Aims and objectives To examine the validity of the Morse Fall Scale by analysing the electronic medical records on fall risk during different phases of hospitalisation. Background Regular monitoring on fall risk with a reliable and valid assessment tool is a key element in the fall prevention. In Korea, the Morse Fall Scale is currently being used in numerous medical institutions, yet it has not been comprehensively evaluated whether it is suitable and valid. Design The study design was a retrospective case–control study. Methods The participants included 151 fallers and 694 nonfallers. Data were extracted from a university hospital implementing Morse Fall Scale in the electronic medical records between October 2010 and June 2011. The nonfallers were selected by the stratified random sampling method among the patients who were in the hospital during the same period as the fallers. The Morse Fall Scale scores during three different time periods of hospital stay were used for analysis: the initial assessment score upon admission, the last and the maximum scores recorded from admission to the fall or discharge. Results With the maximum Morse Fall Scale score and the best cut‐off point of 51, validity indicators showed the highest performance: 0·72 for sensitivity, 0·91 for specificity, 0·63 for positive predictive value, 0·94 for negative predictive value, 0·63 for Youden Index and 0·77 for the area under the receiver operating characteristic curve. Conclusion The Morse Fall Scale showed relatively high predictive performance for the Korean population. Relevance to clinical practice The study results recommend practice change in fall prevention. As the validity was highest when the patient was first classified into the high‐risk group based on the Morse Fall Scale cut‐off score 51, patients classified as high risk should be placed under special nursing interventions until the day of their discharge, regardless of change in the patient state.