Li Juan, Hu Li, Zhao Qinghua, et al. Application ROC curve analysis to explore high risk cut-off point of Morse fall scale. [J]. Journal of Chongqing Medical University 42(5):591-595(2017)
DOI:
Li Juan, Hu Li, Zhao Qinghua, et al. Application ROC curve analysis to explore high risk cut-off point of Morse fall scale. [J]. Journal of Chongqing Medical University 42(5):591-595(2017) DOI: 10.13406/j.cnki.cyxb.001293.
Application ROC curve analysis to explore high risk cut-off point of Morse fall scale
Objective:To discuss the high risk cut-off point of Chinese version of the Morse Fall Scale. Methods:A retrospective study was conducted based on the data of falls in the Nursing Adverse Events Report System from January 2016 to December 2016.A population based 1∶2 matched case-control study was conducted;338 control subjects were those who hospitalized for the same period without falls. ROC curve was used to analyze the accuracy and best high risk cut-off point of the Chinese version of the Morse Fall Scale.Results:The area under ROC curve was 0.835 and 95% confidence intervals were 0.794-0.876. With the Morse Fall Scale score and the best cut-off point of 50
Youden’s index was the largest
at 0.630. The validity indicators were 75.74% for sensitivity
87.28% for specificity
74.85% for positive predictive value(PPV)
87.80% for negative predictive value(NPV)
83.43% for accuracy
5.95 for positive likelihood ratio. Conclusion:The study results show 50 point can be used as the best cut-off value of screening high-risk groups.