Evaluation of the Humpty Dumpty Falls Scale: An Analysis of Electronic Medical Records
10.22650/JKCNR.2019.25.2.142
- Author:
Yun Hee CHO
1
;
Young Ju KIM
Author Information
1. Unit Manager, Department of Nursing, The Catholic University of Korea, Uijeongbu ST. Mary's Hospital, Korea.
- Publication Type:Original Article
- Keywords:
Accidental Falls;
Instrumentation;
Child;
Hospitalized;
Sensitivity and Specificity
- MeSH:
Accidental Falls;
Case-Control Studies;
Child;
Diagnosis;
Electronic Health Records;
Humans;
Inpatients;
Length of Stay;
Logistic Models;
Odds Ratio;
Retrospective Studies;
Risk Assessment;
Risk Factors;
ROC Curve;
Sensitivity and Specificity
- From:
Journal of Korean Clinical Nursing Research
2019;25(2):142-150
- CountryRepublic of Korea
- Language:Korean
-
Abstract:
PURPOSE: The aim of this study was to evaluate the efficiency of the Humpty Dumpty Falls Scale as one of the falls risk assessment tools, and also to evaluate risk factors as predictors of falls in pediatric patient populations. METHODS: In a retrospective, case-control design with data from the electronic medical records of 13 pediatric patients who fell and 1,941 who did not fall before matching and 429 who did not fall after matching by gender, age, diagnosis, and length of stay. RESULTS: All the variables showed no significant differences after matching. At the cutoff score of 13, sensitivity, specificity, negative and positive predictive values were 92.3%, 37.1%, 99.9%, and 0.01%, respectively. The area under the Receiver Operating Characteristics was 0.597. The results from the logistic regression showed that the pediatric inpatient population who had higher risk scores was significantly associated with falls. The odds ratios ranged from 1.31 to 4.71 with 90% confidence interval. CONCLUSION: The saturation impairments criterion as one of the diagnostic parameter was negatively associated with falls, but the relative risk score was higher than the other criteria. Therefore, it seems that the diagnostic parameter seems to be required to verify results through large sample studies.
