Abstract
The Humpty Dumpty Falls Prevention Scale (HDFS) was developed with the goal of preventing injury and reducing falls in children (Gonzalez et al., 2020; Hill-Rodriguez et al., 2009). The aim of this study was to compare sensitivity, specificity, and predictive validity of the original HDFS (Hill-Rodriguez et al., 2009) to the changes proposed to the scale by Sarik et al. (2022), utilizing data from four geographically diverse children's hospitals within the United States of America.
A secondary analysis of aggregate fall data from four US-based hospitals was conducted. All inpatient admissions with accompanying HDFS fall scores from a one-year period were included. Both the original scoring and parameters of the HDFS and the updated scoring and parameters proposed by Sarik et al. (2022) were calculated. Distributions of scores, sensitivity, specificity, and the predictive validity of the two scoring methods were compared.
Of 55,711 individual patient encounters analyzed, 151 falls were reported across all institutions (0.3 % of encounters). The average HDFS score on the updated tool was 12.92 (SD = 1.93), with 77 % of patients classified as "at risk". Sensitivity for the new tool was found to be 85 % while specificity was found to be 23 %. When examining the predictive validity of the cases identified as "at risk", 24 % of cases were classified correctly.
After analysis, the updated HDFS demonstrated superior performance in identifying patients at risk of fall. All institutions currently using the HDFS are encouraged to utilize the updated scale.