The purpose of the study was to validate the selected fall risk assessment scales among hospitalized adult patients from the data of electronic medical records by comparing the sensitivity, specificity,
positive predictive value and negative predictive values of Morse Fall Scale(MFS), Bobath Memorial Hospital Fall Risk Assessment Scale(BMFRAS) and Johns Hopkins Hospital Fall Risk Assessment tool(JHFRAT).
The 120 patients who experienced fall episodes during their hospitalization as reported at Patient Safety Management System from June 2010 to December 2013 were assigned to the fall-group, while another 120 patients who didn''t experience fall episodes with age, sex, clinical departments and wards matched with the fall group were assigned to the control group.
Data were collected from January 2014 for two months while maintaining the consistency of data collection. General characteristics and disease
-related characteristics of the subjects along with the fall related
factors for the fall group were collected from the electronic medical records by using structured data collection forms. Data on fall risk factors were collected by Morse Fall Scale(MFS), Bobath Memorial
Hospital Fall Risk Assessment Scale(BMFRAS) and Johns Hopkins
Hospital Fall Risk Assessment tool(JHFRAT).
Data were entered and analyzed by using SPSSWIN V. 20.0 and Medcalc program was used for comparing the mean difference among AUC of each assessment scale.
The results are as follows.
1) General characteristics of age, sex, clinical departments and wards were matched between the fall and the no-fall groups, and there was no significant group difference on occupation, education level, and the course of hospitalization. Disease-related characteristics of the circulatory system, nervous system, respiratory system, musculoskeletal system, digestive system, urinary tract disease and eye disease also do not show statistically significant differences between the two groups, confirming homogeneous between the groups.
2) The level of physical activity for the fall group was reported as being independent in 43 patients(35.8%), required partial assistance in 42 patients(35.0%), required consistent help in 32 patients(26.7%), and bed-ridden state in 3 patients(2.5%). The symptoms of the fall group include general weakness(49.2%), dizziness(28.3%), sleep disturbances(19.2%) with an average of 2.7 symptoms. The medications taken prior to 24 hours of the falls of the fall group were anti-depressants in 27 patients(22.5%), anti-hypertensives in 49 patients(40.8%), anti-convulsant and pain killers in 33 patients(27.5%), anti-glycemic agents in 26 patients(21.7%) and diuretics in 23 patients(19.2%) with an average of 2.52 medications. The fall accident were frequently occurred in the beds(44.2%) or in their hospital room (59.2%). The patient''s condition was the primary reason for fall in 107 cases(89.2%). Physical injury after a fall was occurred in 66 patients (55%), lacerations in 25 patients(20.8%), abrasions in 22 patients(18.3%), or fractures in 14 patients(11.7%).
3) The average scores of fall risk assessment scales were 60.8 of the fall group and 41.5 of the no-fall group on Morse Fall Scale(MFS); 10.9 for the fall group and 7.6 for the no-fall group on Bobath Memorial Hospital Fall Risk Assessment Scale(BMFRAS); and 11.8 for the fall group and 9.8 for the no-fall group on Johns Hopkins Hospital Fall Risk Assessment tool(JHFRAT). There was statistically significant differences
between the two groups(p<.001) for all three scales.
4) The comparison of the validity tests showed that Morse Fall Scale (MFS) at a cut-off score of 48 had .806 for areas under the curve of ROC(receiver operating characteristic), 76.7% for sensitivity, 77.5% for specificity, 77.3% for positive predictive value, and 76.9% for negative predictive value. Bobath Memorial Hospital Fall Risk Assessment Scale (BMFRAS) at a cut-off score of 10 had .695 for areas under the curve of ROC(receiver operating characteristic), 75.0% for sensitivity, 58.3% for specificity, 64.3% for positive predictive value, and 70.0% for negative predictive value. Finally, Johns Hopkins Hospital Fall Risk Assessment tool(JHFRAT) at a cut-off score of 11 had sensitivity 62.5%, specificity 63.3%, positive predictive value 63.0%, negative predictive value 62.8%. Therefore, the Morse Fall Scale(MFS) with the highest score and the highest discrimination was evaluated to be suitable and reasonable for predicting falls in inpatients admitted to the medical or surgical units of the university hospital.
5) Subgroup analysis was conducted to compare the validity of the fall risk scales each for medical and surgical units. Morse Fall Scale(MFS) at a cut-off score of 48 had sensitivity of 76.9% and a negative predictive value of 77.8% on medical subjects, had sensitivity of 75.0% and a negative predictive value of 69.2% on surgical subjects. Bobath Memorial Hospital Fall Risk Assessment Scale(BMFRAS) at a cut-off score of 10 had sensitivity of 76.9% and a negative predictive value of 70.7% on medical subjects, had sensitivity of 62.5% and a negative predictive value of 66.7% on surgical subjects. Johns Hopkins Hospital Fall Risk Assessment tool(JHFRAT) at a cut-off score of 11 had sensitivity of 64.4% and a negative predictive value of 64.4% on medical subjects, had sensitivity of 50.0% and a negative predictive value of 52.9% on surgical subjects. Therefore, the Morse Fall Scale(MFS) has high validity and was suitable for applying subjects admitted in medical and surgical subjects commonly.
Based on this study results, Morse Fall Scale(MFS) can be suggested as the appropriate tool for assessing fall risk assessment in adult population hospitalized at medical or surgical units of university hospitals. The routine assessment of fall risk factors for the hospitalized patients will be essential for the effective fall prevention through early screening of high risk fall group using Morse Fall Scale(MFS).