Abstracts

Rationale: Patients admitted to the Epilepsy Monitoring Unit (EMU) for diagnostic and presurgical evaluation are at risk for falls, yet pose unique risks and challenges to standard inpatient fall prevention protocols. There is inter-institutional variability in the utilization and application of EMU-specific fall risk assessment scales and fall prevention protocols. The lack of a validated fall risk assessment tool may contribute to the absence of standard fall prevention strategies and variable fall rates among different institutions.

Methods: We performed a retrospective, single-center, comparative analysis of inpatient, nurse-administered fall risk scales performed at Vanderbilt University Medical Center (VUMC) EMU. This study compared the scores of patients with EMU fall and age/sex-matched EMU controls using eight, nursing administered fall-risk assessment tools. Scales included modified Conley, Downtown, Morse Fall Assessment tool, STRATIFY, Heindrick II Fall Risk Model (HFRM II), John Hopkins Fall Risk Assessment Tool (JHFRAT), Schmid, and Hester-Davis scales. A sub-group analysis of epilepsy patients with and without EMU fall was also performed. Comparisons between groups of qualitative fall risk designations were performed using chi-square analysis while the comparison of quantitative score distribution was compared using students t-test.

Results: A total of 56 EMU patients (28 fall patients and 28 sex and age-matched controls) evaluated between July 2017 and September 2020 were included in the study. Their fall risk was retrospectively scored using each of the 8 scales (qualitative and quantitative) based on the available documentation at or prior to EMU admission. EMU falls most often occurred within the first 3 days of EMU admission (p=0.0094) consistent with findings of prior studies³. The most significant predictor of fall in the EMU was a history of fall prior to EMU admission (p < 0.0001). A fall within the EMU was also a significant factor predicting fall following discharge from the EMU (p = 0.011). The JHFRAT was the only scale to demonstrate a significant difference between fall and control groups based on the categorical fall risk designations (p = 0.008). There was a significant difference in the distribution of quantitative scores between the fall and control groups using the Morse (p = 0.012), JHFRAT (p = 0.003), Schmid (p = 0.029) and Hester Davis Scales (p = 0.049). None of the scales demonstrated significant categorical differences in the epilepsy patient subgroup. The modified Conley (p= 0.03) and Morse scale (0.025) demonstrated differences in the distribution of quantitative scores in the epilepsy subgroup.