Research Abstract: Advanced Trainee - Oral Presentation Australian and New Zealand Society for Geriatric Medicine Annual Scientific Meeting 2023

Current cut points of three falls risk assessment tools are inferior to calculated cut points in Geriatric Evaluation and Management units. (#24)

Vivian Lee 1 , Linda Appiah-Kubi 1 , Sara Vogrin 2 3 , Jesse Zanker 2 3 , Claire Long 1 , Joanna Mitropoulos 1
  1. Department of Geriatric Medicine, Western Health, Melbourne, Victoria, Australia
  2. Department of Medicine, Western Health, University of Melbourne, Melbourne, Victoria, Australia
  3. Australian Institute for Musculoskeletal science (AIMSS), Melbourne, Victoria, Australia

Aims

In 2022, Western Health (WH) introduced a new falls risk assessment tool, Western Health St. Thomas’ Risk Assessment Tool (WH-STRATIFY), a modification of The Northern Hospital Modified St Thomas's Risk Assessment Tool (TNH-STRATIFY), replacing its existing Peninsula Health Falls Risk Screening Tool (PH-FRAT). This retrospective observational study aimed to determine the predictive accuracy of the three tools on admission to Geriatric Evaluation Management units (GEM).

 

Methods

Data was collected on 54 consecutive patients who fell during their GEM admission and compared with 62 randomly sampled patients who did not fall between December 2020-June 2021. The three falls risk assessment tools were scored for each patient. The event rate Youden Index (Youden IndexER) was calculated to compare the tools’ predictive accuracy using both original and calculated optimal cut points.  

 

Results

All three tools had low predictive accuracy for falls. TNH-STRATIFY had the highest predictive accuracy (Youden IndexER = 0.20, 95% confidence interval CI = 0.07, 0.34). The PH-FRAT (Youden IndexER = 0.01 and 95% CI = -0.04, 0.05) and WH-STRATIFY (Youden IndexER = 0.00 and 95% CI = -0.04, 0.03) were equivalent and inaccurate. The predictive accuracy of PH-FRAT (Youden IndexER = 0.14 and 95% CI = 0.01, 0.29) and WH-STRATIFY (Youden IndexER = 0.18 and 95% CI = 0.00, 0.35) for falls improved when calculated optimal cut points were applied.

 

Conclusions

The optimal cut points of falls risk assessment tools should be determined and validated in different clinical settings to optimise local predictive accuracy, enabling appropriate falls risk mitigation and resource allocation.