Evaluating Wilson sims tool: Quality improvement on fall risk in UAE psychiatric patients

Abstract

Listen

Translate article

BACKGROUND Falls among patients in psychiatric inpatient healthcare settings present a significant global challenge, despite ongoing efforts to reduce the risks. Falls negatively impact patient safety, post-treatment recovery, and medical costs. AIM This study aimed to evaluate the predictive validity of the Wilson Sims Fall Risk Assessment Tool (WSFRAT), investigate predictors of falls, and determine the optimal cut-off point for the WSFRAT. Additionally, we aimed to assess the predictive validity of clinical judgement in identifying individuals at high risk of falls. METHODS We conducted a psychometric study at a specialized psychiatry hospital in Dubai, United Arab Emirates, using data collected from hospital-wide quality projects between April 16, 2019, and March 31, 2021. Our sample comprised 492 patients. RESULTS Contrary to the recommended cut-off point of 7 in the literature, this study results indicate that the optimal cut-off point for the WSFRAT was 5+. This yielded an accuracy of 87%, a diagnostic odds ratio (DOR) of 0.728, a kappa value of 0.208, a sensitivity of 83%, and a specificity of 87%. Furthermore, the regression analyses identified significant predictors of fall risk, including age, gender, assistive device, WSFRAT, and evaluation duration. Notably, clinical judgement did not significantly predict fall risk (p=0.331). CONCLUSION In conclusion, the present research demonstrates that the WSFRAT is a reliable tool with high sensitivity and specificity for predicting falls in psychiatric inpatient settings. The findings emphasize the importance of employing evidence-based tools and a comprehensive assessment approach to prevent falls. Furthermore, our findings challenge the recommended cut-off point of 7 and highlight the need for further research to confirm the optimal cut-off point. Finally, this investigation revealed that clinical judgement alone is not an effective method for predicting falls in this population.