Prediction of In-Hospital Falls Using NRS, PACD Score and FallRS: A Retrospective Cohort Study

Area of Science:

• Gerontology
• Healthcare Management
• Clinical Medicine

Background:

• In-hospital falls lead to prolonged stays and increased healthcare costs.
• Early identification of patients at risk for falls is essential for implementing preventive measures.
• Developing reliable fall risk assessment tools can improve patient safety and reduce adverse events.

Purpose of the Study:

• To evaluate the predictive performance of the Post-acute care discharge (PACD) score and the Nutritional Risk Screening (NRS) score for in-hospital falls.
• To develop and assess a novel fall risk score (FallRS) for predicting patient falls.
• To compare the accuracy of these scores in identifying patients at risk for falls.

Main Methods:

• A retrospective cohort study was conducted on 19,270 medical in-patients in a Swiss tertiary care hospital (January 2016 - March 2022).
• The predictive ability of the PACD score, NRS score, and the newly developed FallRS was assessed using the area under the curve (AUC).
• Adult patients with a minimum length of stay of two days were included in the analysis.

Main Results:

• A total of 19,270 admissions were analyzed, with 528 admissions (2.74%) experiencing at least one fall.
• The AUC for predicting falls ranged from 0.61 for NRS to 0.69 for PACD score.
• The FallRS demonstrated a slightly higher AUC of 0.70 but was more complex to calculate, with a specificity of 77% and sensitivity of 49% at a cutoff of 13 points.

Conclusions:

• Clinical scores assessing various aspects of patient care offer fair accuracy in predicting in-hospital falls.
• A validated fall risk score can aid in developing effective strategies to minimize falls within healthcare settings.
• Further prospective studies are required to validate these scores against more specific fall prediction tools.