Enhancing Patient Safety by Incorporating Automated Fall-Risk Increasing Drug (FRID) Screening into Fall Prevention Management in the Physiotherapy Out-Patient Department (OPD) of Tuen Mun Hospital (TMH)

Abstract Description

Authors: (including presenting author): :

Chan CY(1), Wong LC(1), Wan S(1), Mak MY(1)

Affiliation: :

(1)Physiotherapy Department, Tuen Mun Hospital, Hospital Authority

Keyword 1: :

Fall Prevention

Keyword 2: :

AI

Keyword 3: :

Patient safety

Keyword 4: :

Risk screening

Keyword 5: :

Innovations

Keyword 6: :

NULL

Introduction: :

Fall risk assessment is a critical component of physiotherapy practice. To further strengthen fall prevention management, we incorporated the FRID screening as part of the fall risk assessment for every patient in the Physiotherapy OPD of TMH since Mar 2025. During early phase of implementation, identification of FRID relied on manual screening of patient medication list from the Clinical Management System (CMS) against a comprehensive FRID information sheet containing more than 120 drugs. This manual process was time-consuming and prone to human error. To enhance efficiency and accuracy, we developed a web-based FRID screening tool, hosted in the HA Teams, to facilitate instant FRID screening since Oct 2025.

Objectives: :

(1)To evaluate the efficiency and accuracy improvement in FRID screening
(2)To determine the clinical impact of improved FRID identification on fall-related adverse events

Methodology: :

With no financial implication, the web-based FRID screening tool was self-developed with coding accelerated by AI and hosted in the HA Teams to allow secure internal access. Patient’s medication list was extracted from the CMS, and processed by the screening tool, which cross-referenced the drugs against the FRID database and automatically identified the presence of FRID. For patients identified with FRID, physiotherapists would provide respective fall prevention management. In Oct 2025, an audit was conducted to assess staff ability to screen 7 FRID in a drug list containing 26 drugs. Number of fall-related adverse events were also monitored.

Result & Outcome: :

Results:
10 physiotherapists were audited in Oct 2025. Manual screening yielded incomplete FRID identification in 100% of audited cases while the web-based FRID screening tool could ensure 100% accuracy. Average screening time was significantly reduced from 2-3 minutes per patient to less than 5 seconds per patient. Since the utilization of automated FRID screening tool, there were no fall incidents in Oct to Dec 2025 whereas there were 2 fall incidents in Jul to Sep 2025. Conclusions:
Automated FRID screening tool enhanced efficiency and accuracy, improved the quality of fall risk assessment and contributed to safer patient care. This innovative approach aligned with HA's smart care initiative and demonstrated the potential of AI-assisted technology in improving clinical decision-making and patient safety.