Examining Nurses' Trust and Acceptance of FAIR, an AI-powered Falls Risk Recommender

NCT07078240 | Not Yet Recruiting DMC

• 1 other identifier: NMRCRTF25jan-004
• Study Type: interventional
• Target: 60
• Locations: 0 countries
• Sites: N/A

Brief Summary

An exploratory mixed-method study will be conducted to test acceptance and trust of an AI-powered falls risk predictor system by inpatient hospital nurses

Conditions

Falls Risk

Keywords

Artificial Intelligence trust; Artificial Intelligence acceptance; Artificial Intelligence

Outcome Measures

Primary Outcomes (2)

• Incidence of FAIR's flag acceptance

  Examination of how often the flags raised by FAIR are accepted by nurses, and whether they are accepted or ignored correctly.

  1 Day of Study

• Time taken to do falls risk assessment

  The time taken by the nurses to perform their falls risk assessment will be recorded

  1 Day of Study

Secondary Outcomes (2)

• Time spent looking at FAIR

  1 Day of Study

• Baseline and Post-Simulation Nurse trust and acceptance of FAIR

  Baseline

Study Arms (2)

Intervention Arm

EXPERIMENTAL

Nurses in the intervention arm will perform will receive training on the nature of AI recommender (FAIR) and how it works, and how they can apply it in their assessment of the patient. After that, they will be introduced to three simulated patients with different conditions and needs - intended to reflect a patient at "low risk of falls", "moderate risk of falls" and "high risk of falls" and asked to read and interpret the FAIR recommendations before making their own falls risk assessment of the patient using mWHeFRA.

Other: Falls risk - Artificial Intelligence Recommender (FAIR)

Control arm

PLACEBO COMPARATOR

Nurses in the control arms will be reinforced on fall risk assessment methods using the modified Western modified Western Health Falls Risk Assessment Tool (mWHeFRA). After that, they will be introduced to three simulated patients with different conditions and needs - intended to reflect a patient at "low risk of falls", "moderate risk of falls" and "high risk of falls" and asked to perform assessments of the patient mWHeFRA.

Other: modified Western Health Falls Risk Assessment Tool (mWHeFRA)

Interventions

Falls risk - Artificial Intelligence Recommender (FAIR) OTHER

FAIR is an alert system built into the hospital's electronic medical record system. It is an adaptation of a machine learning model for fall risk calculation built in another hospital in Singapore. FAIR combines multiple patient-specific variables to identify if a patient is at increased risk of falling during their inpatient stay, marking them as a 'falls risk'. Based on the 'flag' raised, the nurse will be instructed to prioritise her falls risk assessment of the patient (If deemed 'high risk') or to do so subsequently as a lower priority once other pressing patient care issues are resolved (if deemed 'low risk'). That way, it ensures the requirements of each patient receiving a falls risk assessment as scored through mWHeFRA are still met, with FAIR allowing nurses to better prioritise their focus and attention on the patient that most needs the assessment at point of admission,

Intervention Arm

modified Western Health Falls Risk Assessment Tool (mWHeFRA) OTHER

The mWHeFRA is the hospital's standard falls risk assessment tool. All nurses are expected to be proficient in its use to guide their risk assessment of patients

Control arm

Eligibility Criteria

• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)

You may qualify if:

• Practicing nurse involved in falls risk assessments of patients

Sponsors & Collaborators

• Tan Tock Seng Hospital: lead
• Marquette University: collaborator
• Lee Kong Chian School of Medicine, Nanyang Technological University: collaborator

Related Publications (1)

• Schulz PJ, Lwin MO, Kee KM, Goh WWB, Lam TYT, Sung JJY. Modeling the influence of attitudes, trust, and beliefs on endoscopists' acceptance of artificial intelligence applications in medical practice. Front Public Health. 2023 Nov 28;11:1301563. doi: 10.3389/fpubh.2023.1301563. eCollection 2023.

  PMID: 38089040 BACKGROUND

Central Study Contacts

George Glass, PhD Student

CONTACT

+65 6903-5384; GLAS0002@e.ntu.edu.sg

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: NON RANDOMIZED
• Masking: NONE
• Purpose: HEALTH SERVICES RESEARCH
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Deputy Head of Nursing Research

Model Details: Two phases of interventional study. First is a parallel arm study comparing nurse acceptance and trust in intervention vs control arm in simulation lab setting.

Study Record Dates

• First Submitted: June 22, 2025
• First Posted: July 22, 2025
• Study Start (Estimated): January 1, 2027
• Primary Completion (Estimated): December 31, 2027
• Study Completion (Estimated): June 30, 2029
• Last Updated: July 22, 2025

Record last verified: 2025-07

Data Sharing

• IPD Sharing: Will share
• Time Frame: Available from 2026 to 2036
• Access Criteria: Individuals will have to contact me as Primary Investigator for permission to use