Trial Outcomes & Findings for AI-Powered Fall Risk Prediction in Nursing Care (NCT NCT07000981)
NCT ID: NCT07000981
Last Updated: 2025-12-02
Results Overview
Assessment of the patient's fall risk with the Morse Fall Scale It is an effective and simple measurement tool that is frequently used in hospitals in Turkey and used to diagnose potential patient fall risks for the nursing profession. The scale consists of six criteria (secondary diagnosis, presence of a history of falls, mobilization support, presence of intravenous access or heparin use, gait/transfer, and mental status) that diagnose fall risk. According to this assessment tool, if the patient has a score below 25 points, he/she is in the low risk group for falls. If the score is between 25 and 50, the patient is in the medium risk group, and if the score is 51 and above, the patient is in the high risk group. A minimum score of 0 and a maximum score of 125 can be obtained from the scale. This scale allows a systematic determination of the fall risk of patients in clinical settings.
COMPLETED
177 participants
Day 1
2025-12-02
Participant Flow
The study population included patients treated at the Department of Physical Medicine and Rehabilitation at Turgut Özal Medical Center. A priori power analysis indicated that a sample size of 161 participants would be sufficient to detect an effect size of 0.30 with 95% confidence and 5% margin of error. In total, 177 participants were enrolled using non-probability sampling to accommodate potential data loss during the study.
Research Inclusion Criteria * Being 18 years or older * Being able to read and write Turkish * Being able to walk with or without support Study Exclusion Criteria * Not being able to speak or understand Turkish at a sufficient level * Being diagnosed * Being immobile * Being cognitively inadequate for the study
Participant milestones
| Measure |
Cohort Group
All 177 participants completed the fall risk assessment using the Decision Support System.
|
|---|---|
|
Overall Study
STARTED
|
177
|
|
Overall Study
COMPLETED
|
177
|
|
Overall Study
NOT COMPLETED
|
0
|
Reasons for withdrawal
Withdrawal data not reported
Baseline Characteristics
Race and Ethnicity were not collected from any participant.
Baseline characteristics by cohort
| Measure |
Cohort Group
n=177 Participants
The fall risk assessment application used in this study was developed using a methodology. A descriptive information form was created for each participant, including basic information such as age, height, weight, and other data. This information was then integrated into the Morse Fall Scale to assess fall risk. Patients included in the sample were first administered the Morse Fall Scale and then given a gait assessment using a computer vision system consisting of a digital camera that monitored patients walking at a comfortable pace in the clinic corridor. Acceleration along the X, Y, and Z axes during walking was also assessed using an accelerometer carried in the patients' pockets.
Data obtained from this system was analyzed using machine learning methods to estimate the biomechanics of the lower and upper extremities in real time. Inferences were then drawn regarding key gait characteristics such as stride length, cadence, period, and range of motion. This data was then used to train and validate an artificial neural network (ANN) that sought correlations between the extracted features and the Morse Fall Scale score. As a result, the final system was able to predict the fall risk of new cases in real time through gait analysis. The AI-based system developed as part of the study was introduced to the literature under the name "GuArDrop."
|
|---|---|
|
Age, Continuous
|
52.38 years
STANDARD_DEVIATION 16.55 • n=177 Participants
|
|
Sex: Female, Male
Female
|
87 Participants
n=177 Participants
|
|
Sex: Female, Male
Male
|
90 Participants
n=177 Participants
|
|
Region of Enrollment
Turkey
|
177 Participants
n=177 Participants
|
|
BMI
|
27.21 kg/m^2
STANDARD_DEVIATION 5.64 • n=177 Participants
|
PRIMARY outcome
Timeframe: Day 1Assessment of the patient's fall risk with the Morse Fall Scale It is an effective and simple measurement tool that is frequently used in hospitals in Turkey and used to diagnose potential patient fall risks for the nursing profession. The scale consists of six criteria (secondary diagnosis, presence of a history of falls, mobilization support, presence of intravenous access or heparin use, gait/transfer, and mental status) that diagnose fall risk. According to this assessment tool, if the patient has a score below 25 points, he/she is in the low risk group for falls. If the score is between 25 and 50, the patient is in the medium risk group, and if the score is 51 and above, the patient is in the high risk group. A minimum score of 0 and a maximum score of 125 can be obtained from the scale. This scale allows a systematic determination of the fall risk of patients in clinical settings.
Outcome measures
| Measure |
Cohort Group
n=177 Participants
The fall risk assessment application used in this study was developed using a methodology. A descriptive information form was created for each participant, including basic information such as age, height, weight, and other data. This information was then integrated into the Morse Fall Scale to assess fall risk. Patients included in the sample were first administered the Morse Fall Scale and then given a gait assessment using a computer vision system consisting of a digital camera that monitored patients walking at a comfortable pace in the clinic corridor. Acceleration along the X, Y, and Z axes during walking was also assessed using an accelerometer carried in the patients' pockets.
Data obtained from this system was analyzed using machine learning methods to estimate the biomechanics of the lower and upper extremities in real time. Inferences were then drawn regarding key gait characteristics such as stride length, cadence, period, and range of motion. This data was then used to train and validate an artificial neural network (ANN) that sought correlations between the extracted features and the Morse Fall Scale score. As a result, the final system was able to predict the fall risk of new cases in real time through gait analysis. The AI-based system developed as part of the study was introduced to the literature under the name "GuArDrop."
|
|---|---|
|
Fall Risk Categories Based on the Morse Fall Scale
Low Risk (<25)
|
131 Participants
|
|
Fall Risk Categories Based on the Morse Fall Scale
Medium Risk (25-50)
|
41 Participants
|
|
Fall Risk Categories Based on the Morse Fall Scale
High Risk (≥51)
|
5 Participants
|
PRIMARY outcome
Timeframe: Day 1Population: A total of 177 participants were enrolled in the study, out of which 35 were assigned to the test group for evaluating the model's classification performance. This test group was not used in the training or validation phases.
Classification accuracy of the decision support system was evaluated based on the percentage of test units correctly classified. The scale ranges from 0% to 100%, where higher values indicate better performance. This metric reflects the proportion of correctly identified cases by the system during model evaluation.
Outcome measures
| Measure |
Cohort Group
n=35 Participants
The fall risk assessment application used in this study was developed using a methodology. A descriptive information form was created for each participant, including basic information such as age, height, weight, and other data. This information was then integrated into the Morse Fall Scale to assess fall risk. Patients included in the sample were first administered the Morse Fall Scale and then given a gait assessment using a computer vision system consisting of a digital camera that monitored patients walking at a comfortable pace in the clinic corridor. Acceleration along the X, Y, and Z axes during walking was also assessed using an accelerometer carried in the patients' pockets.
Data obtained from this system was analyzed using machine learning methods to estimate the biomechanics of the lower and upper extremities in real time. Inferences were then drawn regarding key gait characteristics such as stride length, cadence, period, and range of motion. This data was then used to train and validate an artificial neural network (ANN) that sought correlations between the extracted features and the Morse Fall Scale score. As a result, the final system was able to predict the fall risk of new cases in real time through gait analysis. The AI-based system developed as part of the study was introduced to the literature under the name "GuArDrop."
|
|---|---|
|
Fall Risk Classification Accuracy of the Decision Support System
|
91.4 Percentage (%)
Standard Error 2.10
|
PRIMARY outcome
Timeframe: Day 1Population: Of the 177 enrolled participants, 35 were allocated to the test group for the evaluation of model performance using standard machine learning metrics. The remaining participants were used for model training and validation.
This outcome measure evaluates the classification performance of a clinical decision support system using standard machine learning metrics: precision, recall, and F1-score. These metrics are based on a scale ranging from 0 to 1. Higher values indicate better classification performance. Precision is defined as the proportion of true positive predictions among all positive predictions. Recall is defined as the proportion of true positive predictions among all actual positives. The F1-score is the harmonic mean of precision and recall.
Outcome measures
| Measure |
Cohort Group
n=35 Participants
The fall risk assessment application used in this study was developed using a methodology. A descriptive information form was created for each participant, including basic information such as age, height, weight, and other data. This information was then integrated into the Morse Fall Scale to assess fall risk. Patients included in the sample were first administered the Morse Fall Scale and then given a gait assessment using a computer vision system consisting of a digital camera that monitored patients walking at a comfortable pace in the clinic corridor. Acceleration along the X, Y, and Z axes during walking was also assessed using an accelerometer carried in the patients' pockets.
Data obtained from this system was analyzed using machine learning methods to estimate the biomechanics of the lower and upper extremities in real time. Inferences were then drawn regarding key gait characteristics such as stride length, cadence, period, and range of motion. This data was then used to train and validate an artificial neural network (ANN) that sought correlations between the extracted features and the Morse Fall Scale score. As a result, the final system was able to predict the fall risk of new cases in real time through gait analysis. The AI-based system developed as part of the study was introduced to the literature under the name "GuArDrop."
|
|---|---|
|
Classification Performance Metrics of the Decision Support System (F1 Score, Precision, Recall)
Precision (AUC)
|
0.864 Proportion
Standard Error 2.58
|
|
Classification Performance Metrics of the Decision Support System (F1 Score, Precision, Recall)
Recall
|
0.914 Proportion
Standard Error 2.10
|
|
Classification Performance Metrics of the Decision Support System (F1 Score, Precision, Recall)
F1-Score
|
0.920 Proportion
Standard Error 2.03
|
Adverse Events
Cohort Group
Serious adverse events
Adverse event data not reported
Other adverse events
Adverse event data not reported
Additional Information
Results disclosure agreements
- Principal investigator is a sponsor employee
- Publication restrictions are in place