Dynamic Nomogram for Predicting the Fall Risk of Stroke Patients: An Observational Study

Area of Science:

• Neurology
• Rehabilitation Medicine
• Geriatrics

Background:

• Standard fall risk assessments are inadequate for stroke patients.
• Stroke rehabilitation requires specialized fall prediction tools.

Purpose of the Study:

• To develop and validate a dynamic nomogram model for predicting fall risk in stroke patients undergoing rehabilitation.
• To identify independent risk factors for falls in this population.

Main Methods:

• An observational study included 488 stroke patients.
• Data on fall risk factors and functional tests were collected.
• Forward stepwise regression and a dynamic nomogram were utilized for model development and analysis.

Main Results:

• A fall incidence rate of 24.4% was observed among 469 patients.
• Key risk factors identified: age (60-69, ≥80), recent fall history, Berg Balance Scale (BBS) score <40, and anxiety.
• The dynamic nomogram showed good differentiation (AUC-ROC 0.756) and calibration.

Conclusions:

• Age, prior falls, anxiety, and low BBS scores are independent predictors of falls in stroke patients during rehabilitation.
• The developed dynamic nomogram model is accurate, reliable, and clinically useful for fall risk prediction.
• The model offers a simple and practical approach to enhance patient safety during stroke recovery.