1 PhD Scholar, Faculty of Physiotherapy, Marwadi University, Rajkot.
2 PhD Guide, Faculty of Physiotherapy, Marwadi University, Rajkot.
Received: 20th Feb, 2026 | Revised: 4th Mar, 2026 | Accepted: 25th Mar, 2026 | Available Online: 10th Apr, 2026
Walking aid prescription is a critical component of stroke rehabilitation, yet it is often based on subjective clinical judgment, leading to variability in decision-making. The present study aimed to develop and validate a data-driven predictive model for walking aid prescription in stroke patients by integrating key clinical and functional variables. An observational analytical study was conducted on 168 stroke patients, wherein balance (Berg Balance Scale), voluntary control, functional independence (Functional Independence Measure), mobility (Timed Up and Go test), spasticity (Modified Ashworth Scale), and fear of falling were assessed. A composite score, termed the Predicted Walking Aid Score (PWAS), was formulated using weighted contributions of these variables. The mean PWAS was 23.30 ± 5.97, with significant variation across walking aid categories. PWAS demonstrated a strong positive correlation with balance (r = 0.74) and functional independence (r = 0.58), and a moderate negative correlation with mobility (r = –0.43). Patients requiring no aid had the highest PWAS scores, followed by those using a cane, while walker users had the lowest scores, indicating good discriminative ability of the model. The findings highlight balance as the most influential determinant of walking aid prescription. The developed model provides a structured and objective framework for clinical decision-making, with potential to improve accuracy, consistency, and patient outcomes in stroke rehabilitation. Further validation in diverse populations is recommended to enhance generalizability.
Keywords: Stroke rehabilitation, Walking aid prescription, Predictive model, Berg Balance Scale, Functional Independence Measure, Timed Up and Go, Mobility, Balance, Physiotherapy.
How to cite this article: Chhatlani R, Kakkad A. A Data-Driven Approach To Walking Aid Prescription In Stroke Patients: Development And Validation Of A Predictive Model. Int J Drug Deliv Technol. 2026;16(4):541-548. DOI: 10.25258/ijddt.16.4.56
Source of support: Nil.
Conflict of interest: The authors declare no conflict of interest.