Falls after stroke are recognized as a serious public health challenge worldwide, contributing to increased morbidity and reduced quality of life among stroke survivors. ^{1, 2} Falls are quite common in this population due to increased cognitive-motor interference (CMI). This interference impacts their ability to effectively integrate balance, walking, and cognitive tasks. ³ Cognitive-motor interference in stroke patients is often characterized by deficits in executive functions, which include challenges with memory, attention, and motor control. ^{1, 4, 5} Notably, difficulties in performing dual tasks—an essential aspect of executive function—are significant risk factors for falls among stroke survivors. ^{6– 8}

Impaired postural balance and low falls efficacy are critical factors that contribute to frequent falls in stroke survivors. ^{9, 10} Recent studies, such as those by Xu (2018) ¹¹ and Sjöholm et al. (2022), ¹ highlight the connection between reduced balance and an increased risk of falling. There is growing interest in assessing cognitive-motor interference through dual-task paradigms; however, most previous observational research has primarily relied on established tools like the Berg Balance Scale (BBS) and the Tinetti Falls Efficacy Scale (FES) to evaluate balance and fear of falling. Evidence regarding the predictive ability of dual-task performance tests for fall risk in stroke survivors remains limited.

Dual-task assessments are shown to be reliable for measuring cognitive-motor performance in stroke survivors, ¹² yet the comparative effectiveness of different dual-task tests in predicting fall risk is underexplored. Studies by Hofheinz and Mibs (2016) ¹³ have validated dual-task tests, such as the Timed Up and Go Test, as effective predictors of fall risk in older adults. However, further research is needed to see if these findings apply to stroke survivors, who have unique cognitive and motor challenges. Emerging research supports the utility of dual-task paradigms like the WWT and SWWT for assessing CMI and identifying fallers. ^{4, 7} These assessments engage higher executive functions that extend beyond automatic walking tasks. ^{14, 15} However, a recent predictive analytics study by Abdollahi et al. (2024) highlighted limitations in current dual-task assessments for fall risk prediction, emphasizing the need for more tailored methods. ¹⁶

While the clinical value of dual-task assessments in identifying fall risk is established, there remains a paucity of research examining how stroke survivors perform on both WWT and SWWT tests and how these outcomes correlate with fall efficacy (FES) and balance (BBS). Assessing dual-task performance alongside established scales may provide deeper insight into fall mechanisms among stroke survivors and facilitate the development of targeted cognitive or behavioral interventions. ¹⁷

Therefore, the objective of this study is to determine whether dual-task performance, specifically the Walking While Talking (WWT) and Stops Walking While Talking (SWWT) tests, is correlated with established fall risk predictors such as the Berg Balance Scale (BBS) and the Falls Efficacy Scale (FES). Additionally, the study aims to compare the predictive ability of WWT and SWWT in identifying fall risk among stroke survivors using these standardized fall prediction tools. Finally, the research seeks to identify which of the dual-task assessments, WWT or SWWT, serves as a superior predictor of fall risk in this population.

We hypothesize that stroke survivors with poor dual-task performance on the Walking While Talking (WWT) and Stops Walking While Talking (SWWT) tests will exhibit significant associations with fall risk, as evaluated by the Berg Balance Scale (BBS) and the Falls Efficacy Scale (FES) scores. Additionally, we anticipate that there will be a significant difference in the predictive accuracy between WWT and SWWT in determining fall risk among stroke survivors.

This study aimed to evaluate the predictive ability of dual-task performance measures, specifically the Walking While Talking (WWT) and Stops Walking While Talking (SWWT) tests, in assessing fall risk among stroke survivors. The findings revealed significant associations between dual-task performance and traditional fall risk measures, emphasizing the value of incorporating these assessments into clinical practice for stroke rehabilitation. Previous research has established the importance of dual-task paradigms in identifying fall risk among older adults and patients with neurological impairments. Abdollahi et al. (2024) reported that changes in gait performance under dual-task conditions were significantly associated with fall risk, particularly in frail elderly populations. ¹⁶ This aligns with our results, where both WWT-S and WWT-C demonstrated strong negative correlations with the Berg Balance Scale (BBS) (r = -0.73 and r = -0.74, respectively), suggesting that slower dual-task completion times indicate poorer balance and an increased risk of falls in stroke survivors. The strong relationship between dual-task performance and balance observed in our study supports the clinical relevance of these assessments.

Conversely, some studies suggest that dual-task assessments offer no significant predictive advantage over single-task evaluations. ²⁹ The effectiveness of dual-task assessments may also vary by age group. ³⁰ However, the high Area Under the Curve (AUC) observed in our ROC analysis (0.911) for dual-task tests highlights their superior predictive capability. This finding indicates that dual-task assessments, particularly both versions (Simple & Complex) of WWT, offer significant clinical value when used alongside traditional measures like BBS and FES. The high level of discrimination supports their inclusion in comprehensive fall risk assessments for stroke survivors. The high sensitivity (97.8%) and specificity (99%) of the WWT tests show they are very effective at accurately identifying fall risk in stroke survivors. While the SWWT test also has value, its lower sensitivity (68.9%) and specificity (91.3%) make it better suited for quick assessments. These results support using dual-task assessments, especially WWT, in stroke rehabilitation to improve fall prevention strategies.

The SWWT test demonstrated significant predictive power for fall risk (p = 0.009) in our logistic regression analysis, aligning with research advocating for its use as an effective assessment tool. Lundin-Olsson et al. (1997) first highlighted SWWT as valuable for identifying individuals at higher risk of falls in elderly cohorts. ²⁸ Our study extends this application to stroke survivors, revealing that participants who stopped walking while talking (SWWT-positive) had a substantially higher risk of falling, reinforcing its utility as an essential part of dual-task evaluation in stroke rehabilitation programs.

The findings supports the clinical value of dual-task performance measures, particularly WWT and SWWT tests, in assessing fall risk among stroke survivors. While previous research has underscored the significance of cognitive-motor interference in fall risk, ^{3, 4} our results enhance this understanding by demonstrating strong correlations with established measures like BBS and FES. This aligns with studies validating dual-task testing as a proxy for evaluating broader functional deficits. ¹² The predictive power of WWT (both versions), shown by their high sensitivity and specificity, highlights their potential use in comprehensive stroke rehabilitation programs. ¹³

Although our study did not assess gait parameters directly, it is essential to recognize how cognitive-motor interference can affect motor control, contributing to an increased fall risk. ⁸ This is consistent with the understanding that proprioceptive and executive function deficits can influence dual-task performance, impacting balance. ¹⁶ Incorporating dual-task assessments can provide a more nuanced approach to identifying fall risk beyond traditional single-task methods.

The relationship between falls efficacy, as measured by the Falls Efficacy Scale (FES), and dual-task performance adds an important dimension to understanding fall risk. Our study found moderate positive correlations between WWT-S and FES (r = 0.67) and WWT-C and FES (r = 0.61), indicating that slower completion times were associated with higher fear of falling. These findings align with studies emphasizing the psychological components of balance confidence and their impact on fall risk. Low confidence in balance, reflected in higher FES scores, has been linked to poorer performance in balance and gait tasks.

This study’s relatively small sample size of 68 participants, may limit the statistical power and generalizability of the findings to a broader population. Additionally, the cross-sectional design limits our ability to establish causative relationships between dual-task performance and fall risk over time. Furthermore, conducting the study in clinical settings may not fully represent the real-life challenges that stroke survivors face when multitasking in their daily lives, which could impact the study’s relevance. Finally, while we used the Berg Balance Scale (BBS) and Falls Efficacy Scale (FES) as our main measures—both of which are valid—they might not cover all aspects of fall risk for this group. For future research, it would be beneficial to conduct larger studies over a longer period and to look into additional assessment tools. This could improve our understanding and applicability of findings across different patient groups.

The clinical implications are significant; while the predictive power of SWWT was substantial, the WWT tests demonstrated higher sensitivity and specificity, suggesting they are more effective at accurately identifying individuals at risk. This contrasts with studies finding SWWT to be less sensitive but still valuable for quick assessments. The high sensitivity (97.8%) and specificity (99%) for WWT-S and WWT-C underscore their robustness as reliable indicators for fall risk. Overall, this study contributes to the growing evidence that dual-task performance assessments are crucial for identifying stroke survivors at risk of falls. Acknowledging both significant and non-significant outcomes ensures a balanced discussion that reflects the complexity of dual-task testing in clinical practice. Our findings highlight the superior predictive power of WWT over SWWT, supporting its use in fall risk screening. Future studies should explore longitudinal designs and larger sample sizes to validate these results further and investigate mechanisms underlying dual-task deficits in stroke rehabilitation.

DL was the principal investigator and enrolled participants for this study. The study was conceptualized by AMJ, AN, and SK. VP conducted the data analysis and led the literature review and critical revision of the manuscript. Supervision of the trial and data collection was carried out by VK, AMJ, and SP, while oversight and guidance during data collection were provided by VP, AN, and SK. RP, VK, and SP made significant contributions to the study design and manuscript preparation. All authors provided critical input and participated in the preparation and final approval of the manuscript.

Approval for the study protocol was obtained on 20/01/2022 from both the Scientific Committee and the Institutional Ethics Committee (Approval #: IEC KMC MLR 01/2022/45) of Kasturba Medical College (KMC) Mangalore, part of the Manipal Academy of Higher Education (MAHE), Manipal. Additionally, the study was registered with The Clinical Trials Registry-India (CTRI/2021/05/033513). The study was conducted in accordance with the ethical principles of the Declaration of Helsinki for research involving human participants Written informed consent was obtained from all participants prior to their inclusion in the study.