Development and evaluation of a home-based balance training program for stroke survivors

Thesis Abstract

Stroke survivors frequently experience impairments in balance that constrain activities of daily living and increase fall risk, particularly after discharge from inpatient rehabilitation. This study addresses the gap in accessible, evidence-based home-based balance interventions by developing, implementing, and evaluating a structured program that can be self-administered with remote supervision. The aim is to determine the effectiveness of a 12-week home-based balance training (HBBT) program on postural stability, functional mobility, and fall risk, and to explore user acceptability and adherence determinants. Specific objectives are (1) to design a progressive, multimodal HBBT protocol integrating static and dynamic balance tasks, perturbation-based elements, and dual-task components; (2) to evaluate the impact of HBBT on postural sway, measured by center-of-pressure metrics (sway area, velocity, and entropy) under eyes-open/closed and firm/foam surface conditions; (3) to assess changes in functional mobility (Berg Balance Scale, Timed Up and Go) and activities of daily living (Modified Barthel Index); (4) to examine fall incidence and fear of falling using the Falls Efficacy Scale-International; (5) to identify adherence, usability, and perceived barriers/facilitators through qualitative interviews; and (6) to test a theoretical model linking self-efficacy, motivation, and adherence to balance outcomes, guided by Social Cognitive Theory and the Theory of Planned Behavior. A parallel mixed-methods design will be employed. The quantitative component will recruit 120 community-dwelling adults aged 50–75 years with a clinical diagnosis of stroke within 6–24 months post-incident, using stratified random sampling from two metropolitan rehabilitation networks. Participants will be randomized to either the HBBT intervention or usual care for 12 weeks, with assessments at baseline, 6 weeks, 12 weeks (end of intervention), and 24 weeks (follow-up). The HBBT program will be delivered via a tablet-enabled home toolkit containing instructional videos, a printable exercise booklet, a wearable inertial sensor for form feedback, and weekly tele-supervision sessions. Primary outcomes include objective postural control (force platform-derived measures) and functional mobility (BBS, TUG). Secondary outcomes encompass ADL performance (MBI), falls incidence, and fear of falling (FES-I). Sample size is powered to detect a moderate between-group difference (Cohen’s d = 0.5) in BBS with 80% power at ? = 0.05, accounting for 20% attrition. Qualitative data will be gathered from a purposive subsample of 30 participants (15 per arm) through semi-structured interviews at the 12-week mark to elucidate user experience, adherence determinants, and contextual facilitators/barriers. Thematic analysis will be conducted following Braun and Clarke’s framework, supported by NVivo software. Quantitative data will be analyzed using repeated-measures ANOVA and linear mixed-effects models to examine differential changes over time between groups, with post-hoc contrasts. Mediation analyses will test the hypothesized pathways from self-efficacy and motivation to adherence and balance outcomes. The integration of quantitative and qualitative findings will follow a convergent parallel design to provide a comprehensive interpretation of effectiveness and implementation feasibility. Expected findings include greater improvements in postural stability (reduced sway, increased entropy indicating healthier strategy), superior gains in BBS and TUG performance, reduced fear of falling, and lower fall incidence in the HBBT group compared with controls. Adherence is anticipated to be mediated by perceived usefulness, ease of use, and ongoing social support from tele-supervision, aligning with Social Cognitive Theory and Theory of Planned Behavior. The study will contribute to knowledge by establishing an empirically validated, scalable home-based balance intervention for stroke survivors, integrating objective biomechanical measures with real-world functional outcomes and implementation determinants. The findings will inform clinical guidelines for post-discharge rehabilitation, policy on remote rehabilitation delivery, and future research on personalized home-based programs. Recommendations include refinement of the digital toolkit for diverse literacy levels, integration with caregiver training modules, and exploration of long-term maintenance strategies beyond 24 weeks.