Digitizing rehabilitation outcomes and assessing data quality in clinical trials: implementing validated scales in REDCap for a stroke RCT

Background: Rehabilitation randomized controlled trials (RCTs) are methodologically challenging due to the complexity of interventions, the heterogeneity of patient profiles, and the need to evaluate multiple functional domains through standardized outcome measures. Manual administration and scoring of these measures increase clinician workload and risk of error, while multicenter settings amplify variability in data collection. To address these challenges, we implemented a comprehensive electronic repository of validated rehabilitation outcome measures within REDCap for the StrokeFit4 trial, a multicenter RCT evaluating robotics-assisted rehabilitation in stroke patients, which started in August 2024 and is expected to complete enrollment in July 2026. Methods: Twenty-eight validated clinical outcome measures were digitized as REDCap instruments, spanning domains such as motor and cognitive function, balance, quality of life, nutritional status, and psychological profile. Whenever possible, automated scoring, branching logic, and data validation rules were implemented to reduce errors and standardize administration across centers. Twenty instruments not subject to copyright restrictions were openly shared on GitHub. To complement REDCap’s built-in data quality features, we developed a Data Quality Dashboard using Python and Streamlit. The dashboard was designed for monthly use to detect missing or abnormal values, filter analyses by clinical centers (Data Access Groups), allow manual issue reporting, and generate structured CSV reports. Results: The implemented REDCap instruments enabled structured, reproducible electronic data capture for all trial events, including baseline, treatment, and follow-up time points. Automated scoring and validation reduced the risk of miscalculation and improved data entry consistency. The Data Quality Dashboard extended REDCap functionality by accounting for longitudinal patient status throughout the trial when identifying missing values, thereby avoiding false positives. It also provided a user-friendly interface for visual inspection of scales and population characteristics, enabling both automated and manual quality checks. Conclusions: The integration of validated REDCap instruments with a dedicated Data Quality Dashboard provided a scalable solution for addressing data collection and monitoring challenges in multicenter rehabilitation RCTs. By openly sharing freely available instruments, we promote harmonization and reuse in future studies. This approach has the potential to enhance methodological rigor, improve data quality, and accelerate the generation of reliable evidence on innovative rehabilitation interventions. © The Author(s) 2026.