Standardizing Domains and Metrics of Stroke Recovery: A Systematic Review

Measuring stroke recovery poses a significant challenge, given the complexity of the recovery process. We aimed to identify a standardized and data-driven set of metrics of stroke rehabilitation in the literature that ensures the inclusion of all recovery domains and subdomains in the literature.

A systematic review was conducted by four reviewers using the PRISMA guidelines on PubMed, MEDLINE, and Embase for stroke recovery articles between 2004 and 2024. The inclusion criteria comprised experimental/observational studies, including ischemic and hemorrhagic stroke. All studies had ≥20 participants who were ≥18 years of age, and had a follow-up of ≥3 months. Outcomes included demographics, geographic origin, stroke mechanism, domains and subdomains, metrics used, and follow-up. A bias assessment was performed using the Newcastle-Ottawa Scale and the Cochrane Risk of Bias 2.0 tool. This study was registered with PROSPERO (CRD42024551753).

Our search included 324 studies with a sample of 85,156 participants. The study identified seven domains (perception, physical and motor function (PF), speech and language (S&L), cognition, activities of daily living (ADL), quality of life (QoL), and social interaction) and 96 constituent subdomains that encompass the complete landscape of the stroke recovery literature identified. The domains of PF and ADL constituted the vast share of the literature, albeit reducing in their relative representation over time, while domains such as perception and QoL have been increasingly studied since 2004. Using the domains, the study identified the set and frequency of all commonly used metrics to measure stroke recovery in the literature, of which the NIHSS (n = 72), BI (n = 55), and mRS (n = 51) were the most commonly used. We identified eighteen standard metrics that ensure the inclusion of all seven domains and 96 subdomains. Summary of Review and Conclusions: The identified set of domains and metrics within this study can help inform further clinical research and decision-making by providing a standardized set of metrics to be used for each domain. This approach ensures lesser represented domains and subdomains are also included during testing, providing a more complete view and measure of stroke recovery.