Abstract: Objective To construct a parametric assessment tool based on the International Classification of Functioning, Disability, and Health (ICF) for community rehabilitation in elderly patients with chronic stroke. Methods Using convenience sampling and maximum variation sampling principles, 207 elderly patients with chronic stroke were recruited from the Jiangsu-Zhejiang-Shanghai region. An ICF questionnaire with 136 categories was developed based on the intersection of the ICF category set for community-dwelling older adults and the extended comprehensive ICF core set for stroke. All items were included in subsequent modeling:an Item Response Theory (IRT) model was constructed to estimate difficulty parameters, while a Bayesian Graphical Modeling technique was applied to build a Graphical Modeling (GM) for estimating intervention benefit parameters. Results The established Rasch model demonstrated good reliability (Cronbach’s α=0.966) and validity.The visualization tool identified categories that matched the patients’ ability levels and offered suitable intervention benefits. Thirty-seven of these categories showed significant benefits. Conclusions Building on the existing comprehensive ICF core sets for strokes, the parametric assessment tool incorporates category difficulties and treatment benefit indices. This tool facilitates quantitative decision-making tailored to the capabilities of community-dwelling elderly patients with stroke.

Key words: aged, chronic stroke, International Classification of Functioning, Disability and Health, Item Response Theory, Bayesian discrete graphical model, rehabilitation