Clinical changes and patient profiles after community-based cardiac rehabilitation: a prospective single-arm cohort study

To describe six-month clinical changes and explore data-derived patient profiles among completers of a community-based cardiac rehabilitation (CR) program, and further assess whether baseline characteristics could predict significant increase in 6-min walk distance (6MWD). This prospective single-arm cohort included 189 participants who completed a 6-month CR program from July 2023 to February 2025. Pre-to-post changes were assessed using paired analyses. K-means clustering was used to explore patient profiles and was accompanied by stability analyses. In the further analysis, 14 prespecified baseline predictors were evaluated using nested cross-validation to predict an Δ6MWD ≥ 30 m. After 6 months of CR, 6MWD increased by 48.93 m and left ventricular ejection fraction (LVEF) increased by 3.84%, while anxiety and depression scores decreased and SF-12 scores increased (all P < 0.001). Cluster analysis identified three profiles comprising 70, 53, and 66 participants; based on different combinations of clinical characteristics, respectively. The complete nested algorithm-selection workflow yielded a pooled out-of-fold area under the curve of 0.754 (95% CI 0.669–0.832). Community-based CR was associated with favorable functional, cardiac, psychological, and quality-of-life changes. Exploratory patient profiles showed different patterns of observed change. This offers a data-driven approach to allocating resources and providing individualized precision CR in community settings. Larger random-controlled multicenter studies are needed to support stratified community follow-up.