This dataset supports the scoping review 'Predictors of participation in, adherence to and outcomes of cardiac rehabilitation in older adults', whose protocol was registered prospectively on the Open Science Framework (https://doi.org/10.17605/OSF.IO/63TUK).
The review mapped the factors reported to predict three things in adults aged 65 years or older: entry into cardiac rehabilitation, persistence with it, and the outcomes achieved. It followed JBI methodology and was reported using PRISMA-ScR. Searches of MEDLINE via PubMed, the Cochrane Library, Europe PMC and the Directory of Open Access Journals identified 7655 records; after de-duplication 5250 were screened, 818 were sought in full text, and 366 studies published between 2005 and 2026 were included. Of these, 356 yielded 1534 charted predictor associations and 10 contributed none.
The deposit contains four data files. The first lists the 386 reports that reached the data-extraction stage, with a column identifying the 366 retained in the analysis set after 20 were excluded because the reported mean or median age proved to be below 65 years. The second is the main file, holding one row for each extracted predictor association, with the study-level descriptors, the predictor and its category, the domain it predicts, the direction and magnitude of the association as reported, and whether the estimate was adjusted. The third records every country value harmonised for reporting, with the original values preserved. The fourth lists the 146 reports eligible at title and abstract whose full text could not be obtained. No row was deleted at any stage, so the analysis set is reconstructed by filtering rather than by trusting a cleaned file.
Four documents accompany the data: a complete record of the ways in which the conduct of the review departed from the registered protocol, the amended protocol itself, the full electronic search strategy, and the completed PRISMA-ScR checklist. The record of deviations reports each departure with what was planned, what was done, the reason and the effect on the review counts, whether or not it favours the review. It includes the two amendments to the eligibility criteria adopted after the yield of the search had been observed, and four further departures identified during extraction, among them the extraction of 36 studies from their abstracts when the full text could not be secured.
The file 00_readme.txt documents every variable, states how to reproduce the published counts, and sets out the known limitations of the data. No individual participant data are included, because the review synthesises previously published aggregate findings.
