Mapping patient outcomes associated with cardiac rehabilitation: a linked data analysis of 7172 patients from the Victorian cardiac outcomes registry (2019–2021)

Background:
Cardiac rehabilitation is a key component of secondary prevention following percutaneous coronary intervention (PCI), yet participation remains suboptimal. Linked health datasets offer an opportunity to better understand attendance patterns and associated outcomes across the care continuum.

Methods
A retrospective, observational cohort study was conducted using linked health data from 13 public hospitals between 2019 and 2021. Cardiac rehabilitation attendance was defined as participation in ≥1 session. Predictors of cardiac rehabilitation attendance and 12-month outcomes were assessed. A dose–response analysis was also performed, categorizing participants by cardiac rehabilitation session frequency as non-attenders, low attendance (1–5 sessions) and high attendance (≥6 sessions).

Results
Adults undergoing PCI (n = 37,191) were identified, of whom 7126 were successfully linked to the Victorian Integrated Non-Admitted Health (VINAH) dataset. Cardiac rehabilitation attendance was observed in 19.3% of the cohort. Key predictors of attendance were STEMI (OR 1.59, 95% CI 1.32–1.93), NSTEMI (OR 1.24, 95% CI 1.05–1.46), rural/regional program location (OR 1.48, 95% CI 1.25–1.75), and length of stay >3 days (OR 1.04, 95% CI 1.02–1.07). At 12 months, cardiac rehabilitation attendees had lower mortality (1.0% vs 4.9%, p < 0.001) and fewer unplanned readmissions (p = 0.038). A dose–response relationship was found for 12-month mortality. A dose–response relationship was evident for mortality, with lowest rates among high-attendance participants (0.6%), compared with low-attendance (1.5%) and non-attenders (2.9%) (p < 0.001).

Conclusions
Linking Victorian datasets is feasible and provides valuable insights. Cardiac rehabilitation attendance is low, yet participation, particularly at higher doses, is associated with significantly improved outcomes.