Department of Health Sciences, University of York, York YO10 5DD, UNITED KINGDOM.
Purpose: Despite the proven benefits of cardiac rehabilitation (CR) it remains underutilised in the percutaneous coronary intervention (PCI) cohort in many healthcare systems. The objective of this thesis is to contribute to the growing area of CR research by systematically reviewing CR utilisation determinants in the literature then validate those determinants against routinely collected clinical data.
Methods: A systematic review was conducted to identify CR utilisation determinants in the literature. Data from the UK National Audit of Cardiac Rehabilitation (NACR) for patients who underwent PCI in 2013 to 2016 was retrieved. Three hierarchical logistic regression models were constructed, using multiple imputation as appropriate, to assess the impact of the identified determinants on CR engagement, uptake and adherence. To account for mode of CR delivery on adherence rates, an online survey was administered to 296 CR programmes across the UK.
Results: During the study period, a total of 149,597 cardiac events were recorded in the NACR dataset. Out of this cohort 70,303 (47%) patients underwent a PCI procedure and a total of 59,807 PCI patients were eligible to receive CR. From the CR eligible cohort, 38,246 (63.9%) patients engaged in CR then 28,263 (73.9%) started and finally 22,173 (78.5%) patients completed the programme. The constructed logistic regression models revealed 19 determinants of CR engagement, 23 determinants of CR uptake and 13 determinants of CR adherence. A total of 167 programmes (56.4%) responded to the survey and the results showed that 104 (62.3%) programmes are delivering CR in group- and home based settings while 61 programmes deliver CR in group-based setting only. The Pearson Chi-square test revealed no significant association between mode of CR delivery and adherence rate (p = 0.53, OR: 0.93, 95% CI: 0.87 to 1.01).
Conclusion: This thesis revealed that current CR programmes are not attractive to those who are most deprived, diabetic and smokers. The research has also shown that CR utilisation is not a single patient decision but is also related to service level factors, over which healthcare systems have more direct control.