Coronary revascularisation in the UK: using routinely collected data to explore case trends, treatment effectiveness and outcome prediction.

  • Katherine Mcallister

    Student thesis: Phd


    Background: Coronary artery disease is a common cause of morbidity and mortality in the UK. Interventional revascularisation procedures for addressing the disease include percutaneous coronary intervention (PCI) and coronary artery bypass grafting (CABG), which respectively seek to open up or bypass blocked arteries to restore blood flow to heart muscle. Rates at which these procedures are carried out have changed in recent years, as have clinical indications for referral. PCI is delivered by interventional cardiologists, while CABG is carried out by cardiothoracic surgeons, necessitating multi-disciplinary decision making. There is both within- and cross-speciality debate as to the optimal treatment strategy in some case types. Evaluation of the care provided is of clinical and political importance, and requires information about how post-procedure event rates per operator and hospital compare with those expected given the composition of patient populations. Methods: Two UK-wide audit databases of PCI and CABG procedures were used to explore a range of clinical outcome questions. The patient populations contained within each database were compared to see how they differed, and also how each had changed in recent years. In CABG patients, comparative effectiveness of two different surgical techniques (single vs bilateral mammary artery grafting) was assessed with respect to both short-term and long-term mortality outcomes. In PCI patients, a risk model to predict 30-day mortality was developed for use in clinical appraisal. Results: In both patient populations there had been changes to the relative frequencies of many characteristics over time. In the CABG population, multivariable analysis showed that patients undergoing single mammary artery grafting had lower odds of all-cause mortality within 30 days of procedure than those receiving bilateral mammary artery grafting, but had worse overall survival in the long term. In the PCI population, the developed risk model demonstrated good calibration and discrimination at predicting 30-day all-cause mortality. Discussion: The studies described above demonstrate that large-scale routinely collected data can be used to gain insights into clinical care quality and delivery. These resources are under-utilised at present; correcting this requires an understanding of the limitations of the data and how the information contained therein relates to actual clinical care.
    Date of Award31 Dec 2015
    Original languageEnglish
    Awarding Institution
    • The University of Manchester
    SupervisorIain Buchan (Supervisor), Matthew Sperrin (Supervisor) & Benjamin Bridgewater (Supervisor)


    • epidemiology
    • comparative effectiveness
    • clinical prediction
    • routine data
    • cardiology
    • clinical audit
    • clinical outcomes research
    • clinical epidemiology
    • coronary artery disease
    • secondary use of data
    • cardiac surgery

    Cite this