Risk adjusted mortality rates : Do they differ if bases on administrative data (hospital standardised mortality ratio) versus a physiological predictive model (APACHE IV ®)?

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

Background: The measurement of, and reporting on clinical outcomes, is an integral part of clinical governance but no consensus has been reached about which measures to use and the validity thereof. Objective: To compare an administrative predictive model (Hospital Standardised Mortality Ratio [HSMR]) with a physiological predictive model (APACHE ®IV) to determine the correlation in the predicted risk adjusted mortality rates. To determine whether stratifying the patients into low (<10%), medium (<50%) or high (>80%) risk bands will lead to more accurate comparisons. Design: Prospective cohort study Setting: 63 critical care units in 34 private acute care facilities across South Africa Methods: Both HSMR and APACHE ®IV are calculated routinely in all participating facilities and the research study will use the data generated. An additional audit process will be implemented to determine and ensure the integrity of the data. Ethics: The healthcare facilities have standard processes in place to ensure confidentiality and the statistician analysing the data is employed by the healthcare group and bound to a confidentiality agreement. Ethics approval has also been obtained by the University of Cape Town ethic committee before the approval of the research proposal.