Assessing the impact of comorbidity measures on outcomes following acute coronary syndrome

Acute coronary syndromes (ACS) are a common acute presentation of cardiovascular disease (CVD) associated with significant morbidity, mortality and societal economic burden. With improvements in medical care and post- ACS survival rates, ACS patients are increasingly living with multiple comorbidities. It is recommended that comorbidity burden be considered in clinical decision-making but there are critical gaps in current knowledge, particularly on the relative merits of available measures of comorbidity. This thesis was designed to address some of these through a series of linked studies. In Part 1 a systematic review of published literature identified five comorbidity measures used to predict the outcomes of ACS patients. The Charlson Comorbidity Index (CCI) appeared the most widely validated and commonly used, but most comorbidity measures demonstrated an association between worse prognosis and greater comorbidity. In Part 2, analyses of largescale US National Inpatient Sample (NIS) data from 2004-2014 quantified the trend over time towards greater comorbidity among admitted ACS patients defined using the CCI and the Elixhauser Comorbidity Score (ECS). These analyses also provided estimates of the poorer outcomes, lower levels of invasive treatment, and longer stay and higher costs among those with higher comorbidity burden. In Part 3, a direct head-to-head comparison of the prognostic performance of CCI and ECS is reported, in which the ECS was found to have superior discrimination and goodness-of-fit in predicting important inhospital adverse outcomes. The CCI and ECS differ in the comorbidities they include, the weights assigned to them, and the accepted cut-points for categorisation. In Part 4, an agreement analysis found relatively low agreement between the two measures when classifying level of comorbidity in ACS patients. This nationwide analysis of more than 7 million ACS hospitalisations emphasizes the importance of objective comorbidity burden assessment to guide to management strategy and reliably assess prognosis at ACS patients. Clinical implications and further areas of research are discussed in detail