Aims: Acute coronary syndrome (ACS) is a costly condition for health service provision yet variation in the delivery of care between hospitals persists. A composite measure of adherence with evidence-based clinical-process indicators (CPIs) could better inform hospital performance reporting and clinical outcomes in the management of ACS.
Methods: Data on 7444 ACS patients from 39 Australian hospitals were used to derive a hospital-specific composite quality score by calculating mean adherence to 14 evidence-based CPIs. Using the generalized estimating equation to account for clustering of patients within hospitals and the GRACE risk score to adjust for differences in presenting risk, we evaluated associations between the hospital-specific composite quality score, in-hospital major adverse events, in-hospital mortality and mortality and readmission for ACS at 6 months.
Results: Hospitals had a mean adherence of 68.3% (SD 21.7) with the composite quality score. There was significant variation between hospital adherence tertile 1 (79%) and tertile 3 (56%), P < 0.0001. With risk adjustment, there was an association between hospitals with a higher composite quality score and reduced in-hospital adverse events (OR: 0.85, CI: 0.71-0.99) and survival at hospital discharge (OR: 0.47; 95% CI: 0.28-0.77). There was trending improvement in survival at 6 months (OR 0.48; CI: 0.20-1.16) and fewer readmissions to hospital for ACS at 6 months (OR 0.79; CI 0.60-1.05).
Conclusion: The association between the quality composite score and reduced in-hospital events and survival at hospital discharge supports the utility of reporting CPIs in routine hospital performance reporting on the management of ACS.
Australia and new zealand clinical trial registration (anzctr): CONCORDANCE Registry ACTRN12614000887673.
Keywords: Acute coronary syndrome; Adherence with evidence based care; Clinical process indicators; Quality composite score.
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2016. For permissions please email: [email protected].