Validation of TIMI and GRACE acute coronary risk scores in Alexandria Governorate and their role in the comparison of quality of care between hospitals.

BACKGROUND Risk stratification in acute coronary syndrome (ACS) aims to identify those patients who might benefit prognostically from further investigation and treatment. In addition, risk stratification models have been used by health authorities and hospitals in quality management activities. AIM The present study aimed at validating the Thrombolysis In Myocardial Infarction (TIMI) and The Global Registry of Acute Coronary Events (GRACE) risk scores for prediction of mortality in patients with ACS in Alexandria governorate. In addition, the study aimed also at using one of the validated risk scores to compare risk adjusted mortality among participating hospitals. MATERIAL AND METHODS The study was conducted at hospitals belonging to 3 different health care organizations in Alexandria. All admitted patients with the diagnosis of ACS throughout a period of 6 months were included in the study (n=606). Discriminatory capacity and calibration of the TIMI and GRACE risk scores for detection of in-hospital mortality and mortality within six months of index admission were assessed. RESULTS The study showed that both TIMI and GRACE risk scores had high c statistics of 0.70 or higher. GRACE scores showed equal or higher c statistics than TIMI scores denoting better discriminatory capacity. TIMI risk score showed good calibration while GRACE risk score showed lower calibration capacity with certain patient categories. The GRACE risk score was used to calculate the standardized in-hospital mortality ratio which was higher than 1 for all participating hospitals indicating higher than expected mortality for ACS patients in these hospitals. CONCLUSIONS AND RECOMMENDATIONS GRACE risk score showed good discriminatory capacity, suggesting that it is suitable for clinical use among ACS patients in Alexandria governorate. It was recommended to use GRACE risk score for risk adjustment in quality management activities.